To address the challenge of detecting small defects in Printed Circuit Boards (PCBs), a YOLO-HDEW model based on the enhanced YOLOv8 architecture is proposed. A high-resolution detection layer is introduced at the P2 feature level to improve sensitivity to small targets. Depthwise Separable Convolution (DSConv) is used for downsampling, reducing parameter complexity. An Edge-enhanced Multi-scale Parallel Attention mechanism (EMP-Attention) is integrated to capture multi-scale and edge features. The EMP mechanism is incorporated into the C2f module to form the C2f-EMP module, and dynamic non-monotonic Wise-IoU (W-IoU) loss is employed to enhance bounding box regression. The model is evaluated on the PKU-Market-PCB, DeepPCB, and NEU-DET datasets, with experimental results showing that YOLO-HDEW achieves 98.1% accuracy, 91.6% recall, 90.3% mAP@0.5, and 61.7% mAP@0.5:0.95, surpassing YOLOv8 by 1.5%, 2.3%, 1.2%, and 1.9%, respectively. Additionally, the model demonstrates strong generalization performance on the DeePCB and NEU-DET datasets. These results indicate that YOLO-HDEW significantly improves detection accuracy while maintaining a manageable model size, offering an effective solution for PCB defect detection.

The increasing demands for high-precision semantic segmentation in applications such as autonomous driving, unmanned aerial vehicles, and robotics has made improving segmentation accuracy a major research focus. In this paper, we propose MPNet, (multi-scale progressive network) a novel semantic segmentation model based on the DeepLabV3+ architecture. First, a lightweight MobileNetV2 was employed as the backbone network, and a new multi-scale feature fusion structure was constructed by integrating the backbone with the centralized feature pyramid network (CFPNet). Then, based on the ASPP module, a progressive dilated atrous spatial pyramid pooling (PDASPP) module was designed to further enhance feature extraction. Extensive experiments were conducted on the Cityscapes and PASCAL VOC 2012 Augmented. The experimental results show that MPNet achieved 75.23% mIoU and 83.54% mPA on the Cityscapes dataset, outperforming DeepLabV3+ by 3.01% and 3.31%, respectively. On the PASCAL VOC 2012 Augmented dataset, MPNet achieved 72.70% mIoU and 83.57% mPA, with improvements of 1.59% and 1.30% over DeepLabV3+, respectively. These results demonstrate that MPNet significantly improves segmentation accuracy while maintaining model complexity under control, providing an effective solution for semantic segmentation in road scene understanding.