Review of Target Detection Algorithm Based on Deep Learning
Download as PDF
DOI: 10.23977/AICT2020011
Author(s)
Xiaofang Liao, Xianfeng Zeng
Corresponding Author
ABSTRACT
In recent years, artificial intelligence (AI) technology has developed rapidly, and personal safety, social safety, and national security have attracted more and more attention. Deep learning is widely used in different kinds of fields, among which target detection has made continuous breakthroughs in image detection or video processing. Target detection should be real-time and accurate, which is the requirement of people for the effect of target detection, while traditional target detection has been difficult to meet its requirements. Target detection algorithm based on deep learning has become the mainstream in this field. This paper mainly introduced two-stage models based on region detection classification: R-CNN, SPP-NET, Fast R-CNN, Faster R-CNN, and the advantages and disadvantages of the target detection algorithm YOLO and SSD based on regression single-stage model, and summarized and prospected the development direction of target detection.
KEYWORDS
Target detection; Deep learning; Computer vision