Share Email Print
cover

Proceedings Paper • new

Multi-scale bilateral-channels CNN for scene classification
Author(s): Lei Yuan; Kuangrong Hao; Xuesong Tang; Xin Cai; Yongsheng Ding
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

A multi-scale binocular-channels convolution neural network (MBCNN) is proposed to solve complex scene classification and achieved a good accuracy. We use a physiological phenomenon called visual crowding to explain the deficiency of the CNN framework and prove the effectiveness of the double flow model. With the help of a novel bilateral-channels network based on global information and local significant information and our multi-scale feature integration method, the proposed MBCNN can reduce the identification obstacle caused by visual crowding in the V1(Information input area) and V4 (High-level information area) area separately. Experiment results verify that the proposed network has better performance on MIT Indoor 67 and Scene 15 classification datasets.

Paper Details

Date Published: 29 October 2018
PDF: 5 pages
Proc. SPIE 10836, 2018 International Conference on Image and Video Processing, and Artificial Intelligence, 108361D (29 October 2018); doi: 10.1117/12.2514957
Show Author Affiliations
Lei Yuan, Donghua Univ. (China)
Kuangrong Hao, Donghua Univ. (China)
Xuesong Tang, Donghua Univ. (China)
Xin Cai, Donghua Univ. (China)
Yongsheng Ding, Donghua Univ. (China)


Published in SPIE Proceedings Vol. 10836:
2018 International Conference on Image and Video Processing, and Artificial Intelligence
Ruidan Su, Editor(s)

© SPIE. Terms of Use
Back to Top