Today, to diagnose gastrointestinal disorders such as gastric cancer, gastrointestinal bleeding, and study of various causes of indigestion, wired endoscopy is being used. Wired endoscopy procedure is painful and uncomfortable for patients. Also, it is not capable of observing the entire digestive tract. To overcome these problems, wireless capsule endoscopy is an option. It is pain-free and capable of viewing the whole small intestine. Wireless capsule endoscopy system still has limitations. These limitations consist of image quality, bandwidth, power consumption, and the consumed silicon area. On the one hand, we need high image quality and proper frame rate for diagnosing from endoscopy images. On the other hand, most power consumption in capsule belongs to the data transmitting unit. Therefore, there is a need for compressing data with a high compression ratio to provide proper frame rate and image quality, during low power consumption. In this thesis, we propose two methods to compress endoscopy images. In the first method, we use local motion estimation and spatial based prediction. We align="left" dir="LTR" Keywords: Image compression, motion estimation, endoscopy capsule, hardware implementation, spatial and temporal redundancies
