Feature based object recognition using discrete wavelet transform

In this paper, an approach for object recognition based on wavelet transform is presented. This approach decomposes the input image into sub-bands by using the multiresolutional analysis, Discrete Wavelet Transform (DWT). As each sub-band in the decomposed image contains useful information about the object, the mean of each sub-band is considered as features. This approach is tested on Columbia Object Image Library (COIL-100) database. All the objects in the COIL-100 database are considered for the classification based on one nearest neighbor classifier. The result shows that maximum recognition accuracy of 85.72% is achieved by the wavelet features.