Title :
Image Recognition Based on Optical Wavelet and Support Vector Machine
Author :
Zhan, Chun ; Wang, Jianmin ; Xiong, Zhaosong
Author_Institution :
Correspondence & Electron. Coll., Normal Univ., Nanchang, China
Abstract :
Image recognition belongs to the nonlinear classification problem, which has a certain difficulty in the process of image recognition. Image recognition based on optical wavelet and support vector machine is proposed in the paper. Optical wavelet is used to extract the features of images and support vector machine is used to create the image recognition model by utilizing the features. In the experiments, 80 images with 9 classes are used to study the effectiveness of the proposed OWSVM method. Image recognition accuracy of the proposed OWSVM method is 96.25 and image recognition accuracy of OWBPNN is 88.75.The experimental results show that image recognition accuracy of the proposed OWSVM method is better than that of OWBPNN.
Keywords :
feature extraction; image classification; image recognition; support vector machines; wavelet transforms; OWBPNN; OWSVM; feature extraction; image recognition; nonlinear classification problem; optical wavelet; support vector machine; Accuracy; Feature extraction; Image recognition; Nonlinear optics; Optical filters; Optical imaging; Support vector machines; classification; image recognition; optical wavelet; support vector machine;
Conference_Titel :
Information Management, Innovation Management and Industrial Engineering (ICIII), 2011 International Conference on
Conference_Location :
Shenzhen
Print_ISBN :
978-1-61284-450-3
DOI :
10.1109/ICIII.2011.364
