Kang An

Rice is a fundamental food source for more than half of the global population, making stable yields and quality improvements vital for food security and sustainable agricultural development. Early infections of rice leaf diseases often exhibit subtle symptoms, while conventional control methods based on empirical judgment and routine pesticide application result in both yield losses and environmental pollution. A Multi-scale closed-loop tuning via spatial frequency collaborative sensitivity (MCCA-YOLO) model has been proposed in this paper with a multiscale closed-loop tuning and spatial frequency collaborative attention mechanism for the early detection and classification of rice crop diseases. MCCA-YOLO incorporates a closed-loop tuning compound network architecture that combines a dual-backbone feature extractor with a spatial frequency enhancement module to achieve system self-verification feedback, reducing transmission errors and enhancing the texture features of leaves. The framework implements a cross-scale weighted fusion and a deformable spatial hybrid attention enhanced bidirectional feature pyramid fusion network for dynamic feature adaptation, effectively accommodating the complex morphology of rice leaf lesions. By conducting comprehensive ablation studies and comparative experiments with existing techniques on the rice plant diseases v8 dataset, the proposed approach achieves a mean average precision (mAP) of 92.2%, outperforming well-established methods, while delivering superior precision (0.915) and recall (0.900). Extensive empirical validation of additional v9 and Rice Leaf Spot Disease (RLSD) datasets for rice plant diseases further demonstrates the model’s outstanding performance.

As an interdisciplinary course, Machine Vision combines AI and digital image processing methods. This paper develops a comprehensive machine vision experiment on forest wildfire detection that organically integrates digital image processing, machine learning and deep learning technologies. Although the research on wildfire detection has made great progress, many experiments are not suitable for students to operate. Also, the detection with high accuracy is still a big challenge. In this paper, we divide the task of forest wildfire detection into two modules, which are wildfire image classification and wildfire region detection. We propose a novel wildfire image classification algorithm based on Reduce-VGGnet, and a wildfire region detection algorithm based on the optimized CNN with the combination of spatial and temporal features. The experimental results show that the proposed Reduce-VGGNet model can reach 91.20% in accuracy, and the optimized CNN model with the combination of spatial and temporal features can reach 97.35% in accuracy. Our framework is a novel way to combine research and teaching. It can achieve good detection performance and can be used as a comprehensive experiment for Machine Vision course, which can provide the support for talent cultivation in machine vision area.

COVID-19 has led to a severe impact on the society and healthcare system, with early diagnosis and effective treatment becoming critical. The Chest X-ray (CXR) is the most time-saving and cost-effective tool for diagnosing COVID-19. However, manual diagnosis through human eyes is time-consuming and tends to introduce errors. With the challenge of a large number of infections and a shortage of medical resources, a fast and accurate diagnosis technique is required. Manual detection is time-consuming, depends on individual experience, and tends to easily introduce errors. Deep learning methods can be used to develop automated detection and computer-aided diagnosis. However, they require a large amount of data, which is not practical due to the limited annotated CXR images. In this research, SDViT, an approach based on transformers, is proposed for COVID-19 diagnosis through image classification. We propose three innovations, namely, self-supervised learning, detail correction path (DCP), and domain transfer, then add them to the ViT Transformer architecture. Based on experimental results, our proposed method achieves an accuracy of 95.2381%, which is better performance compared to well-established methods on the X-ray Image dataset, along with the highest precision (0.952310), recall (0.963964), and F1-score (0.958102). Extensive experiments show that our model achieves the best performance on the synthetic-covid-cxr dataset as well. The experimental results demonstrate the advantages of our design for the classification task of COVID-19 X-ray images.

PCB (printed circuit board) is an extremely important component of all electronic products, which has greatly facilitated human life. Meanwhile, tons of PCBs in the waste streams become a waste of resources, which puts the recycling and reuse of PCBs in urgent need. In the manufacturing and recycling of electronic products, the classification of PCBs, recognition of sub-components, and defect detection have been the key technology. Traditional manual detection and classification are subjective and rely on individuals’ experience. With the development of artificial intelligence, lots of research efforts have been dedicated to the automated detection and recognition of PCBs. In this paper, we propose a transformer-based model, LPViT, for defect detection and classification of PCBs. We conduct the defect detection task on the dataset DeepPCB, which consists of six different types of PCB defects. Defect detection benefits both manufacturing and recycling of PCBs. Among many electronic products, a group of affordable, general-purpose, and small-size PCBs is very popular, which are referred to as micro-PCBs. The classification and recognition of those PCBs will greatly facilitate the recycling and reuse process. We conduct the classification task on a dataset called micro-PCB , which includes 12 types of popular, general-purpose, affordable, and small PCBs. Through comparative experiments, our system demonstrates its advantage in both classification and defect detection tasks.

3D Object detection is a critical mission of the perception system of a self-driving vehicle. Existing bounding box-based methods are hard to train due to the need to remove duplicated detections in the post-processing stage. In this paper, we propose a center point-based deep neural network (DNN) architecture named RCBi-CenterNet that predicts the absolute pose for each detected object in the 3D world space. RCBi-CenterNet is composed of a recursive composite network with a dual-backbone feature extractor and a bi-directional feature pyramid network (BiFPN) for cross-scale feature fusion. In the detection head, we predict a confidence heatmap that is used to determine the position of detected objects. The other pose information, including depth and orientation, is regressed. We conducted extensive experiments on the Peking University/Baidu-Autonomous Driving dataset, which contains more than 60,000 labeled 3D vehicle instances from 5277 real-world images, and each vehicle object is annotated with the absolute pose described by the six degrees of freedom (6DOF). We validated the design choices of various data augmentation methods and the backbone options. Through an ablation study and an overall comparison with the state-of-the-art (SOTA), namely CenterNet, we showed that the proposed RCBi-CenterNet presents performance gains of 2.16%, 2.76%, and 5.24% in Top 1, Top 3, and Top 10 mean average precision (mAP). The model and the result could serve as a credible benchmark for future research in center point-based object detection.