CN113674232A - Image noise estimation method and device, electronic equipment and storage medium - Google Patents
Image noise estimation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113674232A CN113674232A CN202110924967.9A CN202110924967A CN113674232A CN 113674232 A CN113674232 A CN 113674232A CN 202110924967 A CN202110924967 A CN 202110924967A CN 113674232 A CN113674232 A CN 113674232A
- Authority
- CN
- China
- Prior art keywords
- image block
- value
- target
- image
- gray value
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000004927 fusion Effects 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 11
- 238000010586 diagram Methods 0.000 description 9
- 230000009467 reduction Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 4
- 238000012935 Averaging Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 241000023320 Luma <angiosperm> Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- OSWPMRLSEDHDFF-UHFFFAOYSA-N methyl salicylate Chemical compound COC(=O)C1=CC=CC=C1O OSWPMRLSEDHDFF-UHFFFAOYSA-N 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/97—Determining parameters from multiple pictures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The application relates to an image noise estimation method, an image noise estimation device, an electronic device and a storage medium. The method comprises the following steps: acquiring a reference frame image and a target frame image aligned with the reference frame image; dividing the reference frame image and the target frame image to obtain a reference image block and a target image block; determining the energy value of a reference image block and the difference value of the reference image block and the related image block; dividing the gray value range of the pixel point into a plurality of sub-gray value ranges; classifying the energy value and the difference value of the reference image block into the sub-gray value interval; determining a target gray value corresponding to the sub gray value interval based on each energy value of the sub gray value interval, and determining a target noise value corresponding to the target gray value based on each image block difference value of the sub gray value interval; and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value. By adopting the method, the image noise estimation accuracy can be improved.
Description
Technical Field
The present application relates to the field of image processing, and in particular, to an image noise estimation method and apparatus, an electronic device, and a storage medium.
Background
With the development of computer technology, image denoising technology appears, which refers to a process of reducing image noise in a digital image. Image noise refers to unnecessary or unnecessary interference information present in the image data. The presence of noise seriously affects the quality of the image, and therefore, image noise reduction of noise in the image is necessary. The noise estimation has very important guiding function in the image denoising process, and accurate denoising of the image can be realized by estimating the noise contained in the image. In the conventional technology, an original image is usually convolved through a plurality of special filter operators, and an image gradient is extracted, so that the noise of the image is determined.
However, the conventional image noise estimation method is very sensitive to small edges and details of an image, the small edges and the details are also considered as noise, and the estimated noise is usually large, so that the image noise estimation accuracy is low.
Disclosure of Invention
The embodiment of the application provides an image noise estimation method and device, electronic equipment and a computer readable storage medium, which can improve the image noise estimation accuracy.
A method of image noise estimation, the method comprising:
acquiring a reference frame image and a target frame image aligned with the reference frame image;
dividing the reference frame image into a plurality of reference image blocks and dividing the target frame image into a plurality of target image blocks according to the same image division mode;
determining an energy value corresponding to each reference image block, and determining an image block difference value between the reference image block and a related image block; the related image block is a target image block with the same position as the reference image block;
dividing gray value intervals corresponding to pixel points in the reference frame image into preset number of sub-gray value intervals;
dividing the energy value and the image block difference value corresponding to each reference image block into the sub-gray value intervals;
determining a target gray value corresponding to each sub-gray value interval based on each energy value in each sub-gray value interval, and determining a target noise value corresponding to the target gray value based on each image block difference value in each sub-gray value interval;
and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value.
An image noise estimation apparatus, the apparatus comprising:
the device comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring a reference frame image and a target frame image aligned with the reference frame image;
the dividing module is used for dividing the reference frame image into a plurality of reference image blocks and dividing the target frame image into a plurality of target image blocks according to the same image dividing mode;
the determining module is used for determining an energy value corresponding to each reference image block and determining image block difference values of the reference image block and related image blocks; the related image block is a target image block with the same position as the reference image block;
the dividing module is further configured to divide a gray value dereferencing interval corresponding to a pixel point in the reference frame image into a preset number of sub-gray value intervals;
the classifying module is used for classifying the energy value and the image block difference value corresponding to each reference image block into the corresponding sub-gray value interval;
the determining module is configured to determine a target gray value corresponding to each sub-gray value interval based on each energy value in each sub-gray value interval, and determine a target noise value corresponding to the target gray value based on each image block difference value in the sub-gray value interval; and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value.
An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
acquiring a reference frame image and a target frame image aligned with the reference frame image;
dividing the reference frame image into a plurality of reference image blocks and dividing the target frame image into a plurality of target image blocks according to the same image division mode;
determining an energy value corresponding to each reference image block, and determining an image block difference value between the reference image block and a related image block; the related image block is a target image block with the same position as the reference image block;
dividing gray value intervals corresponding to pixel points in the reference frame image into preset number of sub-gray value intervals;
dividing the energy value and the image block difference value corresponding to each reference image block into the sub-gray value intervals;
determining a target gray value corresponding to each sub-gray value interval based on each energy value in each sub-gray value interval, and determining a target noise value corresponding to the target gray value based on each image block difference value in each sub-gray value interval;
and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring a reference frame image and a target frame image aligned with the reference frame image;
dividing the reference frame image into a plurality of reference image blocks and dividing the target frame image into a plurality of target image blocks according to the same image division mode;
determining an energy value corresponding to each reference image block, and determining an image block difference value between the reference image block and a related image block; the related image block is a target image block with the same position as the reference image block;
dividing gray value intervals corresponding to pixel points in the reference frame image into preset number of sub-gray value intervals;
dividing the energy value and the image block difference value corresponding to each reference image block into the sub-gray value intervals;
determining a target gray value corresponding to each sub-gray value interval based on each energy value in each sub-gray value interval, and determining a target noise value corresponding to the target gray value based on each image block difference value in each sub-gray value interval;
and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value.
The image noise estimation method, the device, the electronic equipment and the storage medium obtain a reference frame image and a target frame image aligned with the reference frame image, divide the reference frame image into a plurality of reference image blocks according to the same image division mode, and divide the target frame image into a plurality of target image blocks, so that each reference image block and the corresponding target image block have the same number of pixel points; determining an energy value corresponding to each reference image block, and determining image block difference values of the reference image block and the related image block so as to quantize the difference between the reference image block and the related image block; dividing gray value intervals corresponding to pixel points in the reference frame image into preset number of sub-gray value intervals; dividing the energy value and the image block difference value corresponding to each reference image block into the sub-gray value intervals; determining a target gray value corresponding to the sub-gray value interval based on each energy value in each sub-gray value interval, and determining a target noise value corresponding to the target gray value based on each image block difference value in the sub-gray value interval; and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value. Therefore, the reference frame image and the target frame image are divided into corresponding reference image blocks and target image blocks respectively, the energy value of each reference image block is calculated, the image block difference value between each reference image block and the corresponding image block is calculated, each energy value and each image block difference value are divided into corresponding sub-gray value intervals, the target gray value and the target noise value corresponding to each sub-gray value interval are calculated, the noise value corresponding to each reference image block in the reference frame image is estimated, and the image noise estimation accuracy is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a diagram illustrating an exemplary embodiment of a method for estimating noise in an image;
FIG. 2 is a flow diagram illustrating an embodiment of a method for image noise estimation;
FIG. 3 is a flow chart illustrating an image noise estimation method according to another embodiment;
FIG. 4 is a block diagram of an embodiment of an image noise estimation apparatus;
FIG. 5 is a block diagram of an image noise estimation apparatus according to another embodiment;
FIG. 6 is a diagram illustrating an internal structure of an electronic device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
FIG. 1 is a diagram illustrating an exemplary environment in which the image noise estimation method may be implemented. As shown in fig. 1, the application environment includes a terminal 102 and a server 104. The terminal 102 and the server 104 communicate via a network. The terminal 102 may specifically include a desktop terminal or a mobile terminal. The mobile terminal may specifically include at least one of a mobile phone, a tablet computer, a notebook computer, and the like. The server 104 may be implemented as a stand-alone server or as a server cluster comprised of multiple servers. Those skilled in the art will understand that the application environment shown in fig. 1 is only a part of the scenario related to the present application, and does not constitute a limitation to the application environment of the present application.
The terminal 102 may obtain the reference frame image and the target frame image aligned with the reference frame image from the server 104, and divide the reference frame image into a plurality of reference image blocks and the target frame image into a plurality of target image blocks according to the same image division manner. The terminal 102 may determine an energy value corresponding to each reference image block and determine an image block difference value between the reference image block and the associated image block. The terminal 102 may divide a gray value range corresponding to a pixel point in the reference frame image into a preset number of sub-gray value ranges, and divide an energy value and an image block difference value corresponding to each reference image block into the sub-gray value ranges. The terminal 102 may determine a target gray value corresponding to each sub-gray value interval based on each energy value in each sub-gray value interval, and determine a target noise value corresponding to the target gray value based on each image block difference value in the sub-gray value interval. The terminal 102 may determine a noise value corresponding to each reference image block in the reference frame image based on each target gray value and a target noise value corresponding to each target gray value.
FIG. 2 is a flow diagram of a method for image noise estimation in one embodiment. The image noise estimation method in this embodiment is described by taking the terminal 102 in fig. 1 as an example. As shown in fig. 2, the image noise estimation method includes the following steps:
The reference frame image is an image which is used as a reference in the acquired multi-frame images. The target frame image is an image of the multi-frame image except the reference frame and aligned with the reference frame image. The multi-frame image is a plurality of images acquired for the same shooting scene.
Specifically, the terminal may acquire a reference frame image and a target frame image aligned with the reference frame image.
In one embodiment, a reference frame image and a target frame image aligned with the reference frame image are stored in the server in advance. The terminal can communicate with the server and directly acquire the reference frame image and the target frame image aligned with the reference frame image from the server.
In one embodiment, the terminal may include a camera, and in a specific scene, the terminal may shoot the same target object through the camera to obtain a multi-frame image corresponding to the target object. The terminal can process the multi-frame image to obtain a reference frame image and a target frame image aligned with the reference frame image.
The image division refers to an image processing method for dividing an image into a plurality of image blocks. The same image division method refers to a method of dividing an image by the same pixel length and the same pixel width. For example, the image is divided by 8 × 8, that is, the image is divided by the pixel length of 8 (pixels) and the pixel width of 8 (pixels). It can be understood that each image block obtained by dividing the image by 8 × 8 includes 64 pixel points. The reference image block is an image block in the reference frame image and the target image block is an image block in the target frame image.
Specifically, the terminal may preset an image division mode, and perform image division on the reference frame image according to the image division mode to obtain a plurality of reference image blocks corresponding to the reference frame image. Meanwhile, the terminal can perform image division on the target frame image according to the image division mode to obtain a plurality of target image blocks corresponding to the target frame image.
Step 206, determining an energy value corresponding to each reference image block, and determining an image block difference value between the reference image block and a related image block; the relevant image block is a target image block located at the same position as the reference image block.
The energy value is a value representing the energy contained in the reference image block, and it can be understood that the energy value can represent the brightness of the reference image block. The image block difference value is a value representing the degree of difference between the reference image block and the relevant image block.
Specifically, the terminal may calculate an energy value corresponding to each reference image block by referring to a gray value corresponding to each pixel point in the image block, and calculate an image block difference value between the reference image block and the related image block by referring to a gray value corresponding to each pixel point in the image block.
In one embodiment, the terminal may determine an average value of the gray-scale values corresponding to the pixel points in each reference image block, and use the average value as the energy value corresponding to the reference image block. The terminal can determine the sum of squares of absolute values of differences between the gray values corresponding to the pixel points in the reference image block and the gray values corresponding to the pixel points in the related image block, and the sum of squares of the absolute values of the differences is used as the image block difference value of the reference image block and the related image block.
And 208, dividing a gray value range corresponding to the pixel points in the reference frame image into a preset number of sub-gray value ranges.
The gray value range corresponding to the pixel point is the gray value range corresponding to the pixel point. The sub-gray value interval is a sub-interval in the gray value interval, and it can be understood that the gray value interval includes a plurality of sub-gray value intervals.
Specifically, the terminal can determine a value range of a gray value corresponding to a pixel point in the reference frame image, and use the value range of the gray value corresponding to the pixel point in the reference frame image as a gray value interval corresponding to the pixel point in the reference frame image. The terminal can divide the gray value interval into a preset number of sub-gray value intervals.
For example, if the gray scale value corresponding to the pixel point in the reference frame image is in the range of [0,255], the gray scale value corresponding to the pixel point in the reference frame image is in the range of [0,255 ]. And if the terminal averagely divides the gray value range of [0,255] into N sub-gray value ranges. The N sub-gray value intervals may be represented as [0,256/N), [256/N, 256 × 2/N ], and [256 × N (N-1)/N, 256), respectively. Wherein N is a natural number of 2 or more.
And step 210, classifying the energy value and the image block difference value corresponding to each reference image block into the corresponding sub-gray value interval.
Specifically, the terminal may classify the energy value corresponding to each reference image block into the sub-gray value interval to which the energy value corresponding to the reference image block belongs. Meanwhile, the terminal can classify the image block difference value corresponding to each reference image block into the sub-gray value interval to which the image block difference value corresponding to the reference image block belongs.
And the target gray value is a gray value corresponding to the sub gray value interval. The target noise value is a noise value corresponding to the sub-gray value interval.
Specifically, the terminal may calculate a target grayscale value corresponding to each sub-grayscale value interval based on each energy value in the sub-grayscale value interval. The terminal can calculate a target noise value corresponding to the target gray value based on the difference value of each image block in the sub-gray value interval.
In one embodiment, within the sub-gray value interval, the terminal may reject energy values greater than a preset energy threshold corresponding to the sub-gray value interval; and sequencing the energy values which are not removed, and taking the energy value positioned in the middle of the sequenced sequence as a target gray value. The terminal can use the average value of the difference values of each image block in each sub-gray value interval as the target noise value corresponding to the sub-gray value interval.
Specifically, the terminal may determine, in a linear interpolation manner, a noise value corresponding to each reference image block in the reference frame image based on each target gray value and a target noise value corresponding to each target gray value.
In an embodiment, the terminal may further determine, by means of linear interpolation, a noise value corresponding to each pixel point in the reference frame image based on each target gray value and a target noise value corresponding to each target gray value. Specifically, the terminal may determine a noise value corresponding to each gray value in the gray value range by means of linear interpolation, and determine a gray value corresponding to each pixel point. The terminal can determine the noise value corresponding to each pixel point in the reference frame image based on the gray value corresponding to each pixel point and the noise value corresponding to each gray value in the gray value range.
The image noise estimation method comprises the steps of obtaining a reference frame image and a target frame image aligned with the reference frame image, dividing the reference frame image into a plurality of reference image blocks according to the same image dividing mode, and dividing the target frame image into a plurality of target image blocks, so that each reference image block and the corresponding target image block have the same number of pixel points; determining an energy value corresponding to each reference image block, and determining image block difference values of the reference image block and the related image block so as to quantize the difference between the reference image block and the related image block; dividing gray value intervals corresponding to pixel points in the reference frame image into preset number of sub-gray value intervals; dividing the energy value and the image block difference value corresponding to each reference image block into the sub-gray value intervals; determining a target gray value corresponding to the sub-gray value interval based on each energy value in each sub-gray value interval, and determining a target noise value corresponding to the target gray value based on each image block difference value in the sub-gray value interval; and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value. Therefore, the reference frame image and the target frame image are divided into corresponding reference image blocks and target image blocks respectively, the energy value of each reference image block is calculated, the image block difference value between each reference image block and the corresponding image block is calculated, each energy value and each image block difference value are divided into corresponding sub-gray value intervals, the target gray value and the target noise value corresponding to each sub-gray value interval are calculated, the noise value corresponding to each reference image block in the reference frame image is estimated, and the image noise estimation accuracy is improved.
Meanwhile, the image noise estimation method has strong adaptability to an image domain, and can be used for a raw domain, a YUV domain and the like.
In one embodiment, the step 202 of acquiring the reference frame image and the target frame image aligned with the reference frame image specifically includes: acquiring multi-frame images acquired in the same shooting scene; selecting a reference frame image and at least one target frame image from a plurality of frame images; and referring to the reference frame image, and performing image alignment operation on the target frame image to obtain a target frame image aligned with the reference frame image.
Specifically, the terminal can acquire multi-frame images for the same shooting scene, and then the terminal can acquire the multi-frame images acquired in the same shooting scene. The terminal can select a reference frame image and at least one target frame image from the multi-frame images based on the definition and the brightness of the images. It can be understood that the terminal may select an image with the highest definition and the most suitable brightness from the plurality of images as the reference frame image. The terminal can refer to the reference frame image, and perform image alignment operation on the target frame image to obtain the target frame image aligned with the reference frame image.
In the above embodiment, the multiple frames of images are acquired through the same shooting scene, so that the difference between the multiple frames of images is not too large. By selecting the reference frame image from the multi-frame images, the image serving as the reference has higher definition and more proper brightness. Therefore, the image noise estimation accuracy is further improved.
In an embodiment, the step 206, that is, the step of determining the energy value corresponding to each reference image block and determining the image block difference values of the reference image block and the related image block specifically includes: determining the root mean square of the gray values corresponding to the pixel points in each reference image block, and taking the root mean square as the energy value corresponding to the reference image block; and determining the sum of absolute values of differences between the gray values corresponding to the pixel points in the reference image block and the gray values corresponding to the pixel points in the related image block, and taking the sum of the absolute values of the differences as the image block difference value of the reference image block and the related image block.
The Root Mean Square (RMS) of the gray values can be obtained by summing the squares of all the gray values, averaging them, and then squaring. The Sum of Absolute values (SAD, Sum of Absolute Differences) of the Differences between the gray values can be obtained by calculating the difference between the gray value corresponding to each pixel in the reference image block and the gray value corresponding to each pixel in the related image block, taking the Absolute value, and then adding the Absolute values.
Specifically, the terminal may calculate a root mean square of the gray scale value corresponding to each pixel point in each reference image block, and use the root mean square as the energy value corresponding to the reference image block. The terminal can determine the sum of absolute values of differences between the gray values corresponding to the pixel points in the reference image block and the gray values corresponding to the pixel points in the related image block, and the sum of the absolute values of the differences is used as the image block difference value of the reference image block and the related image block.
In the embodiment, the root mean square is used as the energy value corresponding to the reference image block, and the sum of the absolute values of the differences is used as the image block difference value between the reference image block and the related image block, so that the image noise estimation accuracy is further improved.
In an embodiment, the step 212, that is, determining the target gray scale value corresponding to the sub-gray scale value interval based on each energy value in each sub-gray scale value interval, and determining the target noise value corresponding to the target gray scale value based on the difference value of each image block in the sub-gray scale value interval specifically includes: taking the average value of each energy value in each sub-gray value interval as a target gray value corresponding to the sub-gray value interval; eliminating image block difference values larger than a preset threshold corresponding to the sub-gray value intervals in the sub-gray value intervals; and sorting the image block difference values which are not removed, and taking the image block difference value in the middle of the sorted sequence as a target noise value corresponding to the target gray value.
Specifically, the terminal may perform an averaging operation on each energy value in each sub-gray value interval to obtain an average value of each energy value in each sub-gray value interval. The terminal can use the average value of each energy value in each sub-gray value interval as the target gray value corresponding to the sub-gray value interval. The terminal can compare the image block difference values in the sub-gray value interval with the preset threshold corresponding to the sub-gray value interval respectively to obtain the image block difference values larger than the preset threshold. Furthermore, the terminal can reject the image block difference value larger than the preset threshold value corresponding to the sub-gray value interval in the sub-gray value interval, sort the image block difference values not rejected, and use the image block difference value in the middle of the sorted sequence as the target noise value corresponding to the target gray value.
In the above embodiment, the average value of the energy values in each sub-gray value interval is used as the target gray value corresponding to the sub-gray value interval, so that the target gray value can more accurately represent the gray value of each image block in the corresponding sub-gray value interval. By eliminating the image block difference value larger than the preset threshold corresponding to the sub-gray value interval, the situation that the difference between the reference image block and the related image block is large can be eliminated. And the difference value of the image block in the middle of the sorted sequence is used as the target noise value corresponding to the target gray value, so that the target noise value can more accurately represent the noise value of each reference image block in the corresponding sub-gray value interval.
In an embodiment, the step 214, that is, the step of determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value specifically includes: determining a noise value corresponding to each gray value in a gray value range by a linear interpolation mode based on each target gray value and a target noise value corresponding to each target gray value; taking the energy value corresponding to each reference image block in the reference frame image as the gray value corresponding to each reference image block in the reference frame image; and determining the noise value corresponding to each reference image block in the reference frame image based on the gray value corresponding to each reference image block in the reference frame image and the noise value corresponding to each gray value in the gray value range.
The linear interpolation method is a method of determining a value of an unknown quantity between two known quantities using a straight line connecting the two known quantities.
Specifically, the terminal may determine, based on each target grayscale value and a target noise value corresponding to each target grayscale value, a noise value corresponding to each grayscale value in the grayscale value interval in a linear interpolation manner. The terminal can directly use the energy value corresponding to each reference image block in the reference frame image as the gray value corresponding to each reference image block in the reference frame image. The terminal can index to obtain the noise value corresponding to each reference image block in the reference frame image in the noise value mapping relation corresponding to each gray value in the gray value interval based on the gray value corresponding to each reference image block in the reference frame image.
It can be understood that, if the gray value is used as the abscissa and the noise value is used as the ordinate, for each target gray value, the terminal may connect the coordinate point formed by the target gray value and the corresponding target noise value with the coordinate point corresponding to another adjacent target gray value by using a line segment, the abscissa value corresponding to the line segment is the gray value in the interval corresponding to the two target gray values, and the ordinate value corresponding to the line segment is the noise value in the interval of the target noise values corresponding to the two target gray values. Furthermore, based on each target gray value and the target noise value corresponding to each target gray value, a corresponding line graph can be drawn, and based on the line graph, the noise value corresponding to each gray value in the gray value range can be estimated.
In the above embodiment, based on each target gray value and the target noise value corresponding to each target gray value, the noise value corresponding to each reference image block in the reference frame image may be determined in a linear interpolation manner, so that the accuracy of estimating the noise value corresponding to each reference image block in the reference frame image is improved.
In an embodiment, the image noise estimation method specifically further includes: determining fusion weights between each reference image block and related image blocks in the reference frame image based on the image block difference value and the noise value corresponding to each reference image block in the reference frame image; and performing multi-frame fusion denoising processing on the reference frame image based on the fusion weight between each reference image block and the related image block in the reference frame image to obtain a denoised target image.
The multi-frame fusion noise reduction processing is an image processing method for effectively reducing image noise through synthesis of a plurality of images. The target image is an image output after multi-frame fusion noise reduction processing.
Specifically, the terminal may determine fusion weights between each reference image block and related image blocks in the reference frame image based on image block difference values and noise values corresponding to each reference image block in the reference frame image, and perform multi-frame fusion denoising processing on the reference frame image based on the fusion weights between each reference image block and related image blocks in the reference frame image to obtain a denoised target image.
In the above embodiment, the fusion weight between each reference image block and the related image block in the reference frame image is determined based on the image block difference value and the noise value corresponding to each reference image block in the reference frame image, so that the reference image block and the related image block have more reasonable fusion weight. Through the fusion weight between the reference image block and the related image block, the multi-frame fusion noise reduction processing is carried out on the reference frame image, the noise reduction accuracy rate is improved, and the output target image has higher definition.
In one embodiment, applied to the YUV domain, the energy values include a Y-channel energy value, a U-channel energy value, and a V-channel energy value, and the tile difference values include a Y-channel tile difference value, a U-channel tile difference value, and a V-channel tile difference value. The step of determining the energy value corresponding to each reference image block in step 206 specifically includes: for the Y channel, determining a Y channel energy value corresponding to each reference image block based on the Y value; and taking the Y-channel energy value as a U-channel energy value corresponding to the U channel, and taking the Y-channel energy value as a V-channel energy value corresponding to the V channel. The step of determining the image block difference values of the reference image block and the related image block in step 206 specifically includes: for a Y channel, determining a Y channel image block difference value of a reference image block and a related image block based on a Y value; for the U channel, determining a U channel image block difference value of a reference image block and a related image block based on the U value; and for the V channel, determining the V channel image block difference value of the reference image block and the related image block based on the V value.
Among them, YUV is a color coding method. Y represents brightness (Luma) that is gray value, and U and V represent Chroma (Chroma or Chroma), which is used to describe image color and saturation for specifying the color of a pixel. It can be understood that, in the YUV domain, the pixel value of the pixel point of the image can be represented by a value corresponding to three channels, i.e., a Y value of the Y channel, a U value of the U channel, and a V value of the V channel, and the pixel value can be specifically represented as (Y, U, V). The Y-channel energy value is an energy value corresponding to the Y-channel, the U-channel energy value is an energy value corresponding to the U-channel, and the V-channel energy value is an energy value corresponding to the V-channel. The Y-channel image block difference value is an image block difference value corresponding to the Y-channel, the U-channel image block difference value is an image block difference value corresponding to the U-channel, and the V-channel image block difference value is an image block difference value corresponding to the V-channel.
Specifically, for the Y channel, the terminal may determine a Y channel energy value corresponding to each reference image block based on the Y value, i.e., directly take the Y value as the Y channel energy value. The terminal can multiplex Y-channel energy values in the U-channel and the V-channel, namely, the Y-channel energy values can be directly used as U-channel energy values corresponding to the U-channel and the Y-channel energy values can be directly used as V-channel energy values corresponding to the V-channel. For the Y channel, the terminal may determine a Y channel image block difference value of the reference image block and the related image block based on the Y value, determine a U channel image block difference value of the reference image block and the related image block based on the U value for the U channel, and determine a V channel image block difference value of the reference image block and the related image block based on the V value for the V channel.
In one embodiment, the terminal may determine a noise value corresponding to the Y channel based on the Y channel energy value and the Y channel image block difference value. The terminal can determine a noise value corresponding to the U channel based on the U channel energy value and the U channel image block difference value. The terminal can determine a noise value corresponding to the V channel based on the V channel energy value and the V channel image block difference value.
In the above embodiment, the U channel energy value corresponding to the U channel and the V channel energy value corresponding to the V channel can be quickly determined by multiplexing the Y channel energy values corresponding to the Y channels. And respectively and quickly determining a Y-channel image block difference value corresponding to the Y channel, a U-channel image block difference value corresponding to the U channel and a V-channel image block difference value corresponding to the V channel through the Y value, the U value and the V value which respectively correspond to the Y channel, the U channel and the V channel.
In one embodiment, as shown in fig. 3, the terminal may acquire a plurality of frame images from which a reference frame image and at least one target frame image are desired to be selected. The terminal can take the reference frame image as a reference, and carry out image alignment operation on the target frame image and the reference frame image to obtain an aligned target frame image. The terminal can divide the reference frame image and the target frame image according to the same image division mode to obtain the corresponding reference image block and the corresponding target image block. The terminal can calculate the energy value of each reference image block and calculate the image block difference value of each reference image block and the related image block. The terminal can divide the energy value of each reference image block and the image block difference value of each reference image block and the related image block into the sub-gray value intervals. For each sub-gray value interval, the terminal can calculate a target gray value and a target noise value corresponding to the sub-gray value interval based on all energy values and image block difference values in the sub-gray value interval. The terminal can output the target gray value and the target noise value corresponding to each sub gray value interval. Furthermore, the terminal can determine the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value.
In a specific embodiment, an image noise estimation method is provided, which specifically includes the following processes:
(1) acquiring a plurality of frames of images acquired in the same shooting scene.
(2) A reference frame image and at least one target frame image are selected from the multi-frame images.
(3) And referring to the reference frame image, and performing image alignment operation on the target frame image to obtain a target frame image aligned with the reference frame image.
(4) According to the same image division mode, the reference frame image is divided into a plurality of reference image blocks, and the target frame image is divided into a plurality of target image blocks.
(5) And determining the root mean square of the gray value corresponding to each pixel point in each reference image block.
(6) Determining the sum of absolute values of differences between the gray values corresponding to the pixel points in the reference image block and the gray values corresponding to the pixel points in the related image block; the relevant image block is a target image block located at the same position as the reference image block.
In one embodiment, the image noise estimation method can be further applied to a YUV domain, wherein the root mean square comprises a Y channel root mean square, a U channel root mean square and a V channel root mean square, and the sum of absolute values of differences comprises the sum of absolute values of Y channel differences, the sum of absolute values of U channel differences and the sum of absolute values of V channel differences; and taking the Y channel root mean square as the U channel root mean square corresponding to the U channel, and taking the Y channel root mean square as the V channel root mean square corresponding to the V channel. For the Y channel, determining the sum of absolute values of Y channel differences of the reference image block and the related image block based on the Y value; for the U channel, determining the sum of the absolute values of the U channel differences of the reference image block and the related image block based on the U value; for the V channel, the sum of the absolute values of the V channel differences for the reference image block and the relevant image block is determined based on the V value.
(7) Dividing the gray value range corresponding to the pixel points in the reference frame image into a preset number of sub-gray value ranges.
(8) And dividing the sum of the root mean square and the absolute value of the difference corresponding to each reference image block into the sub-gray value intervals.
(9) And taking the mean value of the root mean square values in each sub-gray value interval as the target gray value corresponding to the sub-gray value interval.
(10) And eliminating the sum of absolute values of the differences larger than the preset threshold corresponding to the sub-gray value interval in the sub-gray value interval.
(11) And sorting the sum of the absolute values of the differences which are not removed, and taking the sum of the absolute values of the differences positioned in the middle of the sorted sequence as a target noise value corresponding to the target gray value.
(12) And determining the noise value corresponding to each gray value in the gray value range by a linear interpolation mode based on each target gray value and the target noise value corresponding to each target gray value.
(13) And taking the root mean square corresponding to each reference image block in the reference frame image as the gray value corresponding to each reference image block in the reference frame image.
(14) And determining the noise value corresponding to each reference image block in the reference frame image based on the gray value corresponding to each reference image block in the reference frame image and the noise value corresponding to each gray value in the gray value range.
(15) And determining fusion weights between each reference image block and the related image block in the reference frame image based on the sum of the absolute values of the differences corresponding to each reference image block in the reference frame image and the noise value.
(16) And performing multi-frame fusion denoising processing on the reference frame image based on the fusion weight between each reference image block and the related image block in the reference frame image to obtain a denoised target image.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 2 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 4, there is provided an image noise estimation apparatus 400, including: an obtaining module 401, a dividing module 402, a determining module 403 and a dividing module 404, wherein:
an obtaining module 401 is configured to obtain a reference frame image and a target frame image aligned with the reference frame image.
The dividing module 402 is configured to divide the reference frame image into a plurality of reference image blocks and divide the target frame image into a plurality of target image blocks according to the same image dividing manner.
A determining module 403, configured to determine an energy value corresponding to each reference image block, and determine an image block difference value between the reference image block and a related image block; the relevant image block is a target image block located at the same position as the reference image block.
The dividing module 402 is further configured to divide a gray value range corresponding to a pixel point in the reference frame image into a preset number of sub-gray value ranges.
The dividing module 404 is configured to divide the energy value and the image block difference value corresponding to each reference image block into the sub-gray value intervals to which the reference image block belongs.
A determining module 403, configured to determine a target gray value corresponding to each sub-gray value interval based on each energy value in each sub-gray value interval, and determine a target noise value corresponding to the target gray value based on each image block difference value in the sub-gray value interval; and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value.
In an embodiment, the obtaining module 401 is further configured to obtain multiple frames of images acquired in the same shooting scene; selecting a reference frame image and at least one target frame image from a plurality of frame images; and referring to the reference frame image, and performing image alignment operation on the target frame image to obtain a target frame image aligned with the reference frame image.
In one embodiment, the determining module 403 is further configured to determine a root mean square of the gray scale values corresponding to the pixel points in each reference image block, and use the root mean square as the energy value corresponding to the reference image block; and determining the sum of absolute values of differences between the gray values corresponding to the pixel points in the reference image block and the gray values corresponding to the pixel points in the related image block, and taking the sum of the absolute values of the differences as the image block difference value of the reference image block and the related image block.
In one embodiment, the determining module 403 is further configured to use an average value of the energy values in each sub-gray value interval as a target gray value corresponding to the sub-gray value interval; eliminating image block difference values larger than a preset threshold corresponding to the sub-gray value intervals in the sub-gray value intervals; and sorting the image block difference values which are not removed, and taking the image block difference value in the middle of the sorted sequence as a target noise value corresponding to the target gray value.
In an embodiment, the determining module 403 is further configured to determine, based on each target gray value and a target noise value corresponding to each target gray value, a noise value corresponding to each gray value in a gray value range by a linear interpolation manner; taking the energy value corresponding to each reference image block in the reference frame image as the gray value corresponding to each reference image block in the reference frame image; and determining the noise value corresponding to each reference image block in the reference frame image based on the gray value corresponding to each reference image block in the reference frame image and the noise value corresponding to each gray value in the gray value range.
In one embodiment, applied to the YUV domain, the energy values include a Y-channel energy value, a U-channel energy value, and a V-channel energy value, and the tile difference values include a Y-channel tile difference value, a U-channel tile difference value, and a V-channel tile difference value. The determining module 403 is further configured to determine, for the Y channel, a Y channel energy value corresponding to each reference image block based on the Y value; taking the Y-channel energy value as a U-channel energy value corresponding to the U channel, and taking the Y-channel energy value as a V-channel energy value corresponding to the V channel; for a Y channel, determining a Y channel image block difference value of a reference image block and a related image block based on a Y value; for the U channel, determining a U channel image block difference value of a reference image block and a related image block based on the U value; and for the V channel, determining the V channel image block difference value of the reference image block and the related image block based on the V value.
Referring to fig. 5, in one embodiment, the image noise estimation apparatus 400 further includes: a noise reduction module 405, wherein:
the denoising module 405 is configured to determine a fusion weight between each reference image block and a related image block in the reference frame image based on an image block difference value and a noise value corresponding to each reference image block in the reference frame image; and performing multi-frame fusion denoising processing on the reference frame image based on the fusion weight between each reference image block and the related image block in the reference frame image to obtain a denoised target image.
The image noise estimation device acquires a reference frame image and a target frame image aligned with the reference frame image, divides the reference frame image into a plurality of reference image blocks according to the same image division mode, and divides the target frame image into a plurality of target image blocks, so that each reference image block and the corresponding target image block have the same number of pixel points; determining an energy value corresponding to each reference image block, and determining image block difference values of the reference image block and the related image block so as to quantize the difference between the reference image block and the related image block; dividing gray value intervals corresponding to pixel points in the reference frame image into preset number of sub-gray value intervals; dividing the energy value and the image block difference value corresponding to each reference image block into the sub-gray value intervals; determining a target gray value corresponding to the sub-gray value interval based on each energy value in each sub-gray value interval, and determining a target noise value corresponding to the target gray value based on each image block difference value in the sub-gray value interval; and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value. Therefore, the reference frame image and the target frame image are divided into corresponding reference image blocks and target image blocks respectively, the energy value of each reference image block is calculated, the image block difference value between each reference image block and the corresponding image block is calculated, each energy value and each image block difference value are divided into corresponding sub-gray value intervals, the target gray value and the target noise value corresponding to each sub-gray value interval are calculated, the noise value corresponding to each reference image block in the reference frame image is estimated, and the image noise estimation accuracy is improved.
The division of each module in the image noise estimation apparatus is only for illustration, and in other embodiments, the image noise estimation apparatus may be divided into different modules as needed to complete all or part of the functions of the image noise estimation apparatus.
For the specific definition of the image noise estimation apparatus, reference may be made to the above definition of the image noise estimation method, which is not described herein again. All or part of the modules in the image noise estimation device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the electronic device, or can be stored in a memory in the electronic device in a software form, so that the processor can call and execute operations corresponding to the modules.
Fig. 6 is a schematic diagram of an internal structure of an electronic device in one embodiment. As shown in fig. 6, the electronic device includes a processor and a memory connected by a system bus. Wherein, the processor is used for providing calculation and control capability and supporting the operation of the whole electronic equipment. The memory may include a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The computer program can be executed by a processor to implement an image noise estimation method provided in the following embodiments. The internal memory provides a cached execution environment for the operating system computer programs in the non-volatile storage medium. The electronic device may be any terminal device such as a mobile phone, a tablet computer, a PDA (Personal Digital Assistant), a Point of Sales (POS), a vehicle-mounted computer, and a wearable device.
Those skilled in the art will appreciate that the architecture shown in fig. 6 is a block diagram of only a portion of the architecture associated with the subject application, and does not constitute a limitation on the servers to which the subject application applies, as a particular server may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
The implementation of each module in the image noise estimation apparatus provided in the embodiment of the present application may be in the form of a computer program. The computer program may be run on a terminal or a server. Program modules constituted by such computer programs may be stored on the memory of the electronic device. Which when executed by a processor, performs the steps of the method described in the embodiments of the present application.
The embodiment of the application also provides a computer readable storage medium. One or more non-transitory computer-readable storage media containing computer-executable instructions that, when executed by one or more processors, cause the processors to perform the steps of the image noise estimation method.
A computer program product comprising instructions which, when run on a computer, cause the computer to perform a method of image noise estimation.
Any reference to memory, storage, database, or other medium used herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM), which acts as external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. An image noise estimation method, characterized in that the method comprises:
acquiring a reference frame image and a target frame image aligned with the reference frame image;
dividing the reference frame image into a plurality of reference image blocks and dividing the target frame image into a plurality of target image blocks according to the same image division mode;
determining an energy value corresponding to each reference image block, and determining an image block difference value between the reference image block and a related image block; the related image block is a target image block with the same position as the reference image block;
dividing gray value intervals corresponding to pixel points in the reference frame image into preset number of sub-gray value intervals;
dividing the energy value and the image block difference value corresponding to each reference image block into the sub-gray value intervals;
determining a target gray value corresponding to each sub-gray value interval based on each energy value in each sub-gray value interval, and determining a target noise value corresponding to the target gray value based on each image block difference value in each sub-gray value interval;
and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value.
2. The method of claim 1, wherein said acquiring a reference frame image and a target frame image aligned with the reference frame image comprises:
acquiring multi-frame images acquired in the same shooting scene;
selecting a reference frame image and at least one target frame image from the multi-frame images;
and referring to the reference frame image, and carrying out image alignment operation on the target frame image to obtain a target frame image aligned with the reference frame image.
3. The method of claim 1, wherein determining the energy value corresponding to each reference image block and determining the image block difference values of the reference image block and the related image block comprises:
determining a root mean square of gray values corresponding to each pixel point in each reference image block, and taking the root mean square as an energy value corresponding to the reference image block;
and determining the sum of absolute values of differences between the gray values corresponding to the pixel points in the reference image block and the gray values corresponding to the pixel points in the related image block, and taking the sum of the absolute values of the differences as the image block difference value of the reference image block and the related image block.
4. The method according to claim 1, wherein the determining a target gray scale value corresponding to each sub-gray scale value interval based on each energy value in each sub-gray scale value interval, and determining a target noise value corresponding to the target gray scale value based on each image block difference value in the sub-gray scale value interval comprises:
taking the average value of each energy value in each sub-gray value interval as a target gray value corresponding to the sub-gray value interval;
eliminating image block difference values larger than a preset threshold corresponding to the sub-gray value intervals in the sub-gray value intervals;
and sorting the image block difference values which are not removed, and taking the image block difference value in the middle of the sorted sequence as a target noise value corresponding to the target gray value.
5. The method according to claim 1, wherein the determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value comprises:
determining a noise value corresponding to each gray value in the gray value range in a linear interpolation mode based on each target gray value and a target noise value corresponding to each target gray value;
taking the energy value corresponding to each reference image block in the reference frame image as a gray value corresponding to each reference image block in the reference frame image;
and determining the noise value corresponding to each reference image block in the reference frame image based on the gray value corresponding to each reference image block in the reference frame image and the noise value corresponding to each gray value in the gray value range.
6. The method of claim 1, further comprising:
determining fusion weights between each reference image block and related image blocks in the reference frame image based on the image block difference value and the noise value corresponding to each reference image block in the reference frame image;
and performing multi-frame fusion denoising processing on the reference frame image based on the fusion weight between each reference image block and the related image block in the reference frame image to obtain a denoised target image.
7. The method of any of claims 1 to 6, applied to the YUV domain, wherein the energy values comprise a Y-channel energy value, a U-channel energy value, and a V-channel energy value, and the tile difference values comprise a Y-channel tile difference value, a U-channel tile difference value, and a V-channel tile difference value; the determining an energy value corresponding to each reference image block includes:
for the Y channel, determining a Y channel energy value corresponding to each reference image block based on the Y value;
taking the Y-channel energy value as a U-channel energy value corresponding to a U channel, and taking the Y-channel energy value as a V-channel energy value corresponding to a V channel;
the determining of the image block difference values of the reference image block and the related image block includes:
for a Y channel, determining Y channel image block difference values of the reference image block and the related image block based on a Y value;
for a U channel, determining a U channel image block difference value of the reference image block and a related image block based on a U value;
and for the V channel, determining the V channel image block difference value of the reference image block and the related image block based on the V value.
8. An image noise estimation apparatus, comprising:
the device comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring a reference frame image and a target frame image aligned with the reference frame image;
the dividing module is used for dividing the reference frame image into a plurality of reference image blocks and dividing the target frame image into a plurality of target image blocks according to the same image dividing mode;
the determining module is used for determining an energy value corresponding to each reference image block and determining image block difference values of the reference image block and related image blocks; the related image block is a target image block with the same position as the reference image block;
the dividing module is further configured to divide a gray value dereferencing interval corresponding to a pixel point in the reference frame image into a preset number of sub-gray value intervals;
the classifying module is used for classifying the energy value and the image block difference value corresponding to each reference image block into the corresponding sub-gray value interval;
the determining module is configured to determine a target gray value corresponding to each sub-gray value interval based on each energy value in each sub-gray value interval, and determine a target noise value corresponding to the target gray value based on each image block difference value in the sub-gray value interval; and determining the noise value corresponding to each reference image block in the reference frame image based on each target gray value and the target noise value corresponding to each target gray value.
9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program, wherein the computer program, when executed by the processor, causes the processor to perform the steps of the method according to any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110924967.9A CN113674232A (en) | 2021-08-12 | 2021-08-12 | Image noise estimation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110924967.9A CN113674232A (en) | 2021-08-12 | 2021-08-12 | Image noise estimation method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113674232A true CN113674232A (en) | 2021-11-19 |
Family
ID=78542485
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110924967.9A Pending CN113674232A (en) | 2021-08-12 | 2021-08-12 | Image noise estimation method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113674232A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115941857A (en) * | 2022-12-30 | 2023-04-07 | 湖南大学 | Defogging circuit and method |
| CN117893781A (en) * | 2024-03-15 | 2024-04-16 | 深圳麦风科技有限公司 | Method, device and storage medium for extracting image difference region |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036471A (en) * | 2013-03-05 | 2014-09-10 | 腾讯科技（深圳）有限公司 | Image noise estimation method and image noise estimation device |
| CN107169939A (en) * | 2017-05-31 | 2017-09-15 | 广东欧珀移动通信有限公司 | Image processing method and related product |
| CN108898567A (en) * | 2018-09-20 | 2018-11-27 | 北京旷视科技有限公司 | Image denoising method, apparatus and system |
| CN110211085A (en) * | 2018-02-28 | 2019-09-06 | 清华大学 | A kind of Quality Measures for Image Fusion and system |
| CN110570445A (en) * | 2019-09-09 | 2019-12-13 | 上海联影医疗科技有限公司 | Image segmentation method, device, terminal and readable medium |
| CN111563517A (en) * | 2020-04-20 | 2020-08-21 | 腾讯科技（深圳）有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN111709894A (en) * | 2020-06-16 | 2020-09-25 | 展讯通信（上海）有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| WO2020207262A1 (en) * | 2019-04-09 | 2020-10-15 | Oppo广东移动通信有限公司 | Image processing method and apparatus based on multiple frames of images, and electronic device |
-
2021
- 2021-08-12 CN CN202110924967.9A patent/CN113674232A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036471A (en) * | 2013-03-05 | 2014-09-10 | 腾讯科技（深圳）有限公司 | Image noise estimation method and image noise estimation device |
| CN107169939A (en) * | 2017-05-31 | 2017-09-15 | 广东欧珀移动通信有限公司 | Image processing method and related product |
| CN110211085A (en) * | 2018-02-28 | 2019-09-06 | 清华大学 | A kind of Quality Measures for Image Fusion and system |
| CN108898567A (en) * | 2018-09-20 | 2018-11-27 | 北京旷视科技有限公司 | Image denoising method, apparatus and system |
| WO2020207262A1 (en) * | 2019-04-09 | 2020-10-15 | Oppo广东移动通信有限公司 | Image processing method and apparatus based on multiple frames of images, and electronic device |
| CN110570445A (en) * | 2019-09-09 | 2019-12-13 | 上海联影医疗科技有限公司 | Image segmentation method, device, terminal and readable medium |
| CN111563517A (en) * | 2020-04-20 | 2020-08-21 | 腾讯科技（深圳）有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN111709894A (en) * | 2020-06-16 | 2020-09-25 | 展讯通信（上海）有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| LEO-PENG: "图像降噪算法——时域降噪算法", pages 1 - 4, Retrieved from the Internet <URL:https://blog.csdn.net/weixin_44580210/article/details/108525305> * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115941857A (en) * | 2022-12-30 | 2023-04-07 | 湖南大学 | Defogging circuit and method |
| CN115941857B (en) * | 2022-12-30 | 2024-04-02 | 湖南大学 | Defogging circuit and method |
| CN117893781A (en) * | 2024-03-15 | 2024-04-16 | 深圳麦风科技有限公司 | Method, device and storage medium for extracting image difference region |
| CN117893781B (en) * | 2024-03-15 | 2024-05-07 | 深圳麦风科技有限公司 | Method, device and storage medium for extracting image difference region |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110660066B (en) | Training method of network, image processing method, network, terminal equipment and medium | |
| CN110110799B (en) | Cell sorting method, cell sorting device, computer equipment and storage medium | |
| US10521885B2 (en) | Image processing device and image processing method | |
| CN113674232A (en) | Image noise estimation method and device, electronic equipment and storage medium | |
| CN111161222B (en) | Printing roller defect detection method based on visual saliency | |
| CN113344910B (en) | Defect labeling image generation method and device, computer equipment and storage medium | |
| US20240005468A1 (en) | Image distortion evaluation method and apparatus, and computer device | |
| CN107622504B (en) | Method and device for processing pictures | |
| CN109214996B (en) | Image processing method and device | |
| JP2021531571A (en) | Certificate image extraction method and terminal equipment | |
| CN113160087B (en) | Image enhancement method, device, computer equipment and storage medium | |
| CN116883336A (en) | Image processing method, device, computer equipment and medium | |
| CN114764775A (en) | Infrared image quality evaluation method, device and storage medium | |
| CN113012061A (en) | Noise reduction processing method and device and electronic equipment | |
| CN111986116A (en) | Video noise reduction method and device, electronic equipment and computer readable storage medium | |
| CN111445487A (en) | Image segmentation method and device, computer equipment and storage medium | |
| CN109784379B (en) | Updating method and device of textile picture feature library | |
| CN109447942B (en) | Image ambiguity determining method, apparatus, computer device and storage medium | |
| CN113744294B (en) | Image processing method and related device | |
| CN111723634A (en) | Image detection method and device, electronic equipment and storage medium | |
| CN113837928A (en) | Object color adjusting method and device, electronic equipment and storage medium | |
| CN111681187B (en) | Method and device for reducing color noise, electronic equipment and readable storage medium | |
| CN111539975B (en) | Method, device, equipment and storage medium for detecting moving object | |
| CN113240042A (en) | Image classification preprocessing method, image classification preprocessing device, image classification equipment and storage medium | |
| CN114049288A (en) | Image generation method and device, electronic equipment and computer-readable storage medium |
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 |