Small object Lentinula Edodes logs contamination detection method based on improved YOLOv7 in edge-cloud computing

Abstract

A small object Lentinus Edodes logs contamination detection method (SRW-YOLO) based on improved YOLOv7 in edge-cloud computing environment was proposed to address the problem of the difficulty in the detection of small object contaminated areas of Lentinula Edodes logs. First, the SPD (space-to-depth)-Conv was used to reconstruct the MP module to enhance the learning of effective features of Lentinula Edodes logs images and prevent the loss of small object contamination information, and improve the detection reliability of resource-limited edge devices. Meanwhile, RepVGG was introduced into the ELAN structure to improve the efficiency and accuracy of inference on the contaminated regions of Lentinula Edodes logs through structural reparameterization. This enables models to run more efficiently in mobile edge computing environments while reducing the burden on cloud computing servers. Finally, the boundary regression loss function was replaced with the WIoU (Wise-IoU) loss function, which focuses more on ordinary-quality anchor boxes and makes the model output results more accurate. In this study, the measures of Precision, Recall, and mAP@0.5 reached 97.63%, 96.43%, and 98.62%, respectively, which are 4.62%, 3.63%, and 2.31% higher compared to those for YOLOv7. Meanwhile, the SRW-YOLO model detects better compared with the current advanced one-stage object detection model, providing an efficient, accurate and practical small object detection solution in mobile edge computing environments and cloud computing scenarios.