Search

Camouflaged object detection is a challenging task that aims to identify objects having similar texture to the surroundings. This paper presents to amplify the subtle texture difference between camouflaged objects and the background for camouflaged object detection by formulating multiple texture-aware refinement modules to learn the texture-aware features in a deep convolutional neural network. The texture-aware refinement module computes the biased co-variance matrices of feature responses to extract the texture information, adopts an affinity loss to learn a set of parameter maps that help to separate the texture between camouflaged objects and the background, and leverages a boundary-consistency loss to explore the structures of object details. We evaluate our network on the benchmark datasets for camouflaged object detection both qualitatively and quantitatively. Experimental results show that our approach outperforms various state-of-the-art methods by a large margin.

Objectives Mild Cognitive Impairment (MCI) is the prodromal stage of Alzheimer's disease (AD), which is a progressive and fatal neurodegenerative disorder. Detection of MCI condition can enable early diagnosis resulting in timely intervention to delay the disease progression. Onset of MCI causes tissue alterations in Corpus Callosum (CC) of the brain. Texture analysis of brain Magnetic Resonance (MR) images aids in characterising these imperceptible changes. In this study, Kernel Density Estimation (KDE) technique is used to analyse the textural variations in CC to detect MCI condition. Materials and method The pre-processed brain MR images are obtained from a public access database. Reaction Diffusion level set is employed to segment CC from sagittal slices of the images. Kernel density estimation method is applied to study the local intensity variations within the segmented CC. Statistical features quantifying these variations are extracted from the KDE values. These features are used to differentiate MCI condition using linear classifiers based on discriminant analysis and support vector machine. The results are compared with conventional Grey Level Co-occurrence Matrix (GLCM) features for validation. Results The KDE-based texture features extracted from CC show significant variation between normal and MCI classes. Results demonstrate that this approach can differentiate MCI condition with high accuracy and specificity of 81.3% and 82.7%, respectively. The KDE-based features perform better when compared with GLCM features for distinguishing MCI. Conclusions The KDE-based texture features are able to capture the subtle changes occurring in CC at the MCI stage. This technique achieves comparable performance to other state-of-the-art methods with reduced number of features. Efficiency of the KDE-based texture analysis confirms that the proposed computer assisted technique can be used for mass screening of MCI, which can aid in handling the disease severity.

Music Information Retrieval (MIR) is the task of extracting high-level information, such as genre, artist or instrumentation from music. Genre classification is an important and rapidly evolving research area of MIR. To date, only a small amount of research work has been done on the automatic genre classification of Nigerian songs. Hence, this study presents a new music dataset, namely the ORIN dataset, consisting of only Nigerian songs. The study dataset contains 478 Nigerian traditional songs from five genres: fuji, juju, highlife, waka and apala. The timbral texture and tempo features were mined from 30-second segments of each song using the Librosa Python library. For genre classification, the ORIN datasets was trained on 4 different classifiers- k-Nearest Neighbhour, Support Vector Machine, eXtreme Gradient Boosting (XGBoost) and Random Forest- with 85–15 train-test splits. The results obtained for the five different genres indicates that XGBoost classifier is a better model, having the highest accuracy of 81.94% and recall of 84.57%. This study uses the global mean (Tree SHAP) method to determine feature importance and impact on the classification model. Further analysis on the individual genres found some nearness in the timbral properties between some of the genres. This analysis was confirmed by human observation.

Over the past decades, video salient object detection has achieved rapid development. However, the effect of edge information on spatial and temporal features extraction has not been explored. The method STEG-Net is proposed to solve the problem, we use the extracted edge information to guide the extrction of spatial and temporal features simultaneously. We combine deep texture information with shallow edge information, which can not only retain the edge of the object but also enhance the global information, leading to the more accurate location of the object. At the same time, the shallow edge information and the deep texture information are complementary. As a result, edge information and salient features could advance each other while extracting features. Finally, we evaluate the 16 state-of-the-art VSOD methods on ViSal and FBMS datasets. At the same time we also evaluate the 13 state-of-the-art VSOD methods on UVSD and MCL datasets. The experiment results indicate that the proposed method outperforms the state-of-the-art approaches.

This work focuses on image anomaly detection by leveraging only normal images in the training phase. Most previous methods tackle anomaly detection by reconstructing the input images with an autoencoder (AE)-based model, and an underlying assumption is that the reconstruction errors for the normal images are small, and those for the abnormal images are large. However, these AE-based methods, sometimes, even reconstruct the anomalies well; consequently, they are less sensitive to anomalies. To conquer this issue, we propose to reconstruct the image by leveraging the structure-texture correspondence. Specifically, we observe that, usually, for normal images, the texture can be inferred from its corresponding structure (e.g., the blood vessels in the fundus image and the structured anatomy in optical coherence tomography image), while it is hard to infer the texture from a destroyed structure for the abnormal images. Therefore, a structure-texture correspondence memory (STCM) module is proposed to reconstruct image texture from its structure, where a memory mechanism is used to characterize the mapping from the normal structure to its corresponding normal texture. As the correspondence between destroyed structure and texture cannot be characterized by the memory, the abnormal images would have a larger reconstruction error, facilitating anomaly detection. In this work, we utilize two kinds of complementary structures (i.e., the semantic structure with human-labeled category information and the low-level structure with abundant details), which are extracted by two structure extractors. The reconstructions from the two kinds of structures are fused together by a learned attention weight to get the final reconstructed image. We further feed the reconstructed image into the two aforementioned structure extractors to extract structures. On the one hand, constraining the consistency between the structures extracted from the original input and that from the reconstructed image would regularize the network training; on the other hand, the error between the structures extracted from the original input and that from the reconstructed image can also be used as a supplement measurement to identify the anomaly. Extensive experiments validate the effectiveness of our method for image anomaly detection on both industrial inspection images and medical images.

We used broadband electroadhesion to reproduce the friction force profile measured as a finger slid across a textured surface. In doing so, we were also able to reproduce with high fidelity the skin vibrations characteristic of that texture; however, we found that this did not reproduce the original perception. To begin, the reproduction felt weak. In order to maximize perceptual similarity between a real texture and its friction force playback, the vibratory magnitude of the latter must be scaled up on average ≈ 3X for fine texture and ≈ 5X for coarse texture samples. This additional gain appears to correlate with perceived texture roughness. Additionally, even with optimal scaling and high fidelity playback, subjects could identify which of two reproductions corresponds to a real texture with only 71 % accuracy, as compared to 95 % accuracy when using real texture alternatives. We conclude that while tribometry and vibrometry data can be useful for texture classification, they appear to contribute only partially to texture perception. We propose that spatially distributed excitation of skin within the fingerpad may play an additional key role, and may thus be able to contribute to high fidelity texture reproduction.

Single image super-resolution (SR), as a basic computer vision task, has been widely studied. However, existing single image SR methods, applied in images with normal light, perform poorly on low-light images. To address this limitation, a method of structure-preserving and color-restoring up-sampling for single low-light image is proposed. Theoretical and experimental analysis indicates that existing methods suffer negative effects due to the suppressed color and the weakened texture in low-light image. Therefore, combined with the Retinex theory, the single low-light image up-sampling model is established for the first time, which avoids the conflict between color and texture by distinguishing reflectance and illumination. Further, we develop a structure-preserving and color-restoring up-sampling network for single low-light image SR. In the network, the reflectance and illumination components are obtained by decomposing the observed image, and then up-sampling of reflectance and enhancement of illumination are performed to complete the primary SR. In addition, the gradient information is reconstructed and fused into the up-sampling process to further enrich the SR texture. Experiments demonstrate that our method obtains competitive qualitative and quantitative evaluation on the produced dataset and real-world images.

Textures contain a wealth of image information and are widely used in various fields such as computer graphics and computer vision. With the development of machine learning, the texture synthesis and generation have been greatly improved. As a very common element in everyday life, wallpapers contain a wealth of texture information, making it difficult to annotate with a simple single label. Moreover, wallpaper designers spend significant time to create different styles of wallpaper. For this purpose, this paper proposes to describe wallpaper texture images by using multi-label semantics. Based on these labels and generative adversarial networks, we present a framework for perception driven wallpaper texture generation and style transfer. In this framework, a perceptual model is trained to recognize whether the wallpapers produced by the generator network are sufficiently realistic and have the attribute designated by given perceptual description; these multi-label semantic attributes are treated as condition variables to generate wallpaper images. The generated wallpaper images can be converted to those with well-known artist styles using CycleGAN. Finally, using the aesthetic evaluation method, the generated wallpaper images are quantitatively measured. The experimental results demonstrate that the proposed method can generate wallpaper textures conforming to human aesthetics and have artistic characteristics., IEEE Transactions on Circuits and Systems for Video Technology

Defects, which arise during the growth and processing of the wood, affect the visual and mechanical properties. To improve the utilization of wood with defects, we propose an image inpainting method that reconstructs the texture at defects. The result of image inpainting is visually very consistent with the surrounding healthy board. According to the result, as an application, the plate that is most similar to the surrounding texture could be matched to fill the missing position due to the removal of defects. First, the image inpainting is divided into two stages: coarse inpainting and detailed inpainting. The coarse generator quickly generates the prediction result of the missing area, and the result is fed to the detailed generator to produce a more refined image. Then, the skip connection is applied to the classical Encoder–Decoder in order to solve the problem of information loss due to upsampling. Next, the attention block is introduced in the detailed generator to pay more attention to the meaningful semantics. Finally, normalizing missing and known regions together would result in a bias in the mean and standard deviation, so Partition Normalization was used instead of Batch Normalization. The feature map is divided into two parts, foreground and background, and the two parts are normalized separately. The experimental results show that the results of image inpainting are still high precision with less time spent on the training network. Code, demo and models are available at: https://github.com/NEFUJoeyChen/TG-Net .

The electrovibration tactile display could render the tactile feeling of different textured surfaces by generating the frictional force through voltage modulation. When a user is sliding his/her finger on the display surface, he/she can feel the frictional texture. However, it is not trivial to prepare and fine-tune the appropriate frictional signals for haptic design and texture simulation. In this paper, we present a deep-learning-based framework to generate the frictional signals from the textured images of fabric materials. The generated frictional signal can be used for the tactile rendering on the electrovibration tactile display. Leveraging GANs (Generative Adversarial Networks), our system could generate the displacement-based data of frictional coefficients for the tactile display to simulate the tactile feedback of different fabric materials. Our experimental results show that the proposed generative model could generate the frictional-coefficient signals visually and statistically close to the ground-truth signals. The following user studies on fabric-texture simulation show that users could not discriminate the generated and the ground-truth frictional signals being rendered on the electrovibration tactile display, suggesting the effectiveness of our deep-frictional-signal-generation model.

Pansharpening aims to fuse a multispectral (MS) image with low spatial resolution and a panchromatic (PAN) image with a high-spatial resolution to produce an image with both high spectral and high spatial resolution. In this study, we propose a variational pansharpening method by exploiting cartoon-texture similarities. After decomposition of the PAN image, the cartoon component always contains the global structure information, while the texture component includes the locally patterned information. This enables that the fused high-spatial resolution MS image can preserve the global and local spatial details (e.g., high-order information) well after leveraging the similarities of cartoon and texture components from PAN and MS images. To explore such cartoon-texture similarities, we describe cartoon similarity as gradient sparsity, formulated as a reweighted total variation term. Meanwhile, we use group low-rank constraint for texture similarity that is presented as repetitive texture patterns. By incorporating a data fidelity term for preserving the spectral information on the basis that the down-sampled fused MS image is consistent with the MS image, we further formulate pansharpening as an optimization problem and solve it efficiently using the alternative direction multiplier method. Extensive experiments have been conducted on a series of satellite data sets, and we also carry out a simulated vegetation coverage change experiment to verify the efficiency of the proposed method in remote sensing. The qualitative and quantitative results demonstrate that our method outperforms the state-of-the-art pansharpening methods in terms of both visual effect and objective metrics.

Photo style transfer aims to change the style of a given photo to a reference style image with the constraint by retaining the broad and faithful conservation of the content of the input image. Most previous algorithms still have challenging issues on how to exactly extract and represent the style of the image to avoid the interruption of human visual perception. In this paper, we present a texture preserving photo style transfer algorithm by separating the input image into texture and structure and then applying the deep structure style transfer network to effectively change the extracted style characteristics of the structure. The texture preserving photo style transfer overcomes the main drawback of the previous approaches like distortion and saturation of the boundary of the objects. The quantitative and qualitative experimental results including user study prove that the proposed photo style transfer is universally applicable comparing to remarkable previous approaches.

In this study, we present a novel method for recognizing the properties of materials by using the returning echoes of self-emitted acoustic signals. This is achieved by mimicking the acoustic recognition mechanisms that are present in several animals like bats and dolphins. To implement this mechanism, we propose an end-to-end machine learning approach that uses convolutional neural networks (CNNs), which can translate spectral cues that are delivered by reflected echoes into meaningful information for the target object. To validate the proposed approach, we conducted two different experiments. In the first experiment, we attempted to learn a set of predefined everyday surfaces using that reflected acoustic signals that are emitted from a distance. As an extension of the first experiment, the second experiment investigated whether echo-acoustic signals can be used for the classification of an acoustically smooth surface (e.g., liquid solutions). The experimental results showed that our system can recognize texture and, density information. In comparison to the conventional machine learning approach that relies on feature-engineering processes, our approach that is based on CNNs demonstrated high accuracies (>90%) in the test datasets. We conclude this paper by discussing the limitations of the proposed approach and exploring the potential avenues to incorporate acoustic recognition functionality into computers.

Purpose We evaluated the diagnostic performance of the texture features of dynamic contrast-enhanced (DCE) MRI for breast cancer diagnosis in which the discriminator was optimized, so that the specificity was maximized via the restriction of the negative predictive value (NPV) to greater than 98%. Methods Histologically proven benign and malignant mass lesions of DCE MRI were enrolled retrospectively. Training and testing sets consist of 166 masses (49 benign, 117 malignant) and 50 masses (15 benign, 35 malignant), respectively. Lesions were classified via MRI review by a radiologist into 4 shape types: smooth (S-type, 34 masses in training set and 8 masses in testing set), irregular without rim-enhancement (I-type, 60 in training and 14 in testing), irregular with rim-enhancement (R-type, 56 in training and 22 in testing), and spicula (16 in training and 6 in testing). Spicula were immediately classified as malignant. For the remaining masses, 298 texture features were calculated using a parametric map of DCE MRI in 3D mass regions. Masses were classified into malignant or benign using two thresholds on a feature pair. On the training set, several feature pairs and their thresholds were selected and optimized for each mass shape type to maximize specificity with the restriction of NPV > 98%. NPV and specificity were computed using the testing set by comparison with histopathologic results and averaged on the selected feature pairs. Results In the training set, 27, 12, and 15 texture feature pairs are selected for S-type, I-type, and R-type masses, respectively, and thresholds are determined. In the testing set, average NPV and specificity using the selected texture features were 99.0% and 45.2%, respectively, compared to the NPV (85.7%) and specificity (40.0%) in visually assessed MRI category-based diagnosis. Conclusion We, therefore, suggest that the NPV of our texture-based features method described performs similarly to or greater than the NPV of the MRI category-based diagnosis.

Texture analysis, as well as its broader category radiomics, describes a variety of techniques for image analysis that quantify the variation in surface intensity or patterns, including some that are imperceptible to the human visual system. Cerebral gliomas have been most rigorously studied in brain tumors using MR-based texture analysis (MRTA) to determine the correlation of various clinical measures with MRTA features. Promising results in cerebral gliomas have been shown in the previous MRTA studies in terms of the correlation with the World Health Organization grades, risk stratification in gliomas, and the differentiation of gliomas from other brain tumors. Multiple MRTA studies in gliomas have repeatedly shown high performance of entropy, a measure of the randomness in image intensity values, of either histogram- or gray-level co-occurrence matrix parameters. Similarly, researchers have applied MRTA to other brain tumors, including meningiomas and pediatric posterior fossa tumors.However, the value of MRTA in the clinical use remains undetermined, probably because previous studies have shown only limited reproducibility of the result in the real world. The low-to-modest generalizability may be attributed to variations in MRTA methods, sampling bias that originates from single-institution studies, and overfitting problems to a limited number of samples.To enhance the reliability and reproducibility of MRTA studies, researchers have realized the importance of standardizing methods in the field of radiomics. Another advancement is the recent development of a comprehensive assessment system to ensure the quality of a radiomics study. These two-way approaches will secure the validity of upcoming MRTA studies. The clinical use of texture analysis in brain MRI will be accelerated by these continuous efforts.

Dynamic texture refers to the content in video sequences that is characterized by spatial repetition and temporal variation, such as swaying foliage and flowing water. It is a great challenge to compress the dynamic textures efficiently in the current prediction/transform hybrid video coding framework. However, these textures have little information for machine vision, and human visual perception is less sensitive to the textures than to the structures. Thus, we propose a spatiotemporal generative adversarial network (GAN) based dynamic texture synthesis method for surveillance video coding. We detect and remove the dynamic texture content at encoder side, which is irrelevant to machine vision. We generate the dynamic texture content using the proposed GAN at decoder side, so that the reconstructed videos can be observed by human without deteriorating perceptual quality. Specifically, we design a GAN network to synthesize dynamic textures by exploiting the correlation between spatial and temporal neighbors; we present a surveillance video coding scheme with the dynamic texture detection/synthesis method; we build a high-quality dynamic texture dataset, and we collect a dynamic texture testing dataset that goes beyond the existing video coding test datasets by focusing on surveillance scenes. The proposed video coding scheme has been implemented on top of the High Efficiency Video Coding (HEVC) reference software. Experiments have been conducted to evaluate the quantitative and qualitative performance of the proposed coding scheme. Our method achieves 7.4% and 7.6% bit-rate savings in low-delay-B and low-delay-P settings, respectively, at similar visual quality levels in comparison with HEVC.

Images captured under low-light conditions are generally characterized by poor illumination, low contrast, and nonignorable large amount of noise. In order to improve the visibility in weak illumination scenes, multiple artificial light sources are used, which leads to severe uneven illumination of the scene. The main challenges of dehazing images with complex illumination are to suppress the boosting of unsightly noise when enhancing contrast and avoid overenhancement in bright glow regions. To circumvent problems above, this letter proposes a generalized enhancement framework, which works well not only in uniform light conditions but also in strongly nonuniform illumination low-light scenes. To achieve this, we first decompose the input hazy image into a structure layer containing low-frequency illumination variance and a texture layer containing large amount of high-frequency details. Sequentially, benefit from two derived masks that are intrinsically similar to weight maps, the proposed framework can perform regional adaptive brightness adjustment on the structure layer according to the distribution of light in the input image. Meanwhile, regions of effective details in the texture layer are assigned higher weights, while regions that belong to noise are suppressed. Finally, adding the enhanced texture layer back to the brightened structure layer, visually appealing results are generated. Experimental results on various scenarios demonstrate the superiority of the proposed framework over state-of-the-art methods in terms of both qualitative and quantitative.

In order to more efficiently mine the features of polarimetric synthetic aperture radar (PolSAR) and establish a more appropriate classification model, this article proposes an improved convolutional autoencoder (ICAE) based on texture feature fusion (TFF-ICAE) for PolSAR terrain classification. Firstly, TFF-ICAE specifically designs a multi-indicator squeeze-and-excitation (MI-SE) block and incorporates it into the CAE network. MI-SE can enhance the essential feature information while suppressing the interference information as much as possible, and it can effectively increase the between-class distance while reducing the within-class distance. Then, TFF-ICAE uses gray level co-occurrence matrix (GLCM) to capture the texture features, and it optimally fuses these texture features and the deep features extracted by ICAE to complete the multi-level feature fusion, elevating the feature representation completeness of the terrain. That is, TFF-ICAE effectively enhances the feature separation capability of different categories while greatly elevating the feature representation completeness. Experiments on the data sets of San Francisco, Oberpfaffenhofen and Flevoland show that the proposed TFF-ICAE respectively achieves overall accuracies of 93.44%, 97.61% and 97.78%, which are at least 0.92%, 1.52% and 0.97% higher than other algorithms. Undoubtedly, the superiority of TFF-ICAE is verified on these data sets.

Fake detection has become an urgent task. Generative adversarial networks (GANs) extended to deep learning has shown its extraordinary ability in the fields of image, audio, and speech. But advanced technology benefits us, it also poses a threat to us when used in Cyber Crime. The Deepfake (common name for face manipulation methods) based on GANs can realize the replacement of different faces. Due to the development of GANs, faces generated by Deepfake can already be visually real. Deepfake can purposely replace any face to a different person, so that a fabricated event may be widely spread because of the convenience of the Internet, causing serious impacts such as personal attacks and cyber crime. Based on cutting-edge research , this paper proposes a intelligence forensic method of Deepfake detection. We first discover the subtle texture differences between real and fake image in image saliency, which shows difference in the texture of faces. To amplify this difference, we exploit guided filter with saliency map as guide map to enhance the texture artifacts caused by the post-processing and display the potential features of forgery. Resnet18 classification network efficiently learns the exposed difference and finally realizes the real and fake detection of face images. We evaluate the performance of the method and experiments verify that the proposed method can achieve the state-of-the-art detection accuracy .

This systematic review aimed to assess the methods used to classify mammographic breast parenchymal features in relation to prediction of future breast cancer including the time from mammogram to diagnosis of breast cancer, and methods for the identification of texture features and selection of features for inclusion in analysis. The databases including Medline (Ovid) 1946-,Embase.com1947-, CINAHL Plus 1937-, Scopus 1823-, Cochrane Library (including CENTRAL), andClinicaltrials.gov. were searched through October 2021 to extract published articles in English describing the relationship of parenchymal texture features with risk of breast cancer. Twenty-eight articles published since 2016 were included in the final review. Of these, 7 assessed texture features from film mammograms images, 3 did not report details of the image used, and the others used full field mammograms from Hologic, GE and other manufacturers. The identification of parenchymal texture features varied from using a predefined list to machine-driven identification. Reduction in number of features chosen for analysis in relation to cancer incidence then varied across statistical approaches and machine learning methods. The variation in approach and number of features identified for inclusion in analysis precluded generating a quantitative summary or meta-analysis of the value of these to improve predicting risk of future breast cancers. This updated overview of the state of the art revealed research gaps; based on these, we provide recommendations for future studies using parenchymal features for mammogram images to make use of accumulating image data, and external validation of prediction models that extend to 5 and 10 years to guide clinical risk management. By following these recommendations, we expect to improve risk classification and risk prediction for women to tailor screening and prevention strategies to level of risk.

In order to improve the visual communication ability of traditional national costume patterns, it is necessary to carry out image texture intelligent matching processing. A traditional national costume embroidery feature recognition algorithm based on a double-layer model is proposed. The traditional national costume pattern texture intelligent information acquisition model under the double-layer model is constructed to carry out texture imaging and feature segmentation of traditional national costume patterns, extract the texture histogram of traditional national clothing pattern and national design language, carry out texture segmentation and automatic matching under the two-layer model according to the histogram distribution, enhance and optimize the texture information of traditional national clothing pattern, extract the edge contour feature points of traditional national clothing pattern, and complete the embroidery feature recognition of traditional national clothing. The experimental results show that the designed recognition algorithm has high integrity and accuracy.

Background At present, the residual film pollution in cotton fields is crucial. The commonly used recycling method is the manual-driven recycling machine, which is heavy and time-consuming. The development of a visual navigation system for the recovery of residual film is conducive, in order to improve the work efficiency. The key technology in the visual navigation system is the cotton stubble detection. A successful cotton stubble detection can ensure the stability and reliability of the visual navigation system. Methods Firstly, it extracts the three types of texture features of GLCM, GLRLM and LBP, from the three types of images of stubbles, residual films and broken leaves between rows. It then builds three classifiers: Random Forest, Back Propagation Neural Network and Support Vector Machine in order to classify the sample images. Finally, the possibility of improving the classification accuracy using the texture features extracted from the wavelet decomposition coefficients, is discussed. Results The experiment proves that the GLCM texture feature of the original image has the best performance under the Back Propagation Neural Network classifier. As for the different wavelet bases, the vertical coefficient texture feature of coif3 wavelet decomposition, combined with the texture feature of the original image, is the feature having the best classification effect. Compared with the original image texture features, the classification accuracy is increased by 3.8%, the sensitivity is increased by 4.8%, and the specificity is increased by 1.2%. Conclusions The algorithm can complete the task of stubble detection in different locations, different periods and abnormal driving conditions, which shows that the wavelet coefficient texture feature combined with the original image texture feature is a useful fusion feature for detecting stubble and can provide a reference for different crop stubble detection.

Three-dimensional Morphable Models (3DMMs) are powerful statistical tools for representing the 3D shapes and textures of an object class. Here we present the most complete 3DMM of the human head to date that includes face, cranium, ears, eyes, teeth and tongue. To achieve this, we propose two methods for combining existing 3DMMs of different overlapping head parts: i. use a regressor to complete missing parts of one model using the other, ii. use the Gaussian Process framework to blend covariance matrices from multiple models. Thus we build a new combined face-and-head shape model that blends the variability and facial detail of an existing face model (the LSFM) with the full head modelling capability of an existing head model (the LYHM). Then we construct and fuse a highly-detailed ear model to extend the variation of the ear shape. Eye and eye region models are incorporated into the head model, along with basic models of the teeth, tongue and inner mouth cavity. The new model achieves state-of-the-art performance. We use our model to reconstruct full head representations from single, unconstrained images allowing us to parameterize craniofacial shape and texture, along with the ear shape, eye gaze and eye color., 18 pages, 18 figures, submitted to Transactions on Pattern Analysis and Machine Intelligence (TPAMI) on the 9th of October as an extension paper of the original oral CVPR paper : arXiv:1903.03785

In view of the polysemy of mural images and the style difference among mural images painted in different dynasties as well as the high energy costs of the traditional manual dynasty classification method, which resorts to mural texts and historical documents, this study proposed an adaptive enhancement capsule network (AECN) for automatic dynasty identification of mural images. Based on the original capsule network, we introduced a preconvolution structure to extract the high-level features of the mural images from Mogao Grottoes, such as color and texture. Then, we added an even activation operation to the layers of the network to enhance the fitting performance of the model. Finally, we performed adaptive modifications on the capsule network to increase the gradient smoothness of the model, based on which to optimize the model and thus to increase its classification precision. With the self-constructed DH1926 data set as the study subject, the proposed model achieved an accuracy of 84.44%, an average precision of 82.36%, an average recall rate of 83.75% and a comprehensive assessment score F1 of 83.96%. Compared with modified convolution neural networks and the original capsule network, the model proposed in study increased all the considered indices by more than 3%. It has a satisfactory fitting performance, which can extract the rich features of mural images at multiple levels and well express their semantic information. Furthermore, it has a higher accuracy and better robustness in the classification of the Mogao Grottoes murals, and therefore is of certain application values and research significance.

To evaluate whether mammographic texture features were associated with second primary contralateral breast cancer (CBC) risk, we created a “texture risk score” using pre-treatment mammograms in a case–control study of 212 women with CBC and 223 controls with unilateral breast cancer. The texture risk score was associated with CBC (odds per adjusted standard deviation = 1.25, 95% CI 1.01–1.56) after adjustment for mammographic percent density and confounders. These results support the potential of texture features for CBC risk assessment of breast cancer survivors.

Background Digital cervical auscultation (CA) has been proposed since the 1950s for screening aspiration among patients with dysphagia. Researchers have investigated the 'external' effects such as bolus viscosity, volume, and head and neck positions. However, the influences of standardized liquid viscosity and food texture on swallowing sounds have not been fully understood due to lacking uniform standardization of bolus preparation. Furthermore, a paucity of the literature recommends proper viscous liquids and foods to start swallowing training or monitor the swallowing progress during the continuum of disease based on acoustic signals. Aims To investigate the effects of eight-level liquids and foods on swallowing sound features based on the International Dysphagia Diet Standardisation Initiative (IDDSI). Methods & procedures We collected swallowing sounds from 30 healthy participants ranging in age from 19 to 60 years and who were self-reporting no history of swallowing disorders. Each participant swallowed liquids and foods regarding different consistency or texture with their head-trunk in a neutral position. Outcomes & results Features of swallowing acoustic signals and the IDDSI flow test as well as food test confirmed the level 3 moderately thick (MO3) was more suitable to categorize into liquids and the level 4 extremely thick (EX4) was more corresponded to the properties of food bolus. We found significant differences in duration of acoustic signals across different liquids and foods except between swallowing level 0 thin liquid and level 1 slightly thick liquid, as well as EX4 and level 5 minced and moist. Our results also demonstrated liquid viscosity significantly impacted the peak intensity of swallowing sounds. Conclusions & implications As an initial exploration of digital CA across eight levels of different liquids and foods according to the IDDSI, we established the baseline findings for future comparisons with other study populations or other various consistent liquids/foods. Although both MO3 and EX4 can be considered as liquid or food boluses with high thickness, MO3 might be suitable as the 'start liquid' for patients with dysphagia; however, the decision still needs to be confirmed by the healthcare provider based on patients' safety and the area of deficit. We also concluded there are influences of varied fluid consistency and food texture on swallowing sounds. Furthermore, future investigations should explore whether changing viscosity levels could either continuously or discretely disturb the swallowing acoustic signals. What this paper adds What is already known on the subject Previous studies have found that the 'external' effects such as bolus viscosity, volume, and head and neck positions. Due to lacking uniform standardization of bolus preparation, there is limited information about the influences of standardized liquid viscosity and food texture on swallowing sounds. What this paper adds to the existing knowledge As an initial exploration, we utilized digital CA with a large sample of viscous liquids and different textures of foods based on the IDDSI to investigate the swallowing sounds. What are the potential or actual clinical implications of this work? This study confirms that the effects of various fluid consistency and food texture on swallowing acoustic signals. However, the findings of this study support the need for further research relating to changing viscosity could either continuously or discretely disturb the swallowing acoustic signals.

Recent advances in image super-resolution (SR) explore the power of deep learning to achieve a better reconstruction performance. The deep reconstruction methods mostly reconstruct the super-resolution image based on a large amount of external data training. In order to make full use of the internal features of the image, we propose a novel network for SR called super-resolution network by fusing internal and external features (SRNIF). Specifically, we use the self-similar image itself as the prior knowledge and propose a fully convolutional deep neural network (called Internal Feature Enhancement module, IFE) to extract the internal features of the image so as to obtain detailed texture of the inputted image. Then, the detailed features extracted by the internal feature enhancement module are embedded into the residual network, and the spatial feature transform (SFT) module is used to form the residual module. A deep residual network based on the fusion of internal and external features is built. The proposed method comes with a strong early reconstruction ability and can create the final high-resolution image step by step. Experimental results show that the method has enhanced the texture characteristics and quality of the reconstructed image, and the visual effect and peak signal-to-noise ratio (PSNR) has been improved.

Virtual reality facilities have matured in recent years; however, cybersickness has yet remained a major issue in the virtual reality content. The researchers in this study attempt to simulate the different focal points and depth of field of human vision with an eye on reducing cybersickness by implementing blurring. This paper introduces a new type of technology—texture blur, which was tested in three experiments with different contents in this study. The participants were asked to verbally answer two questionnaires: a discomfort level questionnaire and a simulator sickness questionnaire (SSQ). The results were analyzed to compare the seriousness of the participants’ cybersickness with and without texture blur in the virtual reality content. The experiment results revealed that the participants with texture blur had lower discomfort level scores. The SSQ results also went in line with the previous questionnaire, showing better results with texture blur. Hence, the researchers concluded that texture blur can effectively reduce users’ cybersickness in virtual reality content.

As a sensor with a wide field of view, the panoramic vision sensor is efficient and convenient in perceiving the characteristic information of the surrounding environment and plays an important role in the experience of artistic design of images. The transformation of visual and other sensory experiences in art design is to integrate sound, image, texture, taste, and smell with each other through reasonable rules, to create more excellent crossborder art design works. To improve the sensory experience that art design works bring to the audience, the combination of vision and other sensory experiences can maximize the advantages of multiple information dissemination methods and combine the omnidirectional visual sensor with the sensory experience of art design images. In the method part, this article introduces the omnidirectional vision sensor, art design image, and sensory experience modes and content and introduces the hyperbolic concave mirror theory and the Micusik perspective projection imaging model. In the experimental part, the experimental environment, experimental objects, and experimental procedures of this article are introduced. In the analysis part, this article analyzes the six aspects of image database dependency test, performance, comparison of different distortion types, false detection rate and missing detection rate, algorithm time-consuming comparison, sensory experience analysis, and feature point screening. Among the feelings of the art design image, for the first image, 87.21% of the audience’s feelings are happy, indicating that the main idea of this image can bring joy to people. In the second image, the audience’s feelings are mostly sad. For the third image, more than half of the audience’s feelings are melancholy. For the fourth image, 69.34% of the audience’s inner feelings are calm. It explains that the difference in the content of art design images can bring different sensory experiences to people.

Communication between human beings has several ways, one of the most known and used is speech, both visual and acoustic perceptions sensory are involved, because of that, the speech is considered as a multi-sensory process. Micro contents are a small pieces of information that can be used to boost the learning process. Deep learning is an approach that dives into deep texture layers to learn fine grained details. The convolution neural network (CNN) is a deep learning technique that can be employed as a complementary model with micro learning to hold micro contents to achieve special process. In This paper a proposed model for lip reading system is presented with proposed video dataset. The proposed model receives micro contents (the English alphabet) in video as input and recognize them, the role of CNN deep learning is clearly appeared to perform two tasks, the first one is feature extraction and the second one is the recognition process. The implementation results show an efficient accuracy recognition rate for various video dataset that contains variety lip reader for many persons with age range from 11 to 63 years old, the proposed model gives high recognition rate reach to 98%.

Visual texture, defined by local image statistics, provides important information to the human visual system for perceptual segmentation. Second-order or spectral statistics (equivalent to the Fourier power spectrum) are a well-studied segmentation cue. However, the role of higher-order statistics (HOS) in segmentation remains unclear, particularly for natural images. Recent experiments indicate that, in peripheral vision, the HOS of the widely adopted Portilla-Simoncelli texture model are a weak segmentation cue compared to spectral statistics, despite the fact that both are necessary to explain other perceptual phenomena and to support high-quality texture synthesis. Here we test whether this discrepancy reflects a property of natural image statistics. First, we observe that differences in spectral statistics across segments of natural images are redundant with differences in HOS. Second, using linear and nonlinear classifiers, we show that each set of statistics individually affords high performance in natural scenes and texture segmentation tasks, but combining spectral statistics and HOS produces relatively small improvements. Third, we find that HOS improve segmentation for a subset of images, although these images are difficult to identify. We also find that different subsets of HOS improve segmentation to a different extent, in agreement with previous physiological and perceptual work. These results show that the HOS add modestly to spectral statistics for natural image segmentation. We speculate that tuning to natural image statistics under resource constraints could explain the weak contribution of HOS to perceptual segmentation in human peripheral vision.

The present article is part of a larger project on Conrad’s less known short fiction, the area of his writing which is largely undervalued, and even deprecated at times. The paper’s aim is to enhance the appreciation of “A Smile of Fortune,” by drawing attention to its “inner texture” as representative of Conrad’s “art of expression,” especially in view of the writer’s own belief in the supremacy of form over content as well as “suggestiveness” over “explicitness” in his fiction. To achieve this aim a New Critical (“close reading”), intertextual and comparative approaches to Conrad’s story have been adopted, involving nineteenth- and twentieth-century American literary texts, i.e., both those preceding and those following the publication of Conrad’s ’Twixt Land and Sea (1912) volume featuring the tale in question. The intertextual reading of “A Smile of Fortune” against Bernard Malamud’s short story “The Magic Barrel,” Herman Melville’s Moby Dick, and William Faulkner’s Absalom, Absalom!, with Light in August as a point of reference, reveals the workings in Conrad’s story of the modernist device of denegation, which, alongside antithesis and oxymoron, seems to be largely responsible for the tale’s contradictions and ambiguities, which should thus be perceived as the story’s asset rather than flaw. The textual evidence of Conrad’s tale, as well as its comparison with three short stories: Nathaniel Hawthorne’s “Rappaccini’s Daughter,” Edgar Allan Poe’s “The Fall of the House of Usher,” and Peter Taylor’s “Venus, Cupid, Folly and Time,” seem to confirm the presence of the implications of the theme of incest in Conrad’s text, heretofore unrecognized in criticism. Overall, the foregoing analysis of “A Smile of Fortune” hopes to account for, if not disentangle, the story’s complex narratological meanderings and seemingly insoluble ambiguities, particularly as regards character and motive, naming Conrad rather than Faulkner the precursor of denegation.

The article presents approaches to the automatic detection of sound repetitions and the measurement of sound coherence in Russian poetic text basing on the syllabocentric concept of the sound texture of the verse, which distinguishes the phonosyllabeme as its main operational unit and determines the main types of relations between the elements of repetition - equiphony and metaphony - and also allows describe the text in its unity of sound-segmental and rhythmic structure. The paper considers the linguistic prerequisites, main algorithms and methods, which form the basis of the PHONOTEXT computer program (web application), which automatically analyzes and visually forms the picture of the sound composition of the text presenting its quantitative characteristics. The basic operations of the program are described in detail, as well as the mathematical formulas used to identify the structure and carry out mathematical measurement of sound repetitions in Russian poetic speech. The article discusses the results of the programs use for the comparative study of poetry and prose, individual works and individual styles of Russian poets. In particular, the data obtained using the Phonotext program allows us to speak of a tendency towards syntagmatic consolidation of the phonosyllabeme and intensive granulation of the sound texture of the text in the individual style of Osip Mandelstam as the evidence of his striving for poetic morphologization of the elements of sound repetition, it also shows a tendency towards the use of the dominant sound theme as a means for cohesion of a complex syntactic and compositional whole, in contrast to Vladimir Nabokovs idiostyle, which is characterized by its orientation towards splitting up the elements of repetition, with a relatively chaotic interaction and spreading of consonances over the text and their low activity as a means of syntagmatic segmentation and consolidation of the text. The prospects for the development of the Phonotext program are described for identifying the relations of segment-sound and prosodic parallelism in speech (relations of equirhythmy and equiphony, including rhyme), which will provide an automatic presentation of a complete picture of the sound composition, the interaction of its main forms in the text and its further functional description.

This study was aimed at evaluating the effect of freeze-drying and lacto-fermentation on the texture parameters of images and sensory attributes of beetroots. The samples were imaged using a flatbed scanner, and textures from images converted to color channels L, a, b, R, G, B, X, Y, Z were computed. The discrimination of raw and processed beetroots was performed using models based on textures selected for each color channel. The sensory quality of processed samples was determined using the attributes related to smell, color, texture and taste. The highest discrimination accuracy of 97.25% was obtained for the model built for color channel b. The accuracies for other channels were equal to 96.25% for channel a, 95.25% for channel R, 95% for channel Y, 94.75% for channel B, 94.5% for channel X, 94% for channel L, 92.5% for channel G, 88.25% for channel Z. In the case of some models, the raw and lacto-fermented beetroots were discriminated with 100% correctness. The freeze-dried and freeze-dried lacto-fermented samples were also the most similar in terms of sensory attributes, such as off-odor, attractiveness color, beetroot color, crunchiness, hardness, bitter taste, overall quality. The results indicated that the image parameters and sensory attributes may be related.

Electromyography (EMG) is an upcoming technique for food texture evaluation which has the potential to come up as a successful comprehensive texture evaluation method. EMG texture evaluation is in its development phase and its relations with conventional texture measurement techniques need to be established and evaluated. The objective of this study was to investigate the relationship of EMG masticatory variables with sensory texture and instrumental texture parameters as measured for different textured foods. Five food samples were purchased from the local market namely, dhokla, paneer, rasgulla, cake and jelly. Food samples were subjected to proximate analysis, sensory textural evaluation and two-bite tests using texture profile analyzer. Nutritional compositions of all the selected food materials were significantly (p ≤ 0.01) different values. Textural parameters like hardness, cohesiveness and gumminess showed highly significant (p ≤ 0.01) variations among different selected foods. The study was conducted on eight human subjects. Eighteen EMG mastication variables were obtained during chewing of these five different foods. Fifteen mastication variables were found to be significantly (p ≤ 0.05) effective in explaining the variation in texture of the selected food samples. Mastication parameters were reduced from eighteen to five in number using cluster analysis. Correlation coefficients between EMG mastication parameters at different stages (early, middle and late), and sensory and instrumental texture parameters were found significant. Principal component analysis was conducted on various mastication parameters which resulted in two meaningful principal components, comprising of mastication time, total muscle activity, cycle time, early burst duration and early muscle activity, could explain a cumulative variance of up to 82%. The first principal component (PC1) scores correlated significantly with sensory texture parameters and instrumental texture parameters namely hardness, cohesiveness, chewiness gumminess and adhesiveness. PC1 showed correlation with sensory cohesiveness, chewiness and adhesiveness at (p ≤ 0.05) while with hardness at (p ≤ 0.10) and with gumminess at (p ≤ 0.20). For the instrumental texture parameters significant correlations were found with cohesiveness and gumminess at (p ≤ 0.05). The study revealed that EMG is capable of comprehensively quantifying the dynamic changes occurring in the mouth during chewing of different textured foods. So the information generated through EMG investigations can complement the sensory and instrumental textural analysis for designing new foods.

The relevance of the topic and problem statement. In the process of formation and development of the original style of button accordion music one of the most important roles is played by the factor of texture, which fully embodies the instrumental specificity of the button accordion, and reflects the originality of forms and varieties of musical fabric in button accordion works. The study of the peculiarities of texture organization, as well as highlighting the main trends in texture formation in modern button accordion music, in particular in the works of Ukrainian composers, is seen as an urgent task to understand the essence of the phenomenon of modern button accordion, and in general — to develop domestic musicology in instrumental art. In previous works of the author of this article, an attempt was made to systematize the textured formations of modern button accordion music, which are determined by the specifics of the design of button accordion keyboards (Stashevskyі, 2021). Continuing further study of the textured organization of modern music for button accordion, there is a need to identify and systematize other typical for this work samples of texture and its components, the structure of which is not due to the structural configuration of sound keys on button accordion keyboards. That is, those that are more universal, but, at the same time, are characterized as typical of modern button accordion art. The purpose of the study is to study the most typical textural elements and combinations of modern button accordion music, as well as their systematization based on the analysis of their internal structural organization. The methodology used in this article is based on the use of methods of theoretical musicology (primarily, analysis of elements of musical language and means of expression), as well as methods of structural and system analysis, generalization and classification. The results. The systematization of typical original texture formulas of modern button accordion creativity carried out in this work provides their division into certain groups and kinds (subgroups) in the middle of these groups, in particular: reverberation (chordal and mixed); stereophony (pulsation, transmission, unison); pedalling arpeggio (rhythmic, non­rhythmic); pedal scale (cluster crescendo); monophonic structural­modular constructions (chromatic by broken intervals; gamma­shaped­arpeggio based on a 5­element fingering group); chord transpositions (interval­algorithmic (chromatics; m3; m2­B2; etc.), thematic­melodic); simplified accompaniment (alternation of tessitura­polar sounds); multi­octave duplications (unison, decomposed­arpeggio). The topicality of the work lies in the fact that for the first time in modern musicology the textured elements of modern button accordion music have received analytical discourse and systematization. The practical significance of the work lies in the possibility of its use in theoretical courses on the study of button accordion art, as well as in the practical work of button accordionists. Conclusions. To sum up, we can state that the considered original textural elements and formulas are typical for modern button accordion music, as they occur in works in many cases. The typical button accordion texture formulas considered in this article, which are often found in the artistic plane of musical works, play an important role in the formation of a kind of stylistic handwriting of modern button accordion speech. At the same time, they, in contrast to the group of determinant texture formulas, reflect a certain universalism of their presentation, as they can be transferred to other instrumental texture (keyboard instruments) conditions and implemented in them.

The aim of this article is to propose a model to automatically predict visual judgement of sparkle and graininess of special effect pigments used in industrial coatings. Many applications in the paint and coatings, printing and plastics industry rely on multi-angle color measurements with the aim of properly characterizing the appearance, i.e., the color and texture of the manufactured surfaces. However, when it comes to surfaces containing effect pigments, these methods are in many cases insufficient and it is particularly texture characterization methods that are needed. There are two attributes related to texture that are commonly used: (1) diffuse coarseness or graininess and (2) sparkle or glint impression. In this paper, we analyzed visual perception of both texture attributes using two different psychophysical studies of 38 samples painted with effect coatings including different effect pigments and 31 test persons. Our previous work has shown a good agreement between a study using physical samples with one that uses high-resolution photographs of these sample surfaces. We have also compared the perceived (1) graininess and (2) sparkle with the performance of two commercial instruments that are capable of capturing both attributes. Results have shown a good correlation between the instruments’ readings and the psychophysical studies. Finally, we implemented computational models predicting these texture attributes that have a high correlation with the instrument readings as well as the psychophysical data. By linear scaling of the predicted data using instruments readings, one can use the proposed model for the prediction of graininess and both static and dynamic sparkle values.

In this article, we investigate the importance of phase for texture discrimination and similarity estimation tasks. We first use two psychophysical experiments to investigate the relative importance of phase and magnitude spectra for human texture discrimination and similarity estimation. The results show that phase is more important to humans for both tasks. We further examine the ability of 51 computational feature sets to perform these two tasks. In contrast with the psychophysical experiments, it is observed that the magnitude data is more important to these computational feature sets than the phase data. We hypothesise that this inconsistency is due to the difference between the abilities of humans and the computational feature sets to utilise phase data. This motivates us to investigate the application of the 51 feature sets to phase-only images in addition to their use on the original data set. This investigation is extended to exploit Convolutional Neural Network (CNN) features. The results show that our feature fusion scheme improves the average performance of those feature sets for estimating humans’ perceptual texture similarity. The superior performance should be attributed to the importance of phase to texture similarity.

Detail-and-geometry richness is essential to bas-relief modelling. However, existing image-based and model-based bas-relief modelling techniques commonly suffer from detail monotony or geometry loss. In this paper, we introduce a new bas-relief modelling framework for detail abundance with visual attention based mask generation and geometry preservation, which benefits from our two key contributions. For detail richness, we propose a novel semantic neural network of normal transfer to enrich the texture styles on bas-reliefs. For geometry preservation, we introduce a normal decomposition scheme based on Domain Transfer Recursive Filter (DTRF). Experimental results demonstrate that our approach is advantageous on producing bas-relief modellings with both fine details and geometry preservation.

Recent studies have suggested that textural characteristics of the intima-media complex (IMC) may be more useful than the intima-media thickness (IMT) in evaluating cardiovascular risk. The primary aim of our study was to investigate the association between texture features of the common carotid IMC and prevalent clinical cardiovascular disease (CVD). The secondary aim was to determine whether IMT and IMC texture features vary between the left and right carotid arteries. The study was performed on 2208 longitudinal-section ultrasound images of the left (L) and right (R) common carotid artery (CCA), acquired from 569 men and 535 women out of which 125 had clinical CVD. L and R sides of the IMC were intensity normalized and despeckled. The IMC was semiautomatically delineated for all images using a semiautomated segmentation system, and 61 different texture features were extracted. The corresponding IMT semiautomated measurements (mean±SD) of the L and R sides were 0.73±0.21 mm/0.69±0.19 mm for the normal population and 0.83±0.17 mm/0.79±0.18 mm for those with CVD. IMC texture features did not differ between the right- and left-hand sides. Several texture features were independent predictors of the presence of CVD. The multivariate logistic regression analysis combining age, IMT, and texture features produced a receiver operating characteristic curve with an area under the curve of 89%. A correct classification rate of 77% for separating the normal subject (NOR) versus CVD subjects was achieved using the support vector machine classifier with a combination of clinical features, IMT, and extracted texture features. Texture features provide additional information on the presence of clinical CVD, which is over and above that provided by conventional risk factors or IMT alone. The value of IMC texture features in the prediction of future cardiovascular events should be tested in prospective studies.

Recently, deep convolutional neural networks have achieved great success for single-image deraining. However, affected by the intrinsic overlapping between rain streaks and background texture patterns, a majority of these methods tend to almost remove texture details in rain-free regions and lead to over-smoothing effects in the recovered background. To generate reasonable rain streak layers and improve the reconstruction quality of the background, we propose a distributed feedback network (DFN) in recurrent structure . A novel feedback block is designed to implement the feedback mechanism. In each feedback block, the hidden state with high-level information (output) will flow into the next iteration to correct the low-level representations (input). By stacking multiple feedback blocks, the proposed network where the hidden states are distributed can extract powerful high-level representations for rain streak layers. Curriculum learning is employed to connect the loss of each iteration and ensure that hidden states contain the notion of output. In addition, a self-ensemble strategy for rain removal task, which can retain the approximate vertical character of rain streaks, is explored to maximize the potential performance of the deraining model. Extensive experimental results demonstrated the superiority of the proposed method in comparison with other deraining methods.

Augmented Reality (AR) applications are one of the current mobile app trends. Its ability to display information as if to make the user merge with the information provided. The specialty of the application is in the marker. One marker that is in great demand is known as markerless, such as 3D Object Tracking. In making this 3D Object-based AR application, one must pass a stage where a real object is taken at points or known as points to be used as markers. The application used at this stage of making markers is known as the vuforia object scanner. From the experience of making AR applications, it was found conditions where markers with a large number of points were not able to display information properly, a small number of points were able to display information well and sometimes the two conditions were opposite or comparable, i.e. a lot of points makes a good marker and a few points make a bad marker. This inconsistent condition becomes a problem in this study. The test results show that the points do not affect the marker's ability to display information, but are influenced by the characteristics of the object used as the marker. Tests conducted on randomly selected objects resulted in 2 parameters to be tested in this study, namely the texture and pattern of the object. Markers with good quality to be used as markers for 3D objects are markers with a hard texture, with a detection angle of 90° and 135° and with a regular pattern

Chinese calligraphy has a strong artistry and appreciation. Through deep learning methods such as image translation methods, Chinese characters lacking in a set of calligraphy fonts can be quickly generated. Because of the clear structure of Chinese characters, the quality of the generated images is required to be high. However, it is difficult for a single generator to generate Chinese characters with clear texture, so we propose a multi-scale generative adversarial network (GAN), which contains two sub-GANs. The small-scale GAN generates low-resolution Chinese character outline, while the large-scale GAN supplements the details of the characters based on it. In addition to quantitatively evaluating the generated characters, we pre-train a vgg-19 network to extract stylistic characteristics of Chinese characters and compare the feature discrepancy between generated and real Chinese characters. From the experiment, using our method, the definition of the generated Chinese characters has indeed improved.

PurposeClothing patterns play a dominant role in costume design and have become an important link in the perception of costume art. Conventional clothing patterns design relies on experienced designers. Although the quality of clothing patterns is very high on conventional design, the input time and output amount ratio is relative low for conventional design. In order to break through the bottleneck of conventional clothing patterns design, this paper proposes a novel way based on generative adversarial network (GAN) model for automatic clothing patterns generation, which not only reduces the dependence of experienced designer, but also improve the input-output ratio.Design/methodology/approachIn view of the fact that clothing patterns have high requirements for global artistic perception and local texture details, this paper improves the conventional GAN model from two aspects: a multi-scales discriminators strategy is introduced to deal with the local texture details; and the self-attention mechanism is introduced to improve the global artistic perception. Therefore, the improved GAN called multi-scales self-attention improved generative adversarial network (MS-SA-GAN) model, which is used for high resolution clothing patterns generation.FindingsTo verify the feasibility and effectiveness of the proposed MS-SA-GAN model, a crawler is designed to acquire standard clothing patterns dataset from Baidu pictures, and a comparative experiment is conducted on our designed clothing patterns dataset. In experiments, we have adjusted different parameters of the proposed MS-SA-GAN model, and compared the global artistic perception and local texture details of the generated clothing patterns.Originality/valueExperimental results have shown that the clothing patterns generated by the proposed MS-SA-GAN model are superior to the conventional algorithms in some local texture detail indexes. In addition, a group of clothing design professionals is invited to evaluate the global artistic perception through a valence-arousal scale. The scale results have shown that the proposed MS-SA-GAN model achieves a better global art perception.

Today, large amount of data are stored in image format. Content based image retrieval from bulk databases has become an interesting research topic in last decade. Most of the recent approaches use joint of texture and color information. In most cases, the color and texture features are concatenated together and equal importance is given to each one. The human visual system, usually pays more attention to the textural properties of objects to recognize. In this paper a new approach is proposed for content based image retrieval based on weighted combination of color and texture features. Firstly, to achieve discriminant features, texture features are extracted using modified local binary patterns (MLBP) and local neighborhood differences patterns (LNDP) and filtered gray level co-occurrence matrix (GLCM). Also, quantization color histogram is used to extract color features. Next, the similarity matching is performed based on canbera distance in color and texture features separatly. Finally, a weighted decision is performed to retrieve most similar database images to the user query. The performance of the proposed approach is evaluated on Corel 1 K and Corel 10k datasets. Results show that proposed approach provide better performance than state-of-the-art methods in terms of precision and recall rate.

Character segmentation from epigraphical images helps the optical character recognizer (OCR) in training and recognition of old regional scripts. The scripts or characters present in the images are illegible and may have complex and noisy background texture. In this paper, we present an automated way of segmenting and extracting characters on digitized inscriptions. To achieve this, machine learning models are employed to discern between correctly segmented characters and partially segmented ones. The proposed method first recursively crops the document by sliding a window across the image from top to bottom to extract the content within the window. This results in a number of small images for classification. The segments are classified into character and non-character class based on the features within them. The model was tested on a wide range of input images having irregular, inconsistently spaced, hand written and inscribed characters.

Objective The aim of this study was to investigate whether the texture features of lung computed tomography images were altered by primary breast cancer without pulmonary metastasis. Methods Texture analysis was performed on the regions of interest of lung computed tomography images from 36 patients with breast cancer and 36 healthy controls. Texture parameters between subjects with different clinical stages and hormone receptor (HR) statuses in patients with breast cancer were analyzed. Results Three texture parameters (mean, SD, and variance) were significantly different between patients with breast cancer and healthy controls and between preoperative and postoperative stages in patients with breast cancer. All 3 parameters showed an increasing trend under the tumor-bearing state. These parameters were significantly higher in the stage III + IV group than in the stage I + II group. The variance parameter was significantly higher in the HR-negative group than in the HR-positive group. Conclusions Texture analysis may serve as a novel additional tool for discovering conventionally invisible changes in the lung tissue of patients with breast cancer.

Because pixel values of foggy images are irregularly higher than those of images captured in normal weather (clear images), it is difficult to extract and express their texture. No method has previously been developed to directly explore the relationship between foggy images and semantic segmentation images. We investigated this relationship and propose a generative adversarial network (GAN) for foggy image semantic segmentation (FISS GAN), which contains two parts: an edge GAN and a semantic segmentation GAN. The edge GAN is designed to generate edge information from foggy images to provide auxiliary information to the semantic segmentation GAN. The semantic segmentation GAN is designed to extract and express the texture of foggy images and generate semantic segmentation images. Experiments on foggy cityscapes datasets and foggy driving datasets indicated that FISS GAN achieved state-of-the-art performance.

The COVID-19 pandemic is still ongoing until 2021 and is likely to continue until an uncertain time. This arises because the spread of the SARS-CoV-2 virus also continued to occur in the community. Of the five points in 5M that has been initiated by the government, the focus of this study is the use of face masks. In this study, an image-based automatic mask detection method using a classification approach is proposed. This method can be used in automated systems to increase public discipline in wearing masks to suppress the spread of the SARS-CoV-2 virus. The classes used in the classification are "with mask" and "without mask". The adjacent evaluation local binary patterns (AELBP) method, which is an extension of the local binary patterns (LBP) method, is used to extract the texture features of each image. Tests were carried out on 2,172 facial images of various sizes, facial accessories, and facial expressions. The test results using the AELBP method show that the accuracy and F-measure are 98.39% and 98.08%, respectively. This result is better than other methods which are also evaluated. In addition, testing of the AELBP method execution time shows that this method is feasible to use on real systems., Pandemi COVID-19 masih berlangsung hingga tahun 2021 dan kemungkinan masih akan terus berlangsung hingga waktu yang belum bisa dipastikan. Ini terjadi karena penyebaran virus SARS-CoV-2 juga masih terus terjadi di tengah-tengah masyarakat. Gerakan 5M yang digagas oleh pemerintah masih banyak dilanggar oleh masyarakat di berbagai tempat. Dari lima poin pada 5M, yang menjadi fokus pada penelitian ini adalah penggunaan masker wajah. Pada penelitian ini, diusulkan metode deteksi masker secara otomatis berbasis citra menggunakan pendekatan klasifikasi. Metode ini dapat digunakan pada sistem otomatis untuk meningkatkan kedisiplinan masyarakat dalam mengenakan masker untuk menekan penyebaran virus SARS-CoV-2. Kelas yang digunakan pada klasifikasi adalah “dengan masker” dan “tanpa masker”. Ciri atau feature yang digunakan adalah tekstur dari setiap wajah. Metode adjacent evaluation local binary patterns (AELBP) yang merupakan pengembangan dari metode local binary patterns (LBP) digunakan untuk mengekstraksi ciri tekstur dari setiap citra. Pengujian dilakukan terhadap 2.172 citra wajah dengan berbagai ukuran, aksesori wajah, dan ekspresi wajah. Pengujian juga dilakukan menggunakan beberapa metode ekstraksi tekstur yang lain sebagai pembanding. Hasil pengujian menunjukkan akurasi yang didapatkan menggunakan metode AELBP sebesar 98,30% dan F-measure sebesar 98,08%. Hasil ini lebih baik daripada metode-metode lain yang juga diuji. Pengujian terhadap waktu eksekusi metode AELBP menunjukkan bahwa proses training dilakukan dalam 0,167 detik dan testing dilakukan dalam 0,016 detik. Hasil ini menunjukkan bahwa metode ini feasible untuk digunakan pada sistem yang nyata.

To separately perform visual and texture analyses of the axial, coronal, and sagittal planes of T2-weighted images and identify the optimal method for differentiating between the normal placenta and placenta accreta spectrum (PAS). Eighty consecutive patients (normal group, n = 50; PAS group, n = 30) underwent preoperative MRI. A scoring system (0–2) was used to evaluate the degree of abnormality observed in visual analysis (bulging, abnormal vascularity, T2 dark band, placental heterogeneity). The axial, coronal, and sagittal planes were manually segmented separately to obtain texture features, and seven combinations were obtained: axial; coronal; sagittal; axial and coronal; axial and sagittal; coronal and sagittal; and axial, coronal, and sagittal. Feature selection using the least absolute shrinkage and selection operator method and model construction using a support vector machine algorithm with k-fold cross-validation were performed. AUC was used to evaluate diagnostic performance. The AUC of visual analysis was 0.75. The model ‘coronal and sagittal’ had the highest AUC (0.98) amongst the seven combinations. The fivefold cross-validation for the model ‘coronal and sagittal’ showed AUCs of 0.85 and 0.97 in training and validation sets, respectively. The AUC of the model ‘coronal and sagittal’ for all subjects was significantly higher than that of visual analysis (0.98 vs. 0.75; p