Image Features Research Articles

MM-IRSTD: Conv Self-Attention-Based Multi-Modal Small and Dim Target Detection in Infrared Dual-Band Images

Infrared multi-band small and dim target detection is an important research direction in the fields of modern remote sensing and military surveillance. However, achieving high-precision detection remains challenging due to the small scale, low contrast of small and dim targets, and their susceptibility to complex background interference. This paper innovatively proposes a dual-band infrared small and dim target detection method (MM-IRSTD). In this framework, we integrate a convolutional self-attention mechanism module and a self-distillation mechanism to achieve end-to-end dual-band infrared small and dim target detection. The Conv-Based Self-Attention module consists of a convolutional self-attention mechanism and a multilayer perceptron, effectively extracting and integrating input features, thereby enhancing the performance and expressive capability of the model. Additionally, this module incorporates a dynamic weight mechanism to achieve adaptive feature fusion, significantly reducing computational complexity and enhancing the model’s global perception capability. During model training, we use a spatial and channel similarity self-distillation mechanism to drive model updates, addressing the similarity discrepancy between long-wave and mid-wave image features extracted through deep learning, thus improving the model’s performance and generalization capability. Furthermore, to better learn and detect edge features in images, this paper designs an edge extraction method based on Sobel. Finally, comparative experiments and ablation studies validate the advancement and effectiveness of our proposed method.

Deep Learning-Driven Sign Language Recognition: A Multimodal Approach for Gesture-to-Text Translation Using CNN-RNN Architectures

Sign language serves as a vital communication tool for individuals with hearing and speech impairments. However, a significant barrier exists when others do not possess a strong understanding of sign language often requiring interpreters to facilitate communication. To reduce the reliance on human interpreters, this research aims to develop an intelligent system capable of recognizing and translating sign language gestures into meaningful grammatically correct sentences. The proposed system processes both images and videos to interpret gestures, converting them into text that can be translated into any desired language. The system leverages state-of-the-art deep learning techniques, such as Convolutional Neural Networks (CNNs) for identifying key features in images and Recurrent Neural Networks (RNNs) for understanding temporal sequences in video data. By employing these advanced neural networks, the model is able to comprehend hand movements, facial expressions and other non-verbal cues to construct coherent sentences. Additionally, the system integrates natural language processing (NLP) to refine the output, ensuring the resulting sentences are grammatically correct. Our approach addresses common challenges such as differentiating between subtle hand gestures and reducing the impact of environmental noise in images. This solution holds the potential to significantly enhance communication for the hearing and speech impairedffering an efficient interpreter-free method of translating sign language into widely spoken languages

Enhanced stimulated Raman and fluorescence imaging by single-frame trained BDN

Hyperspectral and multispectral imaging capture an expanded dimension of information that facilitates discoveries. However, image features are frequently obscured by noise generated from the limited photodamage threshold of the specimen. Although machine learning approaches demonstrate considerable promise in addressing this challenge, they typically require extensive datasets, which can be difficult to obtain. Here, we introduce BiFormer denoising network (BDN), designed to effectively and efficiently extract image features by utilizing both local and global level connections, sparse architectures, and fine-tuning. Experimental results indicate that BDN enhances the quality of stimulated Raman scattering (SRS) images by up to 16-fold in signal-to-noise ratio (SNR), particularly improving subtle features at higher spatial frequencies. Furthermore, BDN is successfully adapted to fluorescence imaging, achieving significant improvements in SNR and order-of-magnitude reduction in exposure time, thereby showcasing its versatility across various imaging modalities. Collectively, BDN exhibits substantial potential for spectroscopic imaging applications in the fields of biomedicine and materials science.

ViT-HHO: Optimized vision transformer for diabetic retinopathy detection using Harris Hawk optimization

Diabetic retinopathy (DR) is a significant cause of vision impairment globally, emphasizing the importance of timely and precise detection to prevent severe consequences. This study presents an optimized Vision Transformer (ViT) model that incorporates Harris Hawk Optimization (HHO) to improve the automated detection of diabetic retinopathy (DR). The ViT architecture utilizes self-attention mechanisms to capture local and global features in retinal images. Additionally, HHO optimizes key hyperparameters to maximize the performance of the model. The proposed ViT-HHO model achieved exceptional performance on the APTOS-2019 and IDRiD datasets. Specifically, it achieved 99.83 % accuracy, 99.78 % sensitivity, 99.85 % specificity, and 99.80 % AUC-ROC on the APTOS-2019 dataset, surpassing traditional CNNs and alternative optimization techniques. The model exhibited strong generalization on the IDRiID dataset, achieving an accuracy of 99.11 % and an AUC-ROC of 99.12 %. The ViT-HHO model demonstrates the potential for enhancing the clinical detection of diabetic retinopathy (DR), providing high precision and reliability.•An optimized Vision Transformer (ViT) model was developed using HHO for improved detection of Diabetic Retinopathy (DR).•The model was validated on the APTOS-2019 and IDRiID datasets, demonstrating superior accuracy and AUC-ROC metrics.•The model's generalization and robustness were demonstrated through comprehensive performance evaluations.

Research on Abnormal State Detection of CZ Silicon Single Crystal Based on Multimodal Fusion

The Czochralski method is the primary technique for single-crystal silicon production. However, anomalous states such as crystal loss, twisting, swinging, and squareness frequently occur during crystal growth, adversely affecting product quality and production efficiency. To address this challenge, we propose an enhanced multimodal fusion classification model for detecting and categorizing these four anomalous states. Our model initially transforms one-dimensional signals (diameter, temperature, and pulling speed) into time–frequency domain images via continuous wavelet transform. These images are then processed using a Dense-ECA-SwinTransformer network for feature extraction. Concurrently, meniscus images and inter-frame difference images are obtained from the growth system’s meniscus video feed. These visual inputs are fused at the channel level and subsequently processed through a ConvNeXt network for feature extraction. Finally, the time–frequency domain features are combined with the meniscus image features and fed into fully connected layers for multi-class classification. The experimental results show that the method can effectively detect various abnormal states, help the staff to make a more accurate judgment, and formulate a personalized treatment plan for the abnormal state, which can improve the production efficiency, save production resources, and protect the extraction equipment.

Robotic MAG welding defects and quality assessment with a defect threshold decision model-driven method

The use of machine vision and deep learning methods for online monitoring and evaluation of welding quality during the welding process and timely detection and correction of welding defects is essential for intelligent welding manufacturing. In this paper, with the goal of online monitoring of the welding quality of the robotic arc welding for the excavator boom, we established a multi-source sensing system for robotic arc welding processes. We have developed various feature extraction algorithms for extracting information about welding pools, arc sounds, currents, voltages, and other features. We proposed to extract welding pool image features using Res2-MobileNetV3 to obtain a 12-dimensional welding pool feature, which enabled good inter-class discrimination among different welding defects. We designed a welding defect threshold decision (DD) strategy model using the responsive machine learning and lightweight network Res2-MobileNetV3 method, achieving a welding defect recognition accuracy of 96.59%. Finally, we tested the accuracy and reliability of the welding defect recognition model on an actual welding scene for the robotic welding of the excavator boom. It provided valuable scientific methods and technological approaches for intelligent welding in complex scenes.

Skin Cancer Image Classification Using Artificial Intelligence Strategies: A Systematic Review

The increasing incidence of and resulting deaths associated with malignant skin tumors are a public health problem that can be minimized if detection strategies are improved. Currently, diagnosis is heavily based on physicians’ judgment and experience, which can occasionally lead to the worsening of the lesion or needless biopsies. Several non-invasive imaging modalities, e.g., confocal scanning laser microscopy or multiphoton laser scanning microscopy, have been explored for skin cancer assessment, which have been aligned with different artificial intelligence (AI) strategies to assist in the diagnostic task, based on several image features, thus making the process more reliable and faster. This systematic review concerns the implementation of AI methods for skin tumor classification with different imaging modalities, following the PRISMA guidelines. In total, 206 records were retrieved and qualitatively analyzed. Diagnostic potential was found for several techniques, particularly for dermoscopy images, with strategies yielding classification results close to perfection. Learning approaches based on support vector machines and artificial neural networks seem to be preferred, with a recent focus on convolutional neural networks. Still, detailed descriptions of training/testing conditions are lacking in some reports, hampering reproduction. The use of AI methods in skin cancer diagnosis is an expanding field, with future work aiming to construct optimal learning approaches and strategies. Ultimately, early detection could be optimized, improving patient outcomes, even in areas where healthcare is scarce.

Non-Destructive Monitoring of Sweet Pepper Samples After Selected Periods of Lacto-Fermentation

Fermented food is characterized by positive health-promoting properties. The objective of this study was to distinguish and assess the changes in the flesh structure of sweet bell pepper samples after specific periods of fermentation in a non-destructive manner. Two cultivars of pepper, red and yellow, were subjected to lacto-fermentation. The experiments lasted 56 days and the samples were taken for analysis at the beginning of the study (0 days) and after 3, 7, 10, 14, 21, 28, and 56 days. The fermentation process was monitored based on image features, which were used to develop machine learning models distinguishing samples before and after various periods of lacto-fermentation (0, 3, 7, 10, 14, 21, 28, and 56 days). The average accuracy of the classification of red bell pepper samples was up to 93% for the model built using IBk (Lazy group). The yellow bell pepper samples were distinguished up to 90% accuracy by the LMT algorithm (Trees group). The performed study allowed us to determine the changes in pepper flesh in terms of image textures during lacto-fermentation.

Unlocking the potential of simulated hyperspectral imaging in agro environmental analysis: a comprehensive study of algorithmic approaches

This study focuses on identifying and evaluating the severity of powdery mildew disease in tomato plants. The uniqueness of this work lies in combining the imaging and advanced deep learning methods to develop a technique that transforms Red Green Blue (RGB) images into Simulated Hyperspectral Images (SHSI) to perform spectral and spatial analysis for precise detection and assessment of powdery mildew severity, thereby enhancing disease management. Furthermore, this research evaluates three advanced pre-trained VGG16 models, ResNet50 and EfficientNet-B7 algorithms for image preprocessing and feature extraction. Extracted features are passed to a neural network generator model to convert RGB image features into SHSIs, providing insights into the spectrum. This method enables the image analysis to perform assessments from SHSIs for health classification using Normalized Difference Vegetation Index (NDVI) values, which are meticulously compared with accurate hyperspectral data using metrics like mean absolute error (MAE) and root mean squared error (RMSE). This strategy enhances precision farming, environmental monitoring, and remote sensing accuracy. Results show that ResNet50’s architecture offers a robust framework for this study’s spectral and spatial analysis, making it a suitable choice over VGG16 and EfficientNet-B7 for transforming RGB images into SHSI. These simulated hyperspectral images offer a scalable and affordable approach for precise assessment of crop disease severity.

A novel monitoring method based on the fusion of sound and image signals for laser welding penetration status

Accurately identifying penetration status is crucial for ensuring high-quality welding. However, relying on a single sensor to monitor the welding status falls short in accuracy. Thus, a novel monitoring method that fuses sound and image is proposed for assessing the laser welding penetration status. First, a microphone and a high-speed camera are employed to collect sound and image signals during welding. Second, to overcome noise challenges in sound signal, a denoising method is proposed that utilizes Grey Wolf Optimization (GWO) to find the optimal parameters for Variational Mode Decomposition (VMD). The time and frequency domain features from the denoised sound and Local Binary Pattern (LBP) features from the image are extracted. These features constitute a high-dimensional dataset, which is reduced using Principal Component Analysis (PCA). Finally, a back propagation neural network (BPNN) is constructed for fusion monitoring. The results demonstrate that fusion monitoring achieves higher accuracy (97.9%) compared to using sound (79.9%) or image (92.3%) features independently. In comparison to other methods like K-Nearest Neighbor (KNN), Naive Bayesian (NB), Support Vector Machine (SVM), and Decision Tree, BPNN fusion monitoring achieves higher accuracies, with increases of 18.09%, 12.79%, 6.53%, and 25.84% respectively. The proposed method helps to promote the development of welding intelligence.

An Improved Random Forest Approach on GAN-Based Dataset Augmentation for Fog Observation

The monitoring of fog density is of great importance in meteorology and its applications in environment, aviation and transportation. Nowadays, vision-based fog estimation from images taken with surveillance cameras has made a great supplementary contribution to the scarcely traditional meteorological fog observation. In this paper, we propose a new Random Forest (RF) approach for image-based fog estimation. In order to reduce the impact of data imbalance on recognition, the StyleGAN2-ADA (generative adversarial network with adaptive discriminator augmentation) algorithm is used to generate virtual images to expand the data of low proportions. Key image features related to fog are extracted, and an RF method, integrated with the hierarchical and k-medoid clustering, is deployed to estimate the fog density. The experiment conducted in Sichuan in February 2024 shows that the improved RF model has achieved an average accuracy of fog density observation of 93%, 6.4% higher than the RF model without data expansion, 3–6% higher than the VGG16, the VGG19, the ResNet50, and the DenseNet169 with or without data expansion. What is more, the improved RF method exhibits a very good convergence as a cost-effective solution.

Perceptual super-resolution in multiple sclerosis MRI.

Magnetic resonance imaging (MRI) is crucial for diagnosing and monitoring of multiple sclerosis (MS) as it is used to assess lesions in the brain and spinal cord. However, in real-world clinical settings, MRI scans are often acquired with thick slices, limiting their utility for automated quantitative analyses. This work presents a single-image super-resolution (SR) reconstruction framework that leverages SR convolutional neural networks (CNN) to enhance the through-plane resolution of structural MRI in people with MS (PwMS). Our strategy involves the supervised fine-tuning of CNN architectures, guided by a content loss function that promotes perceptual quality, as well as reconstruction accuracy, to recover high-level image features. Extensive evaluation with MRI data of PwMS shows that our SR strategy leads to more accurate MRI reconstructions than competing methods. Furthermore, it improves lesion segmentation on low-resolution MRI, approaching the performance achievable with high-resolution images. Results demonstrate the potential of our SR framework to facilitate the use of low-resolution retrospective MRI from real-world clinical settings to investigate quantitative image-based biomarkers of MS.

DPNet: Scene text detection based on dual perspective CNN-transformer.

With the continuous advancement of deep learning, research in scene text detection has evolved significantly. However, complex backgrounds and various text forms complicate the task of detecting text from images. CNN is a deep learning algorithm that automatically extracts features through convolution operation. In the task of scene text detection, it can capture local text features in images, but it lacks global attributes. In recent years, inspired by the application of transformers in the field of computer vision, it can capture the global information of images and describe them intuitively. Therefore, this paper proposes scene text detection based on dual perspective CNN-transformer. The channel enhanced self-attention module (CESAM) and spatial enhanced self-attention module (SESAM) proposed in this paper are integrated into the traditional ResNet backbone network. This integration effectively facilitates the learning of global contextual information and positional relationships of text, thereby alleviating the challenge of detecting small target text. Furthermore, this paper introduces a feature decoder designed to refine the effective text information within the feature map and enhance the perception of detailed information. Experiments show that the method proposed in this paper significantly improves the robustness of the model for different types of text detection. Compared to the baseline, it achieves performance improvements of 2.51% (83.81 vs. 81.3) on the Total-Text dataset, 1.87% (86.07 vs. 84.2) on the ICDAR 2015 dataset, and 3.63% (86.72 vs. 83.09) on the MSRA-TD500 dataset, while also demonstrating better visual effects.

Study on the classification of benign and malignant breast lesions using a multi-sequence breast MRI fusion radiomics and deep learning model

PurposeTo develop a multi-modal model combining multi-sequence breast MRI fusion radiomics and deep learning for the classification of benign and malignant breast lesions, to assist clinicians in better selecting treatment plans. MethodsA total of 314 patients who underwent breast MRI examinations were included. They were randomly divided into training, validation, and test sets in a ratio of 7:1:2. Subsequently, features of T1-weighted images (T1WI), T2-weighted images (T2WI), and dynamic contrast-enhanced MRI (DCE-MRI) were extracted using the convolutional neural network ResNet50 for fusion, and then combined with radiomic features from the three sequences. The following models were established: T1 model, T2 model, DCE model, DCE_T1_T2 model, and DCE_T1_T2_rad model. The performance of the models was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. The differences between the DCE_T1_T2_rad model and the other four models were compared using the Delong test, with a P-value < 0.05 considered statistically significant. ResultsThe five models established in this study performed well, with AUC values of 0.53 for the T1 model, 0.62 for the T2 model, 0.79 for the DCE model, 0.94 for the DCE_T1_T2 model, and 0.98 for the DCE_T1_T2_rad model. The DCE_T1_T2_rad model showed statistically significant differences (P < 0.05) compared to the other four models. ConclusionThe use of a multi-modal model combining multi-sequence breast MRI fusion radiomics and deep learning can effectively improve the diagnostic performance of breast lesion classification.

Noise-assisted hybrid attention networks for low-dose PET and CT denoising.

Positron emission tomography (PET) and computed tomography (CT) play a vital role in tumor-related medical diagnosis, assessment, and treatment planning. However, full-dose PET and CT pose the risk of excessive radiation exposure to patients, whereas low-dose images compromise image quality, impacting subsequent tumor recognition and diseasediagnosis. To solve such problems, we propose a Noise-Assisted Hybrid Attention Network (NAHANet) to reconstruct full-dose PET and CT images from low-dose PET (LDPET) and CT (LDCT) images to reduce patient radiation risks while ensuring the performance of subsequent tumor recognition. NAHANet contains two branches: the noise feature prediction branch (NFPB) and the cascaded reconstruction branch. Among them, NFPB providing noise features for the cascade reconstruction branch. The cascaded reconstruction branch comprises a shallow feature extraction module and a reconstruction module which contains a series of cascaded noise feature fusion blocks (NFFBs). Among these, the NFFB fuses the features extracted from low-dose images with the noise features obtained by NFPB to improve the feature extraction capability. To validate the effectiveness of the NAHANet method, we performed experiments using two public available datasets: the Ultra-low Dose PET Imaging Challenge dataset and Low Dose CT Grand Challengedataset. As a result, the proposed NAHANet achieved higher performance on common indicators. For example, on the CT dataset, the PSNR and SSIM indicators were improved by 4.1dB and 0.06 respectively, and the rMSE indicator was reduced by 5.46 compared with the LDCT; on the PET dataset, the PSNR and SSIM was improved by 3.37dB and 0.02, and the rMSE was reduced by 9.04 compared with theLDPET. This paper proposes a transformer-based denoising algorithm, which utilizes hybrid attention to extract high-level features of low dose images and fuses noise features to optimize the denoising performance of the network, achieving good performance improvements on low-dose CT and PETdatasets.

Concatenated CNN-Based Pneumonia Detection Using a Fuzzy-Enhanced Dataset.

Pneumonia is a form of acute respiratory infection affecting the lungs. Symptoms of viral and bacterial pneumonia are similar. Rapid diagnosis of the disease is difficult, since polymerase chain reaction-based methods, which have the greatest reliability, provide results in a few hours, while ensuring high requirements for compliance with the analysis technology and professionalism of the personnel. This study proposed a Concatenated CNN model for pneumonia detection combined with a fuzzy logic-based image improvement method. The fuzzy logic-based image enhancement process is based on a new fuzzification refinement algorithm, with significantly improved image quality and feature extraction for the CCNN model. Four datasets, original and upgraded images utilizing fuzzy entropy, standard deviation, and histogram equalization, were utilized to train the algorithm. The CCNN's performance was demonstrated to be significantly improved by the upgraded datasets, with the fuzzy entropy-added dataset producing the best results. The suggested CCNN attained remarkable classification metrics, including 98.9% accuracy, 99.3% precision, 99.8% F1-score, and 99.6% recall. Experimental comparisons showed that the fuzzy logic-based enhancement worked significantly better than traditional image enhancement methods, resulting in higher diagnostic precision. This study demonstrates how well deep learning models and sophisticated image enhancement techniques work together to analyze medical images.

Multi-Modal Object Detection Method Based on Dual-Branch Asymmetric Attention Backbone and Feature Fusion Pyramid Network

With the simultaneous acquisition of the infrared and optical remote sensing images of the same target becoming increasingly easy, using multi-modal data for high-performance object detection has become a research focus. In remote sensing multi-modal data, infrared images lack color information, it is hard to detect difficult targets with low contrast, and optical images are easily affected by illuminance. One of the most effective ways to solve this problem is to integrate multi-modal images for high-performance object detection. The challenge of fusion object detection lies in how to fully integrate multi-modal image features with significant modal differences and avoid introducing interference information while taking advantage of complementary advantages. To solve these problems, a new multi-modal fusion object detection method is proposed. In this paper, the method is improved in terms of two aspects: firstly, a new dual-branch asymmetric attention backbone network (DAAB) is designed, which uses a semantic information supplement module (SISM) and a detail information supplement module (DISM) to supplement and enhance infrared and RGB image information, respectively. Secondly, we propose a feature fusion pyramid network (FFPN), which uses a Transformer-like strategy to carry out multi-modal feature fusion and suppress features that are not conducive to fusion during the fusion process. This method is a state-of-the-art process for both FLIR-aligned and DroneVehicle datasets. Experiments show that this method has strong competitiveness and generalization performance.

Grade Classification of Camellia Seed Oil Based on Hyperspectral Imaging Technology.

To achieve the rapid grade classification of camellia seed oil, hyperspectral imaging technology was used to acquire hyperspectral images of three distinct grades of camellia seed oil. The spectral and image information collected by the hyperspectral imaging technology was preprocessed by different methods. The characteristic wavelength selection in this study included the continuous projections algorithm (SPA) and competitive adaptive reweighted sampling (CARS), and the gray-level co-occurrence matrix (GLCM) algorithm was used to extract the texture features of camellia seed oil at the characteristic wavelength. Combined with genetic algorithm (GA) and support vector machine algorithm (SVM), different grade classification models for camellia seed oil were developed using full wavelengths (GA-SVM), characteristic wavelengths (CARS-GA-SVM), and fusing spectral and image features (CARS-GLCM-GA-SVM). The results show that the CARS-GLCM-GA-SVM model, which combined spectral and image information, had the best classification effect, and the accuracy of the calibration set and prediction set of the CARS-GLCM-GA-SVM model were 98.30% and 96.61%, respectively. Compared with the CARS-GA-SVM model, the accuracy of the calibration set and prediction set were improved by 10.75% and 12.04%, respectively. Compared with the GA-SVM model, the accuracy of the calibration set and prediction set were improved by 18.28% and 18.15%, respectively. The research showed that hyperspectral imaging technology can rapidly classify camellia seed oil grades.

Need of Feature Extraction in Coconut Tree Disease Detection: A Review

Agriculture has served as the primary means of sustenance for mankind for thousands of years. Even in the present day, it continues to support approximately 50% of the global population. Plant diseases pose a significant threat to crop production, resulting in substantial losses each year on a global scale. To mitigate the financial impact of these diseases, it is crucial to maintain the health of plants throughout their growth and development. The symptoms of infections are predominantly visible on plant leaves, making them a common indicator for disease detection and identification. However, visually observing and identifying diseases is a challenging task that requires extensive human expertise. To provide farmers with improved assistance in disease detection, image processing techniques combined with computational intelligence or soft computing techniques can be employed. These methods offer a more effective approach to disease detection. By extracting features from the symptoms of a disease, it becomes possible to identify and detect the presence of a disease in plants. Therefore, feature extraction techniques play a vital role in such systems. This paper focuses on reviewing the feature extraction methods, highlighting their advantages and disadvantages. It provides a comprehensive discussion on various image features, including color, texture, and shape, for different types of disorders found in diverse agricultural practices.

Unsafe behavior identification on construction sites by combining computer vision and knowledge graph–based reasoning

PurposeWorkers' unsafe behavior is the main cause of construction safety accidents, thereby highlighting the critical importance of behavior-based management. To compensate for the limitations of computer vision in tackling knowledge-intensive issues, semantic-based methods have gained increasing attention in the field of construction safety management. Knowledge graph provides an efficient and visualized method for the identification of various unsafe behaviors.Design/methodology/approachThis study proposes an unsafe behavior identification framework by integrating computer vision and knowledge graph–based reasoning. An enhanced ontology model anchors our framework, with image features from YOLOv5, COCO Panoptic Segmentation and DeepSORT integrated into the graph database, culminating in a structured knowledge graph. An inference module is also developed, enabling automated the extraction of unsafe behavior knowledge through rule-based reasoning.FindingsA case application is implemented to demonstrate the feasibility and effectiveness of the proposed method. Results show that the method can identify various unsafe behaviors from images of construction sites and provide mitigation recommendations for safety managers by automated reasoning, thus supporting on-site safety management and safety education.Originality/valueExisting studies focus on spatial relationships, often neglecting the diversified spatiotemporal information in images. Besides, previous research in construction safety only partially automated knowledge graph construction and reasoning processes. In contrast, this study constructs an enhanced knowledge graph integrating static and dynamic data, coupled with an inference module for fully automated knowledge-based unsafe behavior identification. It can help managers grasp the workers’ behavior dynamics and timely implement measures to correct violations.