KR950010633A - Image Decoding Apparatus and Method for Adaptive Post Processing - Google Patents

Abstract

The apparatus and method for image decoding are to adaptively post-process the inverse discrete cosine transformed coefficients according to the magnitude of the detected quantization step size during variable length decoding. To this end, when the compressed video signal is applied in the form of a bitstream, the device detects and outputs the quantization step size (QS) and the motion vector (MV), which are variable length-decoded in the opposite direction to the encoding, and are applied as additional information. Inverse quantizes the discrete cosine transform coefficients in the frequency domain by the quantization step size (QS) output from the variable length decoder, and performs the inverse discrete cosine transform into the information in the spatial domain, and moves by the motion vector output from the variable length decoder. A pre-processing unit for restoring the video signal before the discrete cosine transform by compensating for it; a post-processing unit for performing separate post-processing according to the size of the quantization step size QS; And a passage control means for controlling the passage so that the signal output from the preprocessor may be selectively performed by the post processor based on a result of comparing the quantization step size with a threshold value.

Description

Image Decoding Apparatus and Method for Adaptive Post Processing

Since this is an open matter, no full text was included.

1 is a diagram of a conventional video decoding apparatus.

2 is a diagram of an image decoding apparatus according to the present invention.

Claims (5)

An image decoding apparatus for reconstructing a compressed video signal by discrete cosine transform and variable length encoding; When the compressed video signal is applied in the form of a bit stream, the variable length decoding is performed in a direction opposite to that in encoding, and the quantization step size QS and the motion vector MV applied as additional information are detected and output. Variable length decoder 10; Inverse quantization of the discrete cosine transform coefficients of the frequency domain by the quantization step size QS output from the variable length decoder 10, and the inverse discrete cosine transform of the information of the spatial domain, and output from the variable length decoder 10 A preprocessing unit 59 for compensating for the motion by the motion vector and restoring the video signal before the discrete cosine transform; A post-processing unit 70 for performing separate post-processing according to the size of the quantization step size QS; By selectively comparing the signal output from the preprocessor 59 to the quantization step size QS output from the variable length decoder 10 with a predetermined threshold, a separate process performed by the post processor 70 is selectively performed. An image post-decoding apparatus capable of adaptive post-processing, characterized in that it comprises a passage control means (60) for controlling the passage so that it can be performed. The processing line (L1) of claim 1, wherein the post-processing unit 70 bypasses the signal applied through the passage control means 60 when the quantization step size QS is smaller than the predetermined threshold value. And a post-processor (71) for filtering the applied signal by the median or average value if the quantization step size (QS) is greater than the predetermined threshold value. 3. The method of claim 2, wherein the passage control means (60) is configured to output a signal output from the preprocessor (59) if the quantization step size (QS) output from the variable length decoder (10) is smaller than the predetermined threshold value. Adaptive to the processing line (L1), and if the quantization step size (QS) is larger than the predetermined threshold value, the signal output from the pre-processing unit 59 to the post-processor 71, characterized in that adaptive Image decoding device capable of post-processing. The method of claim 2, wherein the passage control means (60) is made of a control switch which is switched and controlled by a result of comparing the quantization step size output from the variable length decoder (10) with the predetermined threshold value. Image decoding device capable of adaptive post-processing. Reconstructs the original signal by performing variable length decoding on the video signal compressed and applied by the discrete cosine transform and the variable length encoding, and then inverse quantizing, inverse discrete cosine transform, and compensating for the motion of the inverse discrete cosine transformed signal. In the image decoding method for performing; A quantization step size detection process for detecting a quantization step size among the additional information of the compressed data during the variable length decoding; the quantization step size by comparing the quantization step size detected in the quantization step size detection process with a predetermined threshold value A first post-processing step for post-processing the inverse discrete cosine transformed reconstruction signal by median filtering if the quantization step size is larger than the threshold as a result of the size detection process; And a second post-processing step for post-processing such that the inverse discrete cosine transformed reconstruction signal is bypassed if the quantization step size is smaller than the threshold as a result of the size detection process. Image decoding method that can be processed. ※ Note: The disclosure is based on the initial application.