CN113709494B - Image decompression method and device for super-resolution reconstruction - Google Patents

Image decompression method and device for super-resolution reconstruction Download PDF

Info

Publication number
CN113709494B
CN113709494B CN202110699794.5A CN202110699794A CN113709494B CN 113709494 B CN113709494 B CN 113709494B CN 202110699794 A CN202110699794 A CN 202110699794A CN 113709494 B CN113709494 B CN 113709494B
Authority
CN
China
Prior art keywords
image
decompressed
height
decompression
super
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110699794.5A
Other languages
Chinese (zh)
Other versions
CN113709494A (en
Inventor
潘文培
宋松言
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Allwinner Technology Co Ltd
Original Assignee
Allwinner Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Allwinner Technology Co Ltd filed Critical Allwinner Technology Co Ltd
Priority to CN202110699794.5A priority Critical patent/CN113709494B/en
Publication of CN113709494A publication Critical patent/CN113709494A/en
Application granted granted Critical
Publication of CN113709494B publication Critical patent/CN113709494B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/44—Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00—General purpose image data processing
    • G06T1/60—Memory management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/182—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a pixel
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention discloses an image decompression method and device for super-resolution reconstruction, wherein the method comprises the following steps: determining a default decompressed macro block height of a target compressed image; determining pixel interleaving information corresponding to a target reconstructed image corresponding to the target compressed image; the pixel interleaving information is used for indicating that reconstructed pixels of at least two image blocks in the target reconstructed image are mapped to the same pixel row in the decompressed image of the target compressed image; and determining the real-time decompression macroblock height and pixel line cache operation of the target compressed image during decompression according to the pixel interleaving information and the default decompression macroblock height. Therefore, the invention can adjust the height of the decompressed macro block and buffer the interlaced pixels according to the interlacing condition of the reconstructed pixels in the image, thereby effectively reducing the data reading bandwidth requirement and the operand when the super-resolution reconstruction is finally carried out, and further realizing the online image decompression and the super-resolution reconstruction.

Description

Image decompression method and device for super-resolution reconstruction
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to an image decompression method and apparatus for super-resolution reconstruction.
Background
In order to realize video transmission with higher image quality, a method commonly used in the industry is to combine an image super-resolution reconstruction technology to relieve the pressure of bandwidth and memory, for example, a common false 4K camera in the market, acquire a low-resolution image from a camera, or downsample a high-resolution image acquired by the camera to save resources, reconstruct the image super-resolution to obtain a high-resolution 4K image only before encoding, and finally encode and record the 4K image. Meanwhile, the low-resolution image in the process is compressed by combining an image compression technology, so that the purposes of saving bandwidth and memory to the greatest extent are achieved.
However, for video encoding and decoding, such as scenes needing to be processed according to blocks, on-line operation is difficult to realize by image compression and super-resolution reconstruction, in the prior art, two modules are usually two independent processes, namely, decompression is performed to obtain a whole image, and then super-resolution reconstruction is performed. For this reason, the image compression is not achieved, so that in the on-chip video image processing scheme, the image compression and the super-resolution reconstruction are not used in cascade at the same time. In a scene requiring block output such as video coding, online decompression and super-resolution reconstruction are difficult to achieve, for the following reasons:
In the image super-resolution reconstruction process, each pixel of the amplified image is required to be mapped into an original image according to the amplification factor to obtain a corresponding sub-pixel position, then integer pixels in the neighborhood of the position are taken, and the pixel value of the sub-pixel position is obtained through interpolation calculation and is used as the final pixel value of the super-resolution image.
However, for a common compression algorithm, there is usually a correlation between compressed macro blocks, that is, independent decoding of macro blocks cannot be achieved, and the decoding of Ren Yihong blocks needs to refer to the domain information. For such scenarios, scaling techniques are generally not applicable. Because when pixel interleaving occurs, adjacent image blocks are mapped to the same integer pixel row, and at this time, since the decoding process of the previous macroblock already includes the integer pixel corresponding to the sub-pixel row, when the reconstructed image of the next mapped block is calculated, two compressed macroblocks need to be decoded simultaneously as input, which means that the bandwidth of the read data is doubled, and the operation amount is doubled, which is contrary to the bandwidth-saving goal expected to be achieved by introducing compression.
In the prior art, aiming at the interleaving problem, a buffer is adopted to buffer interleaving data, but since input data is in units of compressed macro blocks, and interleaving conditions are influenced by macro block sizes and scaling factors, the sizes of the buffer are difficult to determine, and in the worst case, all pixel rows need to be buffered, so that the cost is unacceptable for hardware design. It can be seen that the defects exist in the prior art, and the problem needs to be solved.
Disclosure of Invention
The invention aims to solve the technical problem of providing an image decompression method and device for super-resolution reconstruction, which can adjust the height of a decompressed macro block and cache interleaved pixels during image decompression according to the interleaving condition of reconstructed pixels in an image, so that the data reading bandwidth requirement and the operand during final super-resolution reconstruction can be effectively reduced, and further the online image decompression and super-resolution reconstruction can be realized.
To solve the above technical problem, a first aspect of the present invention discloses an image decompression method for super-resolution reconstruction, the method comprising:
determining a default decompressed macro block height of a target compressed image;
determining pixel interleaving information corresponding to a target reconstructed image corresponding to the target compressed image; the pixel interleaving information is used for indicating that reconstructed pixels of at least two image blocks in the target reconstructed image are mapped to the same pixel row in the decompressed image of the target compressed image;
determining the real-time decompression macroblock height and pixel line cache operation of the target compressed image during decompression according to the pixel interleaving information and the default decompression macroblock height; the real-time decompressed macro block height is lower than the default decompressed macro block height when the pixel interleaving information exists; the pixel line buffering operation is used for buffering the same pixel line when the pixel interleaving information exists and calling the same pixel line when the same pixel line is to be reconstructed later.
As an optional implementation manner, in the first aspect of the present invention, the determining a default decompressed macro block height of the target compressed image includes:
determining the default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor;
and determining a default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor, wherein the method comprises the following steps:
determining the height of a corresponding decompressed macro block according to the size of the target reconstructed image and the image scaling factor;
and rounding the height of the decompressed macro block upwards to obtain the default decompressed macro block height of the target compressed image.
As an optional implementation manner, in the first aspect of the present invention, the determining, according to the pixel interleaving information and the default decompressed macroblock height, a real-time decompressed macroblock height and a pixel line buffering operation of the target compressed image during decompression includes:
performing decompression operation on the target compressed image, and judging whether an interweaved pixel exists between a current decompression corresponding image block and a last decompression corresponding image block in the target reconstructed image;
When the judgment result is negative, determining the height of the decompression macro block corresponding to the current decompression corresponding image block as the default decompression macro block height;
when the judgment result is yes, caching the interlaced pixels, and determining the height of a decompressed macro block corresponding to the current decompressed corresponding image block as the height of the macro block obtained by subtracting the height of the interlaced pixels from the height of the default decompressed macro block; the cached interleaved pixels are used for being called when the current decompressed corresponding image block is subjected to super-resolution reconstruction.
In an optional implementation manner, in the first aspect of the present invention, the interleaved pixels are pixel rows that exist in the decompressed image of the target compressed image and are mapped to the current decompressed corresponding image block and the previous decompressed corresponding image block at the same time when super-resolution reconstruction is performed.
As an optional implementation manner, in the first aspect of the present invention, the buffering the interleaved pixel includes:
the interleaved pixels are buffered using a line buffer.
As an optional embodiment, in the first aspect of the invention, the method further comprises:
obtaining a decompressed image obtained after decompressing the target compressed image;
And performing super-resolution reconstruction on the decompressed image to obtain the target reconstructed image.
As an optional implementation manner, in the first aspect of the present invention, the performing the super-resolution reconstruction on the decompressed image to obtain the target reconstructed image includes:
performing super-resolution reconstruction operation on the decompressed image, and judging whether the macroblock height of the current reconstructed macroblock of the decompressed image is smaller than the default decompressed macroblock height;
if yes, calling the corresponding cached interleaved pixels, and combining the pixels in the current reconstructed macroblock to perform super-resolution reconstruction.
A second aspect of an embodiment of the present invention discloses an image decompression apparatus for super-resolution reconstruction, the apparatus comprising:
the height determining module is used for determining the default decompressed macro block height of the target compressed image;
the interleaving determining module is used for determining pixel interleaving information corresponding to a target reconstruction image corresponding to the target compressed image; the pixel interleaving information is used for indicating that reconstructed pixels of at least two image blocks in the target reconstructed image are mapped to the same pixel row in the decompressed image of the target compressed image;
the operation determining module is used for determining the real-time decompression macro block height and pixel line cache operation of the target compressed image during decompression according to the pixel interleaving information and the default decompression macro block height; the real-time decompressed macro block height is lower than the default decompressed macro block height when the pixel interleaving information exists; the pixel line buffering operation is used for buffering the same pixel line when the pixel interleaving information exists and calling the same pixel line when the same pixel line is to be reconstructed later.
As an optional implementation manner, in the second aspect of the present invention, the specific manner in which the height determining module determines the default decompressed macro block height of the target compressed image includes:
determining the default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor;
and the specific mode of determining the default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor by the height determining module comprises the following steps:
determining the height of a corresponding decompressed macro block according to the size of the target reconstructed image and the image scaling factor;
and rounding the height of the decompressed macro block upwards to obtain the default decompressed macro block height of the target compressed image.
As an optional implementation manner, in the second aspect of the present invention, the operation determining module includes:
the execution judging unit is used for executing decompression operation on the target compressed image and judging whether the current decompression corresponding image block and the last decompression corresponding image block in the target reconstructed image have interweaved pixels or not;
the first determining unit is used for determining the height of the decompressed macro block corresponding to the current decompressed corresponding image block as the default decompressed macro block height when the judging result is negative;
The second determining unit is used for caching the interlaced pixels when the judging result is yes, and determining the height of the decompressed macro block corresponding to the current decompressed corresponding image block as the height of the macro block obtained by subtracting the height of the interlaced pixels from the default decompressed macro block; the cached interleaved pixels are used for being called when the current decompressed corresponding image block is subjected to super-resolution reconstruction.
In a second aspect of the present invention, the interlaced pixels are pixel rows that exist in the decompressed image of the target compressed image and are mapped to the current decompressed corresponding image block and the previous decompressed corresponding image block at the same time when super-resolution reconstruction is performed.
As an optional implementation manner, in the second aspect of the present invention, a specific manner of buffering the interleaved pixel by the second determining unit includes:
the interleaved pixels are buffered using a line buffer.
As an alternative embodiment, in the second aspect of the present invention, the apparatus further includes:
the acquisition module is used for acquiring a decompressed image obtained after decompressing the target compressed image;
and the reconstruction module is used for performing super-resolution reconstruction on the decompressed image to obtain the target reconstructed image.
As an optional implementation manner, in the second aspect of the present invention, a specific manner of performing super-resolution reconstruction on the decompressed image by the reconstruction module to obtain the target reconstructed image includes:
performing super-resolution reconstruction operation on the decompressed image, and judging whether the macroblock height of the current reconstructed macroblock of the decompressed image is smaller than the default decompressed macroblock height;
if yes, calling the corresponding cached interleaved pixels, and combining the pixels in the current reconstructed macroblock to perform super-resolution reconstruction.
In a third aspect, the present invention discloses another image decompression apparatus for super-resolution reconstruction, the apparatus comprising:
a memory storing executable program code;
a processor coupled to the memory;
the processor invokes the executable program code stored in the memory to perform some or all of the steps in the image decompression method for super-resolution reconstruction disclosed in the first aspect of the present invention.
A fourth aspect of the invention discloses a computer storage medium storing computer instructions which, when invoked, are adapted to perform part or all of the steps of the image decompression method for super-resolution reconstruction disclosed in the first aspect of the invention.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
the embodiment of the invention discloses an image decompression method and device for super-resolution reconstruction, wherein the method comprises the following steps: determining an origin coordinate corresponding to a target interpolation point coordinate in an original image, and determining a reference part coordinate of the origin coordinate; performing direction detection on the original image pixel points corresponding to the reference part coordinates to obtain edge directions corresponding to the reference part coordinates; determining a sampling window of the target interpolation point according to the edge direction corresponding to the reference part coordinates; and determining a sampling point set of the target interpolation point from the original image according to the sampling window of the target interpolation point. Therefore, according to the embodiment of the invention, the height of the decompressed macro block can be adjusted and the interlaced pixels can be cached when the image is decompressed according to the interlacing condition of the reconstructed pixels in the image, so that the data reading bandwidth requirement and the operand when the super-resolution reconstruction is finally carried out can be effectively reduced, and further the online image decompression and the super-resolution reconstruction can be realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of an image decompression method for super-resolution reconstruction according to an embodiment of the present invention.
Fig. 2 is a flow chart of another image decompression method for super-resolution reconstruction according to an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an image decompression apparatus for super-resolution reconstruction according to an embodiment of the present invention.
Fig. 4 is a schematic structural diagram of another image decompression apparatus for super-resolution reconstruction according to an embodiment of the present invention.
Fig. 5 is a schematic structural diagram of an image decompression apparatus for super-resolution reconstruction according to an embodiment of the present invention.
Fig. 6 is a schematic diagram showing the presence of pixel interleaving in an image reconstruction according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, apparatus, article, or device that comprises a list of steps or elements is not limited to the list of steps or elements but may, in the alternative, include other steps or elements not expressly listed or inherent to such process, method, article, or device.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
First, before further describing the embodiments of the present invention, the technical details of pixel interleaving existing in reconstruction related to the present invention will be explained, referring to fig. 6, in a specific technical scenario, a target reconstructed image with a macroblock size of 16×16 needs to be output finally, and assuming that the scaling factor is 1.25 times, the macroblock height of the reconstructed base image mapped by the target reconstructed image is actually 12.8, where the reconstructed base image is an image after decompression of the target compressed image. As can be seen from fig. 6, when the image block 1 and the image block 2 in the target reconstructed image are mapped to the mapping block 1 and the mapping block 2 in the reconstructed base image, the pixels partially mapped by the mapping block 1 and the mapping block 2 are located in the sub-pixel row of the reconstructed base image, which is marked by 13, at this time, since the decoding process of the mapping block 1 already includes the integer pixel corresponding to the sub-pixel row, when the reconstructed image of the mapping block 2 is calculated, two compressed image macro blocks (that is, the image blocks in the target compressed image corresponding to the mapping block 1 and the mapping block 2 respectively) need to be decoded simultaneously as input, which means that the bandwidth of the read data is doubled, the operation amount is doubled, and there is a problem.
In order to solve the problems, the invention discloses an image decompression method and an image decompression device for super-resolution reconstruction, which can adjust the height of a decompressed macro block and cache interleaved pixels during image decompression according to the interleaving condition of reconstructed pixels in an image, so that the data reading bandwidth requirement and the operand during final super-resolution reconstruction can be effectively reduced, and further the online image decompression and super-resolution reconstruction can be realized. The following will describe in detail.
Example 1
Referring to fig. 1, fig. 1 is a flowchart of an image decompression method for super-resolution reconstruction according to an embodiment of the present invention. The image decompression method for super-resolution reconstruction described in fig. 1 is applied to an image processing system/an image processing device/an image processing server (wherein the image processing server comprises a local image processing server or a cloud image processing server). As shown in fig. 1, the image decompression method for super-resolution reconstruction may include the operations of:
101. a default decompressed macroblock height of the target compressed image is determined.
In the embodiment of the invention, the default decompression macroblock height can be the initial decompression macroblock height corresponding to the target compressed image, and the default decompression macroblock height can be an initial parameter set manually or a parameter obtained by calculation according to the decompression requirement in real time.
102. And determining pixel interleaving information corresponding to the target reconstructed image corresponding to the target compressed image.
In the embodiment of the invention, the pixel interleaving information is used for indicating that reconstructed pixels of at least two image blocks in the target reconstructed image are mapped to the same pixel row in the decompressed image of the target compressed image. As can be seen from the above description, such pixel interleaving information may cause problems of high bandwidth cost and low efficiency in the decompression process of super-resolution reconstruction, and thus corresponding solutions need to be implemented.
103. And determining the real-time decompression macroblock height and pixel line caching operation of the target compressed image during decompression according to the pixel interleaving information and the default decompression macroblock height.
In the embodiment of the invention, the height of the real-time decompression macro block is lower than the default decompression macro block when the pixel interleaving information exists, and by the operation, the existence of the interleaved pixels between the adjacent macro blocks is indicated when the pixel interleaving information exists, so that the height of the real-time decompression macro block is reduced, the interleaved pixels are ignored, and the secondary decompression conducted by the interleaved pixels is not needed in the subsequent reconstruction as in the prior art.
Optionally, the pixel line buffering operation is used for buffering the same pixel line when pixel interleaving information exists and calling the same pixel line when the same pixel line is to be reconstructed later. Through the arrangement, the cached pixels can be called when needed, compared with the method in the prior art that the cache is adopted to cache the interleaving data under the condition that the interleaving size is not determined, the method in the embodiment is obviously more economical and practical, and has lower cost and higher efficiency.
Therefore, according to the embodiment of the invention, the height of the decompressed macro block can be adjusted and the interlaced pixels are cached when the image is decompressed according to the interlacing condition of the reconstructed pixels in the image, so that the data reading bandwidth requirement and the operand when the super-resolution reconstruction is finally carried out can be effectively reduced, and further the online image decompression and the super-resolution reconstruction can be realized.
As an alternative embodiment, in step 101, determining a default decompressed macro block height of the target compressed image includes:
and determining the default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor.
For example, referring to fig. 6, it is required to output a target reconstructed image with a macroblock size of 16×16, and assuming that the scaling factor is 1.25 times, the macroblock height of the reconstructed base image mapped thereto, i.e., the decompressed image of the target compressed image, is actually 12.8.
Therefore, through the optional implementation manner, the default decompression macroblock height of the corresponding target compressed image can be determined according to the size of the target reconstructed image and the image scaling factor, so that the determined default decompression macroblock height is more accurate, the finally determined real-time decompression macroblock height and pixel line caching operation can effectively reduce the data reading bandwidth requirement and the operation amount when the super-resolution reconstruction is finally carried out, and further the online image decompression and super-resolution reconstruction can be realized.
As an optional implementation manner, in the step, determining the default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor includes:
determining the height of a corresponding decompressed macro block according to the size of the target reconstructed image and the image scaling factor;
and (5) rounding the height of the decompressed macro block upwards to obtain the default decompressed macro block height of the target compressed image.
For example, referring to fig. 6, it is necessary to output a target reconstructed image having a macroblock size of 16×16, and assuming that the scaling factor is 1.25 times, the macroblock height of the decompressed image of the target compressed image mapped thereto is actually 12.8, and the compressed macroblock height is an integer, so that it is rounded up to 13.
Therefore, through the optional implementation manner, the height of the decompressed macro block can be rounded upwards to obtain the default decompressed macro block height of the target compressed image, so that the determined default decompressed macro block height is more accurate, the finally determined real-time decompressed macro block height and pixel line caching operation can effectively reduce the data reading bandwidth requirement and the operation amount when the super-resolution reconstruction is finally carried out, and further the online image decompression and the super-resolution reconstruction can be realized.
As an optional implementation manner, in the step 103, determining the real-time decompressed macro block height and the pixel line buffering operation of the target compressed image during decompression according to the pixel interleaving information and the default decompressed macro block height includes:
performing decompression operation on the target compressed image, and judging whether an interweaved pixel exists between a current decompression corresponding image block and a last decompression corresponding image block in the target reconstructed image;
when the judgment result is negative, determining the height of the decompression macro block corresponding to the current decompression corresponding image block as the default decompression macro block height;
and when the judgment result is yes, caching the interleaved pixels, and determining the height of the decompressed macro block corresponding to the current decompressed corresponding image block as the height of the macro block obtained by subtracting the height of the interleaved pixels from the height of the default decompressed macro block, wherein the cached interleaved pixels are used for being called when the current decompressed corresponding image block is subjected to super-resolution reconstruction.
In the embodiment of the invention, the previous corresponding image block is the previous corresponding image block before the current corresponding image block in the reconstruction sequence.
In the embodiment of the invention, the interlaced pixels are pixel rows which exist in the decompressed image of the target compressed image and are mapped to the current decompressed corresponding image block and the last decompressed corresponding image block at the same time when super-resolution reconstruction is performed.
Therefore, according to the alternative embodiment, when the fact that the interleaving pixels exist in the current decompression corresponding image block and the last decompression corresponding image block in the target reconstructed image is judged, the interleaving pixels are cached, the height of the decompression macro block corresponding to the current decompression corresponding image block is determined to be the default decompression macro block height minus the height of the interleaving pixels, and therefore the data reading bandwidth requirement and the operation amount in the final super-resolution reconstruction can be effectively reduced, and further online image decompression and super-resolution reconstruction can be achieved.
As an alternative embodiment, in the foregoing step, buffering the interleaved pixels includes:
the interleaved pixels are buffered using a line buffer.
The following describes a specific implementation of the method disclosed in this example in connection with the pixel interleaving case shown in fig. 6, specifically, as shown in fig. 6, the output image is output in the order of blocks, assuming that the block size is 16x16, and assuming that the scaling factor is 1.25 times, the height of the mapping block is actually 12.8. Since the compressed macroblock height can only be an integer, it is rounded up to 13.
The scheme improves from an image compression end, designs a decompression algorithm to be changeable in macroblock height, carries out macroblock height self-adaptive adjustment according to macroblock interleaving conditions, and executes the following steps:
1. According to the height of the output image block and the scaling factor, calculating the height of the image block mapped to the original image, wherein the height is assumed to be the first macro block height H0 of the decompression module, namely H0=13;
2. if the current image block is interleaved with the image block vertically below the current image block, the decompressed macroblock height of the lower image block adopts a second macroblock height H1 (h1=h0-1), which is 12; otherwise, the first macroblock height 13 is still employed.
3. When interleaving occurs, a line buffer is used to buffer all pixels corresponding to the interleaved lines for the next image line block. Although the macroblock height of the next image block is reduced, since equal numbers of pixels are buffered in the line buffer, it is ensured that all pixels in the next image block can meet the requirement of interpolation operation.
Example two
Referring to fig. 2, fig. 2 is a flowchart of another image decompression method for super-resolution reconstruction according to an embodiment of the present invention. The image decompression method for super-resolution reconstruction described in fig. 2 is applied to an image processing system/an image processing device/an image processing server (wherein the image processing server comprises a local image processing server or a cloud image processing server). As shown in fig. 2, the image decompression method for super-resolution reconstruction may include the operations of:
201. A default decompressed macroblock height of the target compressed image is determined.
202. And determining pixel interleaving information corresponding to the target reconstructed image corresponding to the target compressed image.
203. And determining the real-time decompression macroblock height and pixel line caching operation of the target compressed image during decompression according to the pixel interleaving information and the default decompression macroblock height.
In the embodiment of the present invention, please refer to the detailed description of the steps 101 to 103 in the first embodiment for the description of the steps 201 to 203, and the description of the embodiment of the present invention is omitted.
204. And obtaining a decompressed image obtained after decompressing the target compressed image.
205. And performing super-resolution reconstruction on the decompressed image to obtain a target reconstructed image.
In the embodiment of the invention, super-resolution reconstruction refers to super-resolution reconstruction, and specifically, the super-resolution reconstruction can be realized by using a neural network algorithm, and the invention is not limited to the super-resolution reconstruction.
Therefore, the embodiment of the invention can acquire the decompressed image obtained by decompressing the target compressed image based on the real-time decompressed macro block height and the pixel line buffer operation, and perform super-resolution reconstruction on the decompressed image to obtain the target reconstructed image, so that the data reading bandwidth requirement and the operand when the super-resolution reconstruction is finally performed can be effectively reduced, and further the online image decompression and the super-resolution reconstruction can be realized.
In an optional embodiment, in step 205, performing the super-resolution reconstruction on the decompressed image to obtain the target reconstructed image includes:
performing super-resolution reconstruction operation on the decompressed image, and judging whether the macroblock height of the current reconstructed macroblock of the decompressed image is smaller than the default decompressed macroblock height;
if yes, the corresponding cached interleaved pixels are called, and super-division reconstruction is carried out by combining the pixels in the current reconstructed macroblock.
Therefore, through the optional implementation manner, when judging that the height of the macro block of the current reconstructed macro block of the decompressed image is smaller than the height of the default decompressed macro block, the corresponding cached interleaved pixels are called, and the super-resolution reconstruction is performed by combining the pixels in the current reconstructed macro block, so that the data reading bandwidth requirement and the operation amount when the super-resolution reconstruction is finally performed can be effectively reduced, and further the online image decompression and the super-resolution reconstruction can be realized.
Example III
Referring to fig. 3, fig. 3 is a schematic structural diagram of an image decompression apparatus for super-resolution reconstruction according to an embodiment of the present invention. The image decompression apparatus for super-resolution reconstruction described in fig. 3 is applied to an image processing system/an image processing device/an image processing server (wherein the image processing server includes a local image processing server or a cloud image processing server). As shown in fig. 3, the image decompression apparatus for super-resolution reconstruction may include:
The height determining module 301 is configured to determine a default decompressed macroblock height of the target compressed image.
In the embodiment of the invention, the default decompression macroblock height can be the initial decompression macroblock height corresponding to the target compressed image, and the default decompression macroblock height can be an initial parameter set manually or a parameter obtained by calculation according to the decompression requirement in real time.
The interlace determining module 302 is configured to determine pixel interlace information corresponding to a target reconstructed image corresponding to a target compressed image.
In the embodiment of the invention, the pixel interleaving information is used for indicating that reconstructed pixels of at least two image blocks in the target reconstructed image are mapped to the same pixel row in the decompressed image of the target compressed image. As can be seen from the above description, such pixel interleaving information may cause problems of high bandwidth cost and low efficiency in the decompression process of super-resolution reconstruction, and thus corresponding solutions need to be implemented.
The operation determining module 303 is configured to determine a real-time decompressed macroblock height and a pixel line buffering operation of the target compressed image during decompression according to the pixel interleaving information and the default decompressed macroblock height.
In the embodiment of the invention, the height of the real-time decompression macro block is lower than the default decompression macro block when the pixel interleaving information exists, and by the operation, the existence of the interleaved pixels between the adjacent macro blocks is indicated when the pixel interleaving information exists, so that the height of the real-time decompression macro block is reduced, the interleaved pixels are ignored, and the secondary decompression conducted by the interleaved pixels is not needed in the subsequent reconstruction as in the prior art.
Optionally, the pixel line buffering operation is used for buffering the same pixel line when pixel interleaving information exists and calling the same pixel line when the same pixel line is to be reconstructed later. Through the arrangement, the cached pixels can be called when needed, compared with the method in the prior art that the cache is adopted to cache the interleaving data under the condition that the interleaving size is not determined, the method in the embodiment is obviously more economical and practical, and has lower cost and higher efficiency.
Therefore, according to the embodiment of the invention, the height of the decompressed macro block can be adjusted and the interlaced pixels are cached when the image is decompressed according to the interlacing condition of the reconstructed pixels in the image, so that the data reading bandwidth requirement and the operand when the super-resolution reconstruction is finally carried out can be effectively reduced, and further the online image decompression and the super-resolution reconstruction can be realized.
As an alternative embodiment, the specific manner in which the height determining module 301 determines the default decompressed macro block height of the target compressed image includes:
and determining the default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor.
For example, referring to fig. 6, it is required to output a target reconstructed image with a macroblock size of 16×16, and assuming that the scaling factor is 1.25 times, the macroblock height of the reconstructed base image mapped thereto, i.e., the decompressed image of the target compressed image, is actually 12.8.
Therefore, through the optional implementation manner, the default decompression macroblock height of the corresponding target compressed image can be determined according to the size of the target reconstructed image and the image scaling factor, so that the determined default decompression macroblock height is more accurate, the finally determined real-time decompression macroblock height and pixel line caching operation can effectively reduce the data reading bandwidth requirement and the operation amount when the super-resolution reconstruction is finally carried out, and further the online image decompression and super-resolution reconstruction can be realized.
As an optional embodiment, the specific manner of determining the default decompressed macro block height of the corresponding target compressed image by the height determining module 301 according to the size of the target reconstructed image and the image scaling factor includes:
determining the height of a corresponding decompressed macro block according to the size of the target reconstructed image and the image scaling factor;
and (5) rounding the height of the decompressed macro block upwards to obtain the default decompressed macro block height of the target compressed image.
For example, referring to fig. 6, it is necessary to output a target reconstructed image having a macroblock size of 16×16, and assuming that the scaling factor is 1.25 times, the macroblock height of the decompressed image of the target compressed image mapped thereto is actually 12.8, and the compressed macroblock height is an integer, so that it is rounded up to 13.
Therefore, through the optional implementation manner, the height of the decompressed macro block can be rounded upwards to obtain the default decompressed macro block height of the target compressed image, so that the determined default decompressed macro block height is more accurate, the finally determined real-time decompressed macro block height and pixel line caching operation can effectively reduce the data reading bandwidth requirement and the operation amount when the super-resolution reconstruction is finally carried out, and further the online image decompression and the super-resolution reconstruction can be realized.
As an alternative embodiment, as shown in fig. 4, the operation determining module 303 includes:
an execution judging unit 3031, configured to execute a decompression operation on the target compressed image, and judge whether an interleaved pixel exists between a current image block corresponding to decompression and a previous image block corresponding to decompression in the target reconstructed image;
a first determining unit 3032, configured to determine, as a default decompressed macroblock height, a decompressed macroblock height corresponding to a current decompressed corresponding image block when the determination result is negative;
a second determining unit 3033, configured to buffer the interleaved pixels when the determination result is yes, and determine a decompressed macroblock height corresponding to the current decompressed corresponding image block as a macroblock height obtained by subtracting the interleaved pixel height from a default decompressed macroblock height; the buffered interleaved pixels are used to be invoked when the corresponding image block is currently decompressed for super-resolution reconstruction.
In the embodiment of the invention, the previous corresponding image block is the previous corresponding image block before the current corresponding image block in the reconstruction sequence.
In the embodiment of the invention, the interlaced pixels are pixel rows which exist in the decompressed image of the target compressed image and are mapped to the current decompressed corresponding image block and the last decompressed corresponding image block at the same time when super-resolution reconstruction is performed.
Therefore, according to the alternative embodiment, when the fact that the interleaving pixels exist in the current decompression corresponding image block and the last decompression corresponding image block in the target reconstructed image is judged, the interleaving pixels are cached, the height of the decompression macro block corresponding to the current decompression corresponding image block is determined to be the default decompression macro block height minus the height of the interleaving pixels, and therefore the data reading bandwidth requirement and the operation amount in the final super-resolution reconstruction can be effectively reduced, and further online image decompression and super-resolution reconstruction can be achieved.
As an optional embodiment, the specific manner in which the second determining unit 3033 caches the interleaved pixels includes:
the interleaved pixels are buffered using a line buffer.
As an alternative embodiment, as shown in fig. 4, the apparatus further includes:
the obtaining module 304 is configured to obtain a decompressed image obtained after decompressing the target compressed image;
the reconstruction module 305 is configured to perform super-resolution reconstruction on the decompressed image to obtain a target reconstructed image.
In the embodiment of the invention, super-resolution reconstruction refers to super-resolution reconstruction, and specifically, the super-resolution reconstruction can be realized by using a neural network algorithm, and the invention is not limited to the super-resolution reconstruction.
Therefore, the embodiment of the invention can acquire the decompressed image obtained by decompressing the target compressed image based on the real-time decompressed macro block height and the pixel line buffer operation, and perform super-resolution reconstruction on the decompressed image to obtain the target reconstructed image, so that the data reading bandwidth requirement and the operand when the super-resolution reconstruction is finally performed can be effectively reduced, and further the online image decompression and the super-resolution reconstruction can be realized.
As an optional implementation manner, in the second aspect of the present invention, the specific manner in which the reconstruction module 305 performs the super-division reconstruction on the decompressed image to obtain the target reconstructed image includes:
performing super-resolution reconstruction operation on the decompressed image, and judging whether the macroblock height of the current reconstructed macroblock of the decompressed image is smaller than the default decompressed macroblock height;
If yes, the corresponding cached interleaved pixels are called, and super-division reconstruction is carried out by combining the pixels in the current reconstructed macroblock.
Therefore, through the optional implementation manner, when judging that the height of the macro block of the current reconstructed macro block of the decompressed image is smaller than the height of the default decompressed macro block, the corresponding cached interleaved pixels are called, and the super-resolution reconstruction is performed by combining the pixels in the current reconstructed macro block, so that the data reading bandwidth requirement and the operation amount when the super-resolution reconstruction is finally performed can be effectively reduced, and further the online image decompression and the super-resolution reconstruction can be realized.
Example IV
Referring to fig. 5, fig. 5 is a schematic diagram illustrating another image decompression apparatus for super-resolution reconstruction according to an embodiment of the present invention. The image decompression apparatus for super-resolution reconstruction described in fig. 5 is applied to an image processing system/image processing device/image processing server (wherein the image processing server includes a local image processing server or a cloud image processing server). As shown in fig. 5, the image decompression apparatus for super-resolution reconstruction may include:
a memory 401 storing executable program codes;
a processor 402 coupled with the memory 401;
Wherein the processor 402 invokes executable program code stored in the memory 401 for performing the steps of the image decompression method for super-resolution reconstruction described in embodiment one or embodiment two.
Example five
The embodiment of the invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program causes a computer to execute the steps of the image decompression method for super-resolution reconstruction described in the first or second embodiment.
Example six
The embodiment of the invention discloses a computer program product, which comprises a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute the steps of the image decompression method for super-resolution reconstruction described in the first or second embodiment.
The apparatus embodiments described above are merely illustrative, wherein the modules illustrated as separate components may or may not be physically separate, and the components shown as modules may or may not be physical, i.e., may be located in one place, or may be distributed over a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above detailed description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course by means of hardware. Based on such understanding, the foregoing technical solutions may be embodied in essence or in a portion contributing to the prior art in the form of a software product that may be stored in a computer readable storage medium including Read-only memory (ROM), random access memory (RandomAccessMemory, RAM), programmable Read-only memory (PROM), erasable programmable Read-only memory (ErasableProgrammableReadOnlyMemory, EPROM), one-time programmable Read-only memory (One-OnlyMemory, OTPROM), electrically erasable programmable Read-only memory (CD-ROM) or other optical disk memory, magnetic disk memory, tape memory, or any other medium that can be used for carrying or storing data that is readable by a computer.
Finally, it should be noted that: the embodiment of the invention discloses an image decompression method and device for super-resolution reconstruction, which are disclosed by the embodiment of the invention only for illustrating the technical scheme of the invention, but not limiting the technical scheme; although the invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that; the technical scheme recorded in the various embodiments can be modified or part of technical features in the technical scheme can be replaced equivalently; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.

Claims (10)

1. An image decompression method for super-resolution reconstruction, the method comprising:
determining a default decompressed macro block height of a target compressed image;
determining pixel interleaving information corresponding to a target reconstructed image corresponding to the target compressed image; the pixel interleaving information is used for indicating that reconstructed pixels of at least two image blocks in the target reconstructed image are mapped to the same pixel row in the decompressed image of the target compressed image;
determining the real-time decompression macroblock height and pixel line cache operation of the target compressed image during decompression according to the pixel interleaving information and the default decompression macroblock height; the real-time decompressed macro block height is lower than the default decompressed macro block height when the pixel interleaving information exists; the pixel line buffering operation is used for buffering the same pixel line when the pixel interleaving information exists and calling the same pixel line when the same pixel line is to be reconstructed later.
2. The image decompression method for super-resolution reconstruction according to claim 1, wherein said determining a default decompressed macroblock height of a target compressed image comprises:
determining the default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor;
and determining a default decompressed macro block height of the corresponding target compressed image according to the size of the target reconstructed image and the image scaling factor, wherein the method comprises the following steps:
determining the height of a corresponding decompressed macro block according to the size of the target reconstructed image and the image scaling factor;
and rounding the height of the decompressed macro block upwards to obtain the default decompressed macro block height of the target compressed image.
3. The image decompression method for super-resolution reconstruction according to claim 2, wherein said determining a real-time decompressed macroblock height and a pixel line buffering operation of the target compressed image at the time of decompression according to the pixel interleaving information and the default decompressed macroblock height comprises:
performing decompression operation on the target compressed image, and judging whether an interweaved pixel exists between a current decompression corresponding image block and a last decompression corresponding image block in the target reconstructed image;
When the judgment result is negative, determining the height of the decompression macro block corresponding to the current decompression corresponding image block as the default decompression macro block height;
when the judgment result is yes, caching the interlaced pixels, and determining the height of a decompressed macro block corresponding to the current decompressed corresponding image block as the height of the macro block obtained by subtracting the height of the interlaced pixels from the height of the default decompressed macro block; the cached interleaved pixels are used for being called when the current decompressed corresponding image block is subjected to super-resolution reconstruction.
4. The image decompression method according to claim 3, wherein the interlaced pixels are pixel rows in the decompressed image of the target compressed image, which are mapped to the current decompressed corresponding image block and the previous decompressed corresponding image block at the same time when performing super resolution reconstruction.
5. The image decompression method for super-resolution reconstruction according to claim 3, wherein said buffering said interleaved pixels comprises:
the interleaved pixels are buffered using a line buffer.
6. The image decompression method for super-resolution reconstruction according to claim 3, further comprising:
Obtaining a decompressed image obtained after decompressing the target compressed image;
and performing super-resolution reconstruction on the decompressed image to obtain the target reconstructed image.
7. The image decompression method for super-resolution reconstruction according to claim 6, wherein said super-resolution reconstructing the decompressed image to obtain the target reconstructed image comprises:
performing super-resolution reconstruction operation on the decompressed image, and judging whether the macroblock height of the current reconstructed macroblock of the decompressed image is smaller than the default decompressed macroblock height;
if yes, calling the corresponding cached interleaved pixels, and combining the pixels in the current reconstructed macroblock to perform super-resolution reconstruction.
8. An image decompression apparatus for super-resolution reconstruction, the apparatus comprising:
the height determining module is used for determining the default decompressed macro block height of the target compressed image;
the interleaving determining module is used for determining pixel interleaving information corresponding to a target reconstruction image corresponding to the target compressed image; the pixel interleaving information is used for indicating that reconstructed pixels of at least two image blocks in the target reconstructed image are mapped to the same pixel row in the decompressed image of the target compressed image;
The operation determining module is used for determining the real-time decompression macro block height and pixel line cache operation of the target compressed image during decompression according to the pixel interleaving information and the default decompression macro block height; the real-time decompressed macro block height is lower than the default decompressed macro block height when the pixel interleaving information exists; the pixel line buffering operation is used for buffering the same pixel line when the pixel interleaving information exists and calling the same pixel line when the same pixel line is to be reconstructed later.
9. An image decompression apparatus for super-resolution reconstruction, the apparatus comprising:
a memory storing executable program code;
a processor coupled to the memory;
the processor invokes the executable program code stored in the memory to perform the image decompression method for super-resolution reconstruction as claimed in any one of claims 1-7.
10. A computer storage medium storing computer instructions which, when invoked, are operable to perform the image decompression method for super-resolution reconstruction of any one of claims 1-7.
CN202110699794.5A 2021-06-23 2021-06-23 Image decompression method and device for super-resolution reconstruction Active CN113709494B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110699794.5A CN113709494B (en) 2021-06-23 2021-06-23 Image decompression method and device for super-resolution reconstruction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110699794.5A CN113709494B (en) 2021-06-23 2021-06-23 Image decompression method and device for super-resolution reconstruction

Publications (2)

Publication Number Publication Date
CN113709494A CN113709494A (en) 2021-11-26
CN113709494B true CN113709494B (en) 2024-04-09

Family

ID=78648186

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110699794.5A Active CN113709494B (en) 2021-06-23 2021-06-23 Image decompression method and device for super-resolution reconstruction

Country Status (1)

Country Link
CN (1) CN113709494B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114071049B (en) * 2022-01-13 2022-05-10 广州匠芯创科技有限公司 Image scaling method, device and storage medium

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101321288A (en) * 2008-05-27 2008-12-10 华为技术有限公司 Reference data loading method, device and video encoder
US8872856B1 (en) * 2008-08-14 2014-10-28 Zenverge, Inc. Macroblock based scaling of images using reduced memory bandwidth
CN104869381A (en) * 2014-02-25 2015-08-26 炬芯（珠海）科技有限公司 Image processing system, method and device
CN108171662A (en) * 2017-12-18 2018-06-15 珠海全志科技股份有限公司 Read the method for image compression data and the anti-distortion method comprising this method
CN110944110A (en) * 2018-09-21 2020-03-31 三星电子株式会社 Image signal processor, operating method thereof and application processor
CN111598779A (en) * 2020-05-14 2020-08-28 Oppo广东移动通信有限公司 Image super-resolution processing method and device, electronic device and storage medium
CN112801852A (en) * 2015-01-22 2021-05-14 谷歌有限责任公司 Virtual linebuffer for image signal processor

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080094406A1 (en) * 2004-08-11 2008-04-24 Koninklijke Philips Electronics, N.V. Stripe-Based Image Data Storage
US7595805B2 (en) * 2006-04-11 2009-09-29 Qualcomm Incorporated Techniques to facilitate use of small line buffers for processing of small or large images
JP2015053644A (en) * 2013-09-09 2015-03-19 オリンパス株式会社 Imaging device

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101321288A (en) * 2008-05-27 2008-12-10 华为技术有限公司 Reference data loading method, device and video encoder
US8872856B1 (en) * 2008-08-14 2014-10-28 Zenverge, Inc. Macroblock based scaling of images using reduced memory bandwidth
CN104869381A (en) * 2014-02-25 2015-08-26 炬芯（珠海）科技有限公司 Image processing system, method and device
CN112801852A (en) * 2015-01-22 2021-05-14 谷歌有限责任公司 Virtual linebuffer for image signal processor
CN108171662A (en) * 2017-12-18 2018-06-15 珠海全志科技股份有限公司 Read the method for image compression data and the anti-distortion method comprising this method
CN110944110A (en) * 2018-09-21 2020-03-31 三星电子株式会社 Image signal processor, operating method thereof and application processor
CN111598779A (en) * 2020-05-14 2020-08-28 Oppo广东移动通信有限公司 Image super-resolution processing method and device, electronic device and storage medium

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Hierarchical dense recursive network for image super-resolution;Kui Jiang;Pattern Recognition;全文 *
视频解码与图像超分辨率重建研究;伊力亚尔・吐尔逊;;电视技术(02);全文 *
超分辨率重建算法综述;李欣;电视技术;全文 *

Also Published As

Publication number Publication date
CN113709494A (en) 2021-11-26

Similar Documents

Publication Publication Date Title
US10542276B2 (en) Data caching method and apparatus for video decoder
US11922599B2 (en) Video super-resolution processing method and apparatus
KR100242880B1 (en) Apparatus for tracing contours of segmented regions
US8787460B1 (en) Method and apparatus for motion vector estimation for an image sequence
CN110166771B (en) Video encoding method, video encoding device, computer equipment and storage medium
JPH11243543A (en) Method and system for detecting screen content causing prediction error and using detected information in low-resolution video decoder
GB2560548A (en) Video data processing system
CN113709494B (en) Image decompression method and device for super-resolution reconstruction
US11983848B2 (en) AI frame engine for mobile edge
CN110913219A (en) Video frame prediction method and device and terminal equipment
JP4340532B2 (en) Error concealment method and apparatus
CN113411581B (en) Motion compensation method, system, storage medium and terminal for video sequence
CN112669328B (en) Medical image segmentation method
CN112449182A (en) Video encoding method, device, equipment and storage medium
KR19980085643A (en) Binary Image Interpolation Method Using Adaptive Threshold of Ambient Pixels
CN113852816A (en) Video frame processing method and device, computer equipment and readable storage medium
CN116233453B (en) Video coding method and device
CN116489368B (en) Image dynamic compression method and image dynamic compression device
CN110572713B (en) Transcoding method and processing terminal for adaptive video bandwidth ratio
CN113610705B (en) Image de-interlacing method and device for super-division reconstruction
US20150098512A1 (en) Image processing apparatus and method thereof
CN114727116A (en) Encoding method and device
JP2000278693A (en) Moving picture expanding method and recording medium recording program executing the same
Sun et al. Reduced-reference image quality assessment through SIFT intensity ratio
CN110996108A (en) Video frame reconstruction method and device and terminal equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant