小波变换在目标识别中的应用  

Application of wavelet transform in target recognition

作　　者：徐键[1] 吴庆宪[1] 姜长生[1] 

机构地区：[1]南京航空航天大学自动化学院,南京210016

出　　处：《电光与控制》2007年第5期136-140,共5页Electronics Optics & Control

基　　金：航空支撑基金资助(05C52007);航空科学基金资助(04D52028)

摘　　要：利用小波变换具有易于消除噪声、运算方便、能够体现图像特征等优点,研究了基于小波变换的目标识别方法。首先采用二次B样条小波滤波器组对样本图像进行小波变换,提取多尺度边缘,然后提取不变矩,以此作为图像的特征向量,最后应用小波神经网络进行分类识别。计算机仿真实验表明,该方法取得了较好的识别效果。Wavelet transform is simple in operation, easy to reduce the image noise and can denote the image characteristic point, thus we put forward a target recognition method based on it. Firstly, wavelet transform was made on sample images through quadric B - spline wavelet filters, and multi - resolution edge was extracted. Then moment invariant of edge image was extracted, which was taken as the characteristic vector. Finally, the wavelet neural network was used to classify and recognize the target. Computer simulation showed that such a method achieved good recognition effect.

关 键 词：目标识别 空战 小波变换 多尺度 不变矩 小波神经网络 

分 类 号：V271.4[航空宇航科学与技术—飞行器设计] TP391.4[自动化与计算机技术—计算机应用技术]