Image compression using Transform Coding is investigated in this paper. The digital image is a matrix of different intensity value elements. In general, there are two different techniques for image compression, the first one is by directly implementing quantization and coding, while the second technique transforms the image from spatial to spectral (transform) domain first and then to be quantized and coded. Discrete Cosine Transform (DCT) proved to be the practical tool for this job. For quantization, two approaches are used, the uniform and nonuniform quantization. Comparison of the two approaches suggested our proposed quantization matrix. Appling the proposed matrix for different images gave better results for all related parameters, compared with the two tested approaches................ Keywords: ............Image Compression, Discrete Cosine Transform, DCT.