In this paper, we present an efficient data compression scheme, the template run-length encoding (TRLE) scheme, for image composition of parallel volume rendering systems. Given an image with 2n?2n pixels, in the TRLE scheme, the image is treated as n ?n blocks and each block has 2 ?2 pixels. Since a pixel can be a blank or non-blank pixel, there are 16 templates in a block. To compress an image, the TRLE scheme uses the templates to encode blocks row by row. Blocks in the same row are encoded as a TRLE_sequence. By packing all TRLE_sequences in a packet, the packet is the compressed partial image that can be sent/received among processors. To evaluate the performance of the TRLE scheme, we compare the proposed scheme with the BR, the RLE, and the BRLC schemes. Since a data compression scheme needs to cooperate with some data communication schemes, in the implementation, the binary-swap (BS), the parallel-pipelined (PP), and the rotate-tiling (RT) data communication schemes are used. By combining the four data compression schemes with the three data communication schemes, we have twelve image composition methods. These twelve methods are implemented on a PC cluster. The data computation time and the data communication time are measured. The experimental results show that the TRLE data compression scheme with the RT data communication scheme outperforms other eleven image composition methods.