Image compression is an essential task for storing images in digital format. In this communication, an improved and hugely memory-efficient block optimization technique is presented that incorporates byte compression and discrete wavelet transform (DWT). Instead of the common method of nulling insignificant DWT coefficients, all the DWT coefficients are stored. The only lossy part comes from block optimization without noticeable degradation in the decompressed images. The method shows huge improvement in compression and reduces image storage space. The results obtained from this technique are compared to JPEG and JPEG2000 standard which shows this can be a fast alternative to other compression methods. © Springer Nature Singapore Pte Ltd. 2019.