Paper Info
Title
Research on the Application of YOLOv3-Tiny Algorithm in the Detection of Circuit Experiment Equipment
Abstract
In the physical circuit experiment teaching of junior high school, the equipment detection link before the experiment can be completed by the target detection algorithm based on embedded devices. In order to solve the problems of slow detection speed and false detection and missed detection in the traditional target detection algorithm on embedded devices, a real-time target detection model based on YOLOv3-Tiny is proposed. By data augmentation of self-made circuit experimental equipment data set, the problem of electronic equipment category imbalance is solved. K-means clustering algorithm is used to obtain the optimal Anchor size and number of circuit experimental equipment data sets, and the parameters of YOLOv3-Tiny algorithm are improved. The test results show that the improved YOLOv3-Tiny algorithm can achieve a detection accuracy of 94.61% and a detection frame rate of 39 frame/s on the NPU processor of the embedded device RK3399Pro development board.
Year
DOI
Venue
2021
10.1145/3469213.3470431
PROCEEDINGS OF 2021 2ND INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND INFORMATION SYSTEMS (ICAIIS '21)
Keywords
DocType
Citations 
Target Detection, YOLOv3-Tiny, embedded NPU processor, K-means clustering algorithm
Conference
0
PageRank 
References 
Authors
0.34
0
3
Name
Order
Citations
PageRank
Mingju Liang100.34
Zhenguo Wang200.34
Yang Wang342.75