Block effect removing method for JPEG (Joint Photographic Experts Group) coding under low bit rate

Abstract

The invention relates to a method for removing block effect occurring when JPEG (Joint Photographic Experts Group) coding is carried out under a low bit rate condition. The block effect removing method comprises the steps of carrying out two-round coding on a JPEG, wherein according to the first round coding, after original data is blocked according to a normal coding process, whether a current block is a smooth block or a textural block or not is judged, DCT is carried out on each block, a dc coefficient and block type information of each block are stored and are written into DDR, and moreover, quantized ac coefficients are written into the DDR through run length coding; and according to second-round JPEG coding, the dc coefficient and block type information of each block are read, whether to carry out edge-preserving filtering operation on the current block or not is decided according to whether the current block is the smooth block or not, and quantization, dc prediction and entropycoding operation are carried out according to the normal coding process. According to the method, a JPEG coding process is not changed, for the smooth blocks under the low bit rate, the block effectis greatly removed, the image quality is greatly improved, and the subjective visual effect is improved.

Description

technical field [0001] The invention relates to the field of image algorithm processing, in particular to a method for how to remove block effects in JPEG encoding under low code rate conditions, so as to achieve the purpose of improving image quality and visual effect. Background technique [0002] JPEG (Joint Photographic Experts Group) is the first international image compression standard. It is widely used in various industries due to its excellent compression performance, simple algorithm implementation, and small amount of calculation. However, since JPEG is a lossy compression format, the higher the compression rate, the worse the image quality, and the more prone to block effects. The block effect of JPEG encoded images is introduced in the encoding process. The root cause is mainly in JPEG encoding. DCT (discrete cosine transform) transform and block-independent quantization are adopted in JPEG coding. In JPEG coding, the DCT transform is first performed on 8x8 bloc...

AI Technical Summary

Problems solved by technology

However, since JPEG is a lossy compression format, the higher the compression rate, the worse the image quality and the more prone to block effects. The block effect of JPEG encoded images is introduced in the encoding process. The root cause is mainly in JPEG encoding. DCT (discrete cosine transform) transform and block-independent quantization are adopted in JPEG coding. In JPEG coding, the DCT transform is first performed on 8x8 blocks, and the image information in the spatial domain is transformed into the frequency domain, so that the energy of the image signal is concentrated in a few DCT coefficients such as DCT DC coefficients, DCT low-frequency coefficients, and then quantize the coefficients after DCT transformation. Since the quantization of all blocks in the image is independent, that is, the quantization of the current block does not depend on surrounding blocks, resulting in DCT of domain blocks. The quantization errors after coefficient quantization are not correlated with each other, and for pixels at the boundary of two adjacent blocks, due to the discontinuity of the DCT coefficient quantization errors, the originally smooth data has a large change at the block boundary, resulting in block effects , especially in smooth blocks at low bit rates, this phenomenon is particularly obvious, and the human eye is very sensitive to it, resulting in poor subjective visual quality

Images