A wavelet-based compression method with fast quality controlling capability for long sequence of capsule endoscopy images

Summary form only given. Wireless capsule endoscopy is a state-of-the-art tool for detecting intestinal problems. The amount of image data generated by this tool is so large that data compression issues must be considered. We propose a fast and accurate quality-controlling algorithm that does not require recursive rate estimation for the compression of capsule endoscopy images. A corresponding distortion can be computed from a user-defined PSNR, and the distortion is used as the threshold for the codebook replenishment mechanism in a wavelet-based adaptive vector quantizer (VQ). This mechanism incorporates a pyramid-based vector structure and progressive SPIHT coding to meet accurately the quality demands from a user; the resulting coding performance is excellent. Furthermore, in our VQ implementation, we adopt a modeling, rather than a training, technique for the generation of the initial codebook (CB), where a pseudo-noise sequence is generated to create such a CB at both the encoder and the decoder. Experimental results show that the proposed method does give a fast and reliable quality control of all reconstructed capsule endoscopy images under test, and the CB modeling produces comparable performance to the one using CB training.