An adaptive 3-D DCT coder for image com pression

Abstract

Discrete Cosine Transform(DCT) coding is a well-established technique for image data com pression. As we know that 2-D compression techniques achieve better energy com pactness than 1-D case in terms of transform coding. Our method is not only based on that assume but also extending the 2-D to 3-D to get more compression efficiency. A segmentation technique based on the local energy magnitude is developed to classify the non-overlapping sub-image into different energy levels. Then the 3-D block collector based on the segmentation results will gather the blocks with the same energy level together to form a 3-D block. Finally, 3-D DCT coder is employed to com press the 3-D block. Our 3-D technique is an adaptive strategy which collecting the sub-blocks with the same energy level together. Because each 3D block for DCT has same signal characteristics, more com pression efficiency can be achieved. From the simulation results, we prove the performance in terms of com pression ratio and decoded distortion.