| 摘要 |
<p>PURPOSE:To remarkably improve the compression efficiency at the time of an encoding by converting data which are increased by a bit per picture element than original data by taking a difference into the number of bit which is the same as original image data in a reversible method. CONSTITUTION:Image data to be encoded are inputted in an image data input part 1. At this time, image data to be an object are defined to be the one having the gradation of N-bit per picture element. Next, the data of objective picture element are predicted at every picture element in a prediction part 2, by using the image data inputted from the input part 1. Next, the difference between the original image data for the objective picture element and the predicted value determined by the prediction part 2 is calculated in a difference calculation part 3. At this time, since the difference with the predicted value of N-bit per picture element is calculated in the same way as the original image data of N-bit per picture element, the difference becomes N+1 bit per picture element. Next, these data of (N+1) bit per picture element are converted into the data of N-bit in a reversible method in a data conversion part 4.</p> |
