期刊
EXPERT SYSTEMS WITH APPLICATIONS
卷 87, 期 -, 页码 304-315出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.eswa.2017.06.025
关键词
Convolutional neural network; Gradient descent; Input-output mapping sensitivity error back; propagation; Face recognition at long distances with; small dataset; Sensitivity in cost function; Deep neural structures
类别
资金
- ICT R&D program of MSIP/IITP [R7124-16-0004]
- National Research Foundation of Korea (NRF) - Korea government (MSIP) [NRF-2016M3C1B6929647]
- Institute for Information & Communication Technology Planning & Evaluation (IITP), Republic of Korea [R7124-16-0004] Funding Source: Korea Institute of Science & Technology Information (KISTI), National Science & Technology Information Service (NTIS)
In this paper, we propose a sensitive convolutional neural network which incorporates sensitivity term in the cost function of Convolutional Neural Network (CNN) to emphasize on the slight variations and high frequency components in highly blurred input image samples. The proposed cost function in CNN has a sensitivity part in which the conventional error is divided by the derivative of the activation function, and subsequently the total error is minimized by the gradient descent method during the learning process. Due to the proposed sensitivity term, the data samples at the decision boundaries appear more on the middle band or the high gradient part of the activation function. This highlights the slight changes in the highly blurred input images enabling better feature extraction resulting in better generalization and improved classification performance in the highly blurred images. To study the effect of the proposed sensitivity term, experiments were performed for the face recognition task on small dataset of facial images at different long standoffs in both night-time and day-time modalities. (c) 2017 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据