1ECE Dept., K.L.N. College of Information Technology, Sivagangai, India
2ECE Dept., K.L.N. College of Information Technology, Sivagangai, India
3ECE Dept., Thiagarajar College of Engineering, Madurai, India
4ECE Dept., K.L.N. College of Engineering, Madurai, India
Synthetic Aperture Radar Image Classification is one of the four problem domains in the field of Automatic Target Recognition. This paper proposes Wavelet transform based Euclidean distance with Shanon Index measurement to classify the SAR image, which consists of three steps including preprocessing, feature extraction and classification. Preprocessing removes the speckle noise present in the image. Daubechies wavelet is used to extract the features by obtaining the approximated image. Finally, the classification process is completed using Euclidian distance with the help of Shannon index measurement. The performance of the existing system is compared with the existing Maximum Likelihood Classifier in terms of accuracy and an accuracy of 95.3% is achieved in classification for the proposed method.
SAR, image classification, feature extraction, Daubechies Wavelet, Euclidean distance and Shannon index measurement
