In this paper we propose a steganography technique based on Stantlet transform and T-codes. The Stantlet transform are wavelet-like transforms that provides better time localization and signal compression than the conventional discrete cosine transforms (DCT) and discrete Haar-wavelet transforms. T-codes are used for better embedding capacity and decoding efficiency. The proposed method first encodes the message using best T-codes. We then obtain four sub-bands HH, HL, LH, and LL from cover image by applying Stantlet transforms to it. The encoded message is embedded in high frequency subbands HH, HL and LH based on 2 to 3 LSB method. The algorithm is compared with Haar-Wavelet approach. The metric for image quality used are PSNR and MSSIM. Our experimental results have shown high embedding rate, good imperceptibility and better run time than the known Haar-Wavelet based technique.