Search Results (162)

The task of pose-guided person image generation has long been confronted with two major challenges: high-frequency texture details tend to blur and be lost during appearance transfer, while the semantic identity of the person is difficult to maintain consistently during pose changes. To address these issues, this paper proposes a diffusion-based generative framework that integrates frequency awareness and global semantic alignment. The framework consists of two core modules: a multi-level fractional-order Gabor frequency-aware network, which accurately extracts and reconstructs high-frequency texture features such as hair strands and fabric wrinkles, enhances image detail fidelity through fractional-order filtering and complex domain modeling; and a global semantic-pose alignment module that utilizes a cross-modal attention mechanism to establish a global mapping between pose features and appearance semantics, ensuring pose-driven semantic alignment and appearance consistency. The collaborative function of these two modules ensures that the generated results maintain structural integrity and natural textures even under complex pose variations and large-angle rotations. The experimental results on the DeepFashion and Market1501 datasets demonstrate that the proposed method outperforms existing state-of-the-art approaches in terms of SSIM, FID, and perceptual quality, validating the effectiveness of the model in enhancing texture fidelity and semantic consistency. Full article

Background: The aim of this study was to develop an artificial intelligence (AI)-based decision support model for evaluating the extraction indication of permanent first molars in pediatric patients using panoramic radiographs, and to investigate the potential contribution of this model to the clinical decision-making process. Methods: This retrospective observational study analyzed 1000 panoramic radiographs obtained from children aged 8–10 years who attended the Clinics of Batman University Faculty of Dentistry for routine dental examination. Among the radiographs meeting the inclusion criteria, a total of 176 panoramic images were selected based on dental maturation assessment using the Demirjian tooth development staging system. Cases in which the permanent second molar was classified as Demirjian stages E–F were labeled as “extraction indication present”, while the remaining stages were labeled as “extraction indication absent”. A balanced dataset was created, consisting of 88 cases in each group. Image features were extracted using Gabor filters and Histogram of Oriented Gradients (HOG). The selected features were analyzed using a Support Vector Machine (SVM) classifier with a radial basis function (RBF) kernel. Model performance was evaluated using accuracy, sensitivity, specificity, F1-score, and area under the receiver operating characteristic curve (ROC–AUC). Results: The proposed Gabor–HOG–SVM-based AI model achieved an overall classification accuracy of 77.78% with an AUC value of 0.77 in distinguishing between “extraction indication present” and “extraction indication absent” cases. For the extraction-indicated group, the sensitivity was 0.81 and the F1-score was 0.79, whereas for the non-indicated group, the sensitivity and F1-score were 0.74 and 0.77, respectively. No statistically significant differences were observed between the groups in terms of age or sex distribution (p > 0.05). Conclusions: This study demonstrates that artificial intelligence-based analysis of panoramic radiographic images can provide an objective and reproducible decision support approach for evaluating extraction indications of permanent first molars in pediatric patients. The proposed model should be considered as an adjunctive tool to reduce observer-dependent variability rather than a replacement for clinical judgment, and its clinical applicability should be further validated through multicenter and multi-parametric studies. Full article

This study proposes a topic-modeling guided framework that enhances image classification by introducing semantic clustering prior to CNN training. Images are processed through two key-point extraction pipelines: Scale-Invariant Feature Transform (SIFT) with Sobel edge detection and Block Gabor Filtering (BGF), to obtain local feature descriptors. These descriptors are clustered using K-means to build a visual vocabulary. Bag of Words histograms then represent each image as a visual document. Latent Dirichlet Allocation is applied to uncover latent semantic topics, generating coherent image clusters. Cluster-specific CNN models, including AlexNet, GoogLeNet, and several ResNet variants, are trained under identical conditions to identify the most suitable architecture for each cluster. Two topic guided integration strategies, the Maximum Proportion Topic (MPT) and the Weight Proportion Topic (WPT), are then used to assign test images to the corresponding specialized model. Experimental results show that both the SIFT-based and BGF-based pipelines outperform non-clustered CNN models and a baseline method using Incremental PCA, K-means, Same-Cluster Prediction, and unweighted Ensemble Voting. The SIFT pipeline achieves the highest accuracy of 95.24% with the MPT strategy, while the BGF pipeline achieves 93.76% with the WPT strategy. These findings confirm that semantic structure introduced through topic modeling substantially improves CNN classification performance. Full article

Alzheimer’s Disease (AD) is an advanced brain illness that affects millions of individuals across the world. It causes gradual damage to the brain cells, leading to memory loss and cognitive dysfunction. Although Magnetic Resonance Imaging (MRI) is widely used in AD diagnosis, the existing studies rely solely on the visual representations, leaving alternative features unexplored. The objective of this study is to explore whether MRI sonification can provide complementary diagnostic information when combined with conventional image-based methods. In this study, we propose a novel dual-stream multimodal framework that integrates 2D MRI slices with their corresponding audio representations. MRI images are transformed into audio signals using a multi-scale, multi-orientation Gabor filtering, followed by a Hilbert space-filling curve to preserve spatial locality. The image and sound modalities are processed using a lightweight CNN and YAMNet, respectively, then fused via logistic regression. The experimental results of the multimodal achieved the highest accuracy in distinguishing AD from Cognitively Normal (CN) subjects at 98.2%, 94% for AD vs. Mild Cognitive Impairment (MCI), and 93.2% for MCI vs. CN. This work provides a new perspective and highlights the potential of audio transformation of imaging data for feature extraction and classification. Full article

Melanoma is highly dangerous and can spread rapidly to other parts of the body. It has an increasing fatality rate among different types of cancer. Timely detection of skin malignancies can reduce overall mortality. Therefore, clinical screening methods require more time and accuracy for diagnosis. An automated, computer-aided system would facilitate earlier melanoma detection, thereby increasing patient survival rates. This paper identifies melanoma images using a Convolutional Neural Network. Skin images are preprocessed using Histogram Equalization and Gabor transforms. A Gabor filter-based Convolutional Neural Network (CNN) classifier trains and classifies the extracted features. We adopt Gabor filters because they are bandpass filters that transform a pixel into a multi-resolution kernel matrix, providing detailed information about the image. This study suggests a method with accuracy, sensitivity, and specificity of 98.58%, 98.66%, and 98.75%, respectively. This research supports SDGs 3 and 4 by facilitating early melanoma detection and enhancing AI-driven medical education. Full article

Heterogeneous remote sensing image change detection (Hete-CD) holds significant research value in military and civilian fields. The existing methods often rely on expert experience to design fixed deep network architectures for cross-modal feature alignment and fusion purposes. However, when faced with diverse land cover types, these methods often lead to blurred change boundaries and structural distortions, resulting in significant performance degradations. To address this, we propose an adaptive adversarial learning-based heterogeneous remote sensing image change detection method based on the differentiable filter combination search (DFCS) strategy to provide enhanced generalizability and dynamic learning capabilities for diverse scenarios. First, a fully reconfigurable self-learning discriminator is designed to dynamically synthesize the optimal convolutional architecture from a library of atomic filters containing basic operators. This provides highly adaptive adversarial supervision to the generator, enabling joint dynamic learning between the generator and discriminator. To further mitigate modality differences in the input stage, we integrate a feature fusion module based on the Gabor and local normalized cross-correlation (G-LNCC) to extract modality-invariant texture and structure features. Finally, a geometric structure-based collaborative supervision (GSCS) loss function is constructed to impose fine-grained constraints on the change map from the perspectives of regions, boundaries, and structures, thereby enforcing physical properties. Comparative experimental results obtained on five public Hete-CD datasets show that our method achieves the best F1 values and overall accuracy levels, especially on the Gloucester I and Gloucester II datasets, achieving F1 scores of 93.7% and 95.0%, respectively, demonstrating the strong generalizability of our method in complex scenarios. Full article

Underwater images often suffer from color distortion, reduced contrast, and blurred details due to the selective absorption and scattering of light by water, which limits the performance of underwater visual tasks. To address these issues, this paper proposes an underwater image enhancement method that integrates multi-channel attenuation analysis and discrete cosine transform (DCT). First, the color statistics of an in situ-captured underwater image are mapped to those of a reference image selected from a well-illuminated natural image dataset with standard color distribution; no pristine underwater image is required. This mapping yields a color transfer image, i.e., an intermediate color-corrected result obtained via statistical matching. Subsequently, this image is fused with an attenuation weight map and the original input to produce the final color-corrected result. Secondly, taking advantage of the median’s resistance to extreme value interference and the Sigmoid function’s flexible control of gray-scale transformation, the gray-scale range is adjusted in different regions through nonlinear mapping to achieve global contrast balance. Finally, considering the visual system’s sensitivity to high-frequency details, a saliency map is extracted using Gabor filtering, and the frequency characteristics are analyzed through block DCT transformation. Adaptive gain is applied to high-frequency details to enhance them. Experiments were conducted on the UIEB, EUVP, and LSUI datasets and compared with existing methods. Through qualitative and quantitative analysis, it was verified that the proposed algorithm not only effectively enhances underwater images but also significantly improves image clarity. Full article

The safety of animal-related agricultural products has been a hot issue. To obtain a multi-feature representation of porcine bodies for detecting their health, visible and infrared imaging is valuable for exploiting multiple images of a porcine body from different modalities. However, the direct registration of visible and infrared porcine body images can easily cause the dislocation of structural information and spatial position, due to different resolutions and spectrums of multi-source images. To overcome the problem, a novel multi-source image feature representation method based on contour angle orientation is proposed and named Gabor-Ordinal-based Contour Angle Orientation (GOCAO). Moreover, a visible and infrared porcine body image registration method is described and named GOCAO-Rough to Fine (GOCAO-R2F). First, contour and texture features of the porcine body are acquired using a Gabor filter with variable scales and an ordinal operation. Second, feature points in contours are obtained by curvature scale space (CSS), and the main orientation of each feature point is determined by GOCAO. Third, modified scale-invariant feature transform (MSIFT) features are received on the main orientation and registered with bilateral matching. Finally, accurate registrations are extracted by R2F. Experimental results show that the proposed registration algorithm accurately matches multi-source images for porcine body multi-feature detection and is capable of achieving lower average root-mean-square error than current registration algorithms. Full article

Graphite nodularity is a key indicator for evaluating the microstructure quality of ductile iron and plays a crucial role in ensuring product quality and enhancing manufacturing efficiency. Existing research often only focuses on a single type of feature and fails to utilize multi-source information in a coordinated manner. Single-feature methods are difficult to comprehensively capture microstructures, which limits the accuracy and robustness of the model. This study proposes a hybrid estimation model for the graphite nodularity of ductile cast iron based on multi-source feature extraction. A comprehensive feature engineering pipeline was established, incorporating geometric, color, and texture features extracted via Hue-Saturation-Value color space (HSV) histograms, gray level co-occurrence matrix (GLCM), Local Binary Pattern (LBP), and multi-scale Gabor filters. Dimensionality reduction was performed using Principal Component Analysis (PCA) to mitigate redundancy. An improved watershed algorithm combined with intelligent filtering was used for accurate particle segmentation. Several machine learning algorithms, including Support Vector Regression (SVR), Multi-Layer Perceptron (MLP), Random Forest (RF), Gradient Boosting Regressor (GBR), eXtreme Gradient Boosting (XGBoost) and Categorical Boosting (CatBoost), are applied to estimate graphite nodularity based on geometric features (GFs) and feature extraction. Experimental results demonstrate that the CatBoost model trained on fused features achieves high estimation accuracy and stability for geometric parameters, with R-squared (R²) exceeding 0.98. Furthermore, introducing geometric features into the fusion set enhances model generalization and suppresses overfitting. This framework offers an efficient and robust approach for intelligent analysis of metallographic images and provides valuable support for automated quality assessment in casting production. Full article

E-textiles, a type of textile integrated with conductive sensors, allows users to freely utilize any area of the body in a convenient and comfortable manner. Thus, interactions with e-textiles are attracting more and more attention, especially for text input. However, the functional aging of e-textiles affects the characteristics and even the quality of the captured signal, presenting serious challenges for character recognition. This paper focuses on studying the behavior of e-textile functional aging and alleviating its impact on text input with an unsupervised domain adaptation technique, named A²TEXT (aging-aware e-textile-based text input). We first designed a deep kernel-based two-sample test method to validate the impact of functional aging on handwriting with an e-textile input. Based on that, we introduced a so-called Gabor domain adaptation technique, which adopts a novel Gabor orientation filter in feature extraction under an adversarial domain adaptation framework. We demonstrated superior performance compared to traditional models in four different transfer tasks, validating the effectiveness of our work. Full article

Semantic segmentation of urban scenes from red–green–blue and thermal infrared imagery enables per-pixel categorization, delivering precise environmental understanding for autonomous driving and urban planning. However, existing methods suffer from inefficient fusion and insufficient boundary accuracy due to modal differences. To address these challenges, we propose a bidirectional dynamic adaptation framework with two complementary networks. The modality-aware network uses dual attention and multi-scale feature integration to balance modal contributions adaptively, improving intra-class semantic consistency and reducing modal disparities. The edge-texture guidance network applies pixel-level and feature-level weighting with Sobel and Gabor filters to enhance inter-class boundary discrimination, improving detail and boundary precision. Furthermore, the framework redefines multi-modal synergy using an adaptive cross-modal mutual learning mechanism. This mechanism employs information-driven dynamic alignment and probability-guided semantic consistency to overcome the fixed constraints of traditional mutual learning. This cohesive orchestration enhances multi-modal fusion efficiency and boundary delineation accuracy. Extensive experiments on the MFNet and PST900 datasets demonstrate the framework’s superior performance in urban road, vehicle, and pedestrian segmentation, surpassing state-of-the-art approaches. Full article

Automated optical inspection (AOI) of wood surfaces is critical for ensuring product quality in the furniture and manufacturing industries; however, existing defect detection systems often struggle to generalize across complex grain patterns and diverse defect types. This study proposes a wood defect recognition model employing a Gabor Convolutional Network (GCN) that integrates convolutional neural networks (CNNs) with Gabor filters. To systematically optimize the network’s architecture and improve both detection accuracy and computational efficiency, the Taguchi method is employed to tune key hyperparameters, including convolutional kernel size, filter number, and Gabor parameters (frequency, orientation, and phase offset). Additionally, image tiling and augmentation techniques are employed to effectively increase the training dataset, thereby enhancing the model’s stability and accuracy. Experiments conducted on the MVTec Anomaly Detection dataset (wood category) demonstrate that the Taguchi-optimized GCN achieves an accuracy of 98.92%, outperforming a baseline Taguchi-optimized CNN by 2.73%. Results confirm that Taguchi-optimized GCNs enhance defect detection performance and computational efficiency, making them valuable for smart manufacturing. Full article

Diabetic retinopathy (DR), a critical ocular disease that can lead to blindness, demands early and accurate diagnosis to prevent vision loss. Current automated DR diagnosis methods face two core challenges: first, subtle early lesions such as microaneurysms are often missed due to insufficient feature extraction; second, there is a persistent trade-off between model accuracy and efficiency—lightweight architectures often sacrifice precision for real-time performance, while high-accuracy models are computationally expensive and difficult to deploy on resource-constrained edge devices. To address these issues, this study presents a novel deep learning framework integrating depthwise separable convolution and a multi-view attention mechanism (MVAM) for efficient DR diagnosis using retinal images. The framework employs multi-scale feature fusion via parallel 3 × 3 and 5 × 5 convolutions to capture lesions of varying sizes and incorporates Gabor filters to enhance vascular texture and directional lesion modeling, improving sensitivity to early structural abnormalities while reducing computational costs. Experimental results on both the diabetic retinopathy (DR) dataset and ocular disease (OD) dataset demonstrate the superiority of the proposed method: it achieves a high accuracy of 0.9697 on the DR dataset and 0.9669 on the OD dataset, outperforming traditional methods such as CNN_eye, VGG, and UNet by more than 1 percentage point. Moreover, its training time is only half that of U-Net (on DR dataset) and VGG (on OD dataset), highlighting its potential for clinical DR screening. Full article

Holy basil (Ocimum tenuiflorum L.) is a medicinal herb rich in bioactive flavonoids with therapeutic properties. Traditional quantification methods rely on time-consuming and destructive extraction processes, whereas hyperspectral imaging provides a rapid, non-destructive alternative by analysing spectral signatures. However, effectively linking hyperspectral data to flavonoid levels remains a challenge for developing early detection tools before harvest. This study integrates deep learning with hyperspectral imaging to quantify flavonoid contents in 113 samples from 26 Thai holy basil cultivars collected across diverse regions of Thailand. Two deep learning architectures, ResNet1D and CNN1D, were evaluated in combination with feature extraction techniques, including wavelet transformation and Gabor-like filtering. ResNet1D with wavelet transformation achieved optimal performance, yielding an area under the receiver operating characteristic curve (AUC) of 0.8246 and an accuracy of 0.7702 for flavonoid content classification. Cross-validation demonstrated the model’s robust predictive capability in identifying antioxidant-rich samples. Samples with the highest predicted flavonoid content were identified, and cultivars exhibiting elevated levels of both flavonoids and phenolics were highlighted across various regions of Thailand. These findings demonstrate the predictive capability of hyperspectral data combined with deep learning for phytochemical assessment. This approach offers a valuable tool for non-destructive quality evaluation and supports cultivar selection for higher phytochemical content in breeding programs and agricultural applications. Full article

To solve the problem of the limited capability of multimodal image feature descriptors constructed by gradient information and the phase consistency principle, a method of cumulative structure feature descriptor construction with rotation invariance is proposed in this paper. Firstly, we extract the direction of multi-scale and multi-direction feature point edges using the Log-Gabor odd-symmetric filter and calculate the amplitude of pixel edges based on the phase consistency principle. Then, the main direction of the key points is determined based on the edge direction feature map, and the coordinates are established according to the main direction to ensure that the feature point descriptor has rotation invariance. Finally, the Log-Gabor odd-symmetric filter calculates the cumulative structural response in the maximum direction and constructs a highly identifiable descriptor with rotation invariance. We select several representative heterogeneous images as test data and compare the matching performance of the proposed algorithm with several excellent descriptors. The results indicate that the descriptor constructed in this paper is more robust than other descriptors for heterosource images with rotation changes. Full article