KR100669905B1 - Blocking artifacts reduction apparatus and method using the characteristics of block boundary area - Google Patents

Abstract

1. TECHNICAL FIELD OF THE INVENTION

The present invention relates to an apparatus and method for removing a blocking phenomenon using block boundary region characteristics.

2. The technical problem to be solved by the invention

The present invention provides an apparatus and method for removing a blocking phenomenon for improving image quality by removing a blocking phenomenon of a block-based encoded image (eg, MPEG-4) by adaptively filtering according to characteristics of a block boundary region. The purpose is to provide

3. Summary of Solution to Invention

According to an aspect of the present invention, there is provided an apparatus for removing a block phenomenon, comprising: block-based encoded image input means for receiving a block-based encoded image from an external device; Block boundary classification means for classifying block boundaries with respect to block boundary regions of the block-based encoded image input from the block-based encoded image input means according to characteristics of each block boundary region and characteristics of adjacent subregions; Blocking phenomenon detection means for detecting a block boundary in which a blocking phenomenon occurs in an area classified by the block boundary classification means into a flat area (flat block); Filtering means for adaptively performing filtering on the area detected by the blocking phenomenon detecting means to remove the blocking phenomenon; A filtering means using a wavelet transform for adaptively performing filtering on a region classified into a complex region (a mixed block or a complex block) by the block boundary classification means to remove a blocking phenomenon; And image acquisition means for acquiring and outputting an image from which the blocking phenomenon is removed by using an image from which the blocking phenomenon is removed by the filtering means and the filtering means using the wavelet transform and a block-based encoded image input from the external device. .

4. Important uses of the invention

The present invention is used in a digital video device using a block-based coding scheme such as MPEG-4.

Blocking, Block Boundary Classification, Adaptive Filtering, Wavelet Transform, Deduction

Description

Blocking artifacts reduction apparatus and method using the characteristics of block boundary area}

1 is a block diagram of an embodiment of an apparatus for removing block phenomenon using a block boundary region characteristic according to the present invention;

2 is a flowchart illustrating a method for removing a blocking phenomenon using a block boundary region characteristic according to the present invention;

3 illustrates pixels for block boundary region classification and flat region filtering;

4 is a diagram illustrating a Weber law used in a blocking phenomenon detection process;

FIG. 5 is a diagram showing experimental results of a Hall monitor image of MPEG-4 encoded CIF size at QP = 12, 48 Kbps, and 7.5 Hz. FIG.

FIG. 6 is a diagram showing experimental results of an MPEG-4 encoded image. FIG.

7 is a diagram illustrating an experiment result for a first frame of an MPEG-4 encoded image.

* Explanation of symbols for the main parts of the drawings

110: block-based coded image input unit 120: block boundary classifier

130: blocking phenomenon determination unit 140: one-dimensional filtering unit

150: filtering unit using wavelet transform 151: one-dimensional wavelet transform unit

152: annual threshold value conversion unit 153: one-dimensional inverse wavelet transform unit

160: image acquisition unit

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a blockage removing apparatus and a method for removing a blockage phenomenon in a digital imaging apparatus using a low bit rate block-based video encoding method such as MPEG-4, and more particularly, to block-based discrete cosine. TECHNICAL FIELD The present invention relates to a blockage removal apparatus and a method for removing a blockage phenomenon using a block boundary region characteristic in a low bit rate video encoding technology using a block-based Discrete Cosine Transform (BDCT).

In general, as various multimedia services including information such as still images and videos are provided, video data compression is applied to various applications such as multimedia communication, digital TV (TV), video on demand (VOD), and video conferencing. Essentially used in the field. In particular, image compression methods using block-based Discrete Cosine Transform (BDCT), which enables high energy compression and fast computation, are known as JPEG, H.261, H.263, MPEG-2, and MPEG. It is commonly used in still image and video encoding methods such as) -4.

However, in the block-based Discrete Cosine Transform (BDCT) coding method, image quality may be degraded by dividing an image into blocks having an 8 × 8 pixel size and independently quantizing the Discrete Cosine Transform (DCT) coefficients of each block. In particular, in the image coded at a low bit rate, blocking artifacts, which are severe image quality deterioration, are generated in the block boundary region.

Recently, many post-processing methods have been studied to efficiently remove such blocking phenomenon, and these methods can be largely divided into an image enhancement method and an image reconstruction method. The image enhancement method includes a filtering method in a spatial domain, a method of estimating discrete cosine transform (DCT) coefficients in a frequency domain, and a method using a wavelet transform. The image reconstruction method includes a Constrained Least-Square (CLS) method, a Projection Onto Convex Set (POCS) theory, and a Maximum A Posteriori (MAP) method. Etc.

Among these methods, the image reconstruction method has a disadvantage in that the execution time is long because it needs to go through an iterative calculation process, and the method using the wavelet transform is performed because it has excellent performance but requires additional calculation for wavelet transform and inverse wavelet transform. There is a disadvantage in that the time is long.

As described above, the method of removing a block phenomenon using wavelet transform has a high performance but a complicated calculation process. Therefore, in order to solve these problems with the existing methods, the present invention proposes a novel blocking effect removal apparatus and method thereof.

That is, the present invention has been proposed to solve the above problem, and adaptively performs filtering according to the characteristics of the block boundary region, thereby removing the blocking phenomenon of the block-based coded image (eg, MPEG-4). SUMMARY OF THE INVENTION An object of the present invention is to provide a blockage removing apparatus and a method for improving image quality.

That is, since the block-based coded image exhibits nonlinear characteristics by independently quantizing the discrete cosine transform (DCT) coefficients of each block, the present invention removes the blocking phenomenon by using a simple adaptive filter and a wavelet transform characteristic, which are nonlinear filters. SUMMARY OF THE INVENTION An object of the present invention is to provide an apparatus and method for removing block effect to improve image quality.

Other objects and advantages of the present invention can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. Also, it will be readily appreciated that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

According to an aspect of the present invention, there is provided an apparatus for removing blocking, comprising: block-based coded image input means for receiving a block-based coded image from an external device; Block boundary classification means for classifying block boundaries with respect to block boundary regions of the block-based encoded image input from the block-based encoded image input means according to characteristics of each block boundary region and characteristics of adjacent subregions; Blocking phenomenon detection means for detecting a block boundary in which a blocking phenomenon occurs in an area classified by the block boundary classification means into a flat area (flat block); Filtering means for adaptively performing filtering on the area detected by the blocking phenomenon detecting means to remove the blocking phenomenon; A filtering means using a wavelet transform for adaptively performing filtering on a region classified into a complex region (a mixed block or a complex block) by the block boundary classification means to remove a blocking phenomenon; And image acquisition means for acquiring and outputting an image from which the blocking phenomenon is removed using the image from which the blocking phenomenon is removed by the filtering means and the filtering means using the wavelet transform and a block-based encoded image input from the external device. .

Meanwhile, the method of the present invention provides a block boundary classification step for classifying a block boundary according to the characteristics of each block boundary region and the characteristics of an adjacent subregion in a block boundary region of a block-based encoded input image. ; Determining whether the classified region is a flat region (flat block) or a complex region (complex block, mixed block); As a result of the determination, in the case of a flat region (flat block), after detecting a block boundary in which a blocking phenomenon occurs in the corresponding region, the first blocking phenomenon removing step of adaptively filtering the detected region to remove the blocking phenomenon ; If it is determined that the complex region (mixed block or complex block) is removed, a second blocking phenomenon is removed by adaptively performing a filtering on the corresponding region using a wavelet transform; And obtaining an image from which the blocking phenomenon is removed by using each image from which the blocking phenomenon is removed and the block-based encoded input image.

The above objects, features and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, whereby those skilled in the art may easily implement the technical idea of the present invention. There will be. In addition, in describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of an embodiment of an apparatus for removing block phenomenon using a block boundary region characteristic according to the present invention.

As illustrated in FIG. 1, an apparatus for removing blocking phenomenon using a block boundary region characteristic according to the present invention includes a block-based encoded image input unit 110 for receiving a block-based encoded image from an external device, and the block-based encoded image input unit. The 8 × 8 block boundary region of the block-based coded image from (110) is classified into flat sub-regions and complex sub-regions according to the characteristics of each block boundary region, and then according to the characteristics of adjacent sub-regions. Block boundary classification unit 120 for classifying a block boundary into a flat block, a complex block, and a mixed block, and a block boundary in which a blocking phenomenon occurs in an area classified into a flat area (flat block) by the block boundary classification unit 120. Blocking detection unit 130 for detecting the detection function, the one-dimensional filtering can be adaptively applied to the area detected by the blocking phenomenon detection unit 130 Adaptive filtering using wavelet transform on a region classified as a complex region (mixed block or complex block) by the one-dimensional filtering unit 140 and the block boundary classifying unit 120 to remove the blocking phenomenon An image obtained by removing the blocking phenomenon from the filtering unit 150 using the wavelet transform, the one-dimensional filtering unit 140 and the filtering unit 150 using the wavelet transform, and a block-based input from the external device. And an image obtaining unit 160 for obtaining and outputting an image from which the blocking phenomenon is finally removed using the encoded image.

Here, the filtering unit 150 using the wavelet transform performs a one-dimensional wavelet transform on a region classified into a complex region (mixed block, complex block) by the block boundary classifier 120 to detect a blocking phenomenon. One-dimensional wavelet transforming unit 151, De-quantizing unit 152 for removing the blocking phenomenon detected by the one-dimensional wavelet transforming unit 151 using soft-thresholding, and the delimiting unit 152 ) Includes a one-dimensional inverse wavelet transform unit 153 for performing the one-dimensional inverse wavelet transform on the image from which the blocking phenomenon is removed and transmitting the removed block image to the image acquisition unit 160.

Looking at the operation and specific embodiments of the components in detail as follows.

In general, the blocking phenomenon is more severe in the flat region than in the complex region of the image. Therefore, if the same filtering method is used for the two regions, the blocking phenomenon cannot be effectively removed. Therefore, accurate block boundary region classification is essential in order to efficiently remove the blocking phenomenon occurring in the block boundary region. In addition, since there is a region in which the blocking phenomenon does not occur among the flat block boundary regions, only the region in which the blocking phenomenon occurs in the flat region should be filtered to reduce the computation amount.

In the present invention, for example, block boundary classification is performed by using statistical characteristics of eight pixel values of an 8 × 8 block boundary region, and a block boundary classified as a flat block is determined whether or not a blocking phenomenon occurs. First, as shown in FIG. 3, the variance of each of the subregions S1 and S2 of the block boundary region is obtained. When the two variances are compared with the threshold T1 and the variance is small, the characteristics of adjacent blocks are considered to be similar to each other. Therefore, the two variances are classified into flat subregions. Subsequently, the adjacent subregions are classified into flat blocks if all of the subregions are flat, complex blocks if all of the subregions are complex, and mixed blocks if the subregions are different from each other.

Subsequently, since the absolute value of the difference between the two pixel values of the block boundary area is compared with the threshold value T2 for the area classified as the flat area, if the two values are the same or large, it is determined that the difference in the brightness of the pixel values between adjacent blocks is large. If the absolute value is small, it is determined that the block boundary does not occur. In this case, T1 = 0.5 × QP is used as an example in the block boundary region classification process. In other words, the threshold T1 was determined using the quantization parameter QP. In general, the contrasting ability of contrast is excellent in bright areas, but the contrasting ability of contrast is slightly decreased in very dark places. However, the ratio of the minimum luminance change ΔL of the luminance difference discrimination with respect to the background luminance L in the brightness range of 10 to 1000 cd / m 2 in everyday life is constant at about 2%. This is called Weber's law. This Weber law is as shown in FIG. The threshold T2 used in the block boundary region classification process is used to determine whether a blocking phenomenon occurs in the flat region. In the present invention, for example, when the luminance value is 127 in an 8-bit image, The decision was made in consideration of the amount of change in luminance that can be judged by time and the amount of calculation in filtering.

과 같이 얻을 수 있다. 여기서,

이고,

는 블록화 강도를 나타내는데, 그 값은 일예로 도 3에 도시된 (p₃-p₄)의 절대치이다. 이처럼 본 발명에서는, 평탄 영역에서의 블록화 현상이 계단 함수 형태로 나타나기 때문에, 블록화 강도를 이용하여 블록 경계 영역에 인접한 8개의 화소값을 조정함으로써 블록화 현상을 제거한다. 즉, 본 발명은 블록 경계 영역에서의 화소값의 불연속성을 줄여줌으로써 블록화 현상을 제거할 수 있다.In addition, since the blocking phenomenon occurring in the flat region is represented by a step function, in the present invention, the region in which the blocking phenomenon occurs among the regions classified as the flat region includes eight pixels adjacent to the block boundary region (for example, four left and right sides). The blocking phenomenon is eliminated by performing adaptive one-dimensional filtering on the pixel). At this time, the filtering result is

Can be obtained as here,

ego,

Denotes the blocking strength, which is, for example, an absolute value of (p ₃ -p ₄ ) shown in FIG. ₃ . As described above, in the present invention, since the blocking phenomenon in the flat region appears in the form of a step function, the blocking phenomenon is eliminated by adjusting eight pixel values adjacent to the block boundary region using the blocking intensity. That is, the present invention can eliminate the blocking phenomenon by reducing the discontinuity of the pixel value in the block boundary region.

와 같이 제거한다. 여기서,

,

는 입력영상,

는 첫 번째 스케일 상의 웨이블릿 계수를 나타낸다. 그리고 수직 블록 경계에서도 전술한 수평 블록 경계에서의 블록화 현상 제거 방식과 같은 방식으로 제거한다.On the other hand, the blocking phenomenon occurring in the complex region is eliminated using the Lipschitz regularity theory using the wavelet transform proposed by Mallat et al. Singular points, such as blocking and actual edges, appear as local modulus maxima in the wavelet transform region on the first scale. In the present invention, by analyzing the Repeats law of the local coefficient maximums on the first scale, the blocking phenomenon having the two structures and the actual edge are distinguished. At this time, the singular points having the same structure as the blocking phenomenon are reduced as the scale of wavelet transform coefficient values within the cone of influence range increases as the scale increases based on the repeats theory. It was confirmed that it has a regular constant. Therefore, by removing adaptively the wavelet transform coefficient values having a negative real value according to the adjacent coefficient distribution, only the blocking phenomenon can be effectively eliminated without blurring the actual edges. In other words, we use soft-thresholding on the first scale for wavelet coefficients with negative repeating constants.

Remove as here,

,

Is the input image,

Denotes the wavelet coefficient on the first scale. In addition, the vertical block boundary is removed in the same manner as the above-described block elimination method at the horizontal block boundary.

In addition, in the mixed region, not only the blocking phenomenon of the block boundary region but also a noise such as a ringing phenomenon exists in the block. Therefore, the wavelet coefficients inside the block as well as the wavelet coefficients in the block boundary region are processed in a manner similar to that performed in the complex region.

Through the above-described processing, an image from which the blocking phenomenon is removed may be acquired and output by adding a block-based encoded image input from the outside or replacing a corresponding image. For example, after the block-based encoded input image is buffered, the blocking process may be removed through the above-described processing, and then a corresponding image among the buffered images may be replaced with an image from which the blocking phenomenon is removed. .

As described above, in the present invention, the block boundary region is classified into flat sub-regions and complex sub-regions according to the characteristics of each block boundary region, and then the block boundary is classified according to the characteristics of adjacent sub-regions. By adaptively filtering according to the characteristics of each block boundary region, the pixels to be filtered, and the quantization size, the blocking phenomenon can be eliminated while preserving the characteristics in the block.

2 is a flowchart illustrating a method of removing a blocking phenomenon using a block boundary region characteristic according to the present invention.

First, a block-based coded image is received from an external device (201).

Subsequently, after classifying the 8 × 8 block boundary region of the received block-based coded image into flat sub-regions and complex sub-regions according to the characteristics of each block boundary region (202), characteristics of adjacent sub-regions The block boundary is classified into a flat block, a complex block, and a mixed block according to the block 203.

Then, it is determined whether the classified region is a flat region (flat block) or a complex region (complex block, mixed block) (204).

As a result of the determination 204, if a flat area (flat block) is detected, a block boundary in which a blocking phenomenon occurs in the corresponding area is detected (205). Thereafter, the blockage phenomenon is removed by adaptively performing one-dimensional filtering on the detected region (206).

On the other hand, if the determination result 204, a complex region (mixed block, complex block), the blocking effect is removed by adaptively filtering the corresponding region using the wavelet transform (207 to 209). That is, after detecting a blocking phenomenon by performing one-dimensional wavelet transform on an area classified as the complex region (mixed block or complex block) (207), the detected blocking phenomenon is soft-thresholding by using soft-thresholding. After removal (208), one-dimensional inverse wavelet transform is performed on the image from which the blocking phenomenon is removed (209).

Thereafter, by using each image from which the blocking phenomenon is removed and a block-based encoded image received from the external device, an image from which the blocking phenomenon is finally removed is obtained and output (210).

Specific embodiments of each operation are as described above, and thus will not be described herein again.

For reference, FIGS. 5 to 7 are graphs showing experimental results of the conventional filtering method (MPEG-4 post-processing filter) and the filtering method according to the present invention, and FIG. 5 is an MPEG (MPEG) with QP = 12, 48 Kbps and 7.5 Hz. Experimental results are shown for the 4-4 encoded CIF sized Hall monitor image, FIG. 6 shows the experimental results for the MPEG-4 encoded image, and FIG. 7 is the MPEG. -4 shows experimental results for the first frame of the encoded image. As such, as shown in FIGS. 5 to 7, it can be seen that the performance of the blocking phenomenon removal method according to the present invention is much superior to the conventional method.

As described above, the method of the present invention may be implemented as a program and stored in a recording medium (CD-ROM, RAM, ROM, floppy disk, hard disk, magneto-optical disk, etc.) in a computer-readable form. Since this process can be easily implemented by those skilled in the art will not be described in more detail.

The present invention described above is capable of various substitutions, modifications, and changes without departing from the technical spirit of the present invention for those skilled in the art to which the present invention pertains. It is not limited by the drawings.

As described above, according to the present invention, the filtering is adaptively performed according to the characteristics of the block boundary region, thereby eliminating the blocking of the block-based coded image (eg, MPEG-4), thereby contributing to the improvement of the image quality of the compressed image. have.

That is, since the block-based coded image exhibits nonlinear characteristics by independently quantizing the discrete cosine transform (DCT) coefficients of each block, the present invention removes the blocking phenomenon by using a simple adaptive filter and a wavelet transform characteristic, which are nonlinear filters. There is an effect that can improve the image quality of the image.

That is, the present invention has an effect that can contribute to the improvement of the quality of a compressed image by removing the blocking phenomenon in a digital video apparatus using a low bit rate block-based video encoding method such as MPEG-4.

Claims (16)

In the blocking phenomenon removal device, Block-based encoded image input means for receiving a block-based encoded image from an external device; Block boundary classification means for classifying block boundaries with respect to block boundary regions of the block-based encoded image input from the block-based encoded image input means according to characteristics of each block boundary region and characteristics of adjacent subregions; Blocking phenomenon detection means for detecting a block boundary in which a blocking phenomenon occurs in an area classified by the block boundary classification means into a flat area (flat block); Filtering means for adaptively performing filtering on the area detected by the blocking phenomenon detecting means to remove the blocking phenomenon; A filtering means using a wavelet transform for adaptively performing filtering on a region classified into a complex region (a mixed block or a complex block) by the block boundary classification means to remove a blocking phenomenon; And Image acquisition means for acquiring and outputting an image from which the blocking phenomenon is removed by using the image from which the blocking phenomenon is removed by the filtering means and the filtering means using the wavelet transform and a block-based encoded image input from the external device. Blocking phenomenon removal device using a block boundary region characteristics comprising a. The method of claim 1, The block boundary classification means, The block boundary region of the block-based encoded image input means from the block-based encoded image input unit is classified into flat sub-regions and complex sub-regions according to the characteristics of each block boundary region, and according to the characteristics of the adjacent sub-regions. Blocking phenomenon removal apparatus using the block boundary region characteristics, characterized in that the block boundaries are classified into flat blocks, complex blocks, and mixed blocks. The method of claim 2, The block boundary classification means, After calculating the variance of the subregions of each block boundary region, if the variance is small compared with the first threshold value, it is determined that the characteristics of the adjacent blocks are similar to each other and classified into flat subregions. In addition, if the variance value is large, it is classified into a complex subregion. Then, if all of the adjacent subregions are all flat subregions, the flat block is different. In the case of, the blocking phenomenon removal apparatus using the block boundary region characteristic, characterized in that classified as a mixed block. The method of claim 3, wherein The first threshold is, Blocking phenomenon removal apparatus using the block boundary area characteristic characterized in that determined using 0.5 × QP (quantization parameter). The method of claim 2, The blocking phenomenon detection means, The absolute value of the difference between the two pixel values of the block boundary area is compared with the second threshold for the area classified as the flat area (flat block) by the block boundary classification means. Blocking phenomenon removal apparatus using the block boundary region characteristics characterized in that it is determined that the difference in brightness is large, the block boundary in which the block phenomenon occurs, and if the absolute value is small, the block boundary in which the block phenomenon does not occur. The method of claim 5, The second threshold is An apparatus for removing blocking phenomenon using a block boundary region characteristic, which is determined in consideration of an intermediate brightness value, that is, a luminance change that can be judged by human vision and a calculation amount when performing filtering. The method of claim 5, The filtering means, As the blocking phenomenon in the flat region appears as a step function, the blocking value is removed by adjusting the pixel value adjacent to the block boundary region by using the block strength to reduce the discontinuity of the pixel value in the block boundary region. Blocking phenomenon removal device using block boundary region characteristics. The method of claim 7, wherein The filtering means, Blocking phenomenon removal apparatus using the block boundary region characteristics, characterized in that to remove the blocking phenomenon by performing the adaptive one-dimensional filtering according to the block strength as shown in the following equation. (Mathematical formula)

(here,

ego,

Is the blocking strength.) The method according to any one of claims 1 to 8, The filtering means using the wavelet transform, A one-dimensional wavelet transform unit for detecting a blocking phenomenon by performing one-dimensional wavelet transform on a region classified into a complex region (mixed block or complex block) by the block boundary classification unit; A deductive unit for removing the blocking phenomenon detected by the one-dimensional wavelet transform unit using soft-thresholding; And One-dimensional inverse wavelet transform unit for performing one-dimensional inverse wavelet transform on the image from which the blocking phenomenon is removed by the de-quantizer and transferring the removed image to the image acquisition means. Blocking phenomenon removal device using a block boundary region characteristics comprising a. The method of claim 9, The one-dimensional wavelet transform unit, As singular points, such as blocking and actual edges, appear as local modulus maxima in the wavelet transform region on the first scale, we interpret the Repeats rule of the local coefficient maximums on the first scale. Blocking phenomenon removal apparatus using the block boundary region characteristics characterized in that the detection of the blocking phenomenon. The method of claim 10, The annual threshold unit, Singular points with the same structure as the blocking phenomenon have a repeating constant constant as the wavelet transform coefficient values within the cone of influence range decrease as the scale increases and have a negative real value, the Repeats constant. Blocking phenomenon removal apparatus using a block boundary region feature characterized in that the blocklet phenomenon is removed by adaptively removing wavelet transform coefficient values having negative real values according to adjacent coefficient distributions. The method of claim 11, The annual threshold unit, Blocking using block boundary region characteristics, which removes wavelet coefficients with negative repeating constants on the first scale using soft-thresholding as shown in Equation below. Phenomenon removal device. (Mathematical formula)

(here,

,

Is the input image,

Denotes the wavelet coefficient on the first scale.) The method of claim 12, The annual threshold unit, A deblocking phenomenon using block boundary region characteristics, characterized in that it deducts not only the wavelet coefficients of the block boundary region but also the wavelet coefficients inside the block. In the blocking phenomenon removal method, A block boundary classification step for classifying a block boundary with respect to a block boundary region of a block-based encoded input image according to characteristics of each block boundary region and characteristics of an adjacent subregion; Determining whether the classified region is a flat region (flat block) or a complex region (complex block, mixed block); As a result of the determination, in the case of a flat region (flat block), after detecting a block boundary in which a blocking phenomenon occurs in the corresponding region, the first blocking phenomenon removing step of adaptively filtering the detected region to remove the blocking phenomenon ; If it is determined that the complex region (mixed block or complex block) is removed, a second blocking phenomenon is removed by adaptively performing a filtering on the corresponding region using a wavelet transform; And Acquiring an image from which the blocking phenomenon is removed by using the respective images from which the blocking phenomenon is removed and the block-based encoded input image. Blocking phenomenon removal method using a block boundary region characteristics comprising a. The method of claim 14, The block classification step, Classifying a block boundary region of the block-based coded input image into flat sub-regions and complex sub-regions according to characteristics of each block boundary region; And Classifying block boundaries into flat blocks, complex blocks, and mixed blocks according to characteristics of subregions adjacent to the subregion; Blocking phenomenon removal method using a block boundary region characteristics comprising a. The method according to claim 14 or 15, The second blocking phenomenon removing step, Detecting a blocking phenomenon by performing one-dimensional wavelet transform on the region classified into the complex region (mixed block or complex block); Removing the detected blocking phenomenon by soft-thresholding; And Performing one-dimensional inverse wavelet transform on the image from which the blocking phenomenon is removed by using the de-quantization; Blocking phenomenon removal method using a block boundary region characteristics comprising a.

Images