| 摘要 |
PROBLEM TO BE SOLVED: To provide an image encoding method that cna reduce the complexity of spatial prediction.SOLUTION: An image encoding method calculates an integer gradient (S11), determines a decimal pixel position corresponding to an intersection point between the line of the integer gradient and the boundary between blocks (S12), predicts a pixel value of the pixel position on the basis of a pixel value which is interpolated at the determined decimal pixel position every pixel position within an encoding target block (S13), encodes the encoding target block by using a prediction block, refers to a division table for a value obtained by performing scaling based on c-th power of 2 every predetermined value to achieve a division result when the integer gradient is calculated, calculates the integer gradient by using the achieved division result, and multiplies the integer gradient generated through the scaling by the coordinate value of a pixel position as a prediction target within the encoding target block, and then shifts the multiplication result to the right by c bits to calculate an interpolated pixel value at the decimal pixel position when the pixel value is predicted. |
