Visual standard model and complex wavelet-based nature images recognition
基于视觉标准模型和复数小波的自然图像识别*

This paper proposed an algorithm based on the dual-tree complex wavelet, which has approximate shift invariance and the good directional selectivity, and the algorithm here can extract features with shift-invariance and scale-invariance and can solve the problem that compute complicatedly in visual standard model proposed by Serre. When tested on Caltech5 database, the algorithm can achieve high recognition rate. Combined with up-bottom attention mechanism, the method could keep the recognition rate, and decreases the computation cost greatly. Experimental results show that the proposed algorithm can comparable to standard model but at a lower computation cost.