Advanced Object Detection Algorithms and Its Application Based on Deep Learning

Abstract

Object detection is one of the most basic and challenging problems in such a field, which has become a paramount topic for scholars in recent years. In the past two decades, object detection has developed rapidly from the beginning to the application in all aspects of life, with the improvement both in detection accuracy and detection speed. In this paper, firstly, we focus on the research progress of object detection algorithm based on deep learning in the light of its technical evolution and application. Secondly, we compare and analyze the two-stage and single-stage detection framework from series algorithms based on R-CNN to series algorithms based on Yolo, and introduce the common data sets and index evaluation as well as the application process of the algorithm in the detection fields of pedestrian, face, text, medical image, sign language, etc. Finally, we predict the prospects of deep learning-based object detection algorithms according to the existing challenges.