An efficient YUV-based image compression algorithm for wireless capsule endoscopy

Abstract

The paper presents an efficient image compression algorithm for wireless capsule endoscopy application. This algorithm works in YUV colour plane and exploits some special features of human endoscopic images. Based on the nature of endoscopic images, a set of transform-quantization pair is proposed that results in high compression while keeping the reconstruction quality over 45dB. Compared to other transform-based algorithms targeted to capsule endoscopy, the proposed scheme performs strongly with a compression ratio of 90% and high reconstruction PSNR and SSIM (over 48dB and 0.999).