Search

Segmentation of whole slide histological images through the classification of tissue types of small fragments is an extremely relevant task in digital pathology, necessary for the development of methods for automatic analysis of whole slide histological images. The extremely large resolution of such images also makes the task of increasing image resolution relevant, which allows storing images at a reduced resolution and increasing it if necessary. Annotating whole slide images by histologists is complex and time-consuming, so it is important to make the most efficient use of the available data, both labeled and unlabeled. In this paper we propose a novel neural network method to simultaneously solve the problems of super-resolution of histological images from ×20 optical magnification to ×40 and classifying image fragments into tissue types at ×20 magnification. The use of a single encoder as well as the proposed neural network training scheme allows to achieve better results on both tasks compared to existing approaches. The PATH-DT-MSU WSS2v2 dataset presented for the first time in this paper was used for training and testing the method. On the test sample, an accuracy value of 0.971 and a balanced accuracy value of 0.916 were achieved in the classification task on 5 tissue types, for the super-resolution task, values of PSNR = 32.26 and SSIM = 0.89 were achieved. The source code of the proposed method is available at: https://github.com/Kukty/WSI_SR_CL. [ABSTRACT FROM AUTHOR]

With the increasing applications of traffic scene image classification in intelligent transportation systems, there is a growing demand for improved accuracy and robustness in this classification task. However, due to weather conditions, time, lighting variations, and annotation costs, traditional deep learning methods still have limitations in extracting complex traffic scene features and achieving higher recognition accuracy. The previous classification methods for traffic scene images had gaps in multi-scale feature extraction and the combination of frequency domain, spatial, and channel attention. To address these issues, this paper proposes a multi-scale and multi-attention model based on Res2Net. Our proposed framework introduces an Adaptive Feature Refinement Pyramid Module (AFRPM) to enhance multi-scale feature extraction, thus improving the accuracy of traffic scene image classification. Additionally, we integrate frequency domain and spatial-channel attention mechanisms to develop recognition capabilities for complex backgrounds, objects of different scales, and local details in traffic scene images. The paper conducts the task of classifying traffic scene images using the Traffic-Net dataset. The experimental results demonstrate that our model achieves an accuracy of 96.88% on this dataset, which is an improvement of approximately 2% compared to the baseline Res2Net network. Furthermore, we validate the effectiveness of the proposed modules through ablation experiments. [ABSTRACT FROM AUTHOR]

Medical image diagnosis using deep learning has shown significant promise in clinical medicine. However, it often encounters two major difficulties in real-world applications: (1) domain shift, which invalidates the trained model on new datasets, and (2) class imbalance problems leading to model biases towards majority classes. To address these challenges, this paper proposes a transfer learning solution, named Dynamic Weighting Translation Transfer Learning (DTTL), for imbalanced medical image classification. The approach is grounded in information and entropy theory and comprises three modules: Cross-domain Discriminability Adaptation (CDA), Dynamic Domain Translation (DDT), and Balanced Target Learning (BTL). CDA connects discriminative feature learning between source and target domains using a synthetic discriminability loss and a domain-invariant feature learning loss. The DDT unit develops a dynamic translation process for imbalanced classes between two domains, utilizing a confidence-based selection approach to select the most useful synthesized images to create a pseudo-labeled balanced target domain. Finally, the BTL unit performs supervised learning on the reassembled target set to obtain the final diagnostic model. This paper delves into maximizing the entropy of class distributions, while simultaneously minimizing the cross-entropy between the source and target domains to reduce domain discrepancies. By incorporating entropy concepts into our framework, our method not only significantly enhances medical image classification in practical settings but also innovates the application of entropy and information theory within deep learning and medical image processing realms. Extensive experiments demonstrate that DTTL achieves the best performance compared to existing state-of-the-art methods for imbalanced medical image classification tasks. [ABSTRACT FROM AUTHOR]

The census and management of hazard-bearing entities, along with the integrity of data quality, form crucial foundations for disaster risk assessment and zoning. By addressing the challenge of feature confusion, prevalent in single remotely sensed image recognition methods, this paper introduces a novel method, Spatially Constrained Deep Learning (SCDL), that combines deep learning with spatial constraint strategies for the extraction of disaster-bearing bodies, focusing on dams as a typical example. The methodology involves the creation of a dam dataset using a database of dams, followed by the training of YOLOv5, Varifocal Net, Faster R-CNN, and Cascade R-CNN models. These models are trained separately, and highly confidential dam location information is extracted through parameter thresholding. Furthermore, three spatial constraint strategies are employed to mitigate the impact of other factors, particularly confusing features, in the background region. To assess the method's applicability and efficiency, Qinghai Province serves as the experimental area, with dam images from the Google Earth Pro database used as validation samples. The experimental results demonstrate that the recognition accuracy of SCDL reaches 94.73%, effectively addressing interference from background factors. Notably, the proposed method identifies six dams not recorded in the GOODD database, while also detecting six dams in the database that were previously unrecorded. Additionally, four dams misdirected in the database are corrected, contributing to the enhancement and supplementation of the global dam geo-reference database and providing robust support for disaster risk assessment. In conclusion, leveraging open geographic data products, the comprehensive framework presented in this paper, encompassing deep learning target detection technology and spatial constraint strategies, enables more efficient and accurate intelligent retrieval of disaster-bearing bodies, specifically dams. The findings offer valuable insights and inspiration for future advancements in related fields. [ABSTRACT FROM AUTHOR]

Since plant diseases occurring during the growth process are a significant factor leading to the decline in both yield and quality, the classification and detection of plant leaf diseases, followed by timely prevention and control measures, are crucial for safeguarding plant productivity and quality. As the traditional convolutional neural network structure cannot effectively recognize similar plant leaf diseases, in order to more accurately identify the diseases on plant leaves, this paper proposes an effective plant disease image recognition method aECA-ResNet34. This method is based on ResNet34, and in the first and the last layers of this network, respectively, we add this paper's improved aECAnet with the symmetric structure. aECA-ResNet34 is compared with different plant disease classification models on the peanut dataset constructed in this paper and the open-source PlantVillage dataset. The experimental results show that the aECA-ResNet34 model proposed in this paper has higher accuracy, better performance, and better robustness. The results show that the aECA-ResNet34 model proposed in this paper is able to recognize diseases of multiple plant leaves very accurately. [ABSTRACT FROM AUTHOR]

In this paper, we discuss the implementation of artificial self-awareness mechanisms and self-reflection abilities in deep neural networks. While the current limitations of research prevent achieving cognitive capabilities on par with natural biological entities, the incorporation of basic self-awareness and self-reflection mechanisms in deep learning architectures offers substantial advantages in tackling specific problems across various scientific fields, including geosciences. In the first section, we outline the foundational architecture of our deep learning approach termed Self-Aware Learning (SAL). The subsequent part of the paper highlights the practical benefits of this machine learning methodology through synthetic tests and applications addressed to automatic classification and image analysis of real petrological data sets. We show how Self-Aware Learning allows enhanced accuracy, reduced overfitting problems, and improved performances compared to other existing methods. [ABSTRACT FROM AUTHOR]

Objectives: The quality of network structure is the key to obtain excellent classification performance. Most of existing deep models are designed by the trial-and-error method. When the same model processes different hyperspectral images (HSIs), different network hyperparameters need to be set artificially to adapt to the data characteristics of different images, greatly increasing the difficulty of network design. Aiming at the problem of deep learning model design in hyperspectral image classification, this paper proposes an automatic network structure search method for HSI classification. Methods: In the proposed method, the differentiable structure search technology is first used to search the network structure on the source HSI data sets, then the deep network model is constructed by stacking cells, and finally the classification performance is evaluated on the target HSIs. It should be pointed out that, the proposed method only conducts one search on the source HSI data sets, and the obtained deep model can be applied to other target HSI classification tasks, which can effectively improve the utilization rate of the model. To improve the generalization ability and classification accuracy of the model, multi-source and multi-resolution HSIs are employed to construct the precollected data set, and the partial channel connection operation is introduced to further improve the search efficiency. Results: Three HSIs inculding Houston, Chikusei and Pavia Center are selected as source data. Four HSIs including University of Pavia, Indian Pines, Salinas and Houston 2018 are selected as the target HSIs. The source data is used to determine the network model and the target HSIs are used to evaluate the classification performance. Experimental results show that the proposed method can automatically search out a universal deep network model suitable for HSI classification task, and this model can achieve better classification performance than the conventional deep learning models. The overall classification accuracies of University of Pavia, Indian Pines, Salinas and Houston 2018 are 98.15%, 98.74%, 97.30% and 74.47%, respectively. Conclusions: Compared with other deep models, the model in this paper is automatically determined by differentiable architecture search algorithm according to the data characteristics of HSIs, effectively improving the classification accuracy of target HSIs while avoiding complex network design. Extensive experiments verify the applicability and effectiveness of the proposed method in HSI classification. [ABSTRACT FROM AUTHOR]

Deep learning technology has made significant progress in agricultural image recognition tasks, but the parameter adjustment of deep models usually requires a lot of manual intervention, which is time-consuming and inefficient. To solve this challenge, this paper proposes an adaptive parameter tuning strategy that combines sine–cosine algorithm with Tent chaotic mapping to enhance sea horse optimization, which improves the search ability and convergence stability of standard sea horse optimization algorithm (SHO). Through adaptive optimization, this paper determines the best parameter configuration in ResNet-50 neural network and optimizes the model performance. The improved ESHO algorithm shows superior optimization effects than other algorithms in various performance indicators. The improved model achieves 96.7% accuracy in the corn disease image recognition task, and 96.4% accuracy in the jade fungus image recognition task. These results show that ESHO can not only effectively improve the accuracy of agricultural image recognition, but also reduce the need for manual parameter adjustment. [ABSTRACT FROM AUTHOR]

Classification with a few samples of training set has been a long-standing issue in the field of polarimetric synthetic aperture radar (PolSAR) image analysis and processing. In fact, one of the most important challenges of the PolSAR image classification task is the number of labelled samples. In essence, training network by utilizing just few numbers of training samples cannot be led to the reliable result since the neural network is so sensitive to the number of training samples. In addition, in the real PolSAR image classification task, the huge number of training samples is not accessible. On the other hand, classification performance using just small number of labelled samples is not an accurate result. So, in this paper, aiming at the small number of training samples of the PolSAR image classification task, a novel ensemble self-supervised feature-learning (ESSFL) model is designed. The designed ESSFL can automatically extract PolSAR features conducive to PolSAR image classification with a small number of training samples. In addition, it can significantly decrease the dependence of neural network algorithms on large labelled samples of training set. First, to utilize the spatial–polarimetric features of PolSAR data perfectly, the EfficientNet-B0 is presented and utilized as the main section of the deep learning (DL) model to extract DL features of PolSAR data. Then, using an optimization function that constrains the cross-correlation matrix of various distortions of each sample to the identity matrix, the designed deep learning model can obtain the effective features of homogeneous samples gathering together and heterogeneous samples separating from each other in a self-supervised manner. Moreover, two ensemble learning (EL) models, feature-level and view-level ensemble, are proposed to increase the feature extraction capability and classification result by jointly using spatial features at different scales and polarimetric information at different bands. Finally, the stack of the obtained features and the main polarimetric information of PolSAR data can be input into classifier for the classification of PolSAR data. In this paper we use two different classifiers, the first one is the Random Forest (RF) classifier and the second one is the Support Vector Machine (SVM) classifier. The advantages of SVM include that it can be used to avoid the difficulties of using linear functions in the high-dimensional feature space, and the optimization problem is transformed into dual convex quadratic programs that are so usable for an actual image classification task. In addition, among all the available classification methods, random forests provide one of the highest accuracies. The random forest technique can also handle big data with numerous variables running into thousands. In fact, it can automatically balance data sets when a class is more infrequent than other classes in the data that is so important for a real PolSAR image classification task. Experimental results on three well-known PolSAR data sets illustrate that the designed ESSFL can extract more discriminant features using the designed deep learning model. In the end, the experiments prove that the designed ESSFL has a significant classification performance compared with different deep learning models in the case of small number of training samples, and also it can achieve a better result in the case of large number of training samples by comparison with the most of the deep learning models. [ABSTRACT FROM AUTHOR]

The histopathological image classification method, based on deep learning, can be used to assist pathologists in cancer recognition in colon histopathology. The popularization of automatic and accurate histopathological image classification methods in this way is of great significance. However, smaller medical institutions with limited medical resources may lack colon histopathology image training sets with reliable labeled information; thus they may be unable to meet the needs of deep learning for many labeled training samples. Therefore, in this paper, the colon histopathological image set with rich label information from a certain medical institution is taken as the source domain; the colon histopathological image set from a smaller medical institution with limited medical resources is taken as the target domain. Considering the potential differences between histopathological images obtained by different institutions, this paper proposes a classification learning framework, namely unsupervised domain adaptation with local structure preservation for colon histopathological image classification, which can learn an adaptive classifier by performing distribution alignment and preserving intra-domain local structure to predict the labels of the colon histopathological images from institutions with lower medical resources. Extensive experiments demonstrate that the proposed framework shows significant improvement in accuracy and specificity of colon histopathological images without reliable labeled information compared to models without unsupervised domain adaptation. Specifically, in an affiliated hospital in Fuyang City, Anhui Province, the classification accuracy of benign and malignant colon histopathological images reaches 96.21%. The results of comparative experiments also show promising classification performance of our method in comparison with other unsupervised domain adaptation methods. [ABSTRACT FROM AUTHOR]

Large deep learning models are impressive, but they struggle when real-time data is not available. Few-shot class-incremental learning (FSCIL) poses a significant challenge for deep neural networks to learn new tasks from just a few labeled samples without forgetting the previously learned ones. This setup can easily leads to catastrophic forgetting and overfitting problems, severely affecting model performance. Studying FSCIL helps overcome deep learning model limitations on data volume and acquisition time, while improving practicality and adaptability of machine learning models. This paper provides a comprehensive survey on FSCIL. Unlike previous surveys, we aim to synthesize few-shot learning and incremental learning, focusing on introducing FSCIL from two perspectives, while reviewing over 30 theoretical research studies and more than 20 applied research studies. From the theoretical perspective, we provide a novel categorization approach that divides the field into five subcategories, including traditional machine learning methods, meta learning-based methods, feature and feature space-based methods, replay-based methods, and dynamic network structure-based methods. We also evaluate the performance of recent theoretical research on benchmark datasets of FSCIL. From the application perspective, FSCIL has achieved impressive achievements in various fields of computer vision such as image classification, object detection, and image segmentation, as well as in natural language processing and graph. We summarize the important applications. Finally, we point out potential future research directions, including applications, problem setups, and theory development. Overall, this paper offers a comprehensive analysis of the latest advances in FSCIL from a methodological, performance, and application perspective. • In-depth survey on FSCIL methods, applications, performance. • Categorizing FSCIL into five subcategories for clear analysis. • Evaluating FSCIL research on benchmarks for strengths, weaknesses. • FSCIL applications in computer vision, NLP, and graph. [ABSTRACT FROM AUTHOR]

The document is an editorial from the Intelligent Data Analysis (IDA) Journal, introducing the third issue of volume 28. The issue includes papers on various theoretical and applied topics in the field of Intelligent Data Analysis. The first half of the issue focuses on theoretical contributions, including papers on multilabel image classification and extended image data analysis. The second half covers papers on adaptive and active learning, data extraction and transformation, recommendation systems, medical applications, code search, and generating magazine layouts. The issue concludes with a paper on a new training dataset for Named Entity Recognition tasks. [Extracted from the article]

For two‐dimensional (2D) layered material heterojunctions, dynamic modulation of band alignments allows for the design of devices with flexible multi‐functional applications. In this paper, a device structure is presented based on a MoTe2/SnSe2 field‐effect transistor. By applying a bias voltage to the electrostatic gate, the gate voltage is adjusted from negative to positive, causing the heterojunction to transition from type‐III band alignment to type‐II band alignment. The working mechanism and device performance of the heterojunctions with different band alignments are investigated. The device exhibited outstanding detection range (from ultraviolet to infrared), detectivity (2.42 × 109 Jones), and speed (1.3 ms). Under a positive gate voltage, a higher ratio of light current to dark current (2×103) is achieved. To further demonstrate the potential of the high‐performance devices, their reliability is confirmed through their performance in image recognition using deep learning. [ABSTRACT FROM AUTHOR]

Environmental applications play a significant role in the ongoing research area of Polarimetric Synthetic Aperture Radar (PolSAR) image classification. In this paper, a new model is proposed for classifying PolSAR images and applied to a part of the Greater Cairo area in the Nile basin, South of Delta, Egypt. First, the proposed model performs data pre-processing by extracting the coherency and covariance elements noted as [T] and [C] matrices, respectively. Second, temporal convolutional networks (TCN) deep learning is used to extract the features from coherency and covariance elements and then train the model. Third, the SoftMax classifier is used to classify the PolSAR image. Finally, the proposed model is tested with evaluation metrics. The obtained results show that the proposed model can achieve high classification performances. [ABSTRACT FROM AUTHOR]

The objectives of this paper are to investigate the trade-offs between a physically constrained neural network and a deep, convolutional neural network and to design a combined ML approach ("VarioCNN"). Our solution is provided in the framework of a cyberinfrastructure that includes a newly designed ML software, GEOCLASS-image (v1.0), modern high-resolution satellite image data sets (Maxar WorldView data), and instructions/descriptions that may facilitate solving similar spatial classification problems. Combining the advantages of the physically-driven connectionist-geostatistical classification method with those of an efficient CNN, VarioCNN provides a means for rapid and efficient extraction of complex geophysical information from submeter resolution satellite imagery. A retraining loop overcomes the difficulties of creating a labeled training data set. Computational analyses and developments are centered on a specific, but generalizable, geophysical problem: The classification of crevasse types that form during the surge of a glacier system. A surge is a glacial catastrophe, an acceleration of a glacier to typically 100–200 times its normal velocity. GEOCLASS-image is applied to study the current (2016-2024) surge in the Negribreen Glacier System, Svalbard. The geophysical result is a description of the structural evolution and expansion of the surge, based on crevasse types that capture ice deformation in six simplified classes. [ABSTRACT FROM AUTHOR]

This paper aims to enhance the existing Automatic Meter Reading (AMR) technologies for utilities in the public services sector, such as water, electricity, and gas, by allowing users to regularly upload images of their meters, which are then automatically processed by machines for digit recognition. We propose an end-to-end AMR approach designed explicitly for unconstrained environments, offering practical solutions to common failures encountered during the automatic recognition process, such as image blur, perspective distortion, partial reflection, poor lighting, missing digits, and intermediate digit states, to reduce the failure rate of automatic meter readings. The system's first stage involves checking the quality of the user-uploaded images through the SVM method and requesting re-uploads for images unsuitable for digit extraction and recognition. The second stage employs deep learning models for digit localization and recognition, automatically detecting and correcting issues such as missing and intermediate digits to enhance the accuracy of automatic meter readings. Our research established a gas meter training dataset comprising 52,000 images, extensively annotated across various degrees, to train the deep learning models for high-precision digit recognition. Experimental results demonstrate that, with the simple SVM model, an accuracy of 87.03% is achieved for the classification of blurry image types. In addition, meter digit recognition (including intermediate digit states) can reach 97.6% (mAP), and the detection and correction of missing digits can be as high as 63.64%, showcasing the practical application value of the system developed in this study. [ABSTRACT FROM AUTHOR]

Land-cover classification based on remote sensing technology has been adopted for decision-making concerning agricultural development, urban planning, and ecosystem protection in arid oasis zones. The semantic segmentation method based on deep learning, as a new paradigm, can effectively overcome the limitations of traditional pixel-based and object-based methods and obtain good classification results from high spatial resolution (HSR) remote sensing images. However, how to extract the exact category boundary and realize the high precision mapping is still a problem. This paper proposes a novel high-resolution feature pyramid attention network (HRFPANet) for land-cover classification. It effectively integrates the advantages of multi-scale feature extraction, attention mechanism, and feature fusion and alleviates boundary inconsistency, roughness, and category fragmentation associated with previous semantic segmentation models. The experimental results show that the mIoU score of HRFPANet is 79.5%, which is 11.5% and 2.6% higher than that of PSPNet and UPerNet, respectively. It proves the proposed model can be used for qualified land-cover mapping in arid oasis zones. Our source code is available at . [ABSTRACT FROM AUTHOR]

This paper proposes a new method for recognizing, extracting, and processing Phase-Resolved Partial Discharge (PRPD) patterns from two-dimensional plots to identify specific defect types affecting electrical equipment without human intervention while retaining the principals that make PRPD analysis an effective diagnostic technique. The proposed method does not rely on training complex deep learning algorithms which demand substantial computational resources and extensive datasets that can pose significant hurdles for the application of on-line partial discharge monitoring. Instead, the developed Cosine Cluster Net (CCNet) model, which is an image processing pipeline, can extract and process patterns from any two-dimensional PRPD plot before employing the cosine similarity function to measure the likeness of the patterns to predefined templates of known defect types. The PRPD pattern recognition capabilities of the model were tested using several manually classified PRPD images available in the existing literature. The model consistently produced similarity scores that identified the same defect type as the one from the manual classification. The successful defect type reporting from the initial trials of the CCNet model together with the speed of the identification, which typically does not exceed four seconds, indicates potential for real-time applications. [ABSTRACT FROM AUTHOR]

Image watermarking often involves the use of handheld devices under non-structured conditions for authentication purposes, particularly in the print-cam process where smartphone cameras are used to capture watermarked printed images. However, these images frequently suffer from perspective distortions, making them unsuitable for automated information detection. To address this issue, Cam-Unet, an end-to-end neural network architecture, is presented to predict the mapping from distorted images to rectified ones, specifically tailored for print-cam challenges applied to ID images. Given the limited availability of large-scale real datasets containing ground truth distortions, we created an extensive synthetic dataset by subjecting undistorted images to print-cam attacks. The proposed network is trained on this dataset, using various data augmentation techniques to improve its generalization capabilities. Accordingly, this paper presents an image watermarking system for the print-cam process. The approach combines Fourier transform-based watermarking with Cam-Unet as perspective distortion correction. Results show that the proposed method outperforms existing watermarking approaches typically employed to counter print-cam attacks and achieves an optimal balance between efficiency and cost-effectiveness. [ABSTRACT FROM AUTHOR]

In recent years, the growth spurt of medical imaging data has led to the development of various machine learning algorithms for various healthcare applications. The MedMNISTv2 dataset, a comprehensive benchmark for 2D biomedical image classification, encompasses diverse medical imaging modalities such as Fundus Camera, Breast Ultrasound, Colon Pathology, Blood Cell Microscope etc. Highly accurate classifications performed on these datasets is crucial for identification of various diseases and determining the course of treatment. This research paper presents a comprehensive analysis of four subsets within the MedMNISTv2 dataset: BloodMNIST, BreastMNIST, PathMNIST and RetinaMNIST. Each of these selected datasets is of diverse data modalities and comes with various sample sizes, and have been selected to analyze the efficiency of the model against diverse data modalities. The study explores the idea of assessing the Vision Transformer Model's ability to capture intricate patterns and features crucial for these medical image classification and thereby transcend the benchmark metrics substantially. The methodology includes pre-processing the input images which is followed by training the ViT-base-patch16-224 model on the mentioned datasets. The performance of the model is assessed using key metrices and by comparing the classification accuracies achieved with the benchmark accuracies. With the assistance of ViT, the new benchmarks achieved for BloodMNIST, BreastMNIST, PathMNIST and RetinaMNIST are 97.90%, 90.38%, 94.62% and 57%, respectively. The study highlights the promise of Vision transformer models in medical image analysis, preparing the way for their adoption and further exploration in healthcare applications, aiming to enhance diagnostic accuracy and assist medical professionals in clinical decision-making. [ABSTRACT FROM AUTHOR]

The identification research of hydrogenation catalyst information has always been one of the most important businesses in the chemical industry. In order to aid researchers in efficiently screening high-performance catalyst carriers and tackle the pressing challenge at hand, it is imperative to find a solution for the intelligent recognition of hydrogenation catalyst images. To address the issue of low recognition accuracy caused by adhesion and stacking of hydrogenation catalysts, An image recognition algorithm of hydrogenation catalyst based on FPNC Net was proposed in this paper. In the present study, Resnet50 backbone network was used to extract the features, and spatially-separable convolution kernel was used to extract the multi-scale features of catalyst fringe. In addition, to effectively segment the adhesive regions of stripes, FPN (Feature Pyramid Network) is added to the backbone network for deep and shallow feature fusion. Introducing an attention module to adaptively adjust weights can effectively highlight the target features of the catalyst. The experimental results showed that the FPNC Net model achieved an accuracy of 94.2% and an AP value improvement of 19.37% compared to the original CenterNet model. The improved model demonstrates a significant enhancement in detection accuracy, indicating a high capability for detecting hydrogenation catalyst targets [ABSTRACT FROM AUTHOR]

Polarimetric features extracted from polarimetric synthetic aperture radar (PolSAR) images contain abundant back-scattering information about objects. Utilizing this information for PolSAR image classification can improve accuracy and enhance object monitoring. In this paper, a deep learning classification method based on polarimetric channel power features for PolSAR is proposed. The distinctive characteristic of this method is that the polarimetric features input into the deep learning network are the power values of polarimetric channels and contain complete polarimetric information. The other two input data schemes are designed to compare the proposed method. The neural network can utilize the extracted polarimetric features to classify images, and the classification accuracy analysis is employed to compare the strengths and weaknesses of the power-based scheme. It is worth mentioning that the polarized characteristics of the data input scheme mentioned in this article have been derived through rigorous mathematical deduction, and each polarimetric feature has a clear physical meaning. By testing different data input schemes on the Gaofen-3 (GF-3) PolSAR image, the experimental results show that the method proposed in this article outperforms existing methods and can improve the accuracy of classification to a certain extent, validating the effectiveness of this method in large-scale area classification. [ABSTRACT FROM AUTHOR]

A neural network model needs to be manually designed for ancient mural dynasty recognition, and this paper proposes an ancient mural dynasty recognition algorithm that is based on a neural architecture search (NAS). First, the structural edge information of mural images is extracted for use by the neural network model in recognizing mural missions. Second, an NAS algorithm that is based on contrast selection (CS) simplifies the architecture search to an incremental CS and then searches for the optimal network architecture on the mural dataset. Finally, the identified optimal network architecture is used for training and testing to complete the mural dynasty recognition task. The results show that the top accuracy of the proposed method on the mural dataset is 88.10%, the recall rate is 87.52%, and the precision rate is 87.69%. Each evaluation index used by the neural network model is superior to that of classical network models such as AlexNet and ResNet-50. Compared with NAS methods such as ASNG and MIGO, the accuracy of mural dynasty recognition is higher by an average of 4.27% when using the proposed method. The proposed method is verified on CIFAR-10, CIFAR-100, ImageNet16-120 and other datasets and achieves a good recognition accuracy in the NAS-bench-201 search space, which averages 93.26%, 70.73% and 45.34%, respectively, on the abovementioned datasets. [ABSTRACT FROM AUTHOR]

The development of medical image classification models necessitates a substantial number of labeled images for model training. In real-world scenarios, sample sizes are typically limited and labeled samples often constitute only a small portion of the dataset. This paper aims to investigate a collaborative similarity learning strategy that optimizes pseudo-labels to enhance model accuracy and expedite its convergence, known as the joint similarity learning framework. By integrating semantic similarity and instance similarity, the pseudo-labels are mutually refined to ensure their quality during initial training. Furthermore, the similarity score is utilized as a weight to guide samples away from misclassification predictions during the classification process. To enhance the model's generalization ability, an adaptive consistency constraint is introduced into the loss function to improve performance on untrained datasets. The model achieved a satisfactory accuracy of 93.65% at 80% labeling ratio, comparable to supervised learning methods' performance. Even with very low labeling ratio (e.g., 5%), the model still attained an accuracy of 74.28%. Comparison with other techniques such as Mean Teacher and FixMatch revealed that our approach significantly outperforms them in medical image classification tasks through improving accuracy by approximately 2%, demonstrating this framework's leadership in medical image classification. [ABSTRACT FROM AUTHOR]

Deep neural network has made surprising achievements in natural language processing, image pattern classification recognition, and other domains in the last few years. It is still tough to apply to hardware-constrained or mobile equipment because of the huge number of parameters, high storage as well as computing costs. In this paper, a new sparse iteration neural network architecture is proposed. First, the pruning method is used to compress the model size and make the network sparse. Then the architecture is iterated on the sparse network model, and the network performance is improved without adding additional parameters. Finally, the hybrid deep learning model was carried out on CV tasks and NLP tasks on ANN, CNN, and Transformer. Compared with the sparse network architecture, we finally found that the accuracy of the MINST, CIFAR10, PASCAL VOC 2012, and SQuAD datasets is improved by 0.47%, 0.64%, 3.75%, and 15.06%, respectively. [ABSTRACT FROM AUTHOR]

In recent years, deep learning methods have achieved remarkable success in hyperspectral image classification (HSIC), and the utilization of convolutional neural networks (CNNs) has proven to be highly effective. However, there are still several critical issues that need to be addressed in the HSIC task, such as the lack of labeled training samples, which constrains the classification accuracy and generalization ability of CNNs. To address this problem, a deep multi-scale attention fusion network (DMAF-NET) is proposed in this paper. This network is based on multi-scale features and fully exploits the deep features of samples from multiple levels and different perspectives with an aim to enhance HSIC results using limited samples. The innovation of this article is mainly reflected in three aspects: Firstly, a novel baseline network for multi-scale feature extraction is designed with a pyramid structure and densely connected 3D octave convolutional network enabling the extraction of deep-level information from features at different granularities. Secondly, a multi-scale spatial–spectral attention module and a pyramidal multi-scale channel attention module are designed, respectively. This allows modeling of the comprehensive dependencies of coordinates and directions, local and global, in four dimensions. Finally, a multi-attention fusion module is designed to effectively combine feature mappings extracted from multiple branches. Extensive experiments on four popular datasets demonstrate that the proposed method can achieve high classification accuracy even with fewer labeled samples. [ABSTRACT FROM AUTHOR]

Cloud and snow in remote sensing images typically block the underlying surface information and interfere with the extraction of available information, so detecting cloud and snow becomes a critical problem in remotely sensed image processing. The current methods for detecting clouds and snow are susceptible to interference from complex background, making it difficult to recover cloud edge details and causing missing and false detection phenomena. To address these issues, a cross-dimensional feature attention aggregation network is suggested to realize the segmentation of clouds and snow. To address the problem of interference induced by the similar spectral characteristics of clouds and snow, the context attention aggregation module is added to conflate feature maps of various dimensions and screen the information. Multi-scale strip convolution module (MSSCM) and its improved version MSSCMs are used to extract edge characteristics at different scales and improve the harsh segmentation border. Also, adding deep feature semantic information extraction module to deep features to guide the classification of the model to avoid the interference of complex background. Finally, a 'los beatles' module is used to replace the traditional linear combination in the decoding stage, and the feature information of different granularity is fused and extracted to enhance the model's detection efficiency. In this paper, experiments are carried out on the public datasets: CSWV, HRC _ WHU and L8 _ SPARCS. The MIOU scores on the three datasets are 89.507 % , 91.674 % and 80.722 % , respectively. Comparative experiment findings demonstrate that the network presented in this article can attain the highest detection accuracy and good detection efficiency with low parameters. [ABSTRACT FROM AUTHOR]

Learning in neural networks with locally-tuned neuron models such as radial Basis Function (RBF) networks is often seen as instable, in particular when multi-layered architectures are used. Furthermore, universal approximation theorems for single-layered RBF networks are very well established; therefore, deeper architectures are theoretically not required. Consequently, RBFs are mostly used in a single-layered manner. However, deep neural networks have proven their effectiveness on many different tasks. In this paper, we show that deeper RBF architectures with multiple radial basis function layers can be designed together with efficient learning schemes. We introduce an initialization scheme for deep RBF networks based on k-means clustering and covariance estimation. We further show how to make use of convolutions to speed up the calculation of the Mahalanobis distance in a partially connected way, which is similar to the convolutional neural networks (CNNs). Finally, we evaluate our approach on image classification as well as speech emotion recognition tasks. Our results show that deep RBF networks perform very well, with comparable results to other deep neural network types, such as CNNs. [ABSTRACT FROM AUTHOR]

Hyperspectral image classification (HSIC) has garnered increasing attention among researchers. While classical networks like convolution neural networks (CNNs) have achieved satisfactory results with the advent of deep learning, they are confined to processing local information. Vision transformers, despite being effective at establishing long-distance dependencies, face challenges in extracting high-representation features for high-dimensional images. In this paper, we present the multiscale efficient attention with enhanced feature transformer (MEA-EFFormer), which is designed for the efficient extraction of spectral–spatial features, leading to effective classification. MEA-EFFormer employs a multiscale efficient attention feature extraction module to initially extract 3D convolution features and applies effective channel attention to refine spectral information. Following this, 2D convolution features are extracted and integrated with local binary pattern (LBP) spatial information to augment their representation. Then, the processed features are fed into a spectral–spatial enhancement attention (SSEA) module that facilitates interactive enhancement of spectral–spatial information across the three dimensions. Finally, these features undergo classification through a transformer encoder. We evaluate MEA-EFFormer against several state-of-the-art methods on three datasets and demonstrate its outstanding HSIC performance. [ABSTRACT FROM AUTHOR]

Breast cancer is one of the cancers with high morbidity and mortality in the world, which is a serious threat to the health of women. With the development of deep learning, the recognition about computer-aided diagnosis technology is getting higher and higher. And the traditional data feature extraction technology has been gradually replaced by the feature extraction technology based on convolutional neural network which helps to realize the automatic recognition and classification of pathological images. In this paper, a novel method based on deep learning and wavelet transform is proposed to classify the pathological images of breast cancer. Firstly, the image flip technique is used to expand the data set, then the two-level wavelet decomposition and reconfiguration technology is used to sharpen and enhance the pathological images. Secondly, the processed data set is divided into the training set and the test set according to 8:2 and 7:3, and the YOLOv8 network model is selected to perform the eight classification tasks of breast cancer pathological images. Finally, the classification accuracy of the proposed method is compared with the classification accuracy obtained by YOLOv8 for the original BreaKHis dataset, and it is found that the algorithm can improve the classification accuracy of images with different magnifications, which proves the effectiveness of combining two-level wavelet decomposition and reconfiguration with YOLOv8 network model. [ABSTRACT FROM AUTHOR]

Agricultural dykelands in Nova Scotia rely heavily on a surface drainage technique called land forming, which is used to alter the topography of fields to improve drainage. The presence of land-formed fields provides useful information to better understand land utilization on these lands vulnerable to rising sea levels. Current field boundaries delineation and classification methods, such as manual digitalization and traditional segmentation techniques, are labour-intensive and often require manual and time-consuming parameter selection. In recent years, deep learning (DL) techniques, including convolutional neural networks and Mask R-CNN, have shown promising results in object recognition, image classification, and segmentation tasks. However, there is a gap in applying these techniques to detecting surface drainage patterns on agricultural fields. This paper develops and tests a Mask R-CNN model for detecting land-formed fields on agricultural dykelands using LiDAR-derived elevation data. Specifically, our approach focuses on identifying groups of pixels as cohesive objects within the imagery, a method that represents a significant advancement over pixel-by-pixel classification techniques. The DL model developed in this study demonstrated a strong overall performance, with a mean Average Precision (mAP) of 0.89 across Intersection over Union (IoU) thresholds from 0.5 to 0.95, indicating its effectiveness in detecting land-formed fields. Results also revealed that 53% of Nova Scotia's dykelands are being used for agricultural purposes and approximately 75% (6924 hectares) of these fields were land-formed. By applying deep learning techniques to LiDAR-derived elevation data, this study offers novel insights into surface drainage mapping, enhancing the capability for precise and efficient agricultural land management in regions vulnerable to environmental changes. [ABSTRACT FROM AUTHOR]

Multiple kernel learning (MKL) is an efficient way to improve hyperspectral image classification with few training samples by integrating spectral and spatial features. Nonetheless, presenting a MKL algorithm to capture significant information and eliminate redundancy is indispensable. In this paper, a novel graph-based MKL method (GMKL) is introduced to enhance the performance of the MKL algorithm and achieve high discriminative features, resulting in maximum separability. To do so, an informative low-rank graph is determined via a new algorithm called block diagonal representation that takes into account both local and global data structures. Then, in lieu of a time-consuming search for finding the optimal combination of the basic kernels, unlike the prior works, an optimal projective direction is acquired using graph theory. Various basic kernels generated from superpixels and attribute profiles are projected to produce a discriminative combination of kernels. Eventually, the optimum kernel is given to the support vector machine classifier. Experiments have been conducted on two popular hyperspectral datasets (Indian Pines and Pavia University). For the datasets, the proposed algorithm gained 90.43% and 99.28% overall accuracy, respectively, using only 1% training samples, which is much better than the state-of-the-art MKL algorithm and deep learning-based approaches with lower computational time. [ABSTRACT FROM AUTHOR]

Breast cancer detection is considered a challenging task for the average experienced radiologist due to the variation of the lesions' size and shape, especially with the existence of high fibro-glandular tissues. The revolution of deep learning and computer vision contributes recently in introducing systems that can provide an automated diagnosis for breast cancer that can act as a second opinion for doctors/radiologists. The most of previously proposed deep learning-based Computer-Aided Diagnosis (CAD) systems mainly utilized Convolutional Neural Networks (CNN) that focuses on local features. Recently, vision transformers (ViT) have shown great potential in image classification tasks due to its ability in learning the local and global spatial features. This paper proposes a fully automated CAD framework based on YOLOv4 network and ViT transformers for mass detection and classification of Contrast Enhanced Spectral Mammography (CESM) images. CESM is an evolution type of Full Field Digital Mammography (FFDM) images that provides enhanced visualization for breast tissues. Different experiments were conducted to evaluate the proposed framework on two different datasets that are INbreast and CDD-CESM that provides both FFDM and CESM images. The model achieved at mass detection a mean Average Precision (mAP) score of 98.69%, 81.52%, and 71.65% and mass classification accuracy of 95.65%, 97.61%, and 80% for INbreast, CE-CESM, and DM-CESM, respectively. The proposed framework showed competitive results regarding the state-of-the-art models in INbreast. It outperformed the previous work in the literature in terms of the F1-score by almost 5% for mass detection in CESM. Moreover, the experiments showed that the CESM could provide more morphological features that can be more informative, especially with the highly dense breast tissues. [ABSTRACT FROM AUTHOR]

Simple Summary: In response to the quantitative demand for the practical deployment of applications and the need to enhance the detection accuracy of wildlife in complex field environments in southwest China and neighboring regions, we have refined the wildlife detection algorithm based on re-parameterized convolution. This refinement is specifically targeted at addressing the challenges posed by the low quality of wildlife images captured by camera traps and the limitations of traditional object detection algorithms in feature extraction capability. To address these issues, we have introduced a series of improvement schemes. As a result of these enhancements, there has been a noteworthy improvement in both the accuracy of wildlife detection and the speed of model inference. This advancement offers a convenient and efficient method for the preliminary detection in the context of automated wildlife monitoring. To autonomously detect wildlife images captured by camera traps on a platform with limited resources and address challenges such as filtering out photos without optimal objects, as well as classifying and localizing species in photos with objects, we introduce a specialized wildlife object detector tailored for camera traps. This detector is developed using a dataset acquired by the Saola Working Group (SWG) through camera traps deployed in Vietnam and Laos. Utilizing the YOLOv6-N object detection algorithm as its foundation, the detector is enhanced by a tailored optimizer for improved model performance. We deliberately introduce asymmetric convolutional branches to enhance the feature characterization capability of the Backbone network. Additionally, we streamline the Neck and use CIoU loss to improve detection performance. For quantitative deployment, we refine the RepOptimizer to train a pure VGG-style network. Experimental results demonstrate that our proposed method empowers the model to achieve an 88.3% detection accuracy on the wildlife dataset in this paper. This accuracy is 3.1% higher than YOLOv6-N, and surpasses YOLOv7-T and YOLOv8-N by 5.5% and 2.8%, respectively. The model consistently maintains its detection performance even after quantization to the INT8 precision, achieving an inference speed of only 6.15 ms for a single image on the NVIDIA Jetson Xavier NX device. The improvements we introduce excel in tasks related to wildlife image recognition and object localization captured by camera traps, providing practical solutions to enhance wildlife monitoring and facilitate efficient data acquisition. Our current work represents a significant stride toward a fully automated animal observation system in real-time in-field applications. [ABSTRACT FROM AUTHOR]

This paper proposes a forward layer-wise learning algorithm for CNNs in classification problems. The algorithm utilizes the Separation Index (SI) as a supervised complexity measure to evaluate and train each layer in a forward manner. The proposed method explains that gradually increasing the SI through layers reduces the input data's uncertainties and disturbances, achieving a better feature space representation. Hence, by approximating the SI with a variant of local triplet loss at each layer, a gradient-based learning algorithm is suggested to maximize it. Inspired by the NGRAD (Neural Gradient Representation by Activity Differences) hypothesis, the proposed algorithm operates in a forward manner without explicit error information from the last layer. The algorithm's performance is evaluated on image classification tasks using VGG16, VGG19, AlexNet, and LeNet architectures with CIFAR-10, CIFAR-100, Raabin-WBC, and Fashion-MNIST datasets. Additionally, the experiments are applied to text classification tasks using the DBPedia and AG's News datasets. The results demonstrate that the proposed layer-wise learning algorithm outperforms state-of-the-art methods in accuracy and time complexity. [ABSTRACT FROM AUTHOR]

Deep learning is a powerful technique to boost application performance in various fields, including face recognition, image classification, natural language understanding, and recommendation system. With the rapid increase in the computing power of mobile devices, developers can embed deep learning models into their apps for building more competitive products with more accurate and faster responses. Although there are several works of adversarial attacks against deep learning models in apps, they all need information about the models' internals (i.e., structures and weights) or need to modify the models. In this paper, we propose an effective black‐box approach by training substitute models to spoof the deep learning systems inside the apps. We evaluate our approach on 10 real‐world deep‐learning apps from Google Play to perform black‐box adversarial attacks. Through the study, we find three factors that can affect the performance of attacks. Our approach can reach a relatively high attack success rate of 66.60% on average. Compared with other adversarial attacks on mobile deep learning models, in terms of the average attack success rates, our approach outperforms its counterparts by 27.63%. [ABSTRACT FROM AUTHOR]

Parkinson disease (PD) is a type of neurodegenerative disorder that affects the motor movement of the patient. But each technique has its own advantages or disadvantages. In gene, speech and handwriting data model, the feature extraction and reduction is an important step for efficient classification. These two steps require proper attention for selection and also require high processing time as compared to other data model like images. Because in image modality, the deep learning algorithm can be applied that can perform all process and automate the classification. As compared to these domains, the signal produces better and best results. Because the electroencephalogram (EEG) signal are taken from the brain using electrodes and it helps to observe the brain signals effectively and immediately as compared to the other data modals. Hence, in this paper, the wavelet transform will be used to decompose the signals and statistical features will be extracted from the transformed signal. Here, the satin bower bird optimization will be used for both type of wavelet selection and feature reduction process for final classification. The reduced feature set will be classified using Ensemble Neural Network type including InceptionV3, DenseNet, MobileNet, Xception, and NasNet) recently proposed for medical image classification. The whole process will be realized using MATLAB R2021a software and its performance will be evaluated in terms of Accuracy and is compared against Automated Tunable Q-wavelet transform performance. The proposed ensemble method, employing EEG signal processing and neural networks, achieved a 97% success rate in discriminating PD datasets, surpassing Convolutional Neural Network (CNN) and Machine Learning (ML) classifications (88% –92%). Utilizing MATLAB R2021a, its superiority over Q-wavelet transform was evident, signifying improved PD dataset discrimination. [ABSTRACT FROM AUTHOR]

Deep learning methods have shown significant potential in computer vision applications within the field of civil engineering. One important area of research is the detection and classification of construction machines using convolutional neural networks (CNNs). However, a major challenge in adopting machine learning models for vision systems lies in hyperparameter optimization. Therefore, the objective of this paper is to propose a meticulous statistical approach to hyperparameter tuning for deep learning in the visual classification of construction machines. It is noteworthy that the method for selecting hyperparameters utilizes statistical concepts, including analysis of variance and the Tukey test. Moreover, three research questions were formulated to address statistical differences in performance when varying the CNN architecture, optimizer and configuration (architecture and optimizer). Two additional research questions were posed to determine the recommended hyperparameters for the task and assess performance differences for non-selected hyperparameters. For this purpose, 18 hyperparameter combinations were analyzed, encompassing three CNN architectures (DenseNet, Mobilenet and VGG16) and six optimizers (adadelta, adagrad, rmsprop, adam, adamax and sgd). The Alberta Construction Image Dataset, consisting of 2850 images of three construction machines (excavator, dump truck and concrete mixer truck), was used in the experiments. The results demonstrate that the hyperparameter configurations exhibit statistically significant differences in image classification. Furthermore, the recommended combination (DenseNet × adagrad) yielded the highest average accuracy results in both binary tests (90.0%) and multiclass classification (77.8%). [ABSTRACT FROM AUTHOR]

This paper introduces the Hybrid Quantum Deep Neural Network (HQDNN), a pioneering model that amalgamates classical Convolutional Neural Network (CNN) architecture with Parameterized Quantum Circuits (PmQC) for image classification tasks. The background elucidates the intricacies of CNN and PmQC operations, emphasizing their respective components and functions. HQDNN comprises quantum convolutional and pooling layers alongside a classical fully connected layer, demonstrating its effectiveness through simulations on the MNIST dataset. Results reveal HQDNN's superior learning ability and classification performance compared to traditional CNN, particularly with the inclusion of a pooling layer. This work contributes to the evolution of quantum machine learning, opening avenues for the practical integration of quantum computing in image classification workflows. Future research may explore scalability and diverse applications of HQDNN, marking a promising convergence of quantum computing and deep learning methodologies. [ABSTRACT FROM AUTHOR]

The precise classification of kidneys and kidney tumors is of great guiding significance for the diagnosis and surgical planning of clinical doctors. Artificial segmentation of kidneys and renal tumors not only increases the workload of doctors, but also has a certain degree of subjectivity. Due to the varying morphology and pathological characteristics of various kidney tumors, conventional imaging analysis methods have certain limitations in their application. This study aims to propose a novel optical imaging device based on deep learning to achieve better results in kidney tumor image segmentation and recognition. The study collected a batch of image data on kidney tumors and trained the data using deep learning algorithms. After many experiments and optimization, a model with excellent performance in renal tumor segmentation and recognition was obtained. Compared with several existing methods, the method proposed in this paper has better segmentation accuracy for kidney tumors. This can not only avoid mutual interference between the network and the resolution network, but also improve the recognition efficiency between the two networks. In addition, the system can also provide reliable data feedback for clinical doctors, helping them continuously correct problems in subsequent visual examinations, thereby achieving higher diagnostic accuracy. The experimental results show that the neural network is effective in image segmentation and quantitative analysis, and can be applied in clinical practice. [ABSTRACT FROM AUTHOR]

The accurate identification of bird image is of great significance to protecting the ecological environment and bird species diversity. To address the issue of low recognition accuracy arising from the similarities in features among different bird species and the susceptibility of shallow edge features to loss, this paper proposes a fine-grained bird image classification model that incorporates hierarchical feature fusion and counterfactual feature selection. The model is based on vision transformer and builds a hierarchical feature fusion module and a counterfactual feature enhancement module. The hierarchical feature fusion module superimposes shallow features rich in fine-grained information into deep features to improve the problem of lack of edge detail information in key features. The counterfactual feature enhancement module selects distinguishing features through counterfactual intervention to reduce classification errors caused by highly similar features of different species of birds. The experimental results show that the method can achieve 91.9 % and 91.4 % accuracy on two available datasets, CUB-200-2011 and NABirds, respectively, which is higher than the current mainstream fine-grained bird recognition algorithms and shows excellent classification performance. [ABSTRACT FROM AUTHOR]

This paper discusses the challenges of the hyperparameter tuning in deep learning models and proposes a green approach to the neural architecture search process that minimizes its environmental impact. The traditional approach of neural architecture search involves sweeping the entire space of possible architectures, which is computationally expensive and time-consuming. Recently, to address this issue, performance predictors have been proposed to estimate the performance of different architectures, thereby reducing the search space and speeding up the exploration process. The proposed approach aims to develop a performance predictor by training only a small percentage of the possible hyperparameter configurations. The suggested predictor can be queried to find the best configurations without training them on the dataset. Numerical examples of image denoising and classification enable us to evaluate the performance of the proposed approach in terms of performance and time complexity. [ABSTRACT FROM AUTHOR]

The recognition of steel microstructure images plays a crucial role in the metallographic analysis process. Although some progress has been made through the application of artificial intelligence algorithms, several challenges remain. First, existing algorithms exhibit weak nonlinear feature extraction capabilities and noticeable limitations. Second, they overlook the intrinsic noise and redundant interference present in microscopic images. To address these issues, this paper investigates the automatic recognition of metallographic tissues by leveraging residual structures in deep neural networks. An enhanced residual network model based on transfer learning is proposed, which utilizes the pre-trained weights from the ImageNet dataset to facilitate learning with small sample data. This network offers higher classification accuracy and higher F1 scores. In addition, a deep residual shrinkage network model based on an attention mechanism is proposed. This model incorporates an attention sub-network into the original residual module and employs a soft threshold function to eliminate redundant features, including noise. The proposed algorithms are evaluated against various convolutional neural networks using 20 types of metallographic test sets. The experimental results showed that both methods have high accuracy rates of 95% and 94.44%, respectively, and F1 scores of 0.9464 and 0.9419. While maintaining the complexity of the model, there has been a significant improvement in accuracy, and the models exhibit strong generalization capabilities. Our research contributes to enhancing production efficiency, strengthening quality control, and improving material performance through computer vision technology. [ABSTRACT FROM AUTHOR]

Ecological garden is one of the important links of urban ecosystem. Accurate and effective analysis of temporal and spatial changes of garden vegetation is conducive to the sustainable development of urban resources and environment. However, the elements of garden vegetation are close to each other, and the current analysis methods can't support the remote sensing image analysis in the ecological garden scene. To solve this problem, this paper introduces a new method for recognition and classification based on deep learning networks. Firstly, radiometric correction and geometric correction are used to preprocess the data collected by unmanned aerial vehicle (UAV); the ecological garden remote sensing image analysis network is constructed based on three-dimensional convolutional neural network (3D-CNN), which adopts double convolution pool structure, To address the issue of pixel misclassification due to similar features in landscape vegetation classification, a deep learning network can be used for high-dimensional feature extraction. On this basis, the multiple label conditional random field (MLCRF) algorithm can be applied for global optimization, effectively solving the misclassification problem caused by similar features of vegetation. This method not only improves classification accuracy but also is suitable for large-scale image classification tasks in complex scenes. The simulation results show that the overall recognition accuracy of the proposed 3D-CNN-MLCRF identification method is 97.83% and the Kappa coefficient is 97.18%, which can accurately and effectively realize the task of ecological garden scene classification. [ABSTRACT FROM AUTHOR]

Variable selection is one of the most important contents of machine learning research and a large number of variable selection methods have been proposed at present. From deep learning perspective, we propose a new variable selection method based on multi-layer artificial neural networks in this paper. The method propagates errors gained form a trained network model through the network from output layer to input layer and the values gained by input layer give information about the relative contribution. The variables with big contribution value are selected, which we call neural inverse propagation (NIP). Specifically, after training a neural network, the gained errors between the actual network outputs and the desired outputs go through the network from top to down and arrive at the input layer. The absolute values gained by input units provide information about the relative importance (contribution) of input units, which is that the larger the value is, the more important the corresponding input variable is. We remove k variables corresponding to the first k minimum absolute value gained by input units to the predictive variables, to perform variable selection. The efficiency of the proposed method is illustrated through its application to classification and regression tasks on a number of synthetic datasets and a real-world image classification dataset. [ABSTRACT FROM AUTHOR]

As a list of remotely sensed data sources is available, the effective processing of remote sensing images is of great significance in practical applications in various fields. This paper proposes a new lightweight network to solve the problem of remote sensing image processing by using the method of deep learning. Specifically, the proposed model employs ShuffleNet V2 as the backbone network, appropriately increases part of the convolution kernels to improve the classification accuracy of the network, and uses the maximum overlapping pooling layer to enhance the detailed features of the input images. Finally, Squeeze and Excitation (SE) blocks are introduced as the attention mechanism to improve the architecture of the network. Experimental results based on several multisource data show that our proposed network model has a good classification effect on the test samples and can achieve more excellent classification performance than some existing methods, with an accuracy of 91%, and can be used for the classification of remote sensing images. Our model not only has high accuracy but also has faster training speed compared with large networks and can greatly reduce computation costs. The demo code of our proposed method will be available at https://github.com/li-zi-qi. [ABSTRACT FROM AUTHOR]

Medical image classification is an essential task in the fields of computer-aided diagnosis and medical image analysis. In recent years, researchers have made extensive work on medical image classification by computer vision techniques. However, most of the current work is based on deep learning methods, which still suffer from expensive hardware resources, long time consuming and a lot of parameters to be optimized. In this paper, a multi-granularity ensemble algorithm for medical image classification based on broad learning system is proposed, which is an end-to-end lightweight model. On the one hand, the proposed method is designed to address the problem of weak image feature learning ability of broad learning system. The convolution module with fixed weights based on transfer learning is introduced as a feature extractor to extract fusion features of medical images. On the other hand, the multi-granularity ensemble framework is proposed, which learn the fusion features of medical images from fine-grained to coarse-grained respectively, and the prediction results at different granularity levels are integrated by ensemble learning. In this way, the bottom local features can be sufficiently considered, while the global features can also be taken into account. The experimental results show that on the MedMNIST dataset (containing 10 sub-datasets), the proposed method can shorten the training time by tens of times while having similar accuracy to deep convolutional neural networks. On the ChestXRay2017 dataset, the proposed method can achieve an accuracy of 92.5%, and the training time is also significantly better than other methods. [ABSTRACT FROM AUTHOR]