Texture Feature Extraction Research Articles

Deep Learning-based Texture Feature Extraction Technique for Face Annotation

Deep Learning-based Texture Feature Extraction Technique for Face Annotation

Vegetation Classification in a Mountain–Plain Transition Zone in the Sichuan Basin, China

Developing an effective vegetation classification method for mountain–plain transition zones is critical for understanding ecological patterns, evaluating ecosystem services, and guiding conservation efforts. Existing methods perform well in mountainous and plain areas but lack verification in mountain–plain transition zones. This study utilized terrain data and Sentinel-1 and Sentinel-2 imagery to extract topographic, spectral, texture, and SAR features as well as the vegetation index. By combining feature sets and applying feature elimination algorithms, the classification performance of one-dimensional convolutional neural networks (1D-CNNs), Random Forest (RF), and Multilayer Perceptron (MLP) was evaluated to determine the optimal feature combinations and methods. The results show the following: (1) multi-feature combinations, especially spectral and topographic features, significantly improved classification accuracy; (2) Recursive Feature Elimination based on Random Forest (RF-RFE) outperformed ReliefF in feature selection, identifying more representative features; (3) all three algorithms performed well, with consistent spatial results. The MLP algorithm achieved the best overall accuracy (OA: 81.65%, Kappa: 77.75%), demonstrating robustness and lower dependence on feature quantity. This study presents an efficient and robust vegetation classification workflow, verifies its applicability in mountain–plain transition zones, and provides valuable insights for small-region vegetation classification under similar topographic conditions globally.

Monitoring the Maize Canopy Chlorophyll Content Using Discrete Wavelet Transform Combined with RGB Feature Fusion

To evaluate the accuracy of Discrete Wavelet Transform (DWT) in monitoring the chlorophyll (CHL) content of maize canopies based on RGB images, a field experiment was conducted in 2023. Images of maize canopies during the jointing, tasseling, and grouting stages were captured using unmanned aerial vehicle (UAV) remote sensing to extract color, texture, and wavelet features and to construct a color and texture feature dataset and a fusion of wavelet, color, and texture feature datasets. Backpropagation neural network (BP), Stacked Ensemble Learning (SEL), and Gradient Boosting Decision Tree (GBDT) models were employed to develop CHL monitoring models for the maize canopy. The performance of these models was evaluated by comparing their predictions with measured CHL data. The results indicate that the dataset integrating wavelet features achieved higher monitoring accuracy compared to the color and texture feature dataset. Specifically, for the integrated dataset, the BP model achieved an R2 value of 0.728, an RMSE of 3.911, and an NRMSE of 15.24%; the SEL model achieved an R2 value of 0.792, an RMSE of 3.319, and an NRMSE of 15.34%; and the GBDT model achieved an R2 value of 0.756, an RMSE of 3.730, and an NRMSE of 15.45%. Among these, the SEL model exhibited the highest monitoring accuracy. This study provides a fast and reliable method for monitoring maize growth in field conditions. Future research could incorporate cross-validation with hyperspectral and thermal infrared sensors to further enhance model reliability and expand its applicability.

Artificial intelligence can help individualize Wilms tumor treatment by predicting tumor response to preoperative chemotherapy.

To create a computer-aided prediction (CAP) system to predict Wilms tumor (WT) responsiveness to preoperative chemotherapy (PC) using pre-therapy contrast-enhanced computed tomography (CECT). A single-center database was reviewed for children <18 years diagnosed with WT and received PC between 2001 and 2021. Patients were excluded if pre- and post-PC CECT were not retrievable. According to the Response Evaluation Criteria in Solid Tumors criteria, volumetric response was considered favorable if PC resulted in ≥30% tumor volume reduction. Histological response was considered favorable if post-nephrectomy specimens had ≥66% necrosis. Four steps were used to create the prediction model: tumor delineation; extraction of shape, texture and functionality-based features; integration of the extracted features and selection of the prediction model with the highest diagnostic performance. K-fold cross-validation allowed the presentation of all data in the training and testing phases. A total of 63 tumors in 54 patients were used to train and test the prediction model. Patients were treated with 4-8 weeks of vincristine/actinomycin-D combination. Favorable volumetric and histologic responses were achieved in 46 tumors (73.0%) and 38 tumors (60.3%), respectively. Among machine learning classifiers, support vector machine had the best diagnostic performance with an accuracy, sensitivity, and specificity of 95.24%, 95.65%, and 94.12% for volumetric and 84.13%, 89.47%, 88% for histologic response prediction. Based on pre-therapy CECT, CAP systems can help identify WT that are less likely to respond to PC with excellent accuracy. These tumors can be offered upfront surgery, avoiding the cons of PC.

Enhancing Model Accuracy of UAV-Based Biomass Estimation by Evaluating Effects of Image Resolution and Texture Feature Extraction Strategy

Enhancing Model Accuracy of UAV-Based Biomass Estimation by Evaluating Effects of Image Resolution and Texture Feature Extraction Strategy

Investigations on segmentation-based fractal texture for texture classification in the presence of Gaussian noise.

Texture is a significant component used for several applications in content-based image retrieval. Any texture classification method aims to map an anonymously textured input image to one of the existing texture classes. Extensive ranges of methods for labeling image texture were proposed earlier. However, computing the performance of these methods in the presence of various degradations is always an open area of discussion. Image noise is always a dominant factor among various image degradation factors, affecting the performance of these methods and making texture classification challenging. Therefore, it is essential to investigate the interpretation of these methods in the presence of prominent degradation factors such as noise. Applications for Segmentation-Based Fractal Texture Features (SFTF) include image classification, texture generation, and medical image analysis. They are beneficial for examining textures with intricate, erratic patterns that are difficult to characterize using conventional statistical techniques accurately. This paper assesses two texture feature extraction methods based on SFTF and statistical moment-based texture features in the presence and absence of Gaussian noise. The SFTF and statistical moments-based handcrafted features are passed to a multilayer feed-forward neural network for classification. These models are evaluated on natural textures from Kylberg Texture Dataset 1.0. The results show the superiority of segmentation-based fractal analysis over other approaches. The average accuracy rates using the SFTF are 99% and 97% in the absence and presence of Gaussian noise, respectively.

Leveraging feature extraction and risk-based clustering for advanced fault diagnosis in equipment.

In the contemporary manufacturing landscape, the advent of artificial intelligence and big data analytics has been a game-changer in enhancing product quality. Despite these advancements, their application in diagnosing failure probability and risk remains underexplored. The current practice of failure risk diagnosis is impeded by the manual intervention of managers, leading to varying evaluations for identical products or similar facilities. This study aims to bridge this gap by implementing advanced data analysis techniques on maintenance data from an aluminum extruder. We have employed text embedding, dimensionality reduction, and feature extraction methods, integrating the K-means algorithm with the Silhouette Score for risk level classification. Our findings reveal that the combination of Word2Vec for embedding and Contractive Auto Encoder for dimensionality reduction and feature extraction yields high-performance results. The optimal cluster count, identified as three, achieved the highest Silhouette Score. Statistical analysis using one-way ANOVA confirmed the significance of these findings with a p-value of 5.3213 × e-6, well within the 5% significance threshold. Furthermore, this study utilized BERTopic for topic modeling to extract principal topics from each cluster, facilitating an in-depth analysis of the clusters in relation to the extruder's characteristics. The outcome of this research offers a novel methodology for facility managers to objectively diagnose equipment failures. By minimizing subjective judgment, this approach is poised to significantly enhance the efficacy of quality assurance systems in manufacturing, leveraging the robust capabilities of artificial intelligence.

Enhanced Brain Tumor Classification Through Optimized Semantic Preserved Generative Adversarial Networks.

Brain tumor is a most dangerous disease and requires accurate diagnosis in a short period to ensure the best treatment. Traditional methods for brain tumor classification (BTC) are quite effective, even though usually resulting in clinical manual analysis, which takes more time and prone to errors. Initially, the input image is collected from Brain Tumor dataset. The gathered image is given to preprocessing. In preprocessing stage, trust-based distributed set-membership filtering (TDSF) is used to remove the noise. The preprocessed output is fed to the quaternion offset linear canonical transform (QOLCT) for Grayscale statistic and Haralick texture features extraction. Then the extracted features are fed to the Semantic-Preserved Generative Adversarial Network (SPGAN) for classifying the brain tumor into Glioma, Meningioma and Pituitary. Finally, Hunger Games Search Optimization (HGSO) is used to enhance the weight parameters of SPGAN. The proposed BTC-SPGAN-HGSO method attains the accuracies of 99.72% for Glioma, 99.65% for Meningioma, 99.52% for Pituitary and lowest MSE values across all tumor types, with 0.45% for Glioma, 0.39% for Meningioma, and 0.5% for Pituitary, which performs better than existing models. The simulation results highlight the effectiveness of the proposed BTC-SPGAN-HGSO approach in improving the accuracy of BTC and assist neurologists and physicians make exact decisions of diagnostic.

Real-time drone detection framework based on advanced texture feature extraction and pattern recognition model using GUI

Real-time drone detection framework based on advanced texture feature extraction and pattern recognition model using GUI

Colon Tumor Discrimination Combining Independent Endoscopic Probe-Based Raman Spectroscopy and Optical Coherence Tomography Modalities with Bayes Rule.

Colorectal cancer is one of the most prevalent forms of cancer globally. The most common routine diagnostic methods are the examination of the interior of the colon during colonoscopy or sigmoidoscopy, which frequently includes the removal of a biopsy sample. Optical methods, such as Raman spectroscopy (RS) and optical coherence tomography (OCT), can help to improve diagnostics and reduce the number of unnecessary biopsies. For in vivo use, we have developed fiber-optic probes, one for single-point Raman measurements and one for volumetric OCT. Here, we present the results of a clinical study using these fiber-optic probes in an ex vivo setting. The goal was to evaluate the beneficial effect of combining these two modalities on the AUC ROC score of the machine learning models for the discrimination of cancerous and healthy tissue. In the initial stage of the investigation, both modalities were validated separately using linear discriminant analysis. RS was subjected to spectral preprocessing, while OCT underwent texture feature extraction. Subsequently, both modalities were integrated using the Bayes rule, resulting in an enhanced area under the curve score of 0.93, representing an improvement over the 0.77 score for Raman spectroscopy and 0.86 for OCT.

Combination of Deep and Statistical Features of the Tissue of Pathology Images to Classify and Diagnose the Degree of Malignancy of Prostate Cancer.

Prostate cancer is one of the most prevalent male-specific diseases, where early and accurate diagnosis is essential for effective treatment and preventing disease progression. Assessing disease severity involves analyzing histological tissue samples, which are graded from 1 (healthy) to 5 (severely malignant) based on pathological features. However, traditional manual grading is labor-intensive and prone to variability. This study addresses the challenge of automating prostate cancer classification by proposing a novel histological grade analysis approach. The method integrates the gray-level co-occurrence matrix (GLCM) for extracting texture features with Haar wavelet modification to enhance feature quality. A convolutional neural network (CNN) is then employed for robust classification. The proposed method was evaluated using statistical and performance metrics, achieving an average accuracy of 97.3%, a precision of 98%, and an AUC of 0.95. These results underscore the effectiveness of the approach in accurately categorizing prostate tissue grades. This study demonstrates the potential of automated classification methods to support pathologists, enhance diagnostic precision, and improve clinical outcomes in prostate cancer care.

Transform invariant low rank texture feature extraction and restoration algorithms for architectural decoration surface patterns

Transform invariant low rank texture feature extraction and restoration algorithms for architectural decoration surface patterns

How to identify patient perception of AI voice robots in the follow-up scenario? A multimodal identity perception method based on deep learning

ObjectivesPost-discharge follow-up stands as a critical component of post-diagnosis management, and the constraints of healthcare resources impede comprehensive manual follow-up. However, patients are less cooperative with AI follow-up calls or may even hang up once AI voice robots are perceived. To improve the effectiveness of follow-up, alternative measures should be taken when patients perceive AI voice robots. Therefore, identifying how patients perceive AI voice robots is crucial. This study aims to construct a multimodal identity perception model based on deep learning to identify how patients perceive AI voice robots. MethodsOur dataset includes 2030 response audio recordings and corresponding texts from patients. We conduct comparative experiments and perform an ablation study. The proposed model employs a transfer learning approach, utilizing BERT and TextCNN for text feature extraction, AST and LSTM for audio feature extraction, and self-attention for feature fusion. ResultsOur model demonstrates superior performance against existing baselines, with a precision of 86.67%, an AUC of 84%, and an accuracy of 94.38%. Additionally, a generalization experimentwas conducted using 144 patients’ response audio recordings and corresponding text data from other departments in the hospital, confirming the model’s robustness and effectiveness. ConclusionOur multimodal identity perception model can identify how patients perceive AI voice robots effectively. Identifying how patients perceive AI not only helps to optimize the follow-up process and improve patient cooperation, but also provides support for the evaluation and optimization of AI voice robots.

Multimodal sentiment analysis based on multiple attention

The development of the Internet makes various types of data widely appear on various social platforms, multimodal data provides a new perspective for sentiment analysis. Although the data types are different, there are information expressing the same sentiment. The existing researches on extracting those information are static, and this means that there is a problem of extracting common information in a fixed amount. Therefore, to address this problem, we proposes a method named multimodal sentiment analysis based on multiple attention(MAMSA). Firstly, this method utilized the adaptive attention interaction module to dynamically determine the amount of information contributed by text and image features in multimodal fusion, and multimodal common representations are extracted through cross modal attention to improve the performance of each modal feature representation. Secondly, using sentiment information as a guide to extract text and image features related to sentiment. Finally, using hierarchical manner to fully learning the internal correlations between sentiment-text association representation, sentiment-image association representation, and multimodal common information to improve the performance of the model. We conducted extensive experiments using two public multimodal datasets, and the experimental results validated the availability of the proposed method.

Hyper‐Parameter Optimization of Kernel Functions on Multi‐Class Text Categorization: A Comparative Evaluation

ABSTRACTIn recent years, machine learning (ML) has witnessed a paradigm shift in kernel function selection, which is pivotal in optimizing various ML models. Despite multiple studies about its significance, a comprehensive understanding of kernel function selection, particularly about model performance, still needs to be explored. Challenges remain in selecting and optimizing kernel functions to improve model performance and efficiency. The study investigates how gamma parameter and cost parameter influence performance metrics in multi‐class classification tasks using various kernel‐based algorithms. Through sensitivity analysis, the impact of these parameters on classification performance and computational efficiency is assessed. The experimental setup involves deploying ML models using four kernel‐based algorithms: Support Vector Machine, Radial Basis Function, Polynomial Kernel, and Sigmoid Kernel. Data preparation includes text processing, categorization, and feature extraction using TfidfVectorizer, followed by model training and validation. Results indicate that Support Vector Machine with default settings and Radial Basis Function kernel consistently outperforms polynomial and sigmoid kernels. Adjusting gamma improves model accuracy and precision, highlighting its role in capturing complex relationships. Regularization cost parameters, however, show minimal impact on performance. The study also reveals that configurations with moderate gamma values achieve better balance between performance and computational time compared to higher gamma values or no gamma adjustment. The findings underscore the delicate balance between model performance and computational efficiency by highlighting the trade‐offs between model complexity and efficiency.

Geometrical and Textural Features Extraction for Honey Plants Pollen Recognition

The aim of the study is to investigate the extraction of features from microscopic images of honey plants pollen for classifying honey based on its botanical origin. Pollens from black locust, linden, lavender, canola and thistle are used. The color image of the pollen grain is converted into a gray image, from which classification features are extracted using popular texture recognition methods - Gabor filter, gray level co-occurrence matrix and local binary patterns. The extracted textural features are then processed with the principal component analysis method for dimensionality reduction and removal of correlated data. Geometric features related to the shape of the pollen grain are extracted from the binarized image. Through linear discriminant analysis, four classifiers are synthesized based on the textural and geometric features. To improve their performance, three hybrid structures mixing textural and geometric features are proposed. A comparative analysis of the performance of all seven linear classifiers is performed using a leave-one-out-cross-validation test. The best success rate obtained is 96%. The efficacy of the proposed algorithms is assessed through simulations conducted using the MATLAB programming language.

EF-net: Accurate edge segmentation for segmenting COVID-19 lung infections from CT images

Despite advances in modern medicine including the use of computed tomography for detecting COVID-19, precise identification and segmentation of lesions remain a significant challenge owing to indistinct boundaries and low degrees of contrast between infected and healthy lung tissues. This study introduces a novel model called the edge-based dual-parallel attention (EDA)-guided feature-filtering network (EF-Net), specifically designed to accurately segment the edges of COVID-19 lesions. The proposed model comprises two modules: an EDA module and a feature-filtering module (FFM). EDA efficiently extracts structural and textural features from low-level features, enabling the precise identification of lesion boundaries. FFM receives semantically rich features from a deep-level encoder and integrates features with abundant texture and contour information obtained from the EDA module. After filtering through a gating mechanism of the FFM, the EDA features are fused with deep-level features, yielding features rich in both semantic and textural information. Experiments demonstrate that our model outperforms existing models including Inf_Net, GFNet, and BSNet considering various metrics, offering better and clearer segmentation results, particularly for segmenting lesion edges. Moreover, superior performance on the three datasets is achieved, with dice coefficients of 98.1, 97.3, and 72.1 %.

Enhancing AppAuthentix recommender systems using advanced machine learning techniques to identify genuine and counterfeit android applications.

Smartphone app expansion needs strict security measures to avoid fraud and danger. This study overcomes this issue by identifying apps differently. This new solution uses convolutional neural network (CNN), natural language processing (NLP), and the strong AppAuthentix Recommender algorithm to secure app stores and boost customer confidence in the digital marketplace. Since the app ecosystem has grown, counterfeit and harmful applications have risen, threatening consumers and app merchants. These risks need advanced technology that can distinguish malware from legitimate apps. A complex prediction model using CNNs for image analysis, NLP for text feature extraction, and the novel AppAuthentix Recommender algorithm to properly identify legitimate and counterfeit mobile applications is the goal of this research. The whole strategy secures app stores and authenticates apps. The urgent need to safeguard app markets and users against unauthorized and hazardous programs sparked this study. Our cutting-edge solutions make mobile app consumers' digital lives safer and app marketplaces more trustworthy. CNN, NLP, and AppAuthentix Recommender yielded amazing results in this investigation. Mobile app authenticity may be estimated with 98.25% accuracy. This technology greatly improves app store security and enables mobile app verification. In conclusion, our work offers a novel way to app identification at a time of rapid mobile app development. CNN, NLP, and AppAuthentix Recommender have dramatically enhanced app store security. These new solutions may boost mobile app security and consumer confidence.

A Novel Method for Technology Roadmapping: Nanorobots

In the dynamic field of robotics engineering, nanorobot technology has witnessed rapid advancements. Developing a technology roadmap is essential for quickly identifying the trends and key technological aspects of nanorobotics from an array of multi-source data. Traditional research methods, such as Delphi surveys, bibliometrics, patent analysis, and patent paper citation analyses, often fail to capture the rich semantic information available. Moreover, these approaches generally provide a unidimensional perspective, which restricts their capacity to depict the complex nature of technological evolution. To overcome these shortcomings, this paper introduces a novel framework that utilizes the ALBERT method combined with multi-source data for critical theme extraction. It integrates varied data sources, including academic papers and patents, to explore the interrelation within the nanorobot technology roadmap. The methodology begins with text feature extraction, clustering algorithms, and theme mining to identify dominant technological themes. Subsequently, it applies semantic similarity measures to connect multiple themes, employing a “multi-layer ThemeRiver map” for a visual representation of these inter-layer connections. The paper concludes with a comprehensive analysis from both the technological research and industrial application perspectives, underscoring the principal developmental themes and insights of nanorobot technology, and projecting its future directions.

Magnetic Resonance Imaging Texture Analysis Based on Intraosseous and Extraosseous Lesions to Predict Prognosis in Patients with Osteosarcoma.

Objectives: To construct an optimal magnetic resonance imaging (MRI) texture model to evaluate histological patterns and predict prognosis in patients with osteosarcoma (OS). Methods: Thirty-four patients underwent pretreatment MRI and were diagnosed as having OS by surgical resection or biopsy between September 2008 and June 2018. Histological patterns and 3-year survival were recorded. Manual segmentation was performed in intraosseous, extraosseous, and entire lesions on T1-weighted, T2-weighted, and contrast-enhanced T1-weighted images to extract texture features and perform principal component analysis. A support vector machine algorithm with 3-fold cross-validation was used to construct and validate the models. The area under the receiver operating characteristic curve (AUC) was calculated to evaluate diagnostic performance in evaluating histological patterns and 3-year survival. Results: Eight patients were chondroblastic and the remaining twenty-six patients were non-chondroblastic patterns. Twenty-seven patients were 3-year survivors, and the remaining seven patients were non-survivors. In discriminating chondroblastic from non-chondroblastic patterns, the model from extraosseous lesions on the T2-weighted images showed the highest diagnostic performance (AUCs of 0.94 and 0.89 in the training and validation sets). The model from intraosseous lesions on the T1-weighted images showed the highest diagnostic performance in discriminating 3-year non-survivors from survivors (AUCs of 0.99 and 0.88 in the training and validation sets) with a sensitivity, specificity, positive predictive value, and negative predictive value of 85.7%, 92.6%, 75.0%, and 96.2%, respectively. Conclusions: The texture models of extraosseous lesions on T2-weighted images can discriminate the chondroblastic pattern from non-chondroblastic patterns, while the texture models of intraosseous lesions on T1-weighted images can discriminate 3-year non-survivors from survivors.