Defenses to Protect Against SQL Injection Attacks

Abstract: This paper proposes use of wavelet transforms in image compression. Wavelet transform is generated from respective orthogonal transform. Image compression using Walsh wavelet, Kekre Wavelet and Slant Wavelet transforms are generated from Walsh transform, Kekre Transform and Slant Transform respectively. Four different size combinations of component orthogonal transforms are used. Size of local component transform is varied as N=32, 16, 8 and 4. Experiments are performed on twenty sample colour images of size 256x256x3. Performance of three wavelet transforms is measured in terms of compression ratio and bit rate. Each wavelet transform gives different performance at different component size. Acceptable image quality is obtained even at higher compression ratio 32. From results it has been observed that Slant wavelet transform gives best performance among all, for component transform size M=N= 16 (16-16) up to compression ratio 8. When compression ratio exceeds 8, component transform size in slant wavelet is M=32, N=8 (32-8) that gives minimum RMSE. Kekre Wavelet gives smallest error with component transform size (32- 8) for all compression ratios up to 32. Walsh wavelet shows better performance with component size 32-8 up to compression ratio 6.4. Using Kekre Transform, Kekre Wavelet Transform matrix of any size can be generated. Hence there is no restriction on image size that it should be of power of two. It has been also observed that at minimum bit rate of 0.25 BPP, acceptable image quality is obtained using all three Wavelet Transforms.

Keywords: Kekre Wavelet Transform, Slant Wavelet Transform, Walsh Wavelet Transform, Bit Rate