An orchard pedestrian detection method based on a YOLOv3 algorithm

Abstract

The invention discloses an orchard pedestrian detection method based on a YOLOv3 algorithm. The method specifically comprises the steps of collecting pedestrian images in an orchard; Preprocessing thecollected images, and constructing a standard pedestrian detection data set; Placing a training set into the modified Darknet-53 network structure to extract pedestrian features; Generating a predicted pedestrian boundary frame by using a K-means clustering method, performing category prediction by using a binary cross entropy loss function, and performing multi-scale fusion prediction by using asimilar FPN network; Finally, removing redundant prediction boundary frames through the Soft-NMS, and outputting final prediction boundaries and categories. The pedestrian detection accuracy is high,the real-time performance is good, the robustness of a training model to a complex background is enhanced by aiming at data augmentation methods such as Random Eraging proposed in an orchard environment, and through the adopted Soft-NMS algorithm, the recall rate of detection can be increased, and the introduced group normalization Group Normalizations can reduce the requirements of a trained model on hardware.

Description

technical field [0001] The invention belongs to the technical field of deep learning and pedestrian detection, and in particular relates to a pedestrian detection method based on the YOLOv3 algorithm when an intelligent robot works in an orchard environment. Background technique [0002] In recent years, with the country's emphasis on the creation of modern agricultural industrial parks and the development of intelligent robots, the use of intelligent unmanned agricultural machinery to spray pesticides on orchards and pick fruits has become increasingly popular. During the operation of unmanned agricultural machinery, real-time detection of surrounding obstacles is required. The primary consideration is the detection of surrounding pedestrians to ensure the safety of pedestrians and vehicles. This paper adopts computer vision method combined with deep learning to detect pedestrians. [0003] The restrictive factors affecting the development of pedestrian detection in the or...

AI Technical Summary

Problems solved by technology

Deep learning detection methods based on candidate regions, such as R-CNN and Faster R-CNN, etc., due to the step-by-step completion of target detection and target positioning, the detection speed is slow and cannot meet real-time detection scenarios

Images