Pedestrian Detection based on YOLOv3 Algorithm
- Resource Type
- Conference
- Authors
- Zheng, Xing; Tian, Cuihua
- Source
- 2022 IEEE 5th International Conference on Automation, Electronics and Electrical Engineering (AUTEEE) Automation, Electronics and Electrical Engineering (AUTEEE), 2022 IEEE 5th International Conference on. :1119-1123 Nov, 2022
- Subject
- Communication, Networking and Broadcast Technologies
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Electrical engineering
Automation
Object detection
Feature extraction
Pedestrian Detection
YOLOv3
SPP
ECA
- Language
- ISSN
- 2831-4549
In order to improve the accuracy of pedestrian detection, this paper proposes a target detection algorithm based on yolov3 to detect the missing detection and small-size pedestrian in the original yolov3. This paper introduces ECA attention mechanism to focus on key features and suppress unnecessary features to increase expressiveness; The SPP network structure is added in the algorithm for backbone feature extraction network, which effectively reduces the model calculation amount. The experiment of the improved YOLOv3 algorithm in the INRIA dataset shows that the accuracy of the proposed algorithm in this paper is improved compared with the original model, and the map value is as high as 92.84%. The experimental results present that the detection preciosn is effectively improved and this model can be applied to the detection of daily pedestrians.