Image processing device and method for improving coding efficiency of quantization matrices

An image processing method comprising:

• decoding encoded data that includes a replacement difference coefficient which is a difference between a replacement coefficient used to replace a (0, 0) coefficient of a 32×32 quantization matrix and a (0, 0) coefficient of an 8×8 quantization matrix, to generate quantized data;
• setting a 32×32 default quantization matrix as the current quantization matrix, when a current quantization matrix is copied and as a condition that a difference value between a value of identification information identifying a reference quantization matrix and a value of identification information identifying the current quantization matrix is equal to 0;
• setting the 32×32 quantization matrix as the current quantization matrix by using a nearest neighboring interpolation process against the 8×8 quantization matrix including the sum of the replacement difference coefficient and the replacement coefficient as the (0, 0) coefficient of the 8×8 quantization matrix and by replacing the (0, 0) coefficient with the replacement coefficient, when the current quantization matrix is not copied;
• and
• dequantizing the quantized data generated in the decoding, using the current quantization matrix, thereby improving coding efficiency.