Image Similarity Matching Retrieval on Synergetic Neural Network

摘要：

In this paper, an image similarity matching retrieval algorithm based on synergetic neural network (SNN) is proposed. It is a novel method with advantages of no pseudo-state and closer to natural self-organization process in the field of image retrieval. It utilizes feature vector extraction, attention parameter selection, order parameter calculation, pseudoinverse matrix and its determinant value comparison to achieve better retrieval effect. Due to the structural characteristic of synergetic neural network, it can save time for iteration and improve efficiency and speed. The experimental results show that this algorithm has fast speediness, strong robustness and high accuracy, and provides greater generality and high real-time performance.

作者: Hui Li Xiuli Ma Wanggen Wan Xueli Zhou

作者单位: School of Communication and Information Engineering,Shanghai University, No. 149 Yanchang Road,Shanghai 200072, China

会议类型: 国际会议

会议名称: 2010 International Conference on Audio,Language and Image Processing(2010年音频、语言与图像处理国际会议 ICALIP 2010)

会议地点: 上海

会议语种:英文

页码: 1566-1571

在线出版日期: 2010-11-23（万方平台首次上网日期，不代表论文的发表时间）