Abstract: A method for compressing image data by classifying colors. Initially, image data is transformed from an RGB coordinate to a Lab coordinate. Subsequently, each pixel of the image from the beginning to an end of the first row data is processed in turn. After finishing the processing of the first row data, the next row data is processed in turn to completely process the image.

Abstract: A method for promoting temporal resolution of sequential images is to firstly choose a first visual frame and split it into a plurality of visual blocks according to a predetermined value, then choose a second visual frame. The method then goes to search the second visual frame for mostly resembling respective visual blocks corresponding to every visual block in the first visual frame and estimate the vector displacement of respective corresponding visual blocks of the visual frames. Further, it goes to rebuild a visual frame, estimate the respective pixel values of respective corresponding visual blocks of the visual frames at a time point, combine the pixel values of visual blocks to create the rebuilt visual frame, apply a median-value-filtering procedure upon the rebuilt visual frame, and apply a spatial low-pass filtering procedure upon the rebuilt visual frame, which has undergone the previous procedure.

Abstract: A method for compressing image data by classifying colors. Initially, image data is transformed from an RGB coordinate to a Lab coordinate. Subsequently, each pixel of the image from the beginning to an end of the first row data is processed in turn. After finishing the processing of the first row data, the next row data is processed in turn to completely process the image.

Abstract: A method for promoting temporal resolution of sequential images is to firstly choose a first visual frame and split it into a plurality of visual blocks according to a predetermined value, then choose a second visual frame. The method then goes to search the second visual frame for mostly resembling respective visual blocks corresponding to every visual block in the first visual frame and estimate the vector displacement of respective corresponding visual blocks of the visual frames. Further, it goes to rebuild a visual frame, estimate the respective pixel values of respective corresponding visual blocks of the visual frames at a time point, combine the pixel values of visual blocks to create the rebuilt visual frame, apply a median-value-filtering procedure upon the rebuilt visual frame, and apply a spatial low-pass filtering procedure upon the rebuilt visual frame, which has undergone the previous procedure.