Image compression Based on Uniform Thresholding and Fixed Point Binary Scaling

Abstract: Image compression is one of the main task in saving storage requirements and conserving transmission bandwidth. In this paper, a image compression based on uniform thresholding and fixed point binary scaling is proposed to reduce the computational complexity for compressing satellite Images. The variable bit rate is obtained using fixed point binary scaling and adjusting slope factor. Some images from LANDSAT 8 are compressed and the image quality metrics bits per pixel(bpp), peak signal to noise ratio(PSNR) and computational time are used for comparison. The proposed method from medium to high bitrates performs better than the discrete cosine transform(DCT) based method in terms of PSNR and computational time.

Keywords: bpp, PSNR, fixed Point, Computational time.