Hiding text using the least significant bit technique to improve cover image in the steganography system

One of the highest priorities in era information technology is to achieve an accurate and effective system for hiding security data. goals steganography imperceptability intruder. So this paper work increase imperceptibility on image, which has weaknesses previous studies, as well avoid statistical attacks such chi-square. A method been proposed that includes calculating color contrasts homogeneous areas image dividing them according contrast exploiting data pixels have a high impact embed two first third bits least significant bit (LSB) amount embedded data, regions (IR) classify selected features extracted advance by using support vector machine (SVM) classifier. Work was done standard images taken from dataset (USC-SIPI) types gray images. The results showed worth through peak signal noise ratio (PSNR) reached 89.5 dB due distribution