Analysis on Two-stage Object Detection based on Convolutional Neural Networkorks

Nowadays, with the development and wide application of computer version technology, object detection becomes a popular study direction in image processing, which is widely applied in various fields, such as robot navigation, intelligent video monitoring, industrial inspection and aerospace. In recent years, with the development of deep learning algorithms, there are much more studies about object detection based on deep learning algorithms, which solves the disadvantages of traditional method on object detection, such as poor generalization ability, low detection accuracy, and slow operational rate. There are two mainly classes about object detection: One-stage object detection algorithm and two-stage object detection algorithms. This paper aims at introducing the reserch status of two-stage object detection based on convolutional neural networks. Besides, The paper analyzes the advantages and disadvantages of each algorithm and makes a comparation with these algorithms. In the end, there will be a prospect for the object detection in the future development.