CN116095508A - Image noise processing method and device, electronic equipment and storage medium - Google Patents
Image noise processing method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116095508A CN116095508A CN202310107590.7A CN202310107590A CN116095508A CN 116095508 A CN116095508 A CN 116095508A CN 202310107590 A CN202310107590 A CN 202310107590A CN 116095508 A CN116095508 A CN 116095508A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- matrix
- acquisition
- acquired
- 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
Images
Landscapes
- Image Processing (AREA)
Abstract
The application provides a processing method and device of image noise, electronic equipment and a storage medium, and a plurality of acquired images are acquired; determining an initial acquisition image data matrix based on the plurality of acquisition images; determining a plurality of target computing matrices corresponding to the initial acquired image data matrices; for each image locus in the initial acquisition image data matrix, iterating by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to the image locus according to the time dimension of image acquisition to obtain a target pixel value corresponding to the image locus; and replacing each image site data in the initial acquisition image data matrix with a corresponding target pixel value, and assigning the corresponding acquisition image to obtain a noise-reduced target image. Thus, the target image with the definition meeting the expectations can be obtained, and the accuracy of image definition processing is improved.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a method and apparatus for processing image noise, an electronic device, and a storage medium.
Background
With the wide use of video acquisition equipment, especially video monitoring equipment, images can be acquired by the video monitoring equipment for analysis in various application scenes, in order to ensure that the analysis is accurate, the images with the definition meeting the expectations are expected to be obtained, however, the definition processing mode of the images still is calculated through theoretical values at the present stage, the images with the definition meeting the expectations cannot be obtained more accurately, and how to obtain the images with the definition meeting the expectations accurately so as to better accurately analyze the images acquired by the video monitoring equipment becomes a problem to be solved urgently.
Disclosure of Invention
In view of this, an object of the present application is to provide a method, an apparatus, an electronic device, and a storage medium for processing image noise, which acquire a plurality of acquired images acquired by a video acquisition device, determine an initial acquired image data matrix of the images, and iterate pixel values on image sites in the images by using a target calculation matrix to obtain target pixel values on each image site in the acquired images, which approach to a desired real value, so as to obtain a target image with a definition that meets the desired, thereby improving accuracy of image definition processing.
In a first aspect, an embodiment of the present application provides a method for processing image noise, where the method includes:
acquiring a target video acquired by video acquisition equipment at a fixed angle, and determining a plurality of acquired images from the target video frame by frame;
determining an initial acquisition image data matrix based on the plurality of acquisition images; wherein the initial acquisition image data matrix comprises a plurality of image site data; the image locus data represents the acquired pixel value of a frame of two-dimensional image;
determining a plurality of target computing matrices corresponding to the initial acquired image data matrix; wherein, the elements in each target calculation matrix represent expected pixel values corresponding to the same pixel position in each frame in the acquired image;
for each image locus in the initial acquisition image data matrix, iterating by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to the image locus according to the time dimension of image acquisition to obtain a target pixel value corresponding to the image locus;
and replacing each image site data in the initial acquisition image data matrix with a corresponding target pixel value, and assigning the corresponding acquisition image to obtain a noise-reduced target image.
In a possible implementation manner, the parameters of the image location data comprise pixel positions of the image location data in the acquired image and image frames corresponding to the image location data; determining the initial acquisition image data matrix by:
for each acquired image, determining pixel values of each pixel position in the acquired image on different frames;
and collecting pixel values on different frames according to the located pixel positions, and arranging according to the corresponding pixel positions to generate the initial acquisition image data matrix.
In one possible implementation, each image site data in the initial acquisition image data matrix corresponds to a target calculation matrix; the determining a plurality of target computing matrices corresponding to the initial acquired image data matrix includes:
and for each image locus data in the initial acquisition image data matrix, carrying out assignment initialization on each element in a pre-constructed calculation matrix based on the acquisition pixel value at the pixel position of the first frame corresponding to the image locus data to obtain a target calculation matrix corresponding to the image locus data.
In one possible implementation manner, for each image location in the initial acquired image data matrix, the performing iteration according to the time dimension of image acquisition by using a plurality of expected pixel values included in a target calculation matrix corresponding to the image location to obtain a target pixel value corresponding to the image location includes:
for each image locus in the initial acquisition image data matrix, assigning the acquisition pixel value corresponding to each frame of the image locus to a target element in the target calculation matrix;
for each image locus in the initial acquisition image data matrix, after the target element in the target calculation matrix is assigned, sorting the elements according to the pixel values corresponding to the elements in the target calculation matrix to obtain the assigned and sorted target calculation matrix;
for each image position point in the initial acquisition image data matrix, after the acquisition pixel values of all frames corresponding to the image position point are assigned to the target calculation matrix, determining a target median value of pixel values corresponding to all elements included in the target calculation matrix after assignment sequencing, and determining the target median value as a target pixel value corresponding to the image position point data.
In one possible implementation, the target elements in the target calculation matrix are determined by:
if the currently assigned image frame is an odd number, determining the element currently ordered at the first position in the target calculation matrix as the target element;
and if the currently assigned image frame is even, determining the element currently ordered in the last bit in the target calculation matrix as the target element.
In a second aspect, an embodiment of the present application further provides a processing apparatus for image noise, where the processing apparatus includes:
the image acquisition module is used for acquiring target videos acquired by the video acquisition equipment at a fixed angle and determining a plurality of acquired images from the target videos frame by frame;
an image data matrix determining module for determining an initial acquisition image data matrix based on the plurality of acquisition images; wherein the initial acquisition image data matrix comprises a plurality of image site data; the image locus data represents the acquired pixel value of a frame of two-dimensional image;
a computing matrix determining module for determining a plurality of target computing matrices corresponding to the initial acquired image data matrix; wherein, the elements in each target calculation matrix represent expected pixel values corresponding to the same pixel position in each frame in the acquired image;
The image iteration module is used for carrying out iteration according to the time dimension of image acquisition by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to each image locus aiming at each image locus in the initial acquisition image data matrix to obtain a target pixel value corresponding to the image locus;
and the image processing module is used for replacing each image locus data in the initial acquisition image data matrix with a corresponding target pixel value and then assigning the corresponding acquisition image to obtain a noise-reduced target image.
In a possible implementation manner, the parameters of the image location data comprise pixel positions of the image location data in the acquired image and image frames corresponding to the image location data; the image data matrix determination module is configured to determine the initial acquisition image data matrix by:
for each acquired image, determining pixel values of each pixel position in the acquired image on different frames;
and collecting pixel values on different frames according to the located pixel positions, and arranging according to the corresponding pixel positions to generate the initial acquisition image data matrix.
In one possible implementation, each image site data in the initial acquisition image data matrix corresponds to a target calculation matrix; the computing matrix determining module is used for determining a plurality of target computing matrices corresponding to the initial acquired image data matrix, and the computing matrix determining module is used for:
and for each image locus data in the initial acquisition image data matrix, carrying out assignment initialization on each element in a pre-constructed calculation matrix based on the acquisition pixel value at the pixel position of the first frame corresponding to the image locus data to obtain a target calculation matrix corresponding to the image locus data.
In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating over the bus when the electronic device is running, the processor executing the machine-readable instructions to perform the steps of the method of processing image noise as claimed in any one of the first aspects.
In a fourth aspect, embodiments of the present application further provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method for processing image noise according to any of the first aspects.
According to the image noise processing method, the image noise processing device, the electronic equipment and the storage medium, target videos acquired by the video acquisition equipment at a fixed angle are acquired, and a plurality of acquired images are determined from the target videos frame by frame; determining an initial acquisition image data matrix based on the plurality of acquisition images; determining a plurality of target computing matrices corresponding to the initial acquired image data matrices; for each image locus in the initial acquisition image data matrix, iterating by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to the image locus according to the time dimension of image acquisition to obtain a target pixel value corresponding to the image locus; and replacing each image site data in the initial acquisition image data matrix with a corresponding target pixel value, and assigning the corresponding acquisition image to obtain a noise-reduced target image. In this way, after a plurality of acquired images acquired by the video acquisition equipment are acquired, an initial acquired image data matrix of the images is determined, the pixel values on the image sites in the images are iterated by utilizing the target calculation matrix, so that target pixel values which are close to expected real values on all the image sites in the acquired images are obtained, further, the target images with the definition meeting the expected are obtained, and the accuracy of image definition processing is improved.
In order to make the above objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting the scope, and that other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for processing image noise according to an embodiment of the present application;
FIG. 2 is a schematic diagram of an initial acquisition image matrix provided in an embodiment of the present application;
FIG. 3 is a schematic diagram of a target computing matrix according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of an image noise processing apparatus according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. Based on the embodiments of the present application, every other embodiment that a person skilled in the art would obtain without making any inventive effort is within the scope of protection of the present application.
First, application scenarios applicable to the present application will be described. The method and the device can be applied to the technical field of image processing.
With the wide use of video acquisition equipment, especially video monitoring equipment, images can be acquired by the video monitoring equipment for analysis in various application scenes, in order to ensure that the analysis is accurate, the images with the definition meeting the expectations are expected to be obtained, however, the definition processing mode of the images still is calculated through theoretical values at the present stage, the images with the definition meeting the expectations cannot be obtained more accurately, and how to obtain the images with the definition meeting the expectations accurately so as to better accurately analyze the images acquired by the video monitoring equipment becomes a problem to be solved urgently.
Based on the above, the embodiment of the application provides a processing method of image noise, so as to improve accuracy of image definition processing.
Referring to fig. 1, fig. 1 is a flowchart of a method for processing image noise according to an embodiment of the present application. As shown in fig. 1, the method for processing image noise provided in the embodiment of the present application includes:
s101, acquiring target videos acquired by video acquisition equipment at a fixed angle, and determining a plurality of acquired images from the target videos frame by frame.
S102, determining an initial acquisition image data matrix based on the plurality of acquisition images; wherein the initial acquisition image data matrix comprises a plurality of image site data; the image locus data characterizes the acquired pixel values of a frame of two-dimensional image.
S103, determining a plurality of target calculation matrixes corresponding to the initial acquisition image data matrixes; wherein, the elements in each target calculation matrix represent expected pixel values corresponding to the same pixel position in each frame in the acquired image.
S104, for each image site in the initial acquisition image data matrix, iterating by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to the image site according to the time dimension of image acquisition to obtain a target pixel value corresponding to the image site.
S105, replacing each image site data in the initial acquisition image data matrix with a corresponding target pixel value, and assigning the corresponding acquisition image to obtain a noise-reduced target image.
According to the image noise processing method, after the plurality of acquired images acquired by the video acquisition equipment are acquired, the initial acquired image data matrix of the images is determined, the pixel values on the image sites in the images are iterated by using the target calculation matrix, so that target pixel values which are close to expected true values on all the image sites in the acquired images are obtained, further, the target images with the definition meeting the expected are obtained, and the accuracy of image definition processing is improved.
Exemplary steps of embodiments of the present application are described below:
s101, acquiring target videos acquired by video acquisition equipment at a fixed angle, and determining a plurality of acquired images from the target videos frame by frame.
In the embodiment of the application, the acquired target video acquired by the video acquisition device is the target video shot by the video acquisition device at a fixed angle, and after the acquired target video is acquired, the target video is intercepted frame by frame to obtain a plurality of acquired images in order to ensure the consistency of image processing.
For example, the video capturing device may be a monitoring device, and the purpose of the embodiments of the present application is to remove noise in a video obtained by the monitoring device, so as to obtain an image and a video with sharpness meeting expectations.
In one possible implementation manner, a plurality of acquired images may be taken from the target video frame by frame, or may be taken from the target video at a preset frame interval, and the specific way of taking the acquired images may be determined according to the video processing requirement, which is not limited herein.
S102, determining an initial acquisition image data matrix based on the plurality of acquisition images; wherein the initial acquisition image data matrix comprises a plurality of image site data; the image locus data characterizes the acquired pixel values of a frame of two-dimensional image.
In the embodiment of the application, after a plurality of acquired images are acquired, a corresponding initial acquired image data matrix can be determined according to pixel values of each image on the same image site, that is, the initial acquired image data matrix includes image site data; the image locus data characterizes the acquired pixel values of a frame of two-dimensional image.
Notably, the initial acquisition image data matrix is a spatial matrix of images.
Specifically, the parameters of the image locus data include the pixel position of the image locus data in the acquired image and the image frame corresponding to the image locus data, and the initial acquired image data matrix is determined by the following steps:
a1: for each acquired image, pixel values for respective pixel locations in the acquired image over different frames are determined.
a2: and collecting pixel values on different frames according to the located pixel positions, and arranging according to the corresponding pixel positions to generate the initial acquisition image data matrix.
In the embodiment of the application, for each acquired image, determining pixel values of each pixel position in the acquired image on different frames; and according to the determined pixel values on different frames at the same pixel position, the pixel values are gathered and arranged according to the corresponding pixel position, and a corresponding initial acquisition image data matrix is generated.
Specifically, referring to fig. 2, fig. 2 is a schematic diagram of an initial acquisition image matrix provided in an embodiment of the present application, in which a base unit is an acquisition value d of a two-dimensional image x,y,t Wherein x is the x-th row data of one frame of collected image data, y is the y-th column data of one frame of collected image data, and t is the t-th frame of collected image data; d, d x,y,t Image locus data (pixel value) representing the x-th row and y-th column in the t-th frame image acquisition data, and image locus data d x,y,t Obeying N (u) x,y,t ,σ 2 ) Wherein u is a gaussian distribution of x,y,t Number of image sitesTheoretical median (physical true value) of the corresponding gaussian distribution; sigma is white noise data interference subject to gaussian distribution, and it can be known that the image locus data d is targeted x,y,t For d x,y,t The mathematical modeling formula of (c) may be represented by the following formula:
d x,y,t ＝u x,y,t +σ；
wherein d x,y,t Representing image locus data (pixel values) of an x-th row and y-th column in t-th frame image acquisition data; u (u) x,y,t Theoretical median (physical true value) of corresponding gaussian distribution of image locus data; σ is white noise data interference that follows a gaussian distribution.
S103, determining a plurality of target calculation matrixes corresponding to the initial acquisition image data matrixes; wherein, the elements in each target calculation matrix represent expected pixel values corresponding to the same pixel position in each frame in the acquired image.
In the embodiment of the application, after determining the corresponding space matrix initial acquisition image data matrix, determining a plurality of target calculation matrices according to each image data position in the initial acquisition image data matrix; and gradually approaching the pixel value in the space matrix initial acquisition image data matrix to the expected value through iterative calculation of the target calculation matrix.
Wherein, the elements in each target calculation matrix represent expected pixel values corresponding to the same pixel position in each frame in the acquired image.
Specifically, each image site data in the initial acquisition image data matrix corresponds to one target calculation matrix, and the step of determining a plurality of target calculation matrices corresponding to the initial acquisition image data matrix includes:
b1: and for each image locus data in the initial acquisition image data matrix, carrying out assignment initialization on each element in a pre-constructed calculation matrix based on the acquisition pixel value at the pixel position of the first frame corresponding to the image locus data to obtain a target calculation matrix corresponding to the image locus data.
In the embodiment of the application, for each image locus data in the initial acquisition image data matrix, according to the acquisition pixel value at the pixel position of the first frame corresponding to the image locus data of the image, each element in the pre-constructed calculation matrix is assigned and initialized, so that a target calculation matrix corresponding to the initialized image locus data is obtained.
In a possible implementation, referring to fig. 3, fig. 3 is a schematic diagram of a target computing matrix provided in the embodiment of the present application, where the target computing matrix is a 3-dimensional matrix of two dimensions of a picture (x, y) plus one computing dimension (k), and the computing unit is
Where x is the x-th line expected data of the image acquisition data, y is the y-th column expected data of the image acquisition data, and k is the k-th calculation dimensional expected data of the image acquisition.
Specifically, the size of the calculation dimension of the target calculation matrix is (2k+1), k is a positive integer determined according to the historical experimental law, and can be generally 10;
representing the desired pixel value of the X-th row and Y-th column image site (pixel location) in the kth computational dimension.
It should be noted that, when the target computing matrix is initialized by the collected pixel value at the pixel position of the first frame, the collected pixel value at the pixel position of the first frame is assigned to each element included in the target computing matrix, and all the elements included in the target computing matrix are assigned to obtain the initialized target computing matrix.
S104, for each image site in the initial acquisition image data matrix, iterating by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to the image site according to the time dimension of image acquisition to obtain a target pixel value corresponding to the image site.
In the embodiment of the present application, after initializing the target calculation matrix, iterative calculation is required according to the target calculation matrix, so as to obtain a target pixel value closer to the expected value.
Specifically, the step of "iterating, for each image location in the initial acquired image data matrix, using a plurality of expected pixel values included in a target calculation matrix corresponding to the image location according to a time dimension of image acquisition to obtain a target pixel value corresponding to the image location" includes:
c1: for each image locus in the initial acquisition image data matrix, assigning the acquisition pixel value corresponding to each frame of the image locus to a target element in the target calculation matrix.
c2: and aiming at each image position point in the initial acquisition image data matrix, after the target element in the target calculation matrix is assigned, sorting the elements according to the pixel values corresponding to the elements in the target calculation matrix to obtain the assigned and sorted target calculation matrix.
c3: for each image position point in the initial acquisition image data matrix, after the acquisition pixel values of all frames corresponding to the image position point are assigned to the target calculation matrix, determining a target median value of pixel values corresponding to all elements included in the target calculation matrix after assignment sequencing, and determining the target median value as a target pixel value corresponding to the image position point data.
In this embodiment, for each image location in the initial acquired image data matrix, the acquired pixel value corresponding to each frame of the image location is assigned to a target element in the target calculation matrix.
In one possible implementation, when assigning the target computing matrix, corresponding pixel values need to be assigned to different elements of the target computing matrix in different image frames according to the target frame.
Specifically, the target elements in the target calculation matrix are determined by:
d1: and if the currently assigned image frame is odd, determining the element currently ordered at the first position in the target calculation matrix as the target element.
d2: and if the currently assigned image frame is even, determining the element currently ordered in the last bit in the target calculation matrix as the target element.
In this embodiment of the present application, if the currently assigned image frame is odd (the frame number t acquired by the device is odd), the image site data d of the xth row and y column in the device acquired data of the tth frame is determined x,y,t Assigning the first data in the computation dimension of the computation sequence corresponding to the target computation matrix
If the currently assigned image frame is even (the frame number t acquired by the equipment is even), the image locus data d of the xth row and the y column in the equipment acquisition data of the tth frame is acquired x,y,t Assignment of (2K+1) th data in the calculation dimension of the calculation sequence corresponding to the target calculation matrix +.>
Further, after the corresponding acquired pixel values are assigned to the target elements, for each image position in the initial acquired image data matrix, after the target elements in the target calculation matrix are assigned, sorting the elements according to the pixel values corresponding to the elements in the target calculation matrix, so as to obtain the target calculation matrix with assigned sorting.
Specifically, if the currently assigned image frame is odd (the frame number t acquired by the device is odd), the image site data d of the xth row and the y column in the device acquisition data of the tth frame is acquired x,y,t Assigning the first data in the computation dimension of the computation sequence corresponding to the target computation matrix
Then, the 1 st data from the calculation dimension is performed in the calculation sequence +.>
To the first(2K+1) data->
Is a one-way bubbling ordering of (2).
If the currently assigned image frame is even (the frame number t acquired by the equipment is even), the image locus data d of the xth row and the y column in the equipment acquisition data of the tth frame is acquired x,y,t Assignment of (2K+1) th data in the computation dimension of the computation sequence corresponding to the target computation matrix 
After that, the calculation of the (2K+1) th data from the calculation dimension is then performed in the calculation sequence>
Data 1->
Is a one-way bubbling ordering of (2).
In one possible implementation, the purpose of the one-way bubble ordering in the iterative computation process is to: to maintain monotonic incrementing of the computed series
The expected median data of the number sequence is calculated, and the interference problem of the extremum data can be eliminated by using the logic replaced by the first polling, so that the iteration efficiency is further improved; meanwhile, a unidirectional bubbling algorithm with the algorithm complexity of O (n) is adopted, so that the scheduling resources of a processor are greatly saved.
Further, after the acquired pixel values of all frames corresponding to each image locus in the initial acquired image data matrix are assigned to the target calculation matrix for each image locus, determining the target median value of the pixel values corresponding to the elements included in the target calculation matrix after the assignment ordering, and gradually approaching the expected median value in the target calculation matrix to the theoretical median value (true value) through the iterative process, so that the calculated median value data can be used
Restoring image site data d in a real image x,y,t， To obtain an image with higher definition having a higher expected value.
Specifically, the target median value may be calculated by the following formula:
wherein u is x,y,t For image site d x,y,t τ is the theoretical median and the calculated series of expected median data
Is the theoretical median and the calculated series of expected median data +.>
Is a variance value of (a).
S105, replacing each image site data in the initial acquisition image data matrix with a corresponding target pixel value, and assigning the corresponding acquisition image to obtain a noise-reduced target image.
In the embodiment of the application, after the target pixel value approaching the true value is determined through the target calculation matrix, the pixel value of each image site of the acquired acquisition image is replaced, and the processed target image with the definition meeting the requirement is obtained.
According to the image noise processing method, target videos acquired by video acquisition equipment at a fixed angle are acquired, and a plurality of acquired images are determined from the target videos frame by frame; determining an initial acquisition image data matrix based on the plurality of acquisition images; determining a plurality of target computing matrices corresponding to the initial acquired image data matrices; for each image locus in the initial acquisition image data matrix, iterating by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to the image locus according to the time dimension of image acquisition to obtain a target pixel value corresponding to the image locus; and replacing each image site data in the initial acquisition image data matrix with a corresponding target pixel value, and assigning the corresponding acquisition image to obtain a noise-reduced target image. In this way, after a plurality of acquired images acquired by the video acquisition equipment are acquired, an initial acquired image data matrix of the images is determined, the pixel values on the image sites in the images are iterated by utilizing the target calculation matrix, so that target pixel values which are close to expected real values on all the image sites in the acquired images are obtained, further, the target images with the definition meeting the expected are obtained, and the accuracy of image definition processing is improved.
Referring to fig. 4, fig. 4 is a schematic structural diagram of an apparatus for processing image noise according to an embodiment of the present application, as shown in fig. 4, the processing apparatus 400 includes:
an image acquisition module 410, configured to acquire a target video acquired by a video acquisition device at a fixed angle, and determine a plurality of acquired images from the target video frame by frame;
an image data matrix determination module 420 for determining an initial acquisition image data matrix based on the plurality of acquisition images; wherein the initial acquisition image data matrix comprises a plurality of image site data; the image locus data represents the acquired pixel value of a frame of two-dimensional image;
a calculation matrix determining module 430, configured to determine a plurality of target calculation matrices corresponding to the initial acquired image data matrix; wherein, the elements in each target calculation matrix represent expected pixel values corresponding to the same pixel position in each frame in the acquired image;
an image iteration module 440, configured to iterate, for each image location in the initial acquired image data matrix, according to a time dimension of image acquisition by using a plurality of expected pixel values included in a target calculation matrix corresponding to the image location, to obtain a target pixel value corresponding to the image location;
The image processing module 450 is configured to replace each image location data in the initial acquired image data matrix with a corresponding target pixel value and assign the corresponding acquired image to obtain a noise-reduced target image.
In a possible implementation manner, the parameters of the image location data comprise pixel positions of the image location data in the acquired image and image frames corresponding to the image location data; the image data matrix determination module 420 is configured to determine the initial acquisition image data matrix by:
for each acquired image, determining pixel values of each pixel position in the acquired image on different frames;
and collecting pixel values on different frames according to the located pixel positions, and arranging according to the corresponding pixel positions to generate the initial acquisition image data matrix.
In one possible implementation, each image site data in the initial acquisition image data matrix corresponds to a target calculation matrix; the calculation matrix determining module 430, when configured to determine a plurality of target calculation matrices corresponding to the initial acquired image data matrix, the calculation matrix determining module 430 is configured to:
And for each image locus data in the initial acquisition image data matrix, carrying out assignment initialization on each element in a pre-constructed calculation matrix based on the acquisition pixel value at the pixel position of the first frame corresponding to the image locus data to obtain a target calculation matrix corresponding to the image locus data.
In one possible implementation manner, when the computing matrix determining module 430 is configured to, for each image location in the initial acquired image data matrix, iterate, using a plurality of expected pixel values included in a target computing matrix corresponding to the image location, according to a time dimension of image acquisition, to obtain a target pixel value corresponding to the image location, the computing matrix determining module 430 is configured to:
for each image locus in the initial acquisition image data matrix, assigning the acquisition pixel value corresponding to each frame of the image locus to a target element in the target calculation matrix;
for each image locus in the initial acquisition image data matrix, after the target element in the target calculation matrix is assigned, sorting the elements according to the pixel values corresponding to the elements in the target calculation matrix to obtain the assigned and sorted target calculation matrix;
For each image position point in the initial acquisition image data matrix, after the acquisition pixel values of all frames corresponding to the image position point are assigned to the target calculation matrix, determining a target median value of pixel values corresponding to all elements included in the target calculation matrix after assignment sequencing, and determining the target median value as a target pixel value corresponding to the image position point data.
In one possible implementation, the computing matrix determining module 430 is configured to determine the target element in the target computing matrix by:
if the currently assigned image frame is an odd number, determining the element currently ordered at the first position in the target calculation matrix as the target element;
and if the currently assigned image frame is even, determining the element currently ordered in the last bit in the target calculation matrix as the target element.
According to the image noise processing device, target videos acquired by video acquisition equipment at a fixed angle are acquired, and a plurality of acquired images are determined from the target videos frame by frame; determining an initial acquisition image data matrix based on the plurality of acquisition images; determining a plurality of target computing matrices corresponding to the initial acquired image data matrices; for each image locus in the initial acquisition image data matrix, iterating by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to the image locus according to the time dimension of image acquisition to obtain a target pixel value corresponding to the image locus; and replacing each image site data in the initial acquisition image data matrix with a corresponding target pixel value, and assigning the corresponding acquisition image to obtain a noise-reduced target image. In this way, after a plurality of acquired images acquired by the video acquisition equipment are acquired, an initial acquired image data matrix of the images is determined, the pixel values on the image sites in the images are iterated by utilizing the target calculation matrix, so that target pixel values which are close to expected real values on all the image sites in the acquired images are obtained, further, the target images with the definition meeting the expected are obtained, and the accuracy of image definition processing is improved.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 5, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.
The memory 520 stores machine-readable instructions executable by the processor 510, and when the electronic device 500 is running, the processor 510 communicates with the memory 520 through the bus 530, and when the machine-readable instructions are executed by the processor 510, the steps of the method for processing image noise in the method embodiment shown in fig. 1 can be executed, and the specific implementation is referred to the method embodiment and will not be described herein.
The embodiment of the present application further provides a computer readable storage medium, where a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of the method for processing image noise in the method embodiment shown in fig. 1 may be executed, and a specific implementation manner may refer to the method embodiment and will not be described herein.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Finally, it should be noted that: the foregoing examples are merely specific embodiments of the present application, and are not intended to limit the scope of the present application, but the present application is not limited thereto, and those skilled in the art will appreciate that while the foregoing examples are described in detail, the present application is not limited thereto. Any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or make equivalent substitutions for some of the technical features within the technical scope of the disclosure of the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method of processing image noise, the method comprising:
acquiring a target video acquired by video acquisition equipment at a fixed angle, and determining a plurality of acquired images from the target video frame by frame;
determining an initial acquisition image data matrix based on the plurality of acquisition images; wherein the initial acquisition image data matrix comprises a plurality of image site data; the image locus data represents the acquired pixel value of a frame of two-dimensional image;
determining a plurality of target computing matrices corresponding to the initial acquired image data matrix; wherein, the elements in each target calculation matrix represent expected pixel values corresponding to the same pixel position in each frame in the acquired image;
for each image locus in the initial acquisition image data matrix, iterating by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to the image locus according to the time dimension of image acquisition to obtain a target pixel value corresponding to the image locus;
and replacing each image site data in the initial acquisition image data matrix with a corresponding target pixel value, and assigning the corresponding acquisition image to obtain a noise-reduced target image.
2. The processing method according to claim 1, wherein the parameters of the image location data include pixel positions of the image location data in the acquired image and image frames corresponding to the image location data; determining the initial acquisition image data matrix by:
for each acquired image, determining pixel values of each pixel position in the acquired image on different frames;
and collecting pixel values on different frames according to the located pixel positions, and arranging according to the corresponding pixel positions to generate the initial acquisition image data matrix.
3. The processing method according to claim 1, wherein each image site data in the initial acquisition image data matrix corresponds to a target calculation matrix; the determining a plurality of target computing matrices corresponding to the initial acquired image data matrix includes:
and for each image locus data in the initial acquisition image data matrix, carrying out assignment initialization on each element in a pre-constructed calculation matrix based on the acquisition pixel value at the pixel position of the first frame corresponding to the image locus data to obtain a target calculation matrix corresponding to the image locus data.
4. A processing method according to claim 3, wherein for each image point in the initial acquisition image data matrix, iterating, by using a plurality of expected pixel values included in a target calculation matrix corresponding to the image point, according to a time dimension of image acquisition to obtain a target pixel value corresponding to the image point, including:
for each image locus in the initial acquisition image data matrix, assigning the acquisition pixel value corresponding to each frame of the image locus to a target element in the target calculation matrix;
for each image locus in the initial acquisition image data matrix, after the target element in the target calculation matrix is assigned, sorting the elements according to the pixel values corresponding to the elements in the target calculation matrix to obtain the assigned and sorted target calculation matrix;
for each image position point in the initial acquisition image data matrix, after the acquisition pixel values of all frames corresponding to the image position point are assigned to the target calculation matrix, determining a target median value of pixel values corresponding to all elements included in the target calculation matrix after assignment sequencing, and determining the target median value as a target pixel value corresponding to the image position point data.
5. The processing method according to claim 4, wherein the target elements in the target calculation matrix are determined by:
if the currently assigned image frame is an odd number, determining the element currently ordered at the first position in the target calculation matrix as the target element;
and if the currently assigned image frame is even, determining the element currently ordered in the last bit in the target calculation matrix as the target element.
6. An image noise processing apparatus, characterized in that the processing apparatus comprises:
the image acquisition module is used for acquiring target videos acquired by the video acquisition equipment at a fixed angle and determining a plurality of acquired images from the target videos frame by frame;
an image data matrix determining module for determining an initial acquisition image data matrix based on the plurality of acquisition images; wherein the initial acquisition image data matrix comprises a plurality of image site data; the image locus data represents the acquired pixel value of a frame of two-dimensional image;
a computing matrix determining module for determining a plurality of target computing matrices corresponding to the initial acquired image data matrix; wherein, the elements in each target calculation matrix represent expected pixel values corresponding to the same pixel position in each frame in the acquired image;
The image iteration module is used for carrying out iteration according to the time dimension of image acquisition by utilizing a plurality of expected pixel values included in a target calculation matrix corresponding to each image locus aiming at each image locus in the initial acquisition image data matrix to obtain a target pixel value corresponding to the image locus;
and the image processing module is used for replacing each image locus data in the initial acquisition image data matrix with a corresponding target pixel value and then assigning the corresponding acquisition image to obtain a noise-reduced target image.
7. The processing device of claim 6, wherein the parameters of the image location data include pixel locations of the image location data in the acquired image and image frames corresponding to the image location data; the image data matrix determination module is configured to determine the initial acquisition image data matrix by:
for each acquired image, determining pixel values of each pixel position in the acquired image on different frames;
and collecting pixel values on different frames according to the located pixel positions, and arranging according to the corresponding pixel positions to generate the initial acquisition image data matrix.
8. The processing device of claim 6, wherein each image site data in the initial acquisition image data matrix corresponds to a target calculation matrix; the computing matrix determining module is used for determining a plurality of target computing matrices corresponding to the initial acquired image data matrix, and the computing matrix determining module is used for:
and for each image locus data in the initial acquisition image data matrix, carrying out assignment initialization on each element in a pre-constructed calculation matrix based on the acquisition pixel value at the pixel position of the first frame corresponding to the image locus data to obtain a target calculation matrix corresponding to the image locus data.
9. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating over the bus when the electronic device is running, the processor executing the machine-readable instructions to perform the steps of the method of processing image noise as claimed in any one of claims 1 to 5.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, performs the steps of the image noise processing method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310107590.7A CN116095508A (en) | 2023-02-01 | 2023-02-01 | Image noise processing method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310107590.7A CN116095508A (en) | 2023-02-01 | 2023-02-01 | Image noise processing method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116095508A true CN116095508A (en) | 2023-05-09 |
Family
ID=86186804
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310107590.7A Pending CN116095508A (en) | 2023-02-01 | 2023-02-01 | Image noise processing method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116095508A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117745727A (en) * | 2024-02-21 | 2024-03-22 | 北京科技大学 | Device and method for monitoring hardness of water stemming liquid filling bag |
-
2023
- 2023-02-01 CN CN202310107590.7A patent/CN116095508A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117745727A (en) * | 2024-02-21 | 2024-03-22 | 北京科技大学 | Device and method for monitoring hardness of water stemming liquid filling bag |
| CN117745727B (en) * | 2024-02-21 | 2024-04-26 | 北京科技大学 | Device and method for monitoring hardness of water stemming liquid filling bag |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tsai et al. | Banet: a blur-aware attention network for dynamic scene deblurring | |
| Chen et al. | Blind image deblurring with local maximum gradient prior | |
| Kruse et al. | Learning to push the limits of efficient FFT-based image deconvolution | |
| Schuler et al. | Learning to deblur | |
| Whyte et al. | Deblurring shaken and partially saturated images | |
| Nan et al. | Deep learning for handling kernel/model uncertainty in image deconvolution | |
| CN109859314B (en) | Three-dimensional reconstruction method, three-dimensional reconstruction device, electronic equipment and storage medium | |
| CN112328715B (en) | Visual positioning method, training method of related model, related device and equipment | |
| WO2014070489A1 (en) | Recursive conditional means image denoising | |
| CN116095508A (en) | Image noise processing method and device, electronic equipment and storage medium | |
| CN113362338B (en) | Rail segmentation method, device, computer equipment and rail segmentation processing system | |
| CN103426190A (en) | Image reconstruction method and system | |
| CN112381742A (en) | Single image motion blur removing method and system | |
| CN111275625B (en) | Image deblurring method and device and electronic equipment | |
| CN114757984A (en) | Scene depth estimation method and device of light field camera | |
| Li et al. | Self-supervised blind motion deblurring with deep expectation maximization | |
| Khan et al. | Arbitrarily shaped point spread function (PSF) estimation for single image blind deblurring | |
| WO2013165595A1 (en) | Point spread function cost function with non-uniform weights | |
| Vo | Attention! stay focus! | |
| Carbajal et al. | Single image non-uniform blur kernel estimation via adaptive basis decomposition | |
| CN112837389B (en) | Object model construction method and device and electronic equipment | |
| Xu et al. | Sharp image estimation from a depth-involved motion-blurred image | |
| CN113627324A (en) | Matching method and device of face images, storage medium and electronic device | |
| Conti et al. | A regularized deep learning approach for image de-blurring | |
| WO2019163244A1 (en) | Image processing method, image processing device, and image processing program |
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 |