Investigation of On-Board Compression techniques for Remote Sensing Satellite Imagery

Document Type : Original Article


1 Ph.D., Egyptian Armed Forces.

2 prof. Dr., Egyptian Armed Forces.

3 B.Sc., Egyptian Armed Forces.


The transmission of image data acquired by remote sensing missions, based on space borne platforms is a major bottleneck, as a result of the limitation of on-board power, the huge volume of transmitted data, and the number of accessible ground receiving stations. In particular these arguments hold true for small satellites that faces additional design constraints in terms of size, mass and cost .To overcome downlink restriction, image compression has to be applied, although the provided data transmission rates are constantly growing, they can't keep up with the exponentially increasing data flood provided by the scanners, the reasons are twofold: First the acquisition rate exceeds the transmission rate. Second the satellite is not in constant visibility of ground receiving station, which effectively limits the downlinkable data volume per orbit. This paper introduces implementation of some of the compression techniques for both lossy and lossless which are used for the On-Board satellite missions, taking into consideration that the introduced compression method should support and enable the real time imaging and transmission. Also, a comparison study between both techniques is introduced based on measures like Compression Ratio (CR), Root Mean Square Error (RMSE) and Peak Signal-to-Noise Ratio (PSNR).The experimental results showed that: Huffman code is the most suitable code for lossless compression technique for satellite images, since it gives a considerable high compression ratio with respect to the other lossless algorithms (Run-Length Encode, Lempel-Ziv-Welch code and Arithmetic code). Huffman code does not require much storage capacity such as LZW or more computational requirements such as arithmetic code. Also, the most suitable lossy compression techniques for satellite images is applying Discrete Wavelet Transform (DWT) on the four Most Significant Bit Planes (MSPB) of the image then applying the arithmetic code over the resultant DWT coefficients. This technique achieves higher average compression ratio (8.28:1) than the two other techniques and at the same time the image quality is accepted.