Search

Due to the excellent results achieved by transformers in computer vision, more and more scholars have introduced transformers into the field of medical image segmentation. However, the use of transformers will make the model's parameters very large, which occupies a large amount of the computer's resources, making them very time-consuming during training. In order to alleviate this disadvantage, this paper explores a flexible and efficient search strategy that can find the best subnet from a continuous transformer network. The method is based on a learnable and uniform L1 sparsity constraint, which contains factors that reflect the global importance of the continuous search space in different dimensions, while the search process is simple and efficient, containing a single round of training. At the same time, in order to compensate for the loss of accuracy caused by the search, a pixel classification module is introduced into the model to compensate for the loss of accuracy in the model search process. Our experiments show that the model in this paper compresses 30% of the parameters and FLOPs used, while also showing a slight increase in the accuracy of the model on the Automatic Cardiac Diagnosis Challenge (ACDC) dataset. [ABSTRACT FROM AUTHOR]

The Multisensor Multiresolution Technique (MMT) is applied to unmixed thermal images from ASTER (90 m), using 30 m resolution images from Landsat 8-9 reflective channels. The technique allows for the retrieval of thermal radiance values of the features identified in the high-resolution reflective images and the generation of a high-resolution radiance image. Different alternatives of application of MMT are evaluated in order to determine the optimal methodology design: performance of the Iterative Self-Organizing Data Analysis Technique (ISODATA) and K-means classification algorithms, with different initiation numbers of clusters, and computation of contributions of each cluster using moving windows with different sizes and with and without weight coefficients. Results show the K-means classification algorithm with five clusters, without matrix weighting, and utilizing a 5 × 5 pixel window for synthetic high-resolution image reconstruction. This approach obtained a maximum R2 of 0.846 and an average R2 of 0.815 across all cases, calculated through the validation of the synthetic high-resolution TIR image generated against a real Landsat 8-9 TIR image from the same area, same date, and co-registered. These values imply a 0.89% improvement regarding the second-best methodology design (K-means with five starting clusters with 7 × 7 moving window) and a 410.25% improvement regarding the worst alternative (K-means with nine initial clusters, weighting, and 3 × 3 moving window). [ABSTRACT FROM AUTHOR]

Introduction: The application of artificial intelligence to facial aesthetics has been limited by the inability to discern facial zones of interest, as defined by complex facial musculature and underlying structures. Although semantic segmentation models (SSMs) could potentially overcome this limitation, existing facial SSMs distinguish only three to nine facial zones of interest. Methods: We developed a new supervised SSM, trained on 669 high‐resolution clinical‐grade facial images; a subset of these images was used in an iterative process between facial aesthetics experts and manual annotators that defined and labeled 33 facial zones of interest. Results: Because some zones overlap, some pixels are included in multiple zones, violating the one‐to‐one relationship between a given pixel and a specific class (zone) required for SSMs. The full facial zone model was therefore used to create three sub‐models, each with completely non‐overlapping zones, generating three outputs for each input image that can be treated as standalone models. For each facial zone, the output demonstrating the best Intersection Over Union (IOU) value was selected as the winning prediction. Conclusions: The new SSM demonstrates mean IOU values superior to manual annotation and landmark analyses, and it is more robust than landmark methods in handling variances in facial shape and structure. [ABSTRACT FROM AUTHOR]

Due to the excellent results achieved by transformers in computer vision, more and more scholars have introduced transformers into the field of medical image segmentation. However, the use of transformers will make the model’s parameters very large, which occupies a large amount of the computer’s resources, making them very time-consuming during training. In order to alleviate this disadvantage, this paper explores a flexible and efficient search strategy that can find the best subnet from a continuous transformer network. The method is based on a learnable and uniform L1 sparsity constraint, which contains factors that reflect the global importance of the continuous search space in different dimensions, while the search process is simple and efficient, containing a single round of training. At the same time, in order to compensate for the loss of accuracy caused by the search, a pixel classification module is introduced into the model to compensate for the loss of accuracy in the model search process. Our experiments show that the model in this paper compresses 30% of the parameters and FLOPs used, while also showing a slight increase in the accuracy of the model on the Automatic Cardiac Diagnosis Challenge (ACDC) dataset.

The Multisensor Multiresolution Technique (MMT) is applied to unmixed thermal images from ASTER (90 m), using 30 m resolution images from Landsat 8-9 reflective channels. The technique allows for the retrieval of thermal radiance values of the features identified in the high-resolution reflective images and the generation of a high-resolution radiance image. Different alternatives of application of MMT are evaluated in order to determine the optimal methodology design: performance of the Iterative Self-Organizing Data Analysis Technique (ISODATA) and K-means classification algorithms, with different initiation numbers of clusters, and computation of contributions of each cluster using moving windows with different sizes and with and without weight coefficients. Results show the K-means classification algorithm with five clusters, without matrix weighting, and utilizing a 5 × 5 pixel window for synthetic high-resolution image reconstruction. This approach obtained a maximum R2 of 0.846 and an average R2 of 0.815 across all cases, calculated through the validation of the synthetic high-resolution TIR image generated against a real Landsat 8-9 TIR image from the same area, same date, and co-registered. These values imply a 0.89% improvement regarding the second-best methodology design (K-means with five starting clusters with 7 × 7 moving window) and a 410.25% improvement regarding the worst alternative (K-means with nine initial clusters, weighting, and 3 × 3 moving window).

Convolutional neural networks (CNNs) and recurrent neural networks (RNNs) have gained improved results in remote sensing image data classification. Multispectral image classification can benefit from the rich spectral information extracted by these models for land cover classification. This paper proposes a classification model called a hierarchical convolutional recurrent neural network (HCRNN) to combine the CNN and RNN modules for pixel-level classification of multispectral remote sensing images. In the HCRNN model, the original 13-band information from Sentinel-2 is transformed into a 1D multispectral sequence using a fully connected layer. It is then reshaped into a 3D multispectral feature matrix. The 2D-CNN features are extracted and used as inputs to the corresponding hierarchical RNN. The feature information at each level is adapted to the same convolution size. This network structure fully leverages the advantages of CNNs and RNNs to extract temporal and spatial features from the spectral data, leading to high-precision pixel-level multispectral remote sensing image classification. The experimental results demonstrate that the overall accuracy of the HCRNN model on the Sentinel-2 dataset reaches 97.62%, which improves the performance by 1.78% compared to the RNN model. Furthermore, this study focused on the changes in forest cover in the study area of Laibin City, Guangxi Zhuang Autonomous Region, which was 7997.1016 km2, 8990.4149 km2, and 8103.0020 km2 in 2017, 2019, and 2021, respectively, with an overall trend of a small increase in the area covered. [ABSTRACT FROM AUTHOR]

Hyperspectral remote sensing technology is a breakthrough technology that integrates imaging and spectral methods, but it also faces issues such as classification accuracy being controlled by dimensionality and spectral variability. Based on this, this study proposes a Golden Sine Spotted Hyena Optimization (GSSHO) algorithm that integrates the Golden Sine method, chaos strategy, and tournament strategy to help hyperspectral remote sensing technology achieve higher work efficiency. The implementation of the algorithm can be roughly divided into three stages. This includes initializing using chaotic strategies, calculating the individual fitness value and selecting the best cluster, and after updating the fitness value, continuing to select the optimal solution until the stop condition of the algorithm is met. Among them, the golden sine algorithm provides a new data update method for the spotted hyena optimization algorithm, avoiding situations where individual data cannot be searched for. The Salinas and Pavian Centre datasets were used in the experiment to validate the effectiveness of the improved spotted hyena optimization algorithm. At the same time, four other common algorithms were selected for performance comparison experiments with the algorithm. The experimental results show that the GSSHO algorithm has excellent dimensionality reduction and convergence capabilities. Compared with the total number of original bands, in both datasets, the two values decreased to the total of 32 and 11, respectively, equivalent to 1/6 and 1/9 of the original data; The fitness function values are 0.0836 and 0.0315 respectively. And compared with the other four algorithms, this algorithm also significantly outperforms other algorithms in all aspects of indicators. Therefore, the spotted hyena optimization algorithm based on the golden sine algorithm and chaotic strategy can make band selection more efficient.

In this paper, we proposed a machine learning and computer vision-based automated apple disease detection and recognition system based on leaf symptoms. The proposed method is composed of three parts: diseased region segmentation, feature extraction, and classification. We have segmented the infected portion of the leaf using L*a*b* space-based color segmentation method. Here, average color markers in a*b* space and the nearest neighbor method have been used for classifying each pixel into either healthy, infected, or background regions. We have extracted two types of features: one is the proposed DWT feature and another is L*a*b* space-based color histogram features. Horizontal feature fusion is performed to create the final feature vector. The feature vectors have been classified using several classifiers keeping Random Forrest as the base classifier. In this paper, the experiment is made on Plant Village dataset, where image of Apple Scab, Black Rot, and Cedar Apple Rust disease are taken for both training and testing our model. The fusion of proposed DWT and color histogram features is a novel approach in detecting and recognizing apple leaf disease, which got an accuracy of 98.63%. [ABSTRACT FROM AUTHOR]

Pixel value ordering (PVO) prediction has become a research hotspot in reversible data hiding community and existing achievements have proved the necessity of predicting more pixels in a block. The increase in predicted pixels is beneficial to fully exploiting the redundancy of the cover image, but a single reference pixel may lead to decreased prediction quality. Given this situation, this paper proposes to develop PVO in correlated pixels construction and introduce morphology as an important theoretical tool. Several largest pixels in a block are first selected as candidate reference pixels which are to be activated in descending order. Except for the last one, each active reference pixel is used to predict all pixels to generate a correlation matrix, on which the dilation operation is carried out to pick out connected pixels that are likely to be strongly correlated with the reference pixel. Whether to activate the next reference pixel depends on the presence of disconnected pixels. Once the last reference pixel is activated, all remaining pixels will be taken as connected pixels. After the first round of embedding ends, several smallest pixels are similarly selected as candidate reference pixels to realize the second round of embedding. Experimental results demonstrate that satisfactory superiority in fidelity is achieved, e.g., the average PSNR for the Kodak image database is 63.85 dB with an embedding capacity of 10,000 bits. [ABSTRACT FROM AUTHOR]

Images captured by cameras are sometimes contaminated either during acquisition or transmission. Therefore, a preprocessing step is required which reduces noise from images. In this paper, a novel and efficient edge preserving universal noise removal algorithm is proposed which exploits both the local and global characteristics of the neighboring non-corrupted pixels. In the proposed algorithm, corrupted pixels are detected by robust outlying ratio (ROR) and replaced with the weighted sum (local characteristics) of the neighboring non-corrupted pixels in 3 × 3 window and these weights are obtained by solving the kernel ridge regression (KRR) which uses the global mean and covariance (global characteristics). Extensive experimental results demonstrate that our algorithm has better noise removal capability in terms of both objective and subjective evaluation as compared to existing denoising algorithms. [ABSTRACT FROM AUTHOR]

Pixelwise classification plays an important role for image interpretation. The remote sensing images, especially polarimetric synthetic aperture radar (PolSAR), have provided wide applications for both military and civilian users regardless of weather or lighting conditions. However, the classification of heterogeneous imagery is still challenging. In this article, we propose a novel deep fuzzy graph convolutional network (DFGCN) for pixelwise classification of PolSAR imagery. Inspired by the fuzzy logic, the imagery (i.e., a reflection of the backscattering of the ground) is represented by a fuzzy graph, whose node is associated with a feature vector, and fuzzy weights denoting the similarity degrees between nodes are defined by using both feature distance and spatial relation. We also construct a new graph residual module stacking multiple residual layers to extend the depth of our fuzzy graph network. Moreover, we present a graph shrinking strategy to reduce the graph size for training and predicting the labels of large-scale PolSAR imagery, which can cut down the computational cost. The experimental results on various real-world PolSAR datasets indicate that DFGCN dramatically enhances classification accuracies on both heterogeneous and homogeneous regions with limited labeled pixels compared with the state-of-the-art methods.

Semantic image segmentation treats the issues involved in the object recognition and image segmentation as a combined task. The chief notion of semantic segmentation is to partition the image into visually uniform regions and to discriminate the class of the partitioned regions. Pixel classification is done over the segmented regions by assigning semantic labels. In general, inference frameworks are fed with the combination of low‐level features and high‐level contextual cues to segment an image. Since these combinations are rarely object consistent, result with minimum classification accuracy because of choosing non‐influencing features and cues to track specific objects. To overcome this problem, a nature‐inspired meta‐heuristic optimization algorithm called Seed Picking Crossover Optimization (SPCO) is proposed to optimize i.e. train the CRF (Conditional Random Field) for choosing relevant feature to segment the object with high accuracy. To meritoriously recognize the objects, a semi‐segmentation process is initially performed using Simple Linear Iterative Clustering (SLIC) algorithm. For pixel transformation and pixel association, Dirichlet process mixture model and CRF are employed. Optimized CRFs are used where the parametric optimization is done using the proposed SPCO algorithm. The proposed work results with 84% on classification accuracy and the performance evaluations are done using MSRC‐21 dataset.

Image segmentation in land cover regions which are overlapping in satellite imagery, is one crucial challenge. To detect true belonging of one pixel becomes a challenging problem while classifying mixed pixels in overlapping regions. In current work, we propose one new approach for image segmentation using a hybrid algorithm of K-Means and Cellular Automata algorithms. This newly implemented unsupervised model can detect cluster groups using hybrid 2-Dimensional Cellular-Automata model based on K-Means segmentation approach. This approach detects different land use land cover areas in satellite imagery by existing K-Means algorithm. Since it is a discrete dynamical system, cellular automaton realizes uniform interconnecting cells containing states. In the second stage of current model, we experiment with a 2-dimensional cellular automata to rank allocations of pixels among different land-cover regions. The method is experimented on the watershed area of Ajoy river (India) and Salinas (California) data set with true class labels using two internal and four external validity indices. The segmented areas are then compared with existing FCM, DBSCAN and K-Means methods and verified with the ground truth. The statistical analysis results also show the superiority of the new method.

Here we report on initial efforts to evaluate enhanced darkfield microscopy (EDFM) and light scattering Vis-NIR hyperspectral imaging (HSI) as a rapid screening tool for the offline analysis of mixed cellulose ester (MCE) filter media used to collect airborne nanoparticulate from work environments. For this study, the materials of interest were nanoscale titanium dioxide (TiO2) and silicon dioxide (SiO2; silica), chosen for their frequent use in consumer products. TiO2 and SiO2 nanoscale particles (NPs) were collected on MCE filter media and were imaged and analyzed via EDFM-HSI. When visualized by EDFM, TiO2 and SiO2 NPs were readily apparent as bright spherical structures against a dark background. Moreover, TiO2 and SiO2 NPs were identified in hyperspectral images. EDFM-HSI images and data were compared to scanning transmission electron microscopy (STEM), a NIST-traceable technique for particle size analysis, and the current gold standard for offline analysis of filter media. As expected, STEM provided more accurate sizing and morphology data when compared to EDFM-HSI, but is not ideal for rapid screening of the presence of NPs of interest since it is a costly, low-throughput technique. In this study, we demonstrate the utility of EDFM-HSI in rapidly visualizing and identifying TiO2 and SiO2 NPs on MCE filters. This screening method may prove useful in expediting time-to-knowledge compared to electron microscopy. Future work will expand this evaluation to other industrially relevant NPs, other filter media types, and real-world filter samples from occupational exposure assessments. [ABSTRACT FROM AUTHOR]

Landslide identification is important for understanding their conditioning factors, and for constructing susceptibility, risk, and vulnerability maps. In remote sensing this can be accomplished manually or through classifiers. This study compares three image classifiers (Maximum Likelihood, Random Forest, and Support Vector Machines (SVM)) used in identifying landslides in Itaóca (São Paulo, Brazil). Two datasets were used: a RapidEye-5 (5 m) image and a Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) (12.5 m). Seven pixel-based classifications were produced, two for each classifier and a binary class that identified only landslides and non-landslides. One classification contained five spectral bands (5B), while the other contained six bands (6B) and included the slope derived from the DEM. The results were validated using Kappa index and F1 score. The SVM 6B classification achieved the best results among the validation indices used herein. It identified a landslide area of 399,325 m². The results contribute to landslide mapping in tropical environments using pixel-based classifiers. However, although the SVM classification was successful, only landslides with larger areas were captured by the algorithms, confirming the importance of conducting further analyses using images with finer spatial resolution. [ABSTRACT FROM AUTHOR]

Abstract Landslide identification is important for understanding their conditioning factors, and for constructing susceptibility, risk, and vulnerability maps. In remote sensing this can be accomplished manually or through classifiers. This study compares three image classifiers (Maximum Likelihood, Random Forest, and Support Vector Machines (SVM)) used in identifying landslides in Itaóca (São Paulo, Brazil). Two datasets were used: a RapidEye-5 (5 m) image and a Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) (12.5 m). Seven pixel-based classifications were produced, two for each classifier and a binary class that identified only landslides and non-landslides. One classification contained five spectral bands (5B), while the other contained six bands (6B) and included the slope derived from the DEM. The results were validated using Kappa index and F1 score. The SVM 6B classification achieved the best results among the validation indices used herein. It identified a landslide area of 399,325 m². The results contribute to landslide mapping in tropical environments using pixel-based classifiers. However, although the SVM classification was successful, only landslides with larger areas were captured by the algorithms, confirming the importance of conducting further analyses using images with finer spatial resolution.

The recent development in deep learning allows us to develop a computer visionbased system for recognition, detection, and localization of nutrients deficiency in fruits. Due to the time constraints, it is important to use an optimized and fast system for fruit quality inspection. In this paper, the input is taken as an image. A deep learning-based method extracts low level and high-level features such as edges, geometrical, statistical, texture, intensity, etc. After validation of the system with the test data, the output is predicted by the system. The processing time is optimized by avoiding fully connected layers which further minimize the requirement of neurons in the network. The convolutional neural network extracts the features of the fruits, Rectified Linear Unit (ReLu) removes the non-fruit pixels. Pooling shrinks, the image by selecting the maximum value of the pixel. The process is repeated until the size of the image is at the desired level. The aim is to identify the objects and recognize them. The foreground region objects are of our interest and being segmented for higher-level image processing. The proposed system attains the accuracy of 99.30 % with the processing time of 3.207 sec. [ABSTRACT FROM AUTHOR]

In today's PC illustration, numerous object locations of videos are quite critical duties to accomplish. Swiftly and reliably recognising and distinguishing the multiple aspects of a video is a crucial attribute for collaborating with one's condition (object). The core issue is that in theory, to ensure that no significant aspect is missing; all aspects of a content in a video must be scanned for elements on various different scales. It requires some investment and effort anyway, to really arrange the substance of a given content region and both time and computational limits that an operator can spend on classification are constrained. Two presumption procedures for accelerating the standard identifier are performed by the proposed method and demonstrate their capability by performing both identification efficiency and velocity. The main enhancement of our group-based classifier focuses on accelerating the grouping of sub features by planning the problem as a selection procedure for consecutive features. The subsequent improvement gives better multiscale features to distinguish objects of all sizes without rescaling the information image from a video. Extracting contents from video is an assortment of successive images with a steady time interim. So video can give more data about contents in it when situations are changing regarding time. Along these lines, physically taking care of contents with features are very unimaginable. In the proposed work, it is suggested that a Group-based Video Content Extraction Classifier (GbCCE) extracts content from a video by extracting relevant features using a group-based classifier. The proposed method is distinct from conventional approaches and the findings indicate that better output is demonstrated by the proposed method. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging makes it possible to obtain object information with fine spectral resolution as well as spatial resolution, which is beneficial to a wide array of applications. However, there is a high correlation among the bands in a hyperspectral image (HSI). Band selection (BS), selecting only some representative bands to describe well the original image, is an appropriate approach to tackle this problem. In this study, the authors propose an efficient greedy‐based unsupervised BS method, namely the maximum simplex volume by orthogonal‐projection BS method. The main contributions are two‐fold: (i) an information‐lossless compressed descriptor in the Euclidean sense that can reduce the amount of redundant information in the band analysis and (ii) an orthogonal‐projection‐based algorithm to find the band points forming the simplex of maximum volume. The experimental results on four real HSIs demonstrate that the proposed method can achieve satisfying pixel classification performances and is computationally fast.

A novel PolSAR image speckle reduction algorithm based on a new definition of similarity coefficient is proposed in this paper. Pixels in image are firstly classified into three types by threshold segmentation which is calculated with the similarity features. Then, weighted filtering is applied on the pixels selected according to their types, power features and similarity properties. Experimental results with measured data collected by NASA/JPL AIRSAR system show that the proposed method is more effective than Lee Filter not only in speckle suppression but also in polarimetric properties and structure feature preservation.

One important task of analyzing remote sensing satellite imagery is to categorize the pixels according to various landcover regions. This unsupervised segmentation task can be posed as a problem of pixel intensity clustering, considering the different spectral bands as different features. Due to the presence of noise and overlapping clusters present in remote sensing images, fuzzy clustering is popularly applied for the segmentation task. However, fuzzy C-means like fuzzy clustering algorithms suffer from random initialization that often causes them to get stuck into some local optimum results. In this article, a game theory-motivated approach based on Shapley value to initialize the cluster centers for FCM clustering has been adopted which provides more stability and improved performance. The proposed method explores and exploits the advantages of both game theory and fuzzy technique for pixel classification. The superiority of the approach has been demonstrated over a numeric satellite image data set as well as different real-life remote sensing satellite images both visually and numerically with statistical support. The results have been compared with several popular centroid-based clustering techniques, viz. K-means, K-means++, fuzzy C-means (FCM) and probabilistic FCM, using J m and I indexes as well as visual cluster plots. [ABSTRACT FROM AUTHOR]

This study analyses the use of a pixel classification model to segment amniotic fluid areas on ultrasound (US) images characterized by noise, blurry edge, artifacts, and low contrast. In contrast with the previous methods, this study constrains a training set of pixels based on neighbourhood information with the rectangle window sampling method used to determine the characteristics of each pixel in its environment specifically. The feature extraction is no longer based on the global characteristics of the object rather by taking the value of each pixel in the object area using the sampling window. This research also combines the local first-order statistical methods and gray level information in the window to obtain the characteristics of each pixel. Furthermore, Random Forest and Decision Tree (C.45) were used to classify each pixel into four classes, namely amniotic fluid, placenta, uterus, and fetal body. The classification performance testing of pixel sampling data showed that the Random forest with 5 × 7 window sizes achieved the highest performance at 99.5% accuracy, precision, and recall, respectively. Furthermore, the proposed model was evaluated using 50 new test US images to segment the amniotic fluid area. According to experimental result, proposed models can produce better segmentation area with an increase in the IoU value by 18.3% or a Jaccard coefficient value rate of 0.183 in the range of 0-1 with the previous state of the art method. Furthermore, the proposed model reduces the error rate and improves accuracy by 6.61% and 84.77%, respectively. [ABSTRACT FROM AUTHOR]

Abstract Time series remote sensing image is an important resource for dynamic monitoring of resources and environment, and its abundant time spectrum information can be used to characterize the dynamic change of vegetation coverage. This paper proposes a comprehensive clustering and pixel classification method for extracting the vegetation dynamics based on time series Landsat normalized difference vegetation index (NDVI). This method uses the time-division algorithm for fitting time-series NDVI firstly. And the Markov random field optimized (MRF) semi-supervised dynamic time warping (DTW) kernel fuzzy c-means clustering was constructed. Then the MRF-optimized semi-supervised DTW-kernel fuzzy c-means clustering was combined with the 1-nearest neighbor (1NN) DTW pixel classification to realize the extraction of vegetation dynamics. Shengli Opencast Coal Mine in The Xilin Gol Grassland was taken as the study area to analyze the applicability of the different classification methods. The results showed the fusion algorithm of the MRF-Semi-GDTW-FCM and 1NN-DTW generates accurate classification results with the overall accuracy of 93.8806% and Kappa coefficient of 0.9267, which were 1.7219, 0.0182, and 20.4080% and 0.2916 higher than the clustering and pixel classification, respectively. Experiments proof that the method proposed in this paper is not only simple but also accurate and effective.

Summary: Image segmentation plays a crucial role in many image processing and understanding applications. Despite the huge number of proposed image segmentation techniques, accurate segmentation remains a significant challenge in image analysis. This article investigates the viability of using echo state network (ESN), a biologically inspired recurrent neural network, as features extractor for efficient color image segmentation. First, an ensemble of initial pixel features is extracted from the original images and injected into the ESN reservoir. Second, the internal activations of the reservoir neurons are used as new pixel features. Third, the new features are classified using a feed forward neural network as a readout layer for the ESN. The quality of the pixel features produced by the ESN is evaluated through extensive series of experiments conducted on real world image datasets. The optimal operating range of different ESN setup parameters for producing competitive quality features is identified. The performance of the proposed ESN‐based framework is also evaluated on a domain‐specific application, namely, blood vessel segmentation in retinal images where experiments are conducted on the widely used digital retinal images for vessel extraction (DRIVE) dataset. The obtained results demonstrate that the proposed method outperforms state‐of‐the‐art general segmentation techniques in terms of performance with an F‐score of 0.92 ± 0.003 on the segmentation evaluation dataset. In addition, the proposed method achieves a comparable segmentation accuracy (0.9470) comparing with reported techniques of segmentation of blood vessels in images of retina and outperform them in terms of processing time. The average time required by our technique to segment one retinal image from DRIVE dataset is 8 seconds. Furthermore, empirically derived guidelines are proposed for adequately setting the ESN parameters for effective color image segmentation. [ABSTRACT FROM AUTHOR]

Semantic image segmentation treats the issues involved in the object recognition and image segmentation as a combined task. The chief notion of semantic segmentation is to partition the image into visually uniform regions and to discriminate the class of the partitioned regions. Pixel classification is done over the segmented regions by assigning semantic labels. In general, inference frameworks are fed with the combination of low‐level features and high‐level contextual cues to segment an image. Since these combinations are rarely object consistent, result with minimum classification accuracy because of choosing non‐influencing features and cues to track specific objects. To overcome this problem, a nature‐inspired meta‐heuristic optimization algorithm called Seed Picking Crossover Optimization (SPCO) is proposed to optimize i.e. train the CRF (Conditional Random Field) for choosing relevant feature to segment the object with high accuracy. To meritoriously recognize the objects, a semi‐segmentation process is initially performed using Simple Linear Iterative Clustering (SLIC) algorithm. For pixel transformation and pixel association, Dirichlet process mixture model and CRF are employed. Optimized CRFs are used where the parametric optimization is done using the proposed SPCO algorithm. The proposed work results with 84% on classification accuracy and the performance evaluations are done using MSRC‐21 dataset. [ABSTRACT FROM AUTHOR]

The present study proposes a workflow to extract from orthomosaics the enormous amount of dry stones used by past societies to construct funeral complexes in the Mongolian steppes. Several different machine learning algorithms for binary pixel classification (i.e. stone vs non-stone) were evaluated. Input features were extracted from high-resolution orthomosaics and digital elevation models (both derived from aerial imaging). Comparative analysis used two colour spaces (RGB and HSV), texture features (contrast, homogeneity and entropy raster maps), and the topographic position index, combined with nine supervised learning algorithms (nearest centroid, naive Bayes, k -nearest neighbours, logistic regression, linear and quadratic discriminant analyses, support vector machine, random forest, and artificial neural network). When features are processed together, excellent output maps, very close to or outperforming current standards in archaeology, are observed for almost all classifiers. The size of the training set can be drastically reduced (to ca. 300 samples) by majority voting, while maintaining performance at the highest level (about 99.5% for all performance scores). Note, however, that if the training set is inadequate or not fully representative, the classification results are poor. That said, the methods applied and tested here are extremely rapid. Extensive mapping, which would have been difficult with traditional, manual, or semi-automatic delineation of stones using a vector graphics editor, now becomes possible. This workflow generally surpasses pedestrian surveys using differential GPS or a total station. [ABSTRACT FROM AUTHOR]

In order to solve the problem of difficult pre-processing of crop video image shadows, a probable learning pixel classification method is proposed to study its processing technology. The algorithm effectively detects the shadow area by performing intelligent video collaborative detection on the shaded parts of the crop video sequence. Firstly, the cloud collaborative detection algorithm that can be widely used in agriculture was proposed. The video key frame was obtained and the background modeling algorithm with strong adaptability to crop illumination was applied to realize real-time detection of the target, so as to construct the crop pixel model. Finally, the proposed algorithm and the constructed model are applied to the processing of shadows of agricultural plant video images for experimental verification. The results show that in video frames 47, 194 and 258, the probable learning pixel classification method can be used to determine the shaded part of each frame, which can greatly improve the detection accuracy of crop shadows. The research in this paper shows that the probability learning pixel classification method can better enhance the shadow robustness and accuracy of crop video images. [ABSTRACT FROM AUTHOR]

The smoke from biomass burning on Kalimantan Island has caused severe environmental problems in Southeast Asia’s primary burning regions and surrounding regions due to the overspread haze. To monitor the biomass burning aerosol plumes on Kalimantan Island, the high-temporal-resolution Himawari-8 satellite data were used in this study. However, studies are limited on smoke detection using satellite remote sensing for Kalimantan Island because of the difficulty caused by frequently occurring clouds and the lack of prior knowledge on applying traditional threshold methods. In this study, we used the multilayer perceptron (MLP) method to identify smoke over Kalimantan Island in August 2015, one of the most severe fire seasons. To prepare sufficient supervision information, a pixel-level labeled dataset was established based on the Himawari-8 data. Based on the labeled dataset, three MLP approaches and two sampling methods were applied to create training samples. A comparison between the detection results for the MLP approaches and classification tree analysis (i.e., CTA) showed that MLP is superior to CTA. The visualization results also showed that the detected smoke areas included those mixed with clouds. Some detected smoke is difficult to identify by the human eye, suggesting that the explanatory dataset built for this study is sufficiently comprehensive. Therefore, the pixel-level labeled dataset and MLP are suitable for regions that are frequently cloud-covered.

As coral reefs continue to degrade globally due to climate change, considerable effort and investment is being put into coral restoration. The production of coral offspring via asexual and sexual reproduction are some of the proposed tools for restoring coral populations and will need to be delivered at scale. Simple, inexpensive, and high-throughput methods are therefore needed for rapid analysis of thousands of coral offspring. Here we develop a machine learning pipeline to rapidly and accurately measure three key indicators of coral juvenile fitness: survival, size, and color. Using machine learning, we classify pixels through an open-source, user-friendly interface to quickly identify and measure coral juveniles on two substrates (field deployed terracotta tiles and experimental, laboratory PVC plastic slides). The method’s ease of use and ability to be trained quickly and accurately using small training sets make it suitable for application with images of species of sexually produced corals without existing datasets. Our results show higher accuracy of survival for slides (94.6% accuracy with five training images) compared to field tiles measured over multiple months (March: 77.5%, June: 91.3%, October: 97.9% accuracy with 100 training images). When using fewer training images, accuracy of area measurements was also higher on slides (7.7% average size difference) compared to tiles (24.2% average size difference for October images). The pipeline was 36× faster than manual measurements. The slide images required fewer training images compared to tiles and we provided cut-off guidelines for training for both substrates. These results highlight the importance and power of incorporating high-throughput methods, substrate choice, image quality, and number of training images for measurement accuracy. This study demonstrates the utility of machine learning tools for scalable ecological studies and conservation practices to facilitate rapid management decisions for reef protection.

GF-3, the first C-band full-polarimetric Synthetic Aperture Radar (SAR) satellite with a space resolution up to 1 m, has multiple strip and scan imaging modes. In this paper, we propose a maritime ship detection algorithm that detects ship targets via pixel classification in a Bayesian framework and employ effective enhancement methods to improve detection performance based on the data characteristics. We compare and analyze the results of detection experiments using the proposed algorithm with those of several Constant False Alarm Rate (CFAR) algorithms. The experimental results verify the effectiveness of the proposed algorithm.

Plant phenology studies rely on long-term monitoring of life cycles of plants. High-resolution unmanned aerial vehicles (UAVs) and near-surface technologies have been used for plant monitoring, demanding the creation of methods capable of locating, and identifying plant species through time and space. However, this is a challenging task given the high volume of data, the constant data missing from temporal dataset, the heterogeneity of temporal profiles, the variety of plant visual patterns, and the unclear definition of individuals’ boundaries in plant communities. In this letter, we propose a novel method, suitable for phenological monitoring, based on convolutional networks (ConvNets) to perform spatio-temporal vegetation pixel classification on high-resolution images. We conducted a systematic evaluation using high-resolution vegetation image datasets associated with the Brazilian Cerrado biome. Experimental results show that the proposed approach is effective, overcoming other spatio-temporal pixel-classification strategies. [ABSTRACT FROM AUTHOR]

Hand segmentation is an integral part of many computer vision applications, especially gesture recognition. Training a classifier to classify pixels into hand or background using skin color as a feature is one of the most popular methods for this purpose. This approach has been highly restricted to simple hand segmentation scenarios since color feature alone provides very limited information for classification. Meanwhile there have been a rise of segmentation methods utilizing deep learning networks to exploit multi-layers of complex features learned from image data. Yet a deep neural network requires a large database for training and a powerful computational machine for operations due to its complexity in computations. In this work, the development of comprehensive features and optimized uses of these features with a randomized decision forest (RDF) classifier for the task of hand segmentation in uncontrolled indoor environments is investigated. Newly designed image features and new implementations are provided with evaluations of their hand segmentation performances. In total, seven image features which extract pixel or neighborhood related properties from color images are proposed and evaluated individually as well as in combination. The behaviours of feature and RDF parameters are also evaluated and optimum parameters for the scenario under consideration are identified. Additionally, a new dataset containing hand images in uncontrolled indoor scenarios was created during this work. It was observed from the research that a combination of features extracting color, texture, neighborhood histogram and neighborhood probability information outperforms existing methods for hand segmentation in restricted as well as unrestricted indoor environments using just a small training dataset. Computations required for the proposed features and the RDF classifier are light, hence the segmentation algorithm is suited for embedded devices equipped with limited power, memory, and computational capacities. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Fuzhou University is the property of Journal of Fuzhou University, Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this paper, a novel possibilistic approach for representing pixelic spatial knowledge is proposed to be used in classification; more specifically in segmentation of low quality images. This approach uses the spatial contextual information at the pixel level in order to produce a local possibility distribution. The similarity between this local possibility distribution representing the contextual pixelic information and the possibility distribution for each of the predetermined thematic classes is measured. This measure is used to assign one of these thematic classes to the pixel. In order to show the potential of the proposed possibilistic approach, synthetic and real images (Melanoma) are classified using the possibilistic similarity. The performance is compared with four relevant classic methods and one recent theory-like method (fuzzy c means). Our context-based possibilistic representation approach outperforms the other methods, in terms of classification recognition rate as well as in stability or robustness behavior when compared to those methods. [ABSTRACT FROM AUTHOR]

Attribute profiles (APs) are among the most prominent spectral-spatial pixel description tools, widely employed with optical, and hyperspectral images in particular. Their use, however, with the SAR data has been very limited. In this letter, we investigate the use of the recently developed feature attribute profiles (FPs) for the pixel-based classification of SAR images. Compared to low-level features such as intensity and amplitude, FPs are shown to provide high-level features incorporating successfully the spatial information of the pixels. Based on the classification results obtained on real TerraSAR-X images, it is shown that the FPs are capable of more accurately classifying the pixels compared to conventional and widely used SAR feature extraction methods such as gray-level co-occurrence matrices, histograms of oriented gradients, local binary patterns, and Gabor filters. [ABSTRACT FROM AUTHOR]

Ａｉｍｉｎｇ ａｔ ｉｓｓｕｅ ｏｆ ｐａｒｔｓ ｒｅｃｏｇｎｉｔｉｏｎ ａｎｄ ａｓｓｅｍｂｌｙ ｍｏｎｉｔｏｒｉｎｇ ｉｎ ａｓｓｅｍｂｌｙ ｍａｉｎｔｅｎａｎｃｅ ａｎｄ ｇｕｉｄａｎｃｅ ｏｆ ｍｅｃｈａｎｉｃａｌ ｐｒｏｄｕｃｔｓ, ａ ｐａｒｔ ｒｅｃｏｇｎｉｔｉｏｎ ａｎｄ ａｓｓｅｍｂｌｙ ｍｏｎｉｔｏｒｉｎｇ ｍｅｔｈｏｄ ｂａｓｅｄ ｏｎ Ｐｉｘｅｌ Ｌｏｃａｌ Ｂｉｎａｒｙ Ｐａｔｔｅｒｎ (ＰＸ-ＬＢＰ) ｗａｓ ｐｒｏｐｏｓｅｄ. Ｆｉｒｓｔｌｙ, ｔｈｅ ｃｌａｓｓｉｃａｌ ＬＢＰ ｏｐｅｒａｔｏｒ ｗａｓ ｍｅｒｇｅｄ ｗｉｔｈ ｔｈｅ ｐｉｘｅｌ ｃｌａｓｓｉｆｉｃａｔｉｏｎ ｔｏ ｐｒｏｐｏｓｅ ａｎ ｉｍｐｒｏｖｅｄ ＬＢＰ ｏｐｅｒａｔｏｒ, ｗｈｉｃｈ ｉｓ ｎａｍｅｄ ＰＸ-ＬＢＰ ｏｐｅｒａｔｏｒ. Ｓｅｃｏｎｄｌｙ, ＰＸ-ＬＢＰ ｆｅａｔｕｒｅｓ ｏｆ ｔｈｅ ｄｅｐｔｈ ｉｍａｇｅｓ ｗｅｒｅ ｅｘｔｒａｃｔｅｄ. Ａｎｄ ｔｒａｉｎｉｎｇ ｓｅｔ ａｎｄ ｔｅｓｔ ｓｅｔ ｗｅｒｅ ｏｂｔａｉｎｅｄ. Ｆｉｎａｌｌｙ, ｔｈｅ ｒａｎｄｏｍｉｚｅｄ ｄｅｃｉｓｉｏｎ ｆｏｒｅｓｔｓ ｃｌａｓｓｉｆｉｅｒ ｗａｓ ｔｒａｉｎｅｄ. Ｔｈｅｎ ｔｈｅ ｐｉｘｅｌ ｃｌａｓｓｉｆｉｃａｔｉｏｎ ｏｆ ｔｈｅ ｄｅｐｔｈ ｉｍａｇｅ ｏｆ ｔｈｅ ｔｅｓｔ ｓｅｔ ｗａｓ ｅｘｅｃｕｔｅｄ ｔｏ ｇｅｔ ｔｈｅ ｐｉｘｅｌ ｐｒｅｄｉｃｔｉｏｎ ｉｍａｇｅ ｏｆ ｄｅｐｔｈ ｉｍａｇｅｓ. Ｔｈｅ ｒｅｃｏｇｎｉｔｉｏｎ ｏｆ ａｓｓｅｍｂｌｙ ｐａｒｔｓ ｗａｓ ｒｅａｌｉｚｅｄ ｂｙ ｃｏｍｐａｒｉｎｇ ｔｈｅ ＲＧＢ ｖａｌｕｅｓ ｏｆ ｐｉｘｅｌ ｐｒｅｄｉｃｔｉｏｎ ｉｍａｇｅ ａｎｄ ｔｈｅ ｃｏｒｒｅｓｐｏｎｄｉｎｇ ｃｏｌｏｒ ｌａｂｅｌ ｉｍ- ａｇｅ. Ｔｈｅ ａｓｓｅｍｂｌｙ ｐｒｏｃｅｓｓ ｍｏｎｉｔｏｒｉｎｇ ｗａｓ ｒｅａｌｉｚｅｄ ｂｙ ａｎａｌｙｚｉｎｇ ｔｈｅ ｎｕｍｂｅｒ ａｎｄ ｐｏｓｉｔｉｏｎ ｏｆ ｐｉｘｅｌｓ ｉｎ ｐｉｘｅｌ ｐｒｅｄｉｃｔｉｏｎ ｉｍａｇｅ ｏｆ ｅａｃｈ ａｓｓｅｍｂｌｙ ｐａｒｔ ｔｏ ｄｅｔｅｒｍｉｎｅ ｔｈｅ ａｓｓｅｍｂｌｙ ｅｒｒｏｒ. Ｅｘｐｅｒｉｍｅｎｔ ｒｅｓｕｌｔｓ ｓｈｏｗ ｔｈａｔ ｔｈｅ ｍｅｔｈｏｄ ｐｒｏｐｏｓｅｄ ｉｎ ｔｈｉｓ ｐａｐｅｒ ｈａｓ ｈｉｇｈ ａｃｃｕｒａｃｙꎬｒｅａｌ-ｔｉｍｅ ａｎｄ ｒｏｂｕｓｔ ａｂｉｌｉｔｙꎬａｎｄ ｃａｎ ｂｅ ｕｓｅｄ ｉｎ ｆｉｅｌｄｓ ｏｆ ａｓｓｅｍｂｌｙ ｍａｉｎｔｅｎａｎｃｅ ｇｕｉｄｉｎｇꎬａｓｓｅｍｂｌｙ ｍｏｎｉｔｏｒｉｎｇ ａｎｄ ａｕｔｏｍａｔｉｃ ａｓｓｅｍｂｌｙ. [ABSTRACT FROM AUTHOR]

The paper aims to evaluate the potential of using aerial images captured from FAE Fixed Wing UAV in order to determine the density of plants and areas affected by pests. Based on the unsupervised classification of the orthophoto map obtained with the help of UAV, the re-sowing plan of the affected areas with a spring culture was created. For this purpose, we use professional equipment to gather images, GPS RTK to provide advanced precision, as well as software such as, Agisoft to compose the orthophoto, ArcMap for georeferencing the orthophoto map and LeoWorks to process and classify the resulting images. The utilized method has allowedfor a way to obtain an estimate of the degree of destruction of the rapeseed culture and to facilitate making the optimal decision with the help of a sowing plan. The usage of this method offset the losses due to pest attacks. [ABSTRACT FROM AUTHOR]

Locating and identifying the components mounted on a printed circuit board (PCB) based on machine vision is an important and challenging problem for automated PCB inspection and automated PCB recycling. In this paper, we propose a PCB semantic segmentation method based on depth images that segments and recognizes components in the PCB through pixel classification. The image training set for the PCB was automatically synthesized with graphic rendering. Based on a series of concentric circles centered at the given depth pixel, we extracted the depth difference features from the depth images in the training set to train a random forest pixel classifier. By using the constructed random forest pixel classifier, we performed semantic segmentation for the PCB to segment and recognize components in the PCB through pixel classification. Experiments on both synthetic and real test sets were conducted to verify the effectiveness of the proposed method. The experimental results demonstrate that our method can segment and recognize most of the components from a real depth image of the PCB. Our method is immune to illumination changes and can be implemented in parallel on a GPU.

Abstract Chaetoceros is a dominant genus of marine planktonic diatoms with worldwide distribution. Due to the difficulty of extracting setae from Chaetoceros images, automatic segmentation of Chaetoceros is still a challenging task. In this paper, we address this difficult task by regarding the whole segmentation process as unsupervised pixel-wise classification without human participation. First, we automatically produce positive (object) and negative (background) samples for follow-up training, by combining the advantages of two image processing algorithms: Grayscale Surface Direction Angle Model (GSDAM) for extracting setae information and Canny for detecting cell edges from low-contrast and strong-noisy microscopic images. Second, we develop pixel-wise training by using the produced samples in the training process of Deep Convolutional Neural Network (DCNN). At last, the trained DCNN is used to label other pixels into object and background for final segmentation. We compare our method with eight mainstream segmentation approaches: Otsu's thresholding, Canny, Watershed, Mean Shift, gPb-owt-ucm, Normalized Cut, Efficient Graph-based method and GSDAM. To objectively evaluate segmentation results, we apply six well-known evaluation indexes. Experimental results on a new Chaetoceros image dataset with human labelled ground truth show that our method outperforms the eight mainstream segmentation methods in terms of both quantitative and qualitative evaluation. [ABSTRACT FROM AUTHOR]