Steganography is a talent of hiding statement by embedding message into an innocuous observing cover media. Using steganography, an underground message is embedded inside apiece of unsuspicious information. Image steganography is a technique of conceal information into a cover image to hide it. DWT based approach is most accepted steganography methods in frequency domain due to its simplicity and hiding ability. A novel method for Image steganography based on Least Significant Bit using X-box mapping where we have used several wavelets having unique data. The embedding part is done by this Steganography algorithm where we use four unique bounds which has be consider in two parts i.e. lower bound and upper bound with values of the cover image. This wavelet provides adequate security to the load because with no significant the transformation rules no one can extract the secret data. Second Approach, we can implement the Ant Colony Optimization Technique for extract the message from the cover image or original image. After that we can classify the design system using Back Propagation Neural Network which has been creating in two modules like Training Module and other one Testing Module. Many information security algorithms have been developed steganography algorithms to enhance information security. One of the most recent algorithms is Neural Network. In this paper it encrypts the secret message to protect it from being accessed by unauthorized users before being hidden. The PSNR of the stego image was estimated to measure the stego images quality. The obtained results demonstrated that using secret key provides good security and PSNR value higher than previous image steganography methods. Similarly, DWT and ACO have been used to embedding and extraction process of secret message will be done using Transformation and optimization technique. So in the paper, Back Propagation Neural Network, DWT and ACO with lower Bound and Upper Bound will be introduced. Lastly, we design the secure system to hide the message in the image using testing module checked the performance parameters like Peak Signal to Noise Ratio, Mean Square Error and Root Signal Error. To compare the performance parameters with the previous approaches like (LSBand X-box mapping) on the basis of PSNR and MSE.
DWT, ACO, PSNR, MSE, BPNN, JPEG, Steganography