Search

Abstract The image object recognition and detection technology are widely used in many scenarios. In recent years, big data has become increasingly abundant, and big data-driven artificial intelligence models have attracted more and more attention. Evolutionary computation has also provided a powerful driving force for the optimization and improvement of deep learning models. In this paper, we propose an image object detection method based on self-supervised and data-driven learning. Differ from other methods, our approach stands out due to its innovative use of multispectral data fusion and evolutionary computation for model optimization. Specifically, our method uniquely combines visible light images and infrared images to detect and identify image targets. Firstly, we utilize a self-supervised learning method and the AutoEncoder model to perform high-dimensional feature extraction on the two types of images. Secondly, we fuse the extracted features from the visible light and infrared images to detect and identify objects. Thirdly, we introduce a model parameter optimization method using evolutionary learning algorithms to enhance model performance. Validation on public datasets shows that our method achieves comparable or superior performance to existing methods.

Harini Sriraman, Saleena Badarudeen, Saransh Vats, Prakash Balasubramanian School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, 600127, IndiaCorrespondence: Prakash Balasubramanian, School of Computer Science Engineering, Vellore Institute of Technology, Chennai, Tamil Nadu, India, Tel +91-044-39931228, Fax +91-044-39932555, Email Prakash.bala@vit.ac.inAbstract: Deep Learning (DL) drives academics to create models for cancer diagnosis using medical image processing because of its innate ability to recognize difficult-to-detect patterns in complex, noisy, and massive data. The use of deep learning algorithms for real-time cancer diagnosis is explored in depth in this work. Real-time medical diagnosis determines the illness or condition that accounts for a patient’s symptoms and outward physical manifestations within a predetermined time frame. With a waiting period of anywhere between 5 days and 30 days, there are currently several ways, including screening tests, biopsies, and other prospective methods, that can assist in discovering a problem, particularly cancer. This article conducts a thorough literature review to understand how DL affects the length of this waiting period. In addition, the accuracy and turnaround time of different imaging modalities is evaluated with DL-based cancer diagnosis. Convolutional neural networks are critical for real-time cancer diagnosis, with models achieving up to 99.3% accuracy. The effectiveness and cost of the infrastructure required for real-time image-based medical diagnostics are evaluated. According to the report, generalization problems, data variability, and explainable DL are some of the most significant barriers to using DL in clinical trials. Making DL applicable for cancer diagnosis will be made possible by explainable DL.Keywords: artificial intelligence, AI, machine learning, DL, CNN, healthcare, real-time diagnosis, classification, image processing, elastography, feedforward neural network

Abstract This study aims to evaluate the applicability of the high-resolution WaveFront Phase Imaging Sensor (WFPI) in eyes with Fuchs’ Endothelial Corneal Dystrophy (FECD) through qualitative and quantitative analysis using a custom-designed Automatic Guttae Detection Method (AGDM). The ocular phase was measured using the t $$\cdot$$ · eyede aberrometer and then was processed to obtain its High-Pass Filter Map (HPFM). The subjects were pathological and healthy patients from the Fundación Jiménez-Díaz Hospital (Madrid, Spain). The AGDM was developed and applied in pupils with 3 and 5 mm of diameter. A set of metrics were extracted and evaluated like the Root-Mean-Square error (RMS), Number of guttae, Guttae Area, and Area of Delaunay Triangulation (DT). Finally, a Support Vector Machine (SVM) model was trained to classify between pathological and healthy eyes. Quantitatively, the HPFM reveals a dark spots pattern according to the ophthalmologist’s description of the slit-lamp examination of guttae distribution. There were significant statistical differences in all the metrics when FECD and Healthy groups were compared using the same pupil size; but comparing both pupil sizes for the same group there were significant differences in most of the variables. This sensor is a value tool to objectively diagnose and monitor this pathology through wavefront phase changes.

Abstract Automatic segmentation of renal tumours during renal cell carcinoma surgery can help doctors accurately locate the tumour region, protect the tissues and organs around the kidneys, enhance surgical efficiency, and reduce the possibility of leakage and misdiagnosis. However, since general polyp endoscopic image segmentation models have many problems when facing the task of renal tumour segmentation, there needs to be more research on the segmentation of endoscopic images of renal tumours. This paper proposes a multi‐scale dynamic segmentation kernel network for endoscopic image segmentation of kidney tumours. First, a spatial receptive field module is proposed to augment the feature information and improve the performance of the whole network. Second, an enhanced cross‐attention module is offered to attenuate the effect of a high‐similarity segmentation background. Finally, a multi‐scale dynamic segmentation kernel module is introduced to gradually refine the segmentation results from small to large sizes to obtain more accurate tumour boundaries. Extensive experiments on the established kidney tumour endoscopic dataset and publicly available endoscopic datasets show that this method exhibits enhanced performance and generalization capabilities compared to existing techniques. On this renal tumour dataset, MDSK‐Net achieved excellent results of 94.1% and 90.1% on mDice and mIoU.

Abstract In order to improve the accuracy of granular image detection during image scaling, an image scaling algorithm combining the interpolation algorithm of protective features and Kalman filtering of neurons is proposed. Firstly, the interpolation algorithm of protective features is de‐designed according to the phenomenon of sudden change of edge grey values, and the difference and trend of grey values in horizontal and vertical directions are used to obtain the pre‐scaled image with maximum contrast. Then the grey value data input of the pre‐scaled image is introduced into the Kalman filter of neurons, and the filtering process is optimized by using the black‐box thinking of neurons, and the pre‐scaled image is smoothed. Finally, the scale transformation is implemented to obtain the final grey value to complete the final scaling, so that the image has both visual experience and features. Comparison with mainstream image scaling algorithms shows that the proposed algorithm can effectively overcome the degradation of granularity image detection accuracy due to image scaling, with point acutance value (PAV) improved by 10%–41%, and the granularity detection error caused by image scaling algorithms reduced from about 0.7% to about 0.1%, with a relative improvement of 85%.

Abstract Head pose estimation is an especially challenging task due to the complexity nonlinear mapping from 2D feature space to 3D pose space. To address the above issue, this paper presents a novel and efficient head pose estimation framework based on particle swarm optimized contrastive learning and multimodal entangled graph convolution network. Firstly, a new network, the region and difference‐aware feature pyramid network (RD‐FPN), is proposed for 2D keypoints detection to alleviate the background interference and enhance the feature expressiveness. Then, particle swarm optimized contrastive learning is constructed to alternatively match 2D and 3D keypoints, which takes the multimodal keypoints matching accuracy as the optimization objective, while considering the similarity of cross‐modal positive and negative sample pairs from contrastive learning as a local contrastive constraint. Finally, multimodal entangled graph convolution network is designed to enhance the ability of establishing geometric relationships between keypoints and head pose angles based on second‐order bilinear attention, in which point‐edge attention is introduced to improve the representation of geometric features between multimodal keypoints. Compared with other methods, the average error of our method is reduced by 8.23%, indicating the accuracy, generalization, and efficiency of our method on the 300W‐LP, AFLW2000, BIWI datasets.

Abstract Transformer possesses a broader perceptual scope, while the Convolutional Neural Network (CNN) excels at capturing local information. In this paper, the authors propose the Multi‐Sclale Feature Aggregation Network (MSFA‐Net) for single‐image dehazing which is fused with the advantages of Transformer and CNN. Our MSFA‐Net is based on the encoder–decoder structure, and there are four main innovations. Firstly, the authors make some improvements to the original Swin Transformer to make it more effective for dehazing tasks, and the authors name it Spatial Information Aggregation Transformer (SIAT). The authors place the SIAT in both encoder and decoder of MSFA‐Net for feature extraction. The authors propose an upsampling module called Efficient Spatial Resolution Recovery (ESRR) which is placed in the decoder part. Compared to commonly used transposed convolutions, the authors’ ESRR module has fewer computational cost. Considering that the haze distribution is always uneven and the information from each channel is different, the authors introduce the Dynamic Multi‐Attention (DMA) module to provide pixel‐wise weights and channel‐wise weights for input features. The authors place the DMA module between the encoder and decoder parts. As the network depth increases, the spatial structural information from the high‐resolution layer tends to degrade. To deal with the problem, the authors propose the Multi‐Scale Feature Fusion (MSFF) module to recover missing spatial structural information. The authors place the MSFF module in both the encoder and decoder parts. Extensive experimental results show that the authors’ proposed dehazing network achieves state‐of‐the‐art dehazing performance with relatively low computational cost.

Abstract Traditional dehazing algorithms have long been limited in their ability to remove fog effectively, especially in preserving details. This study proposes an improved dark channel prior dehazing algorithm designed to overcome the limitations of traditional algorithms by refining the transmission map and estimating atmospheric light intensity. The algorithm fully exploits the anti‐haze characteristics of the Gabor filter to extract multi‐directional texture features. By fusing these features to form a guidance map for guided filtering, it effectively reduces the blurring caused by guided filtering on the image edges, thereby producing a more accurate transmission map. Simultaneously, an enhanced atmospheric light successfully reduces interference from white objects in the image. The experiment phase utilized the publicly available RESIDE dataset for validation. The algorithm achieved a Peak Signal‐to‐Noise Ratio (PSNR) of 22.4 and a Structural Similarity Index (SSIM) of 0.88. These metrics indicate the algorithm's superior dehazing capabilities.

Abstract Metaheuristic is one of the techniques to improve the security of image encryption. However, existing metaheuristic image cryptosystems based on metaheuristic may have convergence difficulties during the optimization process, which cause insecurity and slow convergence. Besides, the time cost of the parallel execution model applied to metaheuristic image cryptosystems is not low enough. Therefore, a parallel many‐objective optimized key generation framework is proposed. Firstly, if four or more security indicators of cryptosystem, which are the results of security test, need to be optimized, the many‐objective optimization algorithm is employed to the proposed framework. With method adjusts the chaotic system parameters as the optimization key, which effectively avoid the convergence difficulty of the encryption key. Secondly, a master‐slave parallel model is improved to metaheuristic encryption. The model allocates the most time‐consuming fitness calculation work to slave nodes, which makes the modified model more reasonable and thus reduces the encryption time. To evaluate the performance of the proposed framework, a specific encryption scheme is constructed, that utilizes a 2D quadric map and many‐objective optimization algorithm based on dominance and decomposition (MOEA/DD) to optimize five security indicators. Experimental results reveal that this scheme has good security performances and less parallel encryption time.

Since low-light images are of poor quality and carry noise, resulting in insufficient contrast and brightness, unclear details in the image, and the problem of high cost of obtaining paired low-light image datasets, a no-reference low-light image enhancement method for feature map enhancement is proposed under the influence of the Mach band effect of biological vision. Firstly, the model uses the enhanced filter block (EFB) to enhance the feature of images and feature maps, which can suppress noise and enhance the feature details to improve the ability of the network to learn features. Then, skip connection is combined with efficient spatial attention (ESA) module to extract global context information and local regional features by fusing enhanced shallow features and deep features, which effectively retains image color information, avoids detail loss, and improves the generalization ability of the network. Finally, the pixel estimation curve is used to adjust the dynamic range of the pixels in the low-light image and enhance the brightness of the low-light image. Experimental results show that the PSNR, SSIM, LPIPS and NIQE of the image processed by the algorithm reach 17.709 dB, 0.657, 0.239 and 3.486 respectively. Compared with the existing mainstream algorithms, it can better achieve the purpose of image enhancement, and effectively improves the image brightness and detail information, while maintaining the naturalness of the image.

In recent years, the single-pixel imaging technique which uses a detector without spatial resolution and spatially modulated illumination patterns to reconstruct an object image has been finding its application for imaging objects in visibility obstructing conditions such as smoke or fog. The unifying feature of the studies published so far is the proof of workability of the methods proposed by the authors to improve the quality of single-pixel imaging images at their chosen scattering medium parameters without revealing the limits of applicability. This work experimentally demonstrates the influence of the number of scattering particles in the medium on the contrast of single-pixel images, and also compares the results with images obtained with a CCD camera, which allows not only to compare imaging methods under varying conditions, but also to evaluate the influence of losses introduced by the presence of a scattering medium on the contrast of single-pixel images. This work uses the classical experimental scheme of single-pixel imaging in which the single-pixel detector and focusing lens were replaced by a CCD camera to obtain images for comparison. A cuvette containing milk solution of different concentrations was placed between the object and the detector. For each concentration, an image of the object was reconstructed using the single-pixel imaging method and then recorded on the CCD camera until the concentration of the scattering agent was reached at which no image could be obtained by either method. The contrast was then calculated for each image obtained. It is shown that the single-pixel imaging method for milk concentrations up to 1/150 has an average contrast of 0.21, which does not decrease as the scattering increases. At the same time for CCD camera the contrast in the absence of scattering is 0.70, and with increasing milk concentration monotonically decreases to 0.07. The main feature of images obtained by single-pixel imaging through scattering media is the preservation of contrast as the concentration of the scattering medium increases indicating that the relationships between all the recorded on a single-pixel detector intensities used in the construction of the correlation function are preserved. A single-pixel image ceases to be reconstructed only when information about the object does not reach the detector due to multiple scattering and absorption produced by a milk solution. The considered features show the prospect of using single-pixel imaging for the construction of remote sensing systems with pattern recognition, as they allow obtaining similar images at different scattering coefficients of the scattering medium.

Karayolu esnek üstyapılarındaki çatlaklar genellikle trafik yükleri ve hava koşullarından kaynaklanır. Bu çatlakların genişlemeden tespit edilip gerekli bakımlarının yapılması, yol konforunun sürekliliğini sağlamanın yanı sıra bakım maliyetlerini de azaltacaktır. Bu çalışma, yoldaki çatlakları gerçek zamanlı ve yüksek doğrulukla tespit etmeyi amaçlamaktadır. Bu bağlamda, Geri Yayımlı Birlikte Evrim yaklaşımıyla İyileştirilmiş Derin Sinir Ağları ve görüntü işleme yöntemleri birlikte kullanılmıştır. Ayrıca, çeşitli sayı ve çözünürlüklerde çatlaklı görsel veriler içeren EdmCrack600, AsphaltCrack, CFD ve CrackSegmentation veri setleri kullanılarak yeni bir veri seti oluşturulmuş ve bu veri seti üzerinde Derin Sinir Ağları tabanlı öğrenme gerçekleştirilmiştir. Modelin doğruluğu, CFD veri seti kullanılarak Kesinlik, Duyarlılık ve F1-Skoru ile değerlendirilmiştir. Değerlendirme sonucunda, önerilen yöntemin saniyede 48 görsel üzerinde çatlak tespit edebildiği ve %92,74 Kesinlik, %88,92 Duyarlılık ve %89,61 F1 Skoru başarı oranlarına ulaştığı gözlemlenmiştir.

Abstract Soil texture is one of the most important elements to consider before planting and tillage. These features affect the product selection and regulate its water permeability. Discrimination of soils by determining soil texture features requires an intense workload and is time-consuming. Therefore, having a powerful tool and knowledge for texture-based soil discrimination could enable rapid and accurate discrimination of soils. This study focuses on presenting new models for 6 different soil sample groups (Soil_1 to Soil_6) based on 12 different machine learning algorithms that can be utilized for various problems. As a result, overall accuracy values were determined as greater than 99.2% (Trilayered Neural Network). The greatest accuracy value was found in Bayes Net (99.83%) and followed by Subspace Discriminant (99.80%). In the Bayes Net algorithm, MCC (Matthews Correlation Coefficient) and F-measure values were obtained as 0.994 and 0.995 for Soil_4 and Soil_6 sample groups while these values were 1.000 for other soil groups. Soil types can visually vary based on their texture, mineral composition, and moisture levels. The variability of this can be influenced by fertilization, precipitation levels, and soil cultivation. It is important to capture the images in soil conditions that are more stable. In conclusion, the present study has proven the feasibility of rapid, non-destructive, and accurate discrimination of soils by image processing-based machine learning.

Abstract Smart elevators provide substantial promise for time and energy management applications by utilizing cutting edge artificial intelligence and image processing technology. In order to improve operating efficiency, this project designs an elevator system that uses the YOLO model for object detection. Compared to traditional methods, our results show a 15% improvement in wait times and a 20% reduction in energy use. Due to the elevator’s increased accuracy and dependability, users’ qualitative feedback shows a high degree of pleasure. These results imply that intelligent elevator systems can make a significant contribution to more intelligent building management. Due to the elevator’s increased accuracy and dependability, users’ qualitative feedback shows a high degree of pleasure. These results imply that intelligent elevator systems can make a significant contribution to more intelligent building management. The successful integration of artificial intelligence (AI) and image processing technologies in elevator systems presents a promising foundation for future research and development. Further advancements in object detection algorithms, such as refining YOLO models for even higher accuracy and real-time adaptability, hold potential to enhance operational efficiency. Integrating smart elevators more deeply into IoT networks and building management systems could enable comprehensive energy management strategies and real-time decision-making. Predictive maintenance models tailored to elevator components could minimize downtime and optimize service schedules, enhancing overall reliability. Additionally, exploring adaptive user interfaces and personalized scheduling algorithms could further elevate user satisfaction by tailoring elevator interactions to individual preferences. Sustainable practices, including energy-efficient designs and integration of renewable energy sources, represent crucial avenues for reducing environmental impact. Addressing security concerns through advanced encryption and access control mechanisms will be essential for safeguarding sensitive data in smart elevator systems.

Abstract Emerging Industry 5.0 designs promote artificial intelligence services and data-driven applications across multiple places with varying ownership that need special data protection and privacy considerations to prevent the disclosure of private information to outsiders. Due to this, federated learning offers a method for improving machine-learning models without accessing the train data at a single manufacturing facility. We provide a self-adaptive framework for federated machine learning of healthcare intelligent systems in this research. Our method takes into account the participating parties at various levels of healthcare ecosystem abstraction. Each hospital trains its local model internally in a self-adaptive style and transmits it to the centralized server for universal model optimization and communication cycle reduction. To represent a multi-task optimization issue, we split the dataset into as many subsets as devices. Each device selects the most advantageous subset for every local iteration of the model. On a training dataset, our initial study demonstrates the algorithm's ability to converge various hospital and device counts. By merging a federated machine-learning approach with advanced deep machine-learning models, we can simply and accurately predict multidisciplinary cancer diseases in the human body. Furthermore, in the smart healthcare industry 5.0, the results of federated machine learning approaches are used to validate multidisciplinary cancer disease prediction. The proposed adaptive federated machine learning methodology achieved 90.0%, while the conventional federated learning approach achieved 87.30%, both of which were higher than the previous state-of-the-art methodologies for cancer disease prediction in the smart healthcare industry 5.0.

Abstract Countermeasures for chaff jamming have drawn great attention in the field of radar target detection and tracking. Current approaches for chaff jamming recognition and suppression exhibit limitations in practical effect, generalisation ability, and hybrid jamming handling. To address the above problems, the authors first transform the traditional 1D signal processing problem into a 2D semantic segmentation task and then solve it from the perspective of the dataset construction and algorithm design. For the dataset construction, the authors use both measured and simulated data to synthesise a more realistic labelled dataset (semi‐realistic dataset), which is also with good diversity due to its adjustable chaff interference background. For the algorithm design, the authors propose a Pre‐Decluttering Dual‐Stage UNet (D2UNet) to recognise and suppress chaff jamming in two stages successively, where the former provides prior attention masks for the latter. To further improve the performance of D2UNet, the authors also design a multi‐stage loss function to achieve progressive training. Extensive experimental results demonstrate that D2UNet delivers remarkable recognition accuracy (99.305%) and suppression performance (41.326 dB peak signal‐to‐jamming ratio, 0.9952 structure similarity index measure) on the semi‐realistic dataset. Its practical effect is further verified on measured data.

Abstract Due to differences in the quantity and size of observed targets, hyperspectral images are characterized by class imbalance. The standard deep learning classification model training scheme optimizes the overall classification error, which may lead to performance imbalance between classes in hyperspectral image classification frameworks. Therefore, a novel factor annealing decoupling compositional training method is proposed in this paper. Without requiring resampling or reweighting, it implicitly modulates the training process, so standard models can sufficiently learn the representation of the minority classes and further be trained as robust classifiers. Specifically, the label‐distribution‐aware margin loss is combined with the error‐rate‐based cross‐entropy loss via combination factor, which considers both imbalanced data representation learning and classifier overall performance. Then, a factor annealing optimization training scheme is designed to adjust the combination factor, which solves the stage division problem of two‐stage decoupling learning. Experimental results on two hyperspectral image datasets demonstrate that, as compared with other competing approaches, the proposed method can continuously and stably optimize the model parameters, achieving improvements in class average metrics and difficult classes without affecting overall classification performance.

Abstract Fires not only cause devastating consequences for human life and property, but also lead to soil erosion in forests. Therefore, it is necessary to design a novel algorithm that can quickly monitor smoke from fires. Most existing smoke segmentation methods do not consider the segmentation accuracy of algorithms under limited computational resources. To address this research gap, this paper proposes a lightweight smoke segmentation algorithm based on DeepLabV3+ that achieves fast inference speed and high accuracy for different sizes smoke. To reduce the number of parameters, the feature extraction network of the DeeplabV3+ algorithm is replaced by MobileNetV2, which enhances the extraction ability of the algorithm in segment smoke images. Then, the Convolutional Block Attention Module (CBAM) is added to the encoder part to enhance the perception of the algorithm for small smoke and effectively alleviates smoke mis‐segmentation. Furthermore, a newly designed loss function is used in the network. The experimental results show that the proposed method has improved by 1.27% in Smoke IoU and 1.21% in mPA compared with other methods. The weight size has been reduced to 10.76% of the original DeepLabV3+, and the inference time is only 33.71ms. Therefore, it is a more suitable early fire detection algorithm for resource‐constrained environments.

Abstract Medical image colouring techniques enable to colourize grey‐scale medical images for assisting doctors in diagnosis. Benefiting from the non‐linear fitting ability of deep neural network, deep medical image colouring techniques have achieved remarkable results. However, existing methods are still facing content and structure feature leakage, unrealistic colouring and poor scale invariability. Thus, this paper, proposes a Transformer‐based medical image colouring algorithm with long‐term dependency to avoid feature leakage of coloured images. To be specific, this method employs two different Transformer encoders to generate and encode feature sequences for grey‐scale medical images and real human colour slice images, respectively. Then, a novel multi‐layer Transformer decoder is used to stylize grey‐scale map image features based on the real physical colour feature sequences. For colouring images at different scales, we implement content‐ aware positional encoding with scale invariance and propose style‐aware positional encoding strategy to take realistic and physical colour prior into account. Extensive experimental results indicate our method has achieved better colourization effects than recent state‐of‐the‐art medical image colourization methods.

Abstract Image steganography is the art of concealing secret information within images to prevent detection. In deep‐learning‐based image steganography, a common practice is to fuse the secret image with the cover image to directly generate the stego image. However, not all features are equally critical for data hiding, and some insignificant ones may lead to the appearance of residual artifacts in the stego image. In this article, a novel network architecture for image steganography with hybrid attention mechanism based on generative adversarial network is introduced. This model consists of three subnetworks: a generator for generate stego images, an extractor for extracting the secret images, and a discriminator to simulate the detection process, which aids the generator in producing more realistic stego images. A specific hybrid attention mechanism (HAM) module is designed that effectively fuses information across channel and spatial domains, facilitating adaptive feature refinement within deep image representations. The experimental results suggest that the HAM module not only enhances the image quality during both the steganography and extraction processes but also improves the model's undetectability. Stego images are mixed with varying levels of noise in the training process, which can further improve robustness. Finally, it is verified that the model outperforms current steganography approaches on three datasets and exhibits good undetectability.

Abstract Accurate detection of algae in microscopic image plays a crucial role in water quality monitoring. However, the existing object detection methods still face challenges in accurately detecting different categories of algae in microscopic image. In order to improve the accuracy, an improved YOLOv5s model is proposed for microalgae detection, by combining a Receptive Field Enhancement (RFE) module, the Wise‐IoU v3 dynamic non‐monotonic focal loss function, and a Dynamic head (Dyhead), which is termed receptive field enhancement wise‐IoU dyhead (RWD)‐you only look once (YOLO). Firstly, to detect microalgae of various scales, the Bottleneck in the C3 module of YOLOv5s is replaced with a more reasonable RFE module. Secondly, Wise‐IoU v3 is applied to enhance detection accuracy by assigning varying weights between high‐quality and low‐quality images. Finally, Dyhead is introduced to enhance the representation capacity of the detection head by integrating three attention mechanisms: scale awareness, spatial awareness, and task awareness. The proposed RWD‐YOLO model significantly enhances the accuracy of algae detection in microscopic image. Specifically, the experimental results on the microalgae dataset show that the RWD‐YOLO achieves an mAP@0.5 of 93.2% and an mAP@0.5:0.95 of 65.1%. Compared to the original YOLOv5s, mAP@0.5 and mAP@0.5:0.95 are improved by 3.7% and 5.7%, respectively.

Abstract Lip‐reading deciphers speech by observing lip movements without relying on audio data. The rapid advancements in deep learning have significantly improved lip‐reading for both English and Chinese; however, research on dialects such as Cantonese remains scarce. Consequently, most Chinese lip‐reading datasets focus on Mandarin, with only a few addressing Cantonese. To bridge this gap, a sentence‐level Cantonese lip‐reading dataset, designated as Cantonese lip‐reading sentences are introduced, comprising over 500 unique speakers and more than 30,000 samples. To ensure alignment with real‐world scenarios, no restrictions are imposed on factors such as gender, age, posture, lighting conditions, or speech rate. A comprehensive description of the pipeline employed is provided for collecting and constructing the dataset and introduce an innovative visual frontend, 3D‐visual attention net. This frontend combines the advantages of convolution and self‐attention mechanisms to extract fine‐grained lip region features. These features are subsequently input into the conformer backend for temporal sequence modelling, achieving comparable performance on Chinese Mandarin lip reading dataset, lip reading sentences 2, lip reading sentences 3, and Cantonese lip‐reading sentences datasets. Benchmark tests on Cantonese lip‐reading sentences demonstrate the challenges it poses, providing a novel research foundation for dialect lip‐reading and fostering the advancement of Cantonese lip‐reading tasks.

Abstract Synthetic aperture radar (SAR) image change detection suffers from poor quality of the difference image and low detection accuracy. Hence, this paper proposes a SAR image change detection method based on a fused difference image and an optimized random forest scheme, termed LRN‐SSARF. Specifically, a fusion operator difference image LRN is proposed, which is generated using a weighted fusion of log‐ratio (LR), ratio (R), and normalized ratio (NoR). This difference image generation method reduces noise's influence. Then, the Otsu algorithm is applied to segment the difference image and select the training samples. The training samples are input into the random forest (RF) model optimised by the sparrow search algorithm (SSA) for training and classification. Finally, the region link is uesd to refine the detection results and generate the final result. The change detection results of six real SAR image scenes highlight that the proposed algorithm has a high detection accuracy, and affords appealing integrity and detailed information about the change regions. Specially, the detection accuracy advantage of the Bangladesh dataset is larger, with the accuracy and Kappa coefficient reaching 98.04% and 92.00%, much higher than the competitor methods.

Abstract Image denoising aims to remove noise from images and improve the quality of images. However, most image denoising methods heavily rely on pairwise training strategies and strict prior knowledge about image structure or noise distribution. While these methods exhibit significant results when handling known types of noise, their generalization performance diminishes when confronted with images containing unknown noise distributions. To address this issue, a two‐stage approach is introduced for enhancing the generalizability of image denoising. The proposed method does not rely on a large amount of paired data or prior knowledge of the noise type and level. Instead, it constructs a denoising pipeline with improved generalizability through an MLP‐based denoiser and generative diffusion prior. Specifically, in the first stage, an initial denoised image is predicted with a structure resembling that of the underlying clean image by introducing an MLP‐based U‐shaped denoising network aided by an implicit structural prior. In the second stage, the generalizability and quality of the denoiser are further enhanced by conditioning the result obtained from the previous stage on the pretrained denoising diffusion null‐space model. Extensive experimentation on multiple datasets demonstrates that this method exhibits better denoising performance and generalizability than other image denoising methods.

Abstract The salient object detection task tries to simulate the human visual system for most eye‐catching objects or regions detection. However, due to the complexity of the visual mechanisms, current methods will suffer from severe performance degradation, leading to inconsistent prediction results for the same regions, when directly adopting a model trained on a fixed resolution to evaluate at other different resolutions. Considering that consistency in predictions is essential for salient object detection, a cross‐scale resolution consistent salient object detection method, called RCNet, is proposed. Specifically, to enhance the model's capacity for generalization across images of varying resolutions and make the model implicitly learn the scale invariance, a multi‐resolution data enhancement module is constructed to generate images with arbitrary resolutions for the same scene. Moreover, to accomplish better multi‐level feature fusion, a cross‐scale fusion module is developed to fuse high‐level semantic features and low‐level detail features. Additionally, to explicitly learn the scale invariance of the salient scores, a hybrid salient consistency loss is formulated on salient object detection with different resolutions. Comprehensive evaluations on five benchmark datasets show that RCNet achieves a highly competitive result.

Abstract Micro‐expressions are spontaneous and unconscious facial movements that reveal individuals’ genuine inner emotions. They hold significant potential in various psychological testing fields. As the face is a 3D deformation object, the emergence of facial expression leads to spatial deformation of the face. However, existing databases primarily offer 2D video sequences, limiting descriptions of 3D spatial information related to micro‐expressions. Here, a new micro‐expression database is proposed, which contains 2D image sequences and corresponding 3D point cloud sequences. These samples were classified using both an objective method based on the facial action coding system and a non‐objective emotion classification method that considers video contents and participants’ self‐reports. A variety of feature extraction techniques are applied to 2D data, including traditional algorithms and deep learning methods. Additionally, a novel local curvature‐based algorithm is developed to extract 3D spatio‐temporal deformation features from the 3D data. The authors evaluated the classification accuracies of these two features individually and their fusion results under leave‐one‐subject‐out (LOSO) and tenfold cross‐validation. The results demonstrate that fusing 3D features with 2D features results in improved recognition performance compared to using 2D features alone.

Abstract Synthetic images facilitate stereo matching. However, synthetic images may suffer from image distortion, domain bias, and stereo mismatch, which would significantly restrict the widespread use of stereo matching models in the real world. The first goal in this paper is to synthesize real‐looking images for minimizing the domain bias between the synthesized and real images. For this purpose, sharpened disparity maps are produced from a mono real image. Then, stereo image pairs are synthesized using these imperfect disparity maps and the single real image in the proposed pipeline. Although the synthesized images are as realistic as possible, the domain styles of the synthesized images are always very different from the real images. Thus, the second goal is to enhance the domain generalization ability of the stereo matching network. For that, the feature extraction layer is replaced with a teacher–student model. Then, a constraint of binocular contrast features is imposed on the output of the model. When tested on the KITTI, ETH3D, and Middlebury datasets, the accuracy of the method outperforms traditional methods by at least 30%. Experiments demonstrate that the approaches are general and can be conveniently embedded into existing stereo networks.

Abstract Autism spectrum disorder (ASD) is a complex psychological syndrome characterized by persistent difficulties in social interaction, restricted behaviours, speech, and nonverbal communication. The impacts of this disorder and the severity of symptoms vary from person to person. In most cases, symptoms of ASD appear at the age of 2 to 5 and continue throughout adolescence and into adulthood. While this disorder cannot be cured completely, studies have shown that early detection of this syndrome can assist in maintaining the behavioural and psychological development of children. Experts are currently studying various machine learning methods, particularly convolutional neural networks, to expedite the screening process. Convolutional neural networks are considered promising frameworks for the diagnosis of ASD. This study employs different pre‐trained convolutional neural networks such as ResNet34, ResNet50, AlexNet, MobileNetV2, VGG16, and VGG19 to diagnose ASD and compared their performance. Transfer learning was applied to every model included in the study to achieve higher results than the initial models. The proposed ResNet50 model achieved the highest accuracy, 92%, compared to other transfer learning models. The proposed method also outperformed the state‐of‐the‐art models in terms of accuracy and computational cost.

Nanomaterials, particularly nanofibers produced through electrospinning, have garnered significant attention due to their unique properties and diverse applications. This research explores the influence of collector design, electrode geometry, and collector speed on the properties of electrospun polyacrylonitrile (PAN) nanofibers. Finite element analysis (FEA) was employed to simulate electric fields, revealing the impact of collector geometry on field intensity. The experimental setup, enclosed in an isolated chamber, employed various collector types and electrode configurations. Scanning electron microscope (SEM) analysis showcased the effect of collector speed on fiber alignment and diameter. Furthermore, FEA simulations elucidated the role of electrode geometry and voltage in shaping the electric field, impacting fiber properties. The study introduces a novel, in-house method for producing highly aligned nanofibers and provides insights into optimizing electrospinning parameters for enhanced fiber properties. A testing protocol is devised to minimize surface damage when conducting mechanical tests on nanofiber films, employing a dynamic mechanical analyzer (DMA). Mechanical testing demonstrated the correlation between alignment and tensile strength. Overall, this research contributes valuable insights for tailoring electrospinning processes for tissue engineering and energy storage.

[Objective] To solve the problem of poor manual detection accuracy of traditional grain and chaff separator and improve production efficiency. [Methods] An image detection method based on machine vision was proposed, which realized the feature recognition and separation of grain rough through multi-stage progressive fusion of different image algorithms. The acquired images were selected in the ROI region and enhanced by Retinex algorithm. The Otsu algorithm was used to segment the image, and then the median filtering wwas combined with morphology to remove the image noise. The improved Canny algorithm was used to detect edge features of binary images, and the position information of the contour of the valley rough image was extracted by combining the Hough transform. Finally, the state estimation of the position information was performed by using the Kalman filter, and the best predicted value of the separated position was obtained, while the position offset error was reduced. [Results] The average detection error of the system was 3.14 mm, a decrease of 1.82 mm compared to before filtering, and the average standard deviation of filtering error was 0.8 mm. [Conclusion] This method can effectively detect the grain rough feature information and improve the separation accuracy.

Abstract In view of the reduced power generation efficiency caused by ash or dirt on the surface of photovoltaic panels, and the problems of heavy workload and low efficiency faced by manual detection, this study proposes a method to detect dust or dust on the surface of photovoltaic cells with the help of image processing technology to timely eliminate hidden dangers and improve power generation efficiency.This paper introduces image processing methods based on mathematical morphology, such as image enhancement, image sharpening, image filtering and image closing operation, which makes the image better highlight the target to be recognized. At the same time, it also solves the problem of uneven image binarization caused by uneven illumination in the process of image acquisition. By using the image histogram equalization, the gray level concentration area of the original image is opened or the gray level is evenly distributed, so that the dynamic range of the pixel gray level is increased, so that the image contrast or contrast is increased, the image details are clear, to achieve the purpose of enhancement. When identifying the target area, the method of calculating the proportion of the dirt area to the whole image area is adopted, and the ratio exceeding a certain threshold is judged as a fault. In addition, the improved A* path planning algorithm is adopted in this study, which greatly improves the efficiency of the unmanned aerial vehicle detection of photovoltaic cell dirt, saves time and resources, reduces operation and maintenance costs, and improves the operation and maintenance level of photovoltaic units.

Abstract Plant diseases significantly threaten global agriculture, impacting crop yield and food security. Nearly 30% of the crop yield is lost due to plant diseases. Efficient identification and classification of plant diseases through computer vision techniques have become imperative for timely intervention. However, popular plant disease datasets often suffer from data imbalance, with certain classes underrepresented, hindering the performance of machine learning models. Traditional data augmentation methods, such as rotation and flipping, are limited in their effectiveness, especially when faced with imbalanced datasets. To address this limitation, we explore advanced data augmentation techniques, including Generative Adversarial Networks (GANs) such as CycleGAN and LeafGAN, which have shown promise in generating synthetic images. However, we propose an innovative approach of Object-based single Style Transfer on a single neural network for augmenting the plant disease dataset. This technique focuses on mitigating data imbalance issues within datasets, which can adversely affect the model’s ability to generalize across diverse classes. The proposed method is compared with state-of-the-art data augmentation techniques, highlighting its superiority in addressing data imbalance issues. Our approach aims to produce more realistic and diverse synthetic images, leading to improved model generalization and accuracy in plant disease classification tasks validated using different classifiers. The efficiency of our approach is validated through extensive experimentation and benchmarking against existing methods.

Abstract Intelligent sorting of coal and gangue is of great significance to the intelligent construction of coal mines as well as green development. In this study, we propose a coal and gangue segmentation method with an improved classical segmentation network Mask R‐CNN, denoted as Multichannel Forward‐Linked Confusion Convolution Module (MFCCM)‐Mask R‐CNN. First, we design a MFCCM to construct the feature extraction network by stacking, second, we design a multiscale high‐resolution feature pyramid network structure to realize multipath fusion of feature information to enhance the position and contour information of the target, and finally, we propose a multiscale Mask head to enhance the diversity of information, and capture the more representative and unique features. Training and testing models using self‐built RGB coal and gangue data sets, the accuracy of the improved algorithm reaches 97.38%, which is an improvement of 1.66% compared to the original model. Compared with other segmentation models Unet, Deeplab V3+, Yoloact, Yolov7, and the model after replacing the backbone network, the MFCCM‐Mask R‐CNN has higher precision and recall, and can more accurately realize the efficient segmentation of coal and gangue.

Based on the literature in recent years, the application and development of machine vision in the detection of food bottle and can packaging defects were introduced, and the hardware structure and detection process of the detection system were explained. The research results based on image processing technology and deep learning method in the field of packaging defect detection of food bottles and cans were reviewed and analyzed. The research also summarized the advantages and disadvantages of classification network model and target detection network model both technologies and prospected their future development, and provide a reference for the innovative development of intelligent packaging inspection.

The aflatoxin produced by peanut mildew is highly carcinogenic, and it seriously affects food safety. In order to accurately and quickly identify moldy peanuts, this project proposes a detection method for moldy peanuts based on machine vision. Firstly, the peanut image was double-sided filtering and noise reduction, and then the image was converted to HSV space. The moldy peanut was recognized and detected by superimposing the mold color range extracted in H and S space and the open processing results of V space. The experimental results showed that the recognition accuracy of this method for moldy peanuts reached 95.3%, and the processing time for a single frame of peanut image was 0.6 seconds. Compared with other algorithms, this method had the advantages of fast speed and high accuracy, which can meet the real-time detection of moldy peanuts. At the same time, the grading processing of peanut mold is also more practical.

Abstract Super-resolution imaging has, for more than seventy years, gradually evolved to produce advanced methods for enhancing the resolution of images beyond the diffraction limits. Notwithstanding its foreseeable practical capabilities, we noted that this technology has received undeserved attention. The present work provides an extensive review of super-resolution imaging since its first conception in 1952, contextualizing studies into four research directions: reviews, advances, applications, and hardware implementations. We have discussed achievements, challenges, and potential opportunities of super-resolution imaging to equip researchers, especially those in their early careers, with knowledge to further advance the technology. Our work may open interesting research avenues that may accelerate realization of the technology into commercial imaging devices.

Abstract Crop diseases can significantly affect various aspects of crop cultivation, including crop yield, quality, production costs, and crop loss. The utilization of modern technologies such as image analysis via machine learning techniques enables early and precise detection of crop diseases, hence empowering farmers to effectively manage and avoid the occurrence of crop diseases. The proposed methodology involves the use of modified MobileNetV3Large model deployed on edge device for real-time monitoring of grape leaf disease while reducing computational memory demands and ensuring satisfactory classification performance. To enhance applicability of MobileNetV3Large, custom layers consisting of two dense layers were added, each followed by a dropout layer, helped mitigate overfitting and ensured that the model remains efficient. Comparisons among other models showed that the proposed model outperformed those with an average train and test accuracy of 99.66% and 99.42%, with a precision, recall, and F1 score of approximately 99.42%. The model was deployed on an edge device (Nvidia Jetson Nano) using a custom developed GUI app and predicted from both saved and real-time data with high confidence values. Grad-CAM visualization was used to identify and represent image areas that affect the convolutional neural network (CNN) classification decision-making process with high accuracy. This research contributes to the development of plant disease classification technologies for edge devices, which have the potential to enhance the ability of autonomous farming for farmers, agronomists, and researchers to monitor and mitigate plant diseases efficiently and effectively, with a positive impact on global food security.

Abstract Currently, deep learning methods for low‐light image enhancement tasks mainly focus on the illumination of images, while neglecting the problems of image noise and feature loss. To address this issue, this paper proposes a novel low‐light image enhancement network called DAF‐Retinex, based on the Retinex‐Net. To address the issue of image noise, different from traditional image denoise methods, this paper utilizes a fully convolutional neural network to denoise the reflection component, additionally, a denoising loss function is introduced to suppress noise. For preserving image details and extracting features, this paper creatively introduces self‐calibrated convolutions into low‐light image enhancement tasks, furthermore, a feature augmented attention block consisting of feature‐guided attention (FGA) is designed for feature learning to effectively enhance image illumination and extract image detail features. Experimental results demonstrate that the proposed algorithm in this paper effectively removes image noise and extracts detailed features, resulting in visually improved outcomes. On public datasets, the average improvement in objective evaluation metrics of image quality such as PSNR, SSIM, and NIQE are 1.13%, 4.12%, and 1.28%, respectively.

Abstract Current research on apron conflict detection is often limited to the interaction between the aircraft as a whole and other objects, making it difficult to accomplish targeted identification of vulnerable and high‐cost aircraft components. However, the implementation of detailed aircraft identification is of great significance to enhance the safety of airport surface operations. Based on the excellent performance of High‐Resolution Network (HRNet) in keypoint detection, a lightweight end‐to‐end keypoint detection network, namely Squeeze and Excitation High‐Resolution Network (SEHRNet), is proposed in this paper to solve the problems of HRNet's slower computation and more redundancy. First, the errors arising from coordinate transformations in the coding and decoding process are solved by an improved feature map coding and decoding process. Second, replace the BasicBlock in HRNet with the Depthwise separable convolutions based on the Squeeze‐and‐Excitation Networks, which drastically cuts the computational cost of the network. Third, the improved Bottleneck module is used to further enhance the capability of feature extraction. Experimental results prove that, based on the aircraft keypoint detection dataset, the SEHRNet proposed in this paper shows stronger applicability compared to the current mainstream networks. Compared with the original HRNet, the improved network has higher accuracy, faster speed, and lighter model for aircraft keypoint detection.

Abstract Accurate detection of underwater objects is a key indicator technology to effectively enhance the field of marine development and application, and is of great importance to various fields including marine military defense and seafood aquaculture. Efficient and rapid detection of underwater targets is a crucial technological challenge in this field. To meet the challenges posed by these issues, this study applies the convolutional omni‐efficient layer aggregation network (CO‐ELAN) module to the detector backbone to improve the ability of the network structure to acquire underwater objects from image information. The module improves the feature representation of gradient branching through a multi‐dimensional dynamic convolution and attention mechanism. In terms of loss calculation, the optimized normalized Wasserstein distance approach is used to predict the box distribution probabilistic modelling method to determine comparable distances to the ground box and obtain better samples of small target labels. Here, an underwater image enhancement algorithm based on white balance and underwater blur fusion is used to obtain clear images that enable improved detector performance. After the verification experiment on the URPC2018 dataset, it is found that the detector has better underwater detection ability compared with other detectors in the complex underwater environment. The proposed method achieves a 2.4% improvement over the YOLOv7 baseline model, while reducing computation costs by 5%.

Abstract The discrepancy in data distribution between training and testing scenarios, as well as the inductive bias of convolutional neural networks towards image styles, reduces the model's generalization ability. Many unsupervised domain generalization methods based on feature decoupling suffer from an initial neglect of explicit decoupling of content and style features, resulting in content features that still contain considerable redundant information, thereby restricting improvements in generalization capability. To tackle this problem, this paper optimizes the learning process of domain‐invariant (content) features into an information compression issue, minimizing redundancy in content features. Furthermore, to enhance decoupled learning, this paper introduces innovative cross‐domain loss functions and image reconstruction modules that explicitly decouple and merge content and style across different domains. Extensive experiments demonstrate the method's significant enhancements over recent cutting‐edge approaches.

Abstract In the field of image processing and machine vision, object tracking is a significant and rapidly developing subfield. The numerous potential applications of object tracking have garnered much attention in recent years. The effectiveness of tracking and detecting moving targets is directly related to the quality of motion detection algorithms. This paper presents a new method for estimating the tracking of objects by linearizing their trajectories. Estimating the movement paths of objects in dynamic and complex environments is one of the fundamental challenges in various fields, such as surveillance systems, autonomous navigation, and robotics. Existing methods, such as the Kalman filter and particle filter, each have their strengths and weaknesses. The Kalman filter is suitable for linear systems but less efficient in nonlinear systems, while the particle filter can better handle system nonlinearity but requires more computations. The main goal of this research is to improve the accuracy and efficiency of estimating the movement paths of objects by combining path linearization techniques with existing advanced methods. In this method, the nonlinear model of the object's path is first transformed into a simpler linear model using linearization techniques. The Kalman filter is then used to estimate the states of the linearized system. This approach simplifies the calculations while increasing the estimation accuracy. In the subsequent step, a particle filter-based method is employed to manage noise and sudden changes in the object's trajectory. This combination of two different methods allows leveraging the advantages of both, resulting in a more accurate and robust estimate. Experimental results show that the proposed method performs better than traditional methods, achieving higher accuracy in various conditions, including those with high noise and sudden changes in the movement path. Specifically, the proposed approach improves movement forecasting accuracy by about 12% compared to existing methods. In conclusion, this research demonstrates that object trajectory linearization can be an effective tool for improving object tracking estimation. Combining this technique with existing advanced methods can enhance the accuracy and efficiency of tracking systems. Consequently, the results of this research can be applied to the development of advanced surveillance systems, self-driving cars, and other applications.

Abstract The ability to predict the trajectory of an autonomous vehicle accurately is crucial for safe and efficient navigation. However, predicting diverse and multimodal futures can be challenging. Recent approaches such as attention and graph neural networks have achieved state‐of‐the‐art performance by considering agent interactions and map contexts. This study focused on multi‐agent prediction using an agent‐centric approach with transformers. This enables parallel computation and a comprehensive understanding of the environment. Two main features are introduced: an adaptive receptive field (ARF) that captures the relevant surroundings for each agent, and perception encoding, which serves as spatial context embeddings. The ARF adapts to the agent's velocity and rotation, focusing attention ahead at high speeds or to the sides when it is slower. Perception encoding divides agents or lanes into levels and encodes the information of each level. This approach enables the efficient encoding of complex spatial relationships. The proposed method combines these advances with transformer modelling for multi‐agent trajectory prediction while ensuring real‐time prediction capabilities. The approach is evaluated on the Argoverse benchmark and better performance than the state‐of‐the‐art baseline is achieved. By addressing challenges such as multimodal outputs and robustness, the study enhances the safety and efficiency of autonomous driving systems by more accurately predicting trajectories.

The phenomenon of fluctuating chili prices can be resolved in stages, one of which is through multiple crosses. However, this cross requires precise methods in the evaluation and selection process, especially regarding fruit characteristics. Image-based phenotyping 4.0 approaches can increase the potential precision of such evaluation genotypes, especially when this approach is combined with multivariate analysis. Therefore, both methods are needed to evaluate and select these cayenne multiple crosses. This research aims to identify the effectiveness of multivariate analysis and image-based explanatory characteristics of fruit phenotypes and to select multiple crosses that can continue to the F2 generation. The research was designed with a randomized complete block design of ten F1 multiple cross-genotypes and four check varieties. Each genotype was repeated three times, so there were 42 experimental units. Based on the results, multivariate was considered adequate in determining image explanatory characters based on fruit phenotype and genotype mapping of the population diversity of multiple crosses of cayenne pepper. The characteristics of fruit height, fruit area, and fruit Intden are image-based explanatory characters that can map the completeness of cayenne pepper fruit between multiple crosses well. This indicates that image-based phenotyping and multivariate analysis can provide more detailed image information of the potential of cayenne fruit from multiple crosses than just based on fruit weight. Therefore, both approaches are recommended for analyzing cayenne paper fruit potential, especially for multiple crosses. In addition, three crosses (MC4, MC8, and MC9) are optimal for the next generation to be recommended and continued.

AbstractThis research aims to characterize low-temperature plasma-treated ramie fibers (Boehmeria nivea L. Gaud), create a mathematical model, and analyze them using image processing. The effects of low-temperature plasma treatment on ramie fibers were investigated using 30 kV output powers, 4-min treatment times, and a 4 cm electrode distance. SEM and FT-IR spectroscopy were used to investigate various aspects of the ramie fiber’s chemical properties and surface topography. This study used a statistical approach to analyze the SEM images. The statistical analysis was used to investigate the surface roughness using the SEM images. FT-IR analysis discovered that fibers that were exposed for four minutes differed from untreated fibers. According to the findings, plasma treatment caused: (1) ramie fiber became more hydrophilic, as indicated by the presence of hydrophilic functional groups such as hydroxyl (O–H), carboxyl (O–C), and carbonyl (C=O) in FT-IR with a deeper T%, (2) ramie fiber mass reduction was accompanied by an increase in surface roughness via SEM testing and image processing, and (3) plasma-treated fiber had a coefficient of mass variation (CV) of 0.7211 higher than untreated fiber. The novelty of this study is the use of image processing in SEM to observe surface roughness and its relationship to the chemical structure of ramie fibers observed with FT-IR. Another novelty of this research is the mathematical model of the interaction of plasma species with the ramie fibers on the plane electrode for the first time.

In recent years, there has been a notable increase in interest regarding intelligent classification techniques rooted in Machine Learning within the domain of medical science. Specifically, machine learning, a pivotal area of artificial intelligence, has been extensively utilized to aid medical professionals in predicting and diagnosing various diseases. This study applies two distinct machine learning algorithms to address a medical diagnosis concern related to circular masses in breast cancer. The dataset encompasses 150 cases of breast cancer patients. The primary objective is to assess and compare the effectiveness of artificial neural networks (ANNs) and linear discriminant analysis (LDA) classifiers based on key criteria: accuracy, sensitivity, specificity, and the kappa coefficient in predicting circular masses within breast cancer. Results indicate that the performance of the ANN classifier surpasses that of the LDA model, achieving an accuracy of 97.7%, sensitivity of 95%, specificity of 100%, and a kappa coefficient of 95.31%. Additionally, the final fitted models unveil the pivotal factors significantly influencing circular masses in breast cancer, highlighting Solidity and Entropy as the most critical variables.

The valuable saffron metabolites have encouraged researchers to study their laboratory production. In this study, two experiments have been designed to optimize saffron corm callogenesis in-vitro conditions and assessed the application of image analysis to address some characteristics of calli without cell destruction. Effect of physiological age of the corm, different types of media, and plant growth regulators (PGRs) with different concentrations as independent variables were investigated on cell growth index, the redness intensity (a*value) of callus and the percentage of black spot as dependent variables. Results showed that the mature corm-derived explants on Gamborg (B5) medium create the best callogenesis and production of red-colored cells with the least black spot. It was also revealed that naphthalene acetic acid (NAA), (2 mg/L) in combination with 6-Benzylaminopurine (BA) (1 mg/L) increased the growth index to 1.14. In contrast, the combination of indole-3-acetic acid (IAA), (2 mg/L) or indole-3-butyric acid (IBA), (10 mg/L) with kinetin (KIN), (1 mg/L) was the most effective PGRs for the production of red cells. The HPLC results confirmed that the red color of the calli was related to crocin presence. It is concluded that although the cultivation of saffron mature corms in the B5 medium is the most suitable option for biomass growth and crocin production, the choice of PGRs varies depending on the purpose of the culture (biomass or crocin production). Moreover, image analysis can be considered as a promising, alternative, un-destructive, and fast method for estimating the cell growth index and crocin content.

The early detection of pests in plants and crops is essential for the production of good quality food. Computer vision techniques can be applied for the early detection of pests and which can minimise the pesticides used on the plants. Among many pests, white flies, mites, aphids and thrips are the most hazardous pests that affect the leaves. This paper presents an automated approach for the detection of different types of pests from leaf images of plants. The images of the plant leaves were acquired using a digital camera. Image pre-processing techniques, such as noise removal, filtering and contrast enhancement, are used for improving the quality of the images. Then, the k-means clustering method and global thresholding were used for segmenting the pests from the infected leaves. Textural features are extracted from those segmented images by statistical feature extraction methods. Artificial Neural Network (ANN) and Binary Support Vector Machine (SVM) classifiers were used to classify the unaffected leaf images from the pest affected leaf images and a multi-SVM classifier was used to identify the different types of pests.

Artificial intelligence has become widespread in image processing tasks. At the same time, the number of vulnerabilities is increasing in systems implementing these artificial intelligence technologies (the attack surface is increasing). The main threats to information security can be implemented by introducing malicious perturbations into the input data, regardless of their type. To detect such attacks, approaches and methods have been developed based, in particular, on the use of an auto-encoder or the analysis of layers of the target neural network. The disadvantage of existing methods, which significantly reduce the scope of their application, is binding to the dataset or model architecture. This paper discusses the issues of expanding the scope (increasing scalability) of methods for detecting L0-optimized perturbations introduced by unconventional pixel attacks. An approach to detecting these attacks using statistical analysis of input data, regardless of the model and dataset, is proposed. It is assumed that the pixels of the perturbation embedded in the image, as a result of the L0-optimized attack, will be considered both local and global outliers. Outlier detection is performed using statistical metrics such as deviation from nearest neighbors and Mahalanobis distance. The evaluation of each pixel (anomaly score) is performed as a product of the specified metrics. A threshold clipping algorithm is used to detect an attack. When a pixel is detected for which the received score exceeds a certain threshold, the image is recognized as distorted. The approach was tested on the CIFAR-10 and MNIST datasets. The developed method has demonstrated high accuracy in detecting attacks. On the CIFAR-10 dataset, the accuracy of detecting onepixel attack (accuracy) was 94.3 %, and when detecting a Jacobian based Saliency Map Attack (JSMA) — 98.3 %. The proposed approach is also applicable in the detection of modified pixels. The proposed approach is applicable for detecting one-pixel attacks and JSMA, but can potentially be used for any L0-optimized distortions. The approach is applicable for color and grayscale images regardless of the dataset. The proposed approach is potentially universal for the architecture of a neural network, since it uses only input data to detect attacks. The approach can be used to detect images modified by unconventional adversarial attacks in the training sample before the model is formed.

Today, advancements in science and technology have spurred the rapid evolution of systems like electronic banking, demanding precise, swift, and secure identification of individuals based on their distinct traits. Among these traits, fingerprints stand out as a dependable means of identification, finding application in realms such as crime investigation and national border control due to their simplicity and heightened security. The qualities inherent in fingerprint-based identification have led to its widespread adoption over other identification methods. This article proposes a hybrid biometric system that integrates the Gabor filter and scale-invariant feature transform features and then uses support vector machine and K-nearest neighbors as classifiers, aiming to notably enhance authentication systems by mitigating issues seen in single-method biometric systems. Also, principal component analysis is used to reduce dimensions and eliminate redundancy. In this article, the famous database FVC2004 is used. Test results highlight the considerable reliability and accuracy of the proposed combined approach compared to systems reliant on a singular biometric method.