Search Results (422)

Neural networks have inductive biases owing to the assumptions associated with the selected learning algorithm, datasets, and network structure. Specifically, convolutional neural networks (CNNs) are known for their tendency to exhibit textural biases. This bias is closely related to image classification accuracy. Aligning the model’s bias with the dataset’s bias can significantly enhance performance in transfer learning, leading to more efficient learning. This study aims to quantitatively demonstrate that increasing shape bias within the network by varying kernel sizes and dilation rates improves accuracy on shape-dominant data and enables efficient learning with less data. Furthermore, we propose a novel method for quantitatively evaluating the balance between texture bias and shape bias. This method enables efficient learning by aligning the biases of the transfer learning dataset with those of the model. Systematically adjusting these biases allows CNNs to better fit data with specific biases. Compared to the original model, an accuracy improvement of up to 9.9% was observed. Our findings underscore the critical role of bias adjustment in CNN design, contributing to developing more efficient and effective image classification models. Full article

Due to the inherent limitations of a single viewpoint, reconstructing 3D human meshes from a single image has long been a challenging task. While deep learning networks enable us to approximate the shape of unseen sides, capturing the texture details of the non-visible side remains difficult with just one image. Traditional methods utilize Generative Adversarial Networks (GANs) to predict the normal maps of the non-visible side, thereby inferring detailed textures and wrinkles on the model’s surface. However, we have identified challenges with existing normal prediction networks when dealing with complex scenes, such as a lack of focus on local features and insufficient modeling of spatial relationships.To address these challenges, we introduce EMAR—Enhanced Multi-scale Attention-Driven Single-Image 3D Human Reconstruction. This approach incorporates a novel Enhanced Multi-Scale Attention (EMSA) mechanism, which excels at capturing intricate features and global relationships in complex scenes. EMSA surpasses traditional single-scale attention mechanisms by adaptively adjusting the weights between features, enabling the network to more effectively leverage information across various scales. Furthermore, we have improved the feature fusion method to better integrate representations from different scales. This enhanced feature fusion allows the network to more comprehensively understand both fine details and global structures within the image. Finally, we have designed a hybrid loss function tailored to the introduced attention mechanism and feature fusion method, optimizing the network’s training process and enhancing the quality of reconstruction results. Our network demonstrates significant improvements in performance for 3D human model reconstruction. Experimental results show that our method exhibits greater robustness to challenging poses compared to traditional single-scale approaches. Full article

Image super-resolution (SR) algorithms based on deep learning yield good visual performances on visible images. Due to the blurred edges and low contrast of infrared (IR) images, methods transferred directly from visible images to IR images have a poor performance and ignore the demands of downstream detection tasks. Therefore, an Inception Dilated Super-Resolution (IDSR) network with multiple branches is proposed. A dilated convolutional branch captures high-frequency information to reconstruct edge details, while a non-local operation branch captures long-range dependencies between any two positions to maintain the global structure. Furthermore, deformable convolution is utilized to fuse features extracted from different branches, enabling adaptation to targets of various shapes. To enhance the detection performance of low-resolution (LR) images, we crop the images into patches based on target labels before feeding them to the network. This allows the network to focus on learning the reconstruction of the target areas only, reducing the interference of background areas in the target areas’ reconstruction. Additionally, a feature-driven module is cascaded at the end of the IDSR network to guide the high-resolution (HR) image reconstruction with feature prior information from a detection backbone. This method has been tested on the FLIR Thermal Dataset and the M3FD Dataset and compared with five mainstream SR algorithms. The final results demonstrate that our method effectively maintains image texture details. More importantly, our method achieves 80.55% mAP, outperforming other methods on FLIR Dataset detection accuracy, and with 74.7% mAP outperforms other methods on M3FD Dataset detection accuracy. Full article

Recent studies have explored the generation of three-dimensional (3D) meshes from single images. A key challenge in this area is the difficulty of improving both the generalization and detail simultaneously in 3D mesh generation. To address this issue, existing methods utilize fixed-resolution mesh features to train networks for generalization. This approach is capable of generating the overall 3D shape without limitations on object categories. However, the generated shape often exhibits a blurred surface and suffers from suboptimal texture resolution due to the fixed-resolution mesh features. In this study, we propose a joint optimization method that enhances geometry and texture by integrating generalized 3D mesh generation with adjustable mesh resolution. Specifically, we apply an inverse-rendering-based remeshing technique that enables the estimation of complex-shaped mesh estimations without relying on fixed-resolution structures. After remeshing, we enhance the texture to improve the detailed quality of the remeshed mesh via a texture enhancement diffusion model. By separating the tasks of generalization, detailed geometry estimation, and texture enhancement and adapting different target features for each specific network, the proposed joint optimization method effectively addresses the characteristics of individual objects, resulting in increased surface detail and the generation of high-quality textures. Experimental results on the Google Scanned Objects and ShapeNet datasets demonstrate that the proposed method significantly improves the accuracy of 3D geometry and texture estimation, as evaluated by the PSNR, SSIM, LPIPS, and CD metrics. Full article

Background/Objectives: Accurate detection and classification of nodules in medical images, particularly rheumatoid nodules, are critical due to the varying nature of these nodules, where their specific type is often unknown before analysis. This study addresses the challenges of multi-class prediction in nodule detection, with a specific focus on rheumatoid nodules, by employing a comprehensive approach to feature extraction and classification. We utilized a diverse dataset of nodules, including rheumatoid nodules sourced from the DermNet dataset and local rheumatologists. Method: This study integrates 62 features, combining traditional image characteristics with advanced graph-based features derived from a superpixel graph constructed through Delaunay triangulation. The key steps include image preprocessing with anisotropic diffusion and Retinex enhancement, superpixel segmentation using SLIC, and graph-based feature extraction. Texture analysis was performed using Gray-Level Co-occurrence Matrix (GLCM) metrics, while shape analysis was conducted with Fourier descriptors. Vascular pattern recognition, crucial for identifying rheumatoid nodules, was enhanced using the Frangi filter. A Hybrid CNN–Transformer model was employed for feature fusion, and feature selection and hyperparameter tuning were optimized using Gray Wolf Optimization (GWO) and Particle Swarm Optimization (PSO). Feature importance was assessed using SHAP values. Results: The proposed methodology achieved an accuracy of 85%, with a precision of 0.85, a recall of 0.89, and an F1 measure of 0.87, demonstrating the effectiveness of the approach in detecting and classifying rheumatoid nodules in both binary and multi-class classification scenarios. Conclusions: This study presents a robust tool for the detection and classification of nodules, particularly rheumatoid nodules, in medical imaging, offering significant potential for improving diagnostic accuracy and aiding in the early identification of rheumatoid conditions. Full article

Accurate segmentation of thyroid nodules in ultrasound images is crucial for the diagnosis of thyroid cancer and preoperative planning. However, the segmentation of thyroid nodules is challenging due to their irregular shape, blurred boundary, and uneven echo texture. To address these challenges, a novel Mamba- and ResNet-based dual-branch network (MRDB) is proposed. Specifically, the visual state space block (VSSB) from Mamba and ResNet-34 are utilized to construct a dual encoder for extracting global semantics and local details, and establishing multi-dimensional feature connections. Meanwhile, an upsampling–convolution strategy is employed in the left decoder focusing on image size and detail reconstruction. A convolution–upsampling strategy is used in the right decoder to emphasize gradual feature refinement and recovery. To facilitate the interaction between local details and global context within the encoder and decoder, cross-skip connection is introduced. Additionally, a novel hybrid loss function is proposed to improve the boundary segmentation performance of thyroid nodules. Experimental results show that MRDB outperforms the state-of-the-art approaches with DSC of 90.02% and 80.6% on two public thyroid nodule datasets, TN3K and TNUI-2021, respectively. Furthermore, experiments on a third external dataset, DDTI, demonstrate that our method improves the DSC by 10.8% compared to baseline and exhibits good generalization to clinical small-scale thyroid nodule datasets. The proposed MRDB can effectively improve thyroid nodule segmentation accuracy and has great potential for clinical applications. Full article

Antimony is one of the world’s scarcest metals and is listed as a Critical Raw Material (CRM) for the European Union. To meet the increasing demand for metals in a sustainable way, one of the strategies that could be implemented would be the recovery of metals as by-products. This would decrease the amount of hazardous materials filling mining dumps. The present study investigates the potential for producing antimony as a by-product at the Olympias separation plant in Northern Greece. This plant works a skarn mineralization that shows interesting amounts of Sb. Boulangerite (Pb₅Sb₄S₁₁) reports on Pb concentrate levels reached 8% in the analyzed product. This pre-enrichment is favorable in terms of boulangerite recovery since it can be separated from galena through froth flotation. Boulangerite distribution in the primary ore is quite heterogeneous in terms of the inclusion relationships and grain size. However, a qualitative assessment shows that the current Pb concentrate grain size is too coarse to successfully liberate a good amount of boulangerite. The use of image analysis and textural assessments is pivotal in determining shape factors and crystal size, which is essential for the targeting of flotation parameters during separation. The extraction of antimony as a by-product is possible through a two-step process; namely, (i) the preliminary concentration of boulangerite, followed by (ii) the hydrometallurgical extraction of the antimony from the boulangerite concentrate. The Olympias enrichment plant could therefore set a positive example by promoting the benefits of targeted Sb extraction as a by-product within similar sulfide deposits within the European territory. Full article

Machine vision and near-infrared light technology are widely used in fruits and vegetable grading, as an important means of agricultural non-destructive testing. The characteristics of fruits and vegetables can easily be automatically distinguished by these two technologies, such as appearance, shape, color and texture. Nondestructive testing is reasonably used for image processing and pattern recognition, and can meet the identification and grading of single features and fusion features in production. Through the summary and analysis of the fruits and vegetable grading technology in the past five years, the results show that the accuracy of machine vision for fruits and vegetable size grading is 70–99.8%, the accuracy of external defect grading is 88–95%, and the accuracy of NIR and hyperspectral internal detection grading is 80.56–100%. Comprehensive research on multi-feature fusion technology in the future can provide comprehensive guidance for the construction of automatic integrated grading of fruits and vegetables, which is the main research direction of fruits and vegetable grading in the future. Full article

This paper introduces a novel approach to pavement material crack detection, classification, and segmentation using advanced deep learning techniques, including multi-scale feature aggregation and transformer-based attention mechanisms. The proposed methodology significantly enhances the model’s ability to handle varying crack sizes, shapes, and complex pavement textures. Trained on a dataset of 10,000 images, the model achieved substantial performance improvements across all tasks after integrating transformer-based attention. Detection precision increased from 88.7% to 94.3%, and IoU improved from 78.8% to 93.2%. In classification, precision rose from 88.3% to 94.8%, and recall improved from 86.8% to 94.2%. For segmentation, the Dice Coefficient increased from 80.3% to 94.7%, and IoU for segmentation advanced from 74.2% to 92.3%. These results underscore the model’s robustness and accuracy in identifying pavement cracks in challenging real-world scenarios. This framework not only advances automated pavement maintenance but also provides a foundation for future research focused on optimizing real-time processing and extending the model’s applicability to more diverse pavement conditions. Full article

Background: In recent years, the healthcare field has experienced significant advancements. New diagnostic techniques, treatments, and insights into the causes of various diseases have emerged. Despite these progressions, cancer remains a major concern. It is a widespread illness affecting individuals of all ages and leads to one out of every six deaths. Lung and colon cancer alone account for nearly two million fatalities. Though it is rare for lung and colon cancers to co-occur, the spread of cancer cells between these two areas—known as metastasis—is notably high. Early detection of cancer greatly increases survival rates. Currently, histopathological image (HI) diagnosis and appropriate treatment are key methods for reducing cancer mortality and enhancing survival rates. Digital image processing (DIP) and deep learning (DL) algorithms can be employed to analyze the HIs of five different types of lung and colon tissues. Methods: Therefore, this paper proposes a refined DL model that integrates feature fusion for the multi-classification of lung and colon cancers. The proposed model incorporates three DL architectures: ResNet-101V2, NASNetMobile, and EfficientNet-B0. Each model has limitations concerning variations in the shape and texture of input images. To address this, the proposed model utilizes a concatenate layer to merge the pre-trained individual feature vectors from ResNet-101V2, NASNetMobile, and EfficientNet-B0 into a single feature vector, which is then fine-tuned. As a result, the proposed DL model achieves high success in multi-classification by leveraging the strengths of all three models to enhance overall accuracy. This model aims to assist pathologists in the early detection of lung and colon cancer with reduced effort, time, and cost. The proposed DL model was evaluated using the LC25000 dataset, which contains colon and lung HIs. The dataset was pre-processed using resizing and normalization techniques. Results: The model was tested and compared with recent DL models, achieving impressive results: 99.8% for precision, 99.8% for recall, 99.8% for F1-score, 99.96% for specificity, and 99.94% for accuracy. Conclusions: Thus, the proposed DL model demonstrates exceptional performance across all classification categories. Full article

A portion of individuals diagnosed with primary central nervous system lymphomas (PCNSL) may experience early relapse or refractory (R/R) disease following treatment. This research explored the potential of MRI-based radiomics in forecasting R/R cases in PCNSL. Forty-six patients with pathologically confirmed PCNSL diagnosed between January 2008 and December 2020 were included in this study. Only patients who underwent pretreatment brain MRIs and complete postoperative follow-up MRIs were included. Pretreatment contrast-enhanced T1WI, T2WI, and T2 FLAIR imaging were analyzed. A total of 107 radiomic features, including 14 shape-based, 18 first-order statistical, and 75 texture features, were extracted from each sequence. Predictive models were then built using five different machine learning algorithms to predict R/R in PCNSL. Of the included 46 PCNSL patients, 20 (20/46, 43.5%) patients were found to have R/R. In the R/R group, the median scores in predictive models such as support vector machine, k-nearest neighbors, linear discriminant analysis, naïve Bayes, and decision trees were significantly higher, while the apparent diffusion coefficient values were notably lower compared to those without R/R (p < 0.05). The support vector machine model exhibited the highest performance, achieving an overall prediction accuracy of 83%, a precision rate of 80%, and an AUC of 0.78. Additionally, when analyzing tumor progression, patients with elevated support vector machine and naïve Bayes scores demonstrated a significantly reduced progression-free survival (p < 0.05). These findings suggest that preoperative MRI-based radiomics may provide critical insights for treatment strategies in PCNSL. Full article

In industrial manufacturing, metal surface defect detection often suffers from low detection accuracy, high leakage rates, and false detection rates. To address these issues, this paper proposes a novel model named DSL-YOLO for metal surface defect detection. First, we introduce the C2f_DWRB structure by integrating the DWRB module with C2f, enhancing the model’s ability to detect small and occluded targets and effectively extract sparse spatial features. Second, we design the SADown module to improve feature extraction in challenging tasks involving blurred images or very small objects. Finally, to further enhance the model’s capacity to extract multi-scale features and capture critical image information (such as edges, textures, and shapes) without significantly increasing memory usage and computational cost, we propose the LASPPF structure. Experimental results demonstrate that the improved model achieves significant performance gains on both the GC10-DET and NEU-DET datasets, with a [email protected] increase of 4.2% and 2.6%, respectively. The improvements in detection accuracy highlight the model’s ability to address common challenges while maintaining efficiency and feasibility in metal surface defect detection, providing a valuable solution for industrial applications. Full article

The fish detection task is an essential component of marine exploration, which helps scientists monitor fish population numbers and diversity and understand changes in fish behavior and habitat. It also plays a significant role in assessing the health of marine ecosystems, formulating conservation measures, and maintaining biodiversity. However, there are two main issues with current fish detection algorithms. First, the lighting conditions underwater are significantly different from those on land. In addition, light scattering and absorption in water trigger uneven illumination, color distortion, and reduced contrast in images. The accuracy of detection algorithms can be affected by these lighting variations. Second, the wide variation of fish species in shape, color, and size brings about some challenges. As some fish have complex textures or camouflage features, it is difficult to differentiate them using current detection algorithms. To address these issues, we propose a fish detection algorithm—FishDet-YOLO—through improvement in the YOLOv8 algorithm. To tackle the complexities of underwater environments, we design an Underwater Enhancement Module network (UEM) that can be jointly trained with YOLO. The UEM enhances the details of underwater images via end-to-end training with YOLO. To address the diversity of fish species, we leverage the Mamba model’s capability for long-distance dependencies without increasing computational complexity and integrate it with the C2f from YOLOv8 to create the Mamba-C2f. Through this design, the adaptability in handling complex fish detection tasks is improved. In addition, the RUOD and DUO public datasets are used to train and evaluate FishDet-YOLO. FishDet-YOLO achieves mAP scores of 89.5% and 88.8% on the test sets of RUOD and DUO, respectively, marking an improvement of 8% and 8.2% over YOLOv8. It also surpasses recent state-of-the-art general object detection and underwater fish detection algorithms. Full article

Accurate understanding of urban land use change information is of great significance for urban planning, urban monitoring, and disaster assessment. The use of Very-High-Resolution (VHR) remote sensing images for change detection on urban land features has gradually become mainstream. However, most existing transfer learning-based change detection models compute multiple deep image features, leading to feature redundancy. Therefore, we propose a Transfer Learning Change Detection Model Based on Change Feature Selection (TL-FS). The proposed method involves using a pretrained transfer learning model framework to compute deep features from multitemporal remote sensing images. A change feature selection algorithm is then designed to filter relevant change information. Subsequently, these change features are combined into a vector. The Change Vector Analysis (CVA) is employed to calculate the magnitude of change in the vector. Finally, the Fuzzy C-Means (FCM) classification is utilized to obtain binary change detection results. In this study, we selected four VHR optical image datasets from Beijing-2 for the experiment. Compared with the Change Vector Analysis and Spectral Gradient Difference, the TL-FS method had maximum increases of 26.41% in the F1-score, 38.04% in precision, 29.88% in recall, and 26.15% in the overall accuracy. The results of the ablation experiments also indicate that TL-FS could provide clearer texture and shape detections for dual-temporal VHR image changes. It can effectively detect complex features in urban scenes. Full article

Reconstructing 3D indoor scenes from 2D images has always been an important task in computer vision and graphics applications. For indoor scenes, traditional 3D reconstruction methods have problems such as missing surface details, poor reconstruction of large plane textures and uneven illumination areas, and many wrongly reconstructed floating debris noises in the reconstructed models. This paper proposes a 3D reconstruction method for indoor scenes that combines neural radiation field (NeRFs) and signed distance function (SDF) implicit expressions. The volume density of the NeRF is used to provide geometric information for the SDF field, and the learning of geometric shapes and surfaces is strengthened by adding an adaptive normal prior optimization learning process. It not only preserves the high-quality geometric information of the NeRF, but also uses the SDF to generate an explicit mesh with a smooth surface, significantly improving the reconstruction quality of large plane textures and uneven illumination areas in indoor scenes. At the same time, a new regularization term is designed to constrain the weight distribution, making it an ideal unimodal compact distribution, thereby alleviating the problem of uneven density distribution and achieving the effect of floating debris removal in the final model. Experiments show that the 3D reconstruction effect of this paper on ScanNet, Hypersim, and Replica datasets outperforms the state-of-the-art methods. Full article