This study investigates the design and performance of a double predictor differential pulse code modulation (DP-DPCM) algorithm for image data compression. A modified DP-DPCM image coding system operates on an image that has been preprocessed into segments of variable size, square blocks. The differential values between the nearby pixels within an image block are reduced. Therefore, we can decrease the distribution range of prediction error as well as reduce the bit rate and quantization levels. Each block is separately encoded by a DP-DPCM system to reduce the effect from the fed-back quantization error. The system performance of this variable block-size DP-DPCM image data compression scheme is about 5dB (or greater) coding gain in SNR than that of a conventional DPCM system for low-rate (R < 3) image compression.