ARG-Mask RCNN: An Infrared Insulator Fault-Detection Network Based on Improved Mask RCNN

Lightweight Transmission Line Fault Detection Method Based on Leaner YOLOv7-Tiny

Aiming to address the issues of parameter complexity and high computational load in existing fault detection algorithms for transmission lines, which hinder their deployment on devices like drones, this study proposes a novel lightweight model called Leaner YOLOv7-Tiny. The primary goal is to swiftly and accurately detect typical faults in transmission lines from aerial images. This algorithm inherits the ELAN structure from YOLOv7-Tiny network and replaces its backbone with depthwise separable convolutions to reduce model parameters. By integrating the SP attention mechanism, it fuses multi-scale information, capturing features across various scales to enhance small target recognition. Finally, an improved FCIoU Loss function is introduced to balance the contribution of high-quality and low-quality samples to the loss function, expediting model convergence and boosting detection accuracy. Experimental results demonstrate a 20% reduction in model size compared to the original YOLOv7-Tiny algorithm. Detection accuracy for small targets surpasses that of current mainstream lightweight object detection algorithms. This approach holds practical significance for transmission line fault detection.

Insulator Defect Detection Based on ML-YOLOv5 Algorithm

To address the challenges of balancing accuracy and speed, as well as the parameters and FLOPs in current insulator defect detection, we propose an enhanced insulator defect detection algorithm, ML-YOLOv5, based on the YOLOv5 network. The backbone module incorporates depthwise separable convolution, and the feature fusion C3 module is replaced with the improved C2f_DG module. Furthermore, we enhance the feature pyramid network (MFPN) and employ knowledge distillation using YOLOv5m as the teacher model. Experimental results demonstrate that this approach achieved a 46.9% reduction in parameter count and a 43.0% reduction in FLOPs, while maintaining an FPS of 63.6. It exhibited good accuracy and detection speed on both the CPLID and IDID datasets, making it suitable for real-time inspection of high-altitude insulator defects.

An Improved Mask R-CNN Micro-Crack Detection Model for the Surface of Metal Structural Parts

Micro-crack detection is an essential task in critical equipment health monitoring. Accurate and timely detection of micro-cracks can ensure the healthy and stable service of equipment. Aiming at improving the low accuracy of the conventional target detection model during the task of detecting micro-cracks on the surface of metal structural parts, this paper built a micro-cracks dataset and explored a detection performance optimization method based on Mask R-CNN. Firstly, we improved the original FPN structure, adding a bottom-up feature fusion path to enhance the information utilization rate of the underlying feature layer. Secondly, we added the methods of deformable convolution kernel and attention mechanism to ResNet, which can improve the efficiency of feature extraction. Lastly, we modified the original loss function to optimize the network training effect and model convergence rate. The ablation comparison experiments shows that all the improvement schemes proposed in this paper have improved the performance of the original Mask R-CNN. The integration of all the improvement schemes can produce the most significant performance improvement effects in recognition, classification, and positioning simultaneously, thus proving the rationality and feasibility of the improved scheme in this paper.

Coarse-to-Fine Localization for Detecting Misalignment State of Angle Cocks

The state of angle cocks determines the air connectivity of freight trains, and detecting their state is helpful to improve the safety of the running trains. Although the current research for fault detection of angle cocks has achieved high accuracy, it only focuses on the detection of the closed state and non-closed state and treats them as normal and abnormal states, respectively. Since the non-closed state includes the fully open state and the misalignment state, while the latter may lead to brake abnormally, it is very necessary to further detect the misalignment state from the non-closed state. In this paper, we propose a coarse-to-fine localization method to achieve this goal. Firstly, the localization result of an angle cock is obtained by using the YOLOv4 model. Following that, the SVM model combined with the HOG feature of the localization result of an angle cock is used to further obtain its handle localization result. After that, the HOG feature of the sub-image only containing the handle localization result continues to be used in the SVM model to detect whether the angle cock is in the non-closed state or not. When the angle cock is in the non-closed state, its handle curve is fitted by binarization and window search, and the tilt angle of the handle is calculated by the minimum bounding rectangle. Finally, the misalignment state is detected when the tilt angle of the handle is less than the threshold. The effectiveness and robustness of the proposed method are verified by extensive experiments, and the accuracy of misalignment state detection for angle cocks reaches 96.49%.

Deep Learning-Based Modified YOLACT Algorithm on Magnetic Resonance Imaging Images for Screening Common and Difficult Samples of Breast Cancer

Computer-aided methods have been extensively applied for diagnosing breast lesions with magnetic resonance imaging (MRI), but fully-automatic diagnosis using deep learning is rarely documented. Deep-learning-technology-based artificial intelligence (AI) was used in this work to classify and diagnose breast cancer based on MRI images. Breast cancer MRI images from the Rider Breast MRI public dataset were converted into processable joint photographic expert group (JPG) format images. The location and shape of the lesion area were labeled using the Labelme software. A difficult-sample mining mechanism was introduced to improve the performance of the YOLACT algorithm model as a modified YOLACT algorithm model. Diagnostic efficacy was compared with the Mask R-CNN algorithm model. The deep learning framework was based on PyTorch version 1.0. Four thousand and four hundred labeled data with corresponding lesions were labeled as normal samples, and 1600 images with blurred lesion areas as difficult samples. The modified YOLACT algorithm model achieved higher accuracy and better classification performance than the YOLACT model. The detection accuracy of the modified YOLACT algorithm model with the difficult-sample-mining mechanism is improved by nearly 3% for common and difficult sample images. Compared with Mask R-CNN, it is still faster in running speed, and the difference in recognition accuracy is not obvious. The modified YOLACT algorithm had a classification accuracy of 98.5% for the common sample test set and 93.6% for difficult samples. We constructed a modified YOLACT algorithm model, which is superior to the YOLACT algorithm model in diagnosis and classification accuracy.