Method and apparatus for fast image encoding for image compression

摘要

A high-speed video encoding method and apparatus are provided. In an image encoding method using spatial prediction based on High Efficiency Video Coding (HEVC), prediction directions having directionality in spatial prediction are divided for each prediction level. From the lowest prediction level to the highest prediction level, performing a first prediction search for calculating a rate-distortion cost for a prediction direction included in each prediction level, and according to the first prediction search at a random prediction level. When the result satisfies a preset early termination condition or a first prediction search is performed at the highest prediction level, the first prediction search is terminated. In addition, a second prediction search for calculating rate-distortion costs is performed for prediction directions that do not have directionality in spatial prediction, and prediction for the final encoding is performed using the results of the first prediction search and the results of the second prediction search. Decide the direction