JP2014023138A - Decoder, error concealment processing device and decoding method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correct an image having a decoding error.SOLUTION: A decoder includes: a decoding processing part for decoding coded data and outputting decoded pixel data, coding parameters, and decoding error information; a frame memory for storing pixel data; a coding parameter storing part for storing a coding parameters; a concealment control part for determining on/off of concealment processing based on the decoding error information; a concealment range calculation part for calculating a concealment range if the result of the on/off determination is on; a concealment range dividing part for dividing a concealment range into divided groups, and furthermore into subgroups; a predicted moving vector calculation part for calculating a predicted moving vector for each of the divided groups/subgroups within a concealment range; and a concealed image generation part for generating a concealed image from a predicted moving vector and pixel data stored in the frame memory, if the result of the on/off determination is on.

Description

  The present invention relates to a decoding device, an error concealment processing device, and a decoding method.

  H. When reproducing a video that has been compression-encoded using an encoding method such as H.264 or MPEG-2, it is necessary to decode it to the original format. However, there is an error in the encoded data due to electrical noise on the transmission path. If mixed, the image quality of the decoded video deteriorates, for example, the decoded video is mixed with noise or falls into a state where decoding is impossible.

  As a countermeasure against this image quality degradation, there is a conventional decoding device that generates a corrected image of an errored image and performs error concealment processing to reduce the influence of the error. For example, each video frame is classified into multiple groups according to the nature of the received video material and the degree of transmission error, and it is determined for each group whether video decoding has been processed normally. Has been proposed that applies an error concealment processing method for creating a video to be newly displayed using a previously displayed decoded video or a predetermined specific video in the frame buffer ( For example, Patent Document 1).

  Also, the decoded image data is divided into one or a plurality of divided areas, a motion feature value indicating the motion of the reproduced image is calculated for each divided area from the decoded image data, and the divided areas that cannot be decoded due to an error are calculated. A decoding device is proposed that applies an error concealment processing method that performs error correction by duplicating pixel data from decoded image data based on a correction destination vector using the calculated motion feature value as a correction vector (for example, a patent) Reference 2).

JP-A-11-164296 JP 2008-187446 A

  However, in the decoding device of Patent Document 1, the motion of the video is not taken into consideration, and when there is motion in the target video, the previous decoded video has been inserted by the error concealment process and the region that has been successfully decoded. There was a problem that a large gap occurred between the areas.

  Further, in the decoding device of Patent Document 2, although the motion of the video is taken into consideration, a corrected image is generated by applying vectors having different sizes and angles little by little to adjacent macroblocks. Many edges due to vector differences appear at the boundary of the block, resulting in a problem that the subjective image quality is degraded.

  The present invention has been made to solve the above-described problems, and provides a decoding apparatus that performs concealment processing by generating a corrected image that suppresses the occurrence of an edge while considering the motion of a video and generates less uncomfortable feeling. The purpose is to do.

  A decoding device according to the present invention decodes encoded data and outputs decoded pixel data, an encoding parameter, and decoding error information, and pixel data output by the decoding processor in units of pictures. A frame memory to be stored; a coding parameter storage unit that stores the encoding parameter in units of pictures; a concealment management unit that determines ON / OFF of concealment processing for each picture based on the decoding error information; When the ON / OFF determination result of the concealment process is ON, a conceal range calculation unit for calculating a conceal range, and a plurality of types of sub group subdivision patterns in which sub groups obtained by subdividing the divided group into which the conceal range is divided are set And a divided pattern storage unit for storing the concealment range in the divided group Further, based on the plurality of subgroup subdivision patterns, a concealed range dividing unit that divides the subgroup, and an estimated motion vector that represents the divided group for each divided group / subgroup within the concealed range are calculated. An estimated motion vector calculation unit that performs the concealment processing ON / OFF determination result, and a concealed image generation unit that generates a concealed image from the estimated motion vector and the pixel data stored in the frame memory. Prepare.

  According to the decoding apparatus of the present invention, the entire slice in which an error is detected is set as a concealed range, the concealed range is divided into groups of a plurality of macroblocks, and the place where the motion is complicated is subdivided into subgroups, and error pictures Calculate the estimated motion vector for each group / subgroup within the concealed range from the motion vector of the same-position macroblock of the decoded picture that is close in time and refer to the pixel at the position corresponding to the estimated motion vector Since the corrected image is generated from the surface and the error concealment process is performed, it is possible to obtain a concealed image with less sense of discomfort while maintaining the motion in the video.

It is a block diagram which shows an example of the decoding apparatus which concerns on Embodiment 1 of this invention. It is a flowchart which shows operation | movement of the decoding apparatus which concerns on Embodiment 1 of this invention. It is explanatory drawing of an example of the concealment range setting which concerns on Embodiment 1 of this invention. It is explanatory drawing of an example of the group division | segmentation of the concealment range which concerns on Embodiment 1 of this invention. It is explanatory drawing of an example of the subgroup subdivision pattern of the divided | segmented group which concerns on Embodiment 1 of this invention. It is explanatory drawing of an example of calculation of the estimated motion vector which concerns on Embodiment 1 of this invention. It is a block diagram which shows an example of the decoding apparatus which concerns on Embodiment 2 of this invention. It is a flowchart (difference) which shows an example of operation | movement of the decoding apparatus which concerns on Embodiment 2 of this invention. It is a conceptual diagram explaining an example of the error concealment process of the decoding apparatus which concerns on Embodiment 2 of this invention. It is a block diagram which shows an example of the decoding apparatus which concerns on Embodiment 3 of this invention. It is a flowchart (difference) which shows an example of operation | movement of the decoding apparatus which concerns on Embodiment 3 of this invention. It is a block diagram which shows an example of the decoding apparatus which concerns on Embodiment 4 of this invention. It is a flowchart (difference) which shows an example of operation | movement of the decoding apparatus which concerns on Embodiment 4 of this invention. It is a block diagram which shows an example of the decoding apparatus which concerns on Embodiment 5 of this invention. It is explanatory drawing of an example of the production | generation of the concealment image of the decoding apparatus which concerns on Embodiment 5 of this invention.

  The present invention relates to a decoding device, an error concealment processing device, and a decoding method. Hereinafter, an example of the decoding device and the decoding method according to the present invention will be described.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a decoding apparatus equipped with an error concealment processing unit according to Embodiment 1 of the present invention.

  In the figure, the decoding device includes a decoding processing unit 1 and an error concealment processing unit 2a. The decoding processing unit 1 decodes the encoded data and outputs decoded pixel data, encoding parameters, and decoding error information. The error concealment processing unit 2a receives the pixel data, the encoding parameter, and the decoding error information output from the decoding processing unit 1, outputs the decoded image when normal decoding is performed, and outputs these when an error occurs. Based on the input, a concealed image in which the video disturbed by the decoding error is corrected is generated and output.

  The error concealment processing unit 2a will be described in further detail. The frame memory 21 stores the pixel data output from the decoding processing unit 1 in units of pictures. The encoding parameter storage unit 22 stores the encoding parameters output from the decoding processing unit 1 in units of pictures. The concealment management unit 23 determines ON / OFF of the concealment processing for each picture based on the decoding error information output from the decoding processing unit 1. The concealment range calculation unit 24 calculates the concealment range when the ON / OFF determination result of the concealment process determined by the concealment management unit 23 is ON. The division pattern storage unit 25 stores a plurality of types of subgroup subdivision patterns in which subgroups obtained by subdividing the division group into which the concealment range is divided are set. The conceal range dividing unit 26 divides the conceal range calculated by the conceal range calculation unit 24 into divided groups, and further divides the conceal range into sub groups based on a plurality of subgroup subdivision patterns stored in the divided pattern storage unit 25. . The estimated motion vector calculation unit 27a calculates an estimated motion vector representing the divided group for each divided group / subgroup within the concealed range. The concealed image generation unit 28 generates a concealed image from the estimated motion vector calculated by the estimated motion vector calculation unit 27a and the pixel data stored in the frame memory when the ON / OFF determination result of the concealment process is ON.

  FIG. 2 is a flowchart showing an example of the operation of the decoding apparatus equipped with the error concealment processing unit according to Embodiment 1 of the present invention.

  In the figure, in step ST101, the concealment management unit 23 determines ON / OFF of the concealment process for each picture based on the decoding error information. In step ST102, the conceal range calculation unit 24 calculates the conceal range based on the conceal ON / OFF determination. In step ST103, the concealed range dividing unit 26 divides the concealed range into divided groups. In step ST104, the concealment range dividing unit 26 selects one subgroup subdivision pattern for each group based on the plurality of subgroup subdivision patterns, and divides it into subgroups corresponding to the pattern. In step ST105, the estimated motion vector calculation unit 27a calculates an estimated motion vector representing each divided group / subgroup for each divided group / subgroup within the concealed range. In step ST106, the concealed image generation unit 28 generates an output image from the estimated motion vector and the pixel data of the reference surface based on the reference surface information. If the concealment is OFF in step ST101, the concealed image generation unit 28 outputs an image composed of normally decoded pixel data without generating a concealed image in step ST106.

  Next, further detailed processing will be described.

  First, when the encoded data encoded by the encoding device is input, the decoding processing unit 1 performs a decoding process to generate a decoded image, and outputs pixel data of the decoded image to the error concealment processing unit 2a. . In addition, the encoding parameters used in the decoding process are output to the error concealment processing unit 2a. The decoding processing unit 1 outputs the decoding error information detected during the decoding process to the error concealment processing unit 2a. Here, the encoding parameter is, for example, that the encoding method is H.264. In the case of H.264, the slice configuration and picture type in the screen, the prediction mode for each macroblock, the motion vector, and the like. The decoding error information is, for example, the presence or absence of decoding error detection and the decoding error detection position in the image.

  The pixel data output from the decoding processing unit 1 to the error concealment processing unit 2a is stored in the frame memory 21 for each picture, and the encoding parameter is stored in the encoding parameter storage unit 22 for each picture. Is notified to the concealment management unit 23 as needed.

  The concealment management unit 23 constantly monitors the decoding error information from the decoding processing unit 1, calculates a range of influence when a decoding error is detected, determines conceal ON / OFF for each picture, and uses a reference surface for concealment Is calculated and set.

  For example, H.M. In the case of an encoding scheme in which a next picture is generated by referring to another picture such as H.264, the influence of the decoding error is not only the picture in which the decoding error is detected but also the picture in which the decoding error is detected. Propagated to the picture to be. Therefore, the concealment management unit 23 determines that the picture having the decoding error detected and the picture having the decoding error detected as the reference plane is “conceal ON”, and the others are “conceal OFF”.

  In addition, the reference plane used for concealment is a picture used to supplement information lost due to an error, and the image close to time is characterized by high correlation. Select a picture that is close and successfully decoded. For example, when the picture type to be concealed is an I (intra coding) / P (forward reference coding) picture, one normally decoded I / P picture whose display order is immediately before the concealment target picture is concealed. When the concealment target is a B (forward reference / backward reference encoding) picture, two I / P pictures that are normally decoded immediately before and after the concealment target picture are used as the concealment reference plane. .

  When the concealment management unit 23 determines that “conceal ON”, the conceal range calculation unit 24 calculates the start position and end position of the slice including the macroblock in which the decoding error is detected from the decoding error information and the slice information, and the slice The whole is the concealment range. For example, a description will be given with reference to FIG. 3 which is an explanatory diagram of an example of concealment range setting according to Embodiment 1 of the present invention. H. When an error is mixed in encoded data encoded by a variable length coding method such as H.264, the position where the error is actually mixed is different from the position where the error is detected at the time of decoding, and decoding is performed before the error detection position. The image may be distorted. Therefore, by setting the entire slice, which is a unit of encoding, as the concealment range, a range including all the places where the video may be disturbed is set as the target of the concealment process.

  When the concealed range calculation unit 24 determines the concealed range, the concealed range dividing unit 26 divides the concealed range into a group composed of a plurality of macroblocks (MB) and each group into a plurality of subgroups. Here, the division is performed in the following two stages.

  First, as a first step, the concealment range is divided into groups of a certain size (for example, 4 MB × 4 MB). For example, a description will be given with reference to FIG. 4 which is an explanatory diagram of an example of group division of the concealment range according to the first embodiment of the present invention. Since an image within the concealment range includes a plurality of movements, the concealment process reflecting the movement is performed for each location by dividing the concealment range.

  Next, as a second stage, for each group divided in the first stage, it is evaluated whether subdivision into sub-groups is effective. If it is determined to be effective, subdivision is performed. For example, a description will be given with reference to FIG. 5 which is an explanatory diagram of an example of a subgroup subdivision pattern of divided groups according to Embodiment 1 of the present invention. A plurality of subgroup subdivision patterns to be applied to a group obtained by dividing the concealment range are prepared, and the inside of the group is subdivided into subgroups. Here, an example of the subgroup subdivision pattern for the group of 4 MB × 4 MB divided in the first stage is shown, but the shape and the number of patterns to be subdivided can be arbitrarily set and have the same size as the group 1 You may include a pattern with only one subgroup. The plurality of subgroup subdivision patterns used here are stored and stored in the division pattern storage unit 25 and are referred to as appropriate by the concealment range division unit 26.

  In the concealment process for each group, the degree of variation in image motion within each subgroup is evaluated for all of the plurality of subgroup subdivision patterns, and the subgroup subdivision pattern with the least variation is employed. Here, a small variation in motion means a state in which the motion is uniform. For example, using the motion vector of the macroblock on the reference plane at the same position as the macroblocks that make up the subgroup, calculate the variance of the motion vector in the subgroup, and use this variance as the evaluation value to subdivide all subgroups Subdivision is performed by adopting a pattern in which the maximum value of the variance of the motion vectors of the subgroup is the smallest among the patterns (a divided pattern with less variation of motion vectors in the subgroup). By doing this, concealment processing is performed in large subgroup units for places where movement is simple, and concealment processing is performed in subgroup units that are subdivided only in the portion where the movement is complicated. Can do. Note that a priority may be provided for a subgroup subdivision pattern for determination when there are a plurality of subgroup subdivision patterns having the same maximum evaluation value.

  When the concealed range dividing unit 26 determines a concealed range dividing method, the estimated motion vector calculating unit 27a calculates an estimated motion vector for each group / subgroup. For example, a description will be given with reference to FIG. 6, which is an explanatory diagram of an example of calculation of an estimated motion vector according to Embodiment 1 of the present invention. The estimated motion vector is calculated using the motion vector possessed by the macroblock in the reference plane at the same position as the macroblock in the group. As a method of calculating the estimated motion vector, for example, an average value, median value (median) of motion vectors, or a vector with the highest frequency (mode) taking a histogram of motion vectors may be adopted. The calculation method is not limited to a predetermined statistical value.

  In the concealed image generating unit 28, when the estimated motion vector calculating unit 27a completes the calculation of the estimated motion vector, a decoded image that is normally decoded except for the concealed range is set, and a concealed range that cannot be normally decoded is handled. A concealed image is generated and output by setting a corrected image obtained from the pixel data of the reference plane based on the estimated motion vector calculated for each group / subgroup.

  As described above, according to the decoding apparatus and decoding method according to Embodiment 1 of the present invention, the entire slice in which an error is detected is set as a concealed range, and the concealed range is divided into groups of a plurality of macroblocks. The congested place is subdivided into subgroups, and estimated motion vectors are calculated for each group / subgroup within the concealed range from the motion vectors of the same-position macroblocks of successfully decoded pictures that are temporally close to the error picture. Therefore, the error concealment process is performed by generating the corrected image by acquiring the pixel at the position corresponding to the estimated motion vector from the reference plane, so that the concealed image with less sense of incongruity maintaining the motion in the video can be obtained. Can be obtained.

Embodiment 2. FIG.
In the first embodiment described above, the corrected motion image is generated by calculating the estimated motion vector. Next, in the second embodiment, a description will be given of correcting an estimated motion vector when the estimated motion vector points outside the screen.

  FIG. 7 is a block diagram showing a decoding apparatus equipped with an error concealment processing unit according to Embodiment 2 of the present invention.

  In the figure, the decoding device includes a decoding processing unit 1 and an error concealment processing unit 2b. The decryption processing unit 1 has the same function and operates as the decryption processing unit 1 described in FIG. 1 in the first embodiment of the present invention, so that the description thereof is omitted. Further, the frame memory 21, the encoding parameter storage unit 22, the concealment management unit 23, the conceal range calculation unit 24, the division pattern storage unit 25, the conceal range division unit 26, and the concealment image generation unit 28 that constitute the error concealment processing unit 2b. 1 has the same function as that of the components having the same reference numerals attached to the internal configuration of the error concealment processing unit 2a described in FIG. 1 in the first embodiment of the present invention, and the description thereof is omitted. To do. Therefore, the estimated motion vector calculation unit 27b and the out-of-screen reference estimated motion vector detection unit 29 will be described.

  The estimated motion vector calculation unit 27b calculates an estimated motion vector in the same manner as the estimated motion vector calculation unit 27a described in FIG. 1 in the first embodiment of the present invention, and then the off-screen reference estimation motion vector detection unit 29 performs the calculation. The estimated motion vector corrected based on the group information referring outside the screen of the detected estimated motion vector is sent to the concealed image generating unit. The out-of-screen reference estimated motion vector detection unit 29 checks whether the estimated motion vector calculated by the estimated motion vector calculation unit 27b refers to the outside of the screen for each group / subgroup, and refers to the outside of the screen. In this case, the estimated motion vector calculation unit 27b is notified of group information referring to the outside of the screen.

  FIG. 8 is a flowchart showing an example of the operation of the decoding apparatus equipped with the error concealment processing unit according to the second embodiment of the present invention, and the error according to the first embodiment of the present invention shown in FIG. The difference with the flowchart which shows operation | movement of the decoding apparatus carrying the concealment process part is shown.

  In the figure, steps ST101 to ST105 and step ST106 (not shown) perform the same processing as the flowchart showing the operation of the decoding apparatus according to Embodiment 1 of the present invention shown in FIG. Here, the following steps ST201 and ST202 are added between steps ST105 and ST106.

  In step ST201, the out-of-screen reference estimated motion vector detection unit 29 detects whether there is a reference to the outside of the screen in the estimated motion vector calculated in step ST105. When there is no reference to the outside of the screen in step ST201, the estimated motion vector is not corrected and the process proceeds to step ST106. When it is detected in step ST201 that there is a reference to the outside of the screen, in step ST202, the estimated motion vector calculation unit 27b corrects the estimated motion vector in consideration of the reference to the outside of the screen. After that, in step ST106, the concealed image generation unit 28 generates an output image from the estimated motion vector calculated in step ST105 or the estimated motion vector corrected in step ST202 and the reference plane pixel data based on the reference plane information. . If the concealment is OFF in step ST101, the concealed image generation unit 28 outputs an image composed of normally decoded pixel data without generating a concealed image in step ST106.

  FIG. 9 is a conceptual diagram illustrating an example of error concealment processing of the decoding apparatus according to Embodiment 2 of the present invention. Detailed processing will be described with reference to FIG.

  In the figure, it is assumed that decoded images are displayed in the order of P1 picture that is P (forward reference coding) picture, P2 picture, and I3 picture that is I (intra coding) picture, and P1 picture and I3 picture are decoded normally. The P2 picture shows a case where there is a decoding error. Usually, the decoded image is disturbed before and after the decoding error, but here the object is drawn as it is in the same state as in normal decoding. Note that an I3 picture that is an intra-coded picture does not have a reference plane, and encoding is completed in one sheet. Therefore, the I3 picture does not have a motion vector that refers to another picture, and is affected by decoding errors of other pictures. Not receive.

  Here, since the decoding processing unit 1 detects the decoding error, calculates the concealed range, divides the concealed range, and calculates the estimated motion vector, it is the same as the processing described in the first embodiment. The description is omitted.

(1) Calculation of Estimated Motion Vector First, when the estimated motion vector calculation unit 27b continues the same motion as the immediately preceding P1 picture in which the display order is normally decoded in the forward direction with respect to the concealed range, for example. Then, an estimated motion vector of the P2 picture is calculated from the motion vector of the P1 picture.

(2) Detection of off-screen reference estimated motion vector Next, the off-screen reference estimated motion vector detection unit 29 determines whether or not the estimated motion vector calculated by the estimated motion vector calculation unit 27b refers to the outside of the screen. In the case of referring to the outside of the screen, the estimated motion vector calculating unit 27b is notified of the group information referring to the outside of the screen. Here, the out-of-screen reference estimated motion vector detection unit 29 detects that the leftmost group of the estimated motion vector calculated by the estimated motion vector calculation unit 27b indicates an off-screen reference, and the leftmost The group information is notified to the estimated motion vector calculation unit 27b.

(3) Correction of out-of-screen reference estimated motion vector When the estimated motion vector indicates an out-of-screen reference, it is not possible to generate a concealed image in which pixel data is acquired and corrected using the P1 picture as a reference plane. A concealment process is performed using the I3 picture whose order is backward in time as a reference plane. Here, for example, assuming that the I3 picture moves the same as the P1 picture immediately before the error picture P2, the estimated motion vector is calculated using the same motion vector as the P1 picture immediately before the error picture. However, since the reference direction from the I3 picture to the error picture P2 is opposite to the reference direction from the error picture P2 to the I3 picture, the direction of the estimated motion vector is set to be inverted.
The estimated motion vector calculation unit 27b corrects the estimated motion vector based on the group information that refers to the outside of the screen notified from the out-of-screen reference estimated motion vector detection unit 29. Here, the direction of the estimated motion vector of the leftmost group notified from the off-screen reference estimated motion vector detection unit 29 is reversed.

(4) Generation of Concealed Image The concealed image generation unit 28 is based on the estimated motion vector (when there is no reference outside the screen) calculated by the estimated motion vector calculation unit 27b or the corrected estimated motion vector (when there is a reference outside the screen). Thus, a concealed image of the P2 picture is generated with reference to the picture on the reference plane. Here, based on the estimated motion vector, the concealed image of the P2 picture is generated by appropriately referring to the P1 picture whose display order is temporally forward and the I3 picture of the backward direction as reference planes.

  FIG. 9 shows a case where the motion vector of the P1 picture has one motion vector for each group. However, each group is subdivided into subgroup units shown in FIG. Is set, and the concealed image may be generated with reference to the image of the reference surface.

  In addition, when the display times of two pictures are separated, the magnitude of the estimated motion vector is proportionally corrected according to the time interval. Note that, instead of this time interval, the estimated motion vector may be corrected according to the number of frames between two pictures for a video assuming a motion at a constant frame rate. In addition, an estimated motion vector corresponding to the number of frames between two pictures may be corrected in consideration of the acceleration of a motion target estimated from a plurality of past frames.

  As described above, according to the decoding apparatus and decoding method according to Embodiment 2 of the present invention, as in the decoding apparatus and decoding method according to Embodiment 1 of the present invention, the entire slice from which an error has been detected is concealed. The concealed area is divided into groups consisting of a plurality of macroblocks, and the locations where the motion is complicated are subdivided into subgroups, and the motion of the co-located macroblocks in the normally decoded picture that is close in time to the error picture An estimated motion vector is calculated for each group / subgroup within the concealed range from the vector, and a pixel at a position corresponding to the estimated motion vector is acquired from the reference plane to generate a corrected image and perform error concealment processing. Therefore, it is possible to obtain a concealed image with less sense of incongruity while maintaining the movement in the video.

  Further, according to the decoding apparatus and decoding method according to Embodiment 2 of the present invention, when the estimated motion vector obtained using the reference plane whose display order is the forward direction indicates an off-screen reference, the estimated motion The vector calculation unit 27b changes to use the backward reference plane and corrects the motion vector to be in the opposite direction, so that the estimated motion vector is concealed without image loss even in the group at the screen edge. be able to.

Embodiment 3 FIG.
In the second embodiment described above, when the estimated motion vector points outside the screen, the estimated motion vector is reversed and the concealed image is generated by referring from the backward direction. Next, the third embodiment of the present invention corrects an estimated motion vector when the estimated motion vector points outside the screen and a scene change has occurred in a backward I (intra-coded) picture. Will be described.

  FIG. 10 is a block diagram showing a decoding apparatus equipped with an error concealment processing unit according to Embodiment 3 of the present invention.

  In the figure, the decoding device includes a decoding processing unit 1 and an error concealment processing unit 2c. The decryption processing unit 1 has the same function and operates as the decryption processing unit 1 described in FIG. 7 in the second embodiment of the present invention, so that the description thereof is omitted. In addition, the frame memory 21, the encoding parameter accumulation unit 22, the concealment management unit 23, the conceal range calculation unit 24, the division pattern storage unit 25, the conceal range division unit 26, and the concealment image generation unit 28 that constitute the error concealment processing unit 2c. The out-of-screen reference estimation motion vector detection unit 29 has the same functions as those of the components having the same reference numerals attached to the internal configuration of the error concealment processing unit 2b described in FIG. 7 in the second embodiment of the present invention. Therefore, the description thereof is omitted. Therefore, the estimated motion vector calculation unit 27c and the scene change detection unit 30 will be described.

  The scene change detection unit 30 detects whether or not a scene change has occurred in the decoded image and constantly monitors it, and notifies the estimated motion vector calculation unit 27c that a scene change has occurred. The estimated motion vector calculation unit 27c calculates an estimated motion vector in the same manner as the estimated motion vector calculation unit 27b described in FIG. 7 in the second embodiment of the present invention, and then the off-screen reference estimation motion vector detection unit 29 performs the calculation. The notified estimated motion vector is corrected based on the detection result of the scene change detection unit 30 and sent to the concealed image generation unit 28.

  FIG. 11 is a flowchart showing an example of the operation of the decoding apparatus equipped with the error concealment processing unit according to the third embodiment of the present invention, and the error according to the first embodiment of the present invention shown in FIG. The difference with the flowchart which shows operation | movement of the decoding apparatus carrying the concealment process part is shown.

  In the figure, steps ST101 to ST105 and step ST106 (not shown) perform the same processing as the flowchart showing the operation of the decoding apparatus according to Embodiment 1 of the present invention shown in FIG. Steps ST201 and ST202 perform the same processing as the flowchart showing the operation of the decoding apparatus according to Embodiment 2 of the present invention shown in FIG. Here, step ST301 is added between steps ST201 and ST202, and step ST302 is added between steps ST301 and ST106.

  In step ST201, the out-of-screen reference estimated motion vector detection unit 29 detects whether there is a reference to the outside of the screen in the estimated motion vector calculated in step ST105. When there is no reference to the outside of the screen in step ST201, the estimated motion vector is not corrected and the process proceeds to step ST106. When it is detected in step ST201 that there is a reference to the outside of the screen, in step ST301, the scene change detection unit 30 checks the detection status of whether or not a scene change has occurred. When a scene change occurs in step ST301, in step ST302, the estimated motion vector calculation unit 27c corrects the estimated motion vector in consideration of the off-screen reference and the scene change. When no scene change occurs in step ST301, the estimated motion vector calculation unit 27c corrects the estimated motion vector in step ST202. Thereafter, in step ST106, the concealed image generation unit 28 calculates the estimated motion vector calculated in step ST105 or the estimated motion vector corrected in either step ST202 or step ST302, and pixel data of the reference surface based on the reference surface information. To generate an output image. If concealment is OFF in step ST101, in step ST106, the concealed image generating unit 28 does not generate a concealed image, but outputs an image composed of normally decoded pixel data.

  Next, detailed processing will be described. The decoding processing unit 1 detects a decoding error, calculates a concealed range, divides the concealed range, calculates an estimated motion vector, and detects (2) off-screen reference estimated motion vector in FIG. The process up to the point when the motion vector detection unit 29 notifies the estimated motion vector calculation unit 27c of the group information that refers to the outside of the screen is the same as the process described in the second embodiment, and thus the description thereof is omitted. (3) The correction of the out-of-screen reference estimated motion vector will be described.

(3) Correction of the estimated motion vector outside the screen The estimated motion vector calculation unit 27c, when the estimated motion vector indicates a reference outside the screen, is described as P (forward reference encoding), as described in the second embodiment. Since a concealed image in which pixel data is acquired and corrected using the picture (P1 picture) as a reference plane cannot be generated, a temporally backward I (intra-encoded) picture (I3 picture) in the display order is used as the reference plane. The reference direction from the I3 picture to the P2 picture in error and the reference direction from the P2 picture to the I3 picture are opposite to each other, so that the direction of the estimated motion vector is reversed.

  Here, the estimated motion vector calculation unit 27c confirms the scene change detection status in the backward I3 picture from the scene change detection unit 30 before the correction for inverting the direction of the estimated motion vector, and the scene change is detected. If it is, correction for setting the zero vector is performed without performing correction for inverting the direction of the estimated motion vector. In this correction, for example, the concealment process is performed using the same-position pixel data of the P2 picture as it is. This is because, when a scene change occurs in the backward I3 picture, the correlation between the P2 picture in error and the backward I3 picture is considered to be low, so the backward picture cannot be used for the reference plane. Therefore, when the estimated motion vector calculated by the estimated motion vector calculation unit 27c indicates an off-screen reference and a scene change has occurred in the backward I picture, the group has both forward reference and backward reference. Since it is not effective, the same-position pixel data of the P2 picture is used. The P2 picture used here includes an error, but the same-position pixel data used for the concealment process may not have an error. In this case, even if an error has occurred, the P2 picture is used as it is.

  In this way, when the estimated motion vector indicates an off-screen reference, different correction processing is executed, which includes correction for inverting the direction of the estimated motion vector and correction for setting the zero vector depending on whether or not a scene change has occurred. Further, it is possible to perform a two-stage correction in which a correction for inverting the direction of the estimated motion vector is performed and then a correction for setting the zero vector is performed again.

  As described above, according to the decoding device and the decoding method according to Embodiment 3 of the present invention, an error is detected as in the decoding device and the decoding method according to Embodiment 1 and Embodiment 2 of the present invention. The entire slice is defined as a concealed range, and the concealed range is divided into groups consisting of a plurality of macroblocks. The place where the motion is complicated is subdivided into subgroups, and the same as the correctly decoded picture that is temporally similar to the error picture. An estimated motion vector is calculated for each group / subgroup within the concealment range from the motion vector of the position macroblock, and a corrected image is generated by acquiring a pixel at a position corresponding to the estimated motion vector from the reference plane, thereby generating an error concealment. Process is performed so that a concealed image with less discomfort can be obtained while maintaining the movement in the video. Door can be.

  Moreover, according to the decoding apparatus and decoding method which concern on Embodiment 3 of this invention, using the reference surface where a display order is a front direction similarly to the decoding apparatus and decoding method which concern on Embodiment 2 of this invention When the obtained estimated motion vector indicates a reference outside the screen, the estimated motion vector calculation unit 27c is changed to use the backward reference plane, and the motion vector is corrected so as to be in the opposite direction, Concealment can be performed without omission of an image even in a group at the edge of the screen.

  In addition, according to the decoding apparatus and decoding method according to Embodiment 3 of the present invention, the estimated motion vector that is provided with the scene change detection unit 30 and that is obtained by using the reference plane whose display order is the forward direction is the off-screen reference. When a scene change occurs in a backward I (intra coding) picture, the same-position pixel data of a P (forward reference coding) picture including an error is used as it is. Therefore, it is possible to prevent the generation of a concealed image with reference to an image whose correlation is lost due to a scene change.

Embodiment 4 FIG.
In the first embodiment described above, the corrected motion image is generated by calculating the estimated motion vector. Next, the fourth embodiment shows an embodiment in which the necessity of correction is determined by re-evaluating the calculated estimated motion vector, and the estimated motion vector is corrected when correction is necessary.

  FIG. 12 is a block diagram showing a decoding apparatus equipped with an error concealment processing unit according to Embodiment 4 of the present invention.

  In the figure, the decoding device includes a decoding processing unit 1 and an error concealment processing unit 2d. The decryption processing unit 1 has the same function and operates as the decryption processing unit 1 described in FIG. 1 in the first embodiment of the present invention, so that the description thereof is omitted. In addition, the frame memory 21, the encoding parameter storage unit 22, the concealment management unit 23, the conceal range calculation unit 24, the division pattern storage unit 25, the conceal range division unit 26, and the concealment image generation unit 28 that constitute the error concealment processing unit 2 d. 1 has the same function as that of the components having the same reference numerals attached to the internal configuration of the error concealment processing unit 2a described in FIG. 1 in the first embodiment of the present invention, and the description thereof is omitted. To do. Therefore, the estimated motion vector calculation unit 27d and the estimated motion vector evaluation unit 31 will be described.

  The estimated motion vector calculation unit 27d needs to be corrected by calculating the estimated motion vector in the same manner as the estimated motion vector calculation unit 27a described in FIG. 1 in the first embodiment of the present invention and re-evaluating the estimated motion vector. In such a case, the estimated motion vector is corrected. The estimated motion vector evaluation unit 31 evaluates the necessity of correction of the estimated motion vector calculated by the estimated motion vector calculation unit 27d by evaluating the synchrony of the estimated motion vector of each divided group and the estimated motion vectors of the surrounding divided groups. Determine.

  FIG. 13 is a flowchart showing an example of the operation of the decoding apparatus equipped with the error concealment processing unit according to the second embodiment of the present invention, and the error according to the first embodiment of the present invention shown in FIG. The difference with the flowchart which shows operation | movement of the decoding apparatus carrying the concealment process part is shown.

  In the figure, steps ST101 to ST105 and step ST106 (not shown) perform the same processing as the flowchart showing the operation of the decoding apparatus according to Embodiment 1 of the present invention shown in FIG. Here, steps ST401 to ST403 are added between steps ST105 and ST106.

  In step ST401, the estimated motion vector evaluation unit 31 evaluates the necessity of correction in the estimated motion vector calculated in step ST105. In step ST402, it is determined whether or not the estimated motion vector needs to be corrected. When it is determined in step ST402 that correction is not necessary, the estimated motion vector is not corrected and the process proceeds to step ST106. When it is determined in step ST402 that correction is necessary, the estimated motion vector is corrected in step ST403. Thereafter, in step ST106, the concealed image generation unit 28 generates an output image from the estimated motion vector calculated in step ST105 or the estimated motion vector corrected in step ST403, and pixel data of the reference surface based on the reference surface information. . If the concealment is OFF in step ST101, the concealed image generation unit 28 outputs an image composed of normally decoded pixel data without generating a concealed image in step ST106.

  Next, detailed processing will be described. Since the decoding processing unit 1 detects the decoding error, calculates the concealed range, divides the concealed range, and calculates the estimated motion vector, it is the same as the processing described in the first embodiment, and therefore the description thereof. Is omitted.

  The estimated motion vector evaluation unit 31 re-evaluates the calculated estimated motion vector after the estimated motion vector calculation unit 27d calculates all the estimated motion vectors of each group, and the estimated motion vector determined to be corrected. Is corrected and reflected. For example, a difference between motion vectors of groups around group A is calculated, and if the calculated difference is within a threshold, it is determined that group A is moving in the same manner as the surroundings, and the estimated motion of group A The vector is corrected to the one calculated from the estimated motion vectors of the surrounding groups. The calculation method may be any method such as an average value, mode value (mode), and median value (median) of estimated motion vectors of surrounding groups. Moreover, you may carry out a weighted average by arbitrary weighting.

  As described above, according to the decoding apparatus and decoding method according to Embodiment 4 of the present invention, as in the decoding apparatus and decoding method according to Embodiment 1 of the present invention, the entire slice from which an error has been detected is concealed. The concealed area is divided into groups consisting of a plurality of macroblocks, and the locations where the motion is complicated are subdivided into subgroups, and the motion of the co-located macroblocks in the normally decoded picture that is close in time to the error picture An estimated motion vector is calculated for each group / subgroup within the concealed range from the vector, and a pixel at a position corresponding to the estimated motion vector is acquired from the reference plane to generate a corrected image and perform error concealment processing. Therefore, it is possible to obtain a concealed image with less sense of incongruity while maintaining the movement in the video.

  Further, according to the decoding apparatus and decoding method according to Embodiment 4 of the present invention, after all the estimated motion vectors of each group are calculated, the estimated motion vector evaluation unit 31 deviates from the estimated motion vectors of the surrounding groups. If so, correction is made so as to synchronize, so that it is possible to prevent generation of concealed images having different reference positions locally.

  Note that the estimated motion vector evaluation unit 31 according to the fourth embodiment of the present invention is applied to the estimation of the estimated motion vector obtained by the decoding apparatus and the decoding method according to the second or third embodiment of the present invention and added. Correction processing may be performed, or the initial estimated motion vector may be first evaluated and corrected, and then corrected by the decoding apparatus and decoding method according to Embodiment 2 or 3 of the present invention. A concealed image may be generated based on the estimated motion vector corrected in this way.

Embodiment 5 FIG.
FIG. 14 is a block diagram showing a decoding apparatus equipped with an error concealment processing unit according to Embodiment 5 of the present invention.

  In the figure, the decoding device comprises a decoding processing unit 1 and an error concealment processing unit 2e. The decryption processing unit 1 has the same function and operates as the decryption processing unit 1 described in FIG. 1 in the first embodiment of the present invention, so that the description thereof is omitted. Further, the frame memory 21, the encoding parameter accumulation unit 22, the concealment management unit 23, the concealment range calculation unit 24, and the division pattern storage unit 25 that constitute the error concealment processing unit 2e are shown in FIG. Since it has the same function as the component of the same code | symbol attached | subjected to the internal structure of the error concealment process part 2a demonstrated in (2), it abbreviate | omits the description. Therefore, the 1st concealment process execution part (1st concealment process execution part) 41a, the 2nd concealment process execution part (2nd concealment process execution part) 41b, and the concealment result determination part 42 are demonstrated.

  The first concealment processing execution unit 41a and the second concealment processing execution unit 41b are different from each other in the concealment range as in the predetermined concealment processing method described in the first to fourth embodiments, for example. A concealment process is performed and a concealed image is output. The concealment result determination unit 42 compares the concealment results of the first concealment processing execution unit 41a and the second concealment processing execution unit 41b with the decoded image, and determines the concealed image by the concealment processing method having the smallest difference. Generate.

  Next, detailed processing will be described. Since the decoding process unit 1 detects the decoding error, calculates the concealment range, and divides the concealment range into groups and subgroups, the process is the same as that described in the first embodiment. To do.

  The error concealment processing unit 2e includes a plurality of concealment processing execution units (concealment processing execution units). Here, there are two, but the first concealment processing execution unit 41a and the second concealment processing execution unit 41b are Of the four concealment processing methods described in the first to fourth embodiments, different first and second concealed images are generated by one of the two methods. At this time, a plurality of concealment processing methods are executed for one or a plurality of slices that have been successfully decoded, and a concealed image is obtained. The concealment result determination unit 42 compares the first and second concealed images applied to the successfully decoded slices with the corresponding successfully decoded slices, and determines the concealment processing method with the smallest difference as the error slice. To generate a concealed image.

  FIG. 15 is an explanatory diagram showing an example of generating a concealed image in the decoding apparatus according to Embodiment 5 of the present invention.

  In the figure, picture A is a picture that has been successfully decoded, and picture B is a picture (initially referred to as B0) in which an error is included in slice 2. Here, concealed pictures B1 and B2 are generated by applying the first and second concealment processing methods using picture A as a reference plane. For slices 1, 3, and 4 that do not contain an error in picture B, the evaluation is performed by comparing the slices 1, 3, and 4 in pictures B 1 and B 2 with differences. When the difference between the pictures B0 and B1 is smaller than the difference between the pictures B0 and B2, the concealment processing method 1 is adopted, and the slice 2 is also concealed by the concealment processing method 1 to obtain a final concealed image.

  The slices 1, 3, and 4 that do not include the error of the picture B have been described as being compared and evaluated with the slices 1, 3, and 4 of the pictures B1 and B2, respectively, but the error of the picture B is not included. All slices may not be used for evaluation, but only a part of slices may be used. For example, the determination is performed on one or both of the immediately adjacent slice and the immediately preceding slice adjacent to the slice containing the error. Similarly, when an error is included in a plurality of slices separated in the same picture, each error slice may be determined by the preceding and succeeding slices, and the entire picture may not be concealed by the same concealment processing method. When a plurality of slices are used for evaluation, evaluation may be performed by calculating an evaluation value with weighting based on the positional relationship with the error slice.

  Here, the decoding apparatus according to Embodiment 5 of the present invention has been described as including two concealment processing execution units. However, three or more concealment processing execution units may be provided to generate a concealed image. For example, when the concealment processing execution unit applies the third embodiment of the present invention, the connection omitted from the encoding parameter storage unit 22 to the scene change detection unit 30 assumed inside the concealment processing execution unit. Shall be added as appropriate.

  Moreover, although the structure which shares the common element of each concealment process execution part of Embodiment 1 to Embodiment 4 of this invention by having a several concealment process execution part was shown, for example in FIG. Even if it becomes redundant, each concealment processing execution unit may be provided independently. In addition, the concealment processing execution unit may employ other concealment processing methods as well as the concealment processing methods according to the first to fourth embodiments of the present invention.

  As described above, according to the decoding device and the decoding method according to Embodiment 5 of the present invention, a plurality of concealment processing methods are executed on a slice that has been successfully decoded, and Since the concealment processing method having the smallest difference is calculated and the concealment image is generated, the concealment processing method most suitable for the concealment of the current image is selected and the concealment image is generated. Can do.

  As described above, according to the decoding apparatus and the decoding method according to the embodiment of the present invention, the entire slice in which an error is detected is set as a concealed range, and the concealed range is divided into groups of a plurality of macroblocks, and motion is complicated The sub-group is subdivided into sub-groups, and the estimated motion vector is calculated for each group / sub-group within the concealed range from the motion vector of the co-located macroblock of the successfully decoded picture that is temporally close to the error picture, Since the error concealment process is performed by generating the corrected image by acquiring the pixel at the position according to the estimated motion vector from the reference plane, it is possible to obtain a concealed image with less discomfort while maintaining the motion in the video Can do.

  Note that the decoding device according to the embodiment of the present invention may be configured as a device corresponding to a decoding processing unit and an error concealment processing unit. For example, in the block diagram showing an example of the decoding apparatus according to Embodiment 1 of the present invention shown in FIG. 1, the encoded data is decoded and the decoded pixel data, the encoding parameter, and the decoding error information are output. When the decoding processing unit 1 is provided as a decoding processing device, the pixel data, the encoding parameter, and the decoding error information output from the decoding processing device are input, and the decoded image is output when normal decoding is performed. If an error occurs, the error concealment processing unit 2a that generates and outputs a concealed image obtained by correcting a video distorted by a decoding error based on these inputs may be provided as an error concealment processing device in a single configuration. it can.

DESCRIPTION OF SYMBOLS 1 Decoding processing part, 2a, 2b, 2c, 2d, 2e Error concealment processing part, 21 Frame memory, 22 Encoding parameter storage part, 23 Conceal management part, 24 Conceal range calculation part, 25 Division pattern storage part, 26 Conceal Range dividing unit, 27a, 27b, 27c, 27d Estimated motion vector calculating unit, 28 Concealed image generating unit, 29 Out-of-screen reference estimated motion vector detecting unit, 30 Scene change detecting unit, 31 Estimated motion vector evaluating unit, 41a First A concealment process execution part (first concealment process execution part), 41b a second concealment process execution part (second concealment process execution part), 42 a concealment result determination part.

Claims (11)

A decoding processing unit that decodes the encoded data and outputs pixel data after decoding, encoding parameters, and decoding error information;
A frame memory for storing pixel data output by the decoding processing unit in units of pictures;
An encoding parameter storage unit for storing the encoding parameter in units of pictures;
Based on the decoding error information, a concealment management unit that determines ON / OFF of concealment processing for each picture;
A concealment range calculation unit for calculating a concealment range when the ON / OFF determination result of the concealment process is ON;
A divided pattern storage unit for storing a plurality of types of subgroup subdivision patterns in which subgroups obtained by subdividing the subgroups obtained by dividing the conceal range are set;
Dividing the concealed range into the divided groups, and further dividing the concealed range into the subgroups based on the plurality of subgroup subdivision patterns; and
An estimated motion vector calculating unit that calculates an estimated motion vector representing the divided group / subgroup for each divided group / subgroup within the concealed range;
A decoding apparatus comprising: a concealed image generation unit configured to generate a concealed image from the estimated motion vector and pixel data stored in the frame memory when an ON / OFF determination result of the concealment process is ON. An estimated out-of-screen reference motion vector detection unit that detects whether or not the estimated motion vector calculated by the estimated motion vector calculation unit refers to outside the screen;
The decoding apparatus according to claim 1, wherein the estimated motion vector calculation unit corrects the estimated motion vector when the estimated motion vector refers to an outside of the screen. A scene change detection unit that detects whether or not a scene change has occurred in the decoding process of the encoded data;
The said estimated motion vector calculation part performs the correction process of the said estimated motion vector based on the presence or absence of generation | occurrence | production of the said scene change, when the said estimated motion vector is referring the outside of a screen. Decoding device. An estimated motion vector evaluation unit that determines the necessity of correction by evaluating an estimated motion vector of each divided group and an estimated motion vector of a surrounding divided group for the estimated motion vector calculated by the estimated motion vector calculation unit;
The decoding apparatus according to claim 1, wherein the estimated motion vector calculation unit corrects the estimated motion vector based on the determination result of the necessity of correction. A decoding processing unit that decodes the encoded data and outputs pixel data after decoding, encoding parameters, and decoding error information;
A frame memory for storing pixel data output by the decoding processing unit in units of pictures;
An encoding parameter storage unit for storing the encoding parameter in units of pictures;
Based on the decoding error information, a concealment management unit that determines ON / OFF of concealment processing for each picture;
A concealment range calculation unit for calculating a concealment range when the ON / OFF determination result of the concealment process is ON;
A conceal range dividing unit that divides the divided group into the subgroups based on a plurality of subgroup subdivision patterns that divide the concealed range into the divided groups and set subgroups obtained by subdividing the divided groups;
A plurality of concealment processing execution units that generate concealment images by different concealment processes;
When the ON / OFF determination result of the concealment process is ON, the decoding range that is normally decoded other than the concealment range and the plurality of concealment processing execution units corresponding to the decoding range applied the different concealment processing. A decoding apparatus comprising: a concealment result determination unit that compares the respective decoding ranges and applies a concealment process having the smallest difference to a decoding range that has not been normally decoded to generate a concealed image. An error concealment processing device connected to a decoding processing device that decodes encoded data and outputs decoded pixel data, encoding parameters, and decoding error information,
A frame memory for storing the pixel data in units of pictures;
An encoding parameter storage unit for storing the encoding parameter in units of pictures;
Based on the decoding error information, a concealment management unit that determines ON / OFF of concealment processing for each picture;
A concealment range calculation unit for calculating a concealment range when the ON / OFF determination result of the concealment process is ON;
A divided pattern storage unit for storing a plurality of types of subgroup subdivision patterns in which subgroups obtained by subdividing the subgroups obtained by dividing the conceal range are set;
Dividing the concealed range into the divided groups, and further dividing the concealed range into the subgroups based on the plurality of subgroup subdivision patterns; and
An estimated motion vector calculating unit that calculates an estimated motion vector representing the divided group / subgroup for each divided group / subgroup within the concealed range;
An error concealment processing apparatus comprising: a concealed image generation unit configured to generate a concealed image from the estimated motion vector and the pixel data stored in the frame memory when an ON / OFF determination result of the concealment process is ON. Decoding encoded data, outputting decoded pixel data, encoding parameters, and decoding error information;
Storing the pixel data in a storage means in units of pictures;
Accumulating the encoding parameters in a storage unit in units of pictures;
Determining concealment ON / OFF for each picture based on the decoding error information;
Calculating a concealment range when an ON / OFF determination result of the concealment process is ON;
Dividing the concealed range into the divided groups, and dividing the divided groups into the subgroups based on a plurality of subgroup subdivision patterns in which subgroups obtained by subdividing the divided groups are set;
Calculating an estimated motion vector representing a split group / subgroup for each split group / subgroup within the concealed range;
A decoding method comprising: generating a concealed image from the estimated motion vector and pixel data stored in the storage means when an ON / OFF determination result of the concealment process is ON. Detecting whether the estimated motion vector calculated in the step of calculating the estimated motion vector refers to the outside of the screen;
The decoding method according to claim 7, further comprising a step of correcting the estimated motion vector when the estimated motion vector refers to outside the screen. Detecting the presence or absence of a scene change in the decoding process of the encoded data,
9. The step of correcting the estimated motion vector performs correction processing of the estimated motion vector based on whether or not the scene change has occurred when the estimated motion vector refers to outside the screen. Decoding method as described. For the estimated motion vector calculated in the step of calculating the estimated motion vector, determining the necessity of correction by evaluating the estimated motion vector of each divided group and the estimated motion vector of the surrounding divided groups;
8. The decoding method according to claim 7, further comprising a step of correcting the estimated motion vector based on the determination result of whether or not correction is necessary. Decoding encoded data, outputting decoded pixel data, encoding parameters, and decoding error information;
Storing the pixel data in a storage means in units of pictures;
Accumulating the encoding parameters in a storage unit in units of pictures;
Determining concealment ON / OFF for each picture based on the decoding error information;
Calculating a concealment range when an ON / OFF determination result of the concealment process is ON;
Dividing the concealed range into the divided groups, and dividing the divided groups into the subgroups based on a plurality of subgroup subdivision patterns in which subgroups obtained by subdividing the divided groups are set;
Performing a plurality of concealment processes for generating a concealment image with different concealment processes;
When the ON / OFF determination result of the concealment process is ON, the step of executing the plurality of concealment processes corresponding to the decoded range other than the concealed range and the decoded range is the different concealment process And a step of generating a concealed image by comparing concealment processing with the smallest difference to an error decoding range that has not been normally decoded.

Images