This article utilizes edge oriented convolution modules,pixel attention mechanisms,and reparameterization tech-niques to improve the image resolution of the super-resolution reconstruction algorithm,resulting in superior representation of image feature details.Utilizing the YOLOv4 object detection algorithm and combining the Focus structure,bidirectional feature pyramid network,and lightweight sub channel attention mechanism,the accuracy of object detection in medium and low resolu-tion images is improved.Through experimental research,the lightweight object detection algorithm based on super-resolution reconstruction has a good detection effect on image targets,effectively improves the detection accuracy of images,and has cer-tain reference significance for improving the detection accuracy of small targets in images.
维普
万方数据