CN109697699A - Image processing method, device, equipment and storage medium - Google Patents
Image processing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN109697699A CN109697699A CN201711002231.6A CN201711002231A CN109697699A CN 109697699 A CN109697699 A CN 109697699A CN 201711002231 A CN201711002231 A CN 201711002231A CN 109697699 A CN109697699 A CN 109697699A
- Authority
- CN
- China
- Prior art keywords
- reference image
- pixel
- image block
- target
- block
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 20
- 238000012545 processing Methods 0.000 claims description 28
- 238000000034 method Methods 0.000 claims description 27
- 239000013598 vector Substances 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 5
- 241000208340 Araliaceae Species 0.000 claims 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 2
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 2
- 235000008434 ginseng Nutrition 0.000 claims 2
- 230000000694 effects Effects 0.000 abstract description 7
- 238000005303 weighing Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000005259 measurement Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000003252 repetitive effect Effects 0.000 description 2
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
Abstract
This application provides a kind of image processing method, device, equipment and storage medium, image processing method comprises determining that the reference image block Ni centered on reference image vegetarian refreshments i；It determines the target pixel points j in the contiguous range of the reference image vegetarian refreshments i, and determines the target image block Nj centered on the target pixel points j；Determine the similitude between the reference image block Ni and the target image block Nj；Weighing factor value when determining that the target pixel points j denoises the reference image vegetarian refreshments i according to the similitude；According to the weighted value of the reference image vegetarian refreshments i, the reference image vegetarian refreshments i is denoised.Therefore, image processing method provided by the present application can obtain preferably denoising effect.
Description
Technical field
This application involves field field of image processings, and more particularly, to a kind of image processing method, device, set
Standby and storage medium.
Background technique
Image suffers from the interference of noise signal in acquisition, transmission and recording process, such as Gaussian noise, impulsive noise
Deng.The presence of noise can seriously affect the visual effect of image and reduce the interpretation capability to image object information, image denoising
Task is handled as low layer pictures, has been widely studied and has proposed many effective denoising methods.It is flat from the neighborhood of early stage
The denoising method of the rarefaction representation image based on transform domains such as wavelet transformations and based on excessively complete dictionary finally is denoised, then
To the method for local auto-adaptive.These methods are all partial approaches, and so-called partial approach is exactly in neighborhood where using current pixel
All pixels go to estimate the pixel to realize denoising, therefore its estimated value will receive the influence of other pixels neighborhood Nei.Denoising
Process includes following two step: 1, finding pixel similar with pixel to be processed in the picture.2, more multiphase is found as far as possible
Pixel like pixel, and after being denoised using them.
In the content of the prior art, pixel similar with pixel to be processed is found in the picture, due to noise
It influences, so that the similarity of the pixel found and pixel to be handled is affected；More similar pixel points are found as far as possible,
And the pixel after being denoised using them, due to using the local neighborhood of pixel to be handled, so that the similar picture found
The quantity of vegetarian refreshments can be impacted, and then influences denoising effect.
Summary of the invention
The application provides a kind of image processing method, device, equipment and storage medium, can obtain preferably denoising effect
Fruit.
In a first aspect, providing a kind of image processing method, comprising: determine centered on reference image vegetarian refreshments i with reference to figure
As block Ni；Determine the target pixel points j in the contiguous range of the reference image vegetarian refreshments i, and centered on the target pixel points j
Determine the target image block Nj；Determine the similitude between the reference image block Ni and the target image block Nj；According to
The similitude determines the weighted value influenced when the target pixel points j denoises the reference image vegetarian refreshments i；According to the reference
The weighted value of pixel i denoises the reference image vegetarian refreshments i.
With reference to first aspect, in the first possible implementation of the first aspect, the determination reference picture
Similitude between block Ni and the target image block Nj, comprising: according to the reference image block Ni and the target image block
Euclidean distance between Nj determines the similitude between the reference image block Ni and the target image block Nj, and using following
Formula calculates the Euclidean distance between the reference image block Ni and the target image block Nj:Wherein, u (Ni) is the gray value vectors of the pixel in the reference image block Ni, u
It (Nj) is the gray value vectors of the pixel in the target reference block Nj, Dist (i, j) indicates the reference image block Ni and institute
State the Euclidean distance between target image block Nj.
With reference to first aspect and its above-mentioned implementation, in the second possible implementation of the first aspect, described
The weighted value influenced when determining that the target pixel points j denoises the reference image vegetarian refreshments i according to the similitude, comprising: make
The weighted value is determined with following formula:Wherein, w (i, j) expression is denoised to pixel i
When the weighted value contributed of pixel j, h is similarity weight weight parameter.
With reference to first aspect, in a third possible implementation of the first aspect, described according to the reference pixel
The weighted value of point i denoises the reference image vegetarian refreshments i, including, the denoising of the target pixel points i is estimated using following formula
Estimated value afterwards:Wherein De (i) indicates the estimated value after pixel i denoising,
Ω (i) indicates region of search centered on pixel i, and u (j) indicates the value at pixel j, C (i) be weight normalization because
Son, expression formula are as follows: C (i)=∑j∈Ω(i)w(i,j)。
With reference to first aspect, in a fourth possible implementation of the first aspect, the reference image block Ni and institute
The pixel size for stating target image Nj is identical.
Second aspect, provides a kind of image processing apparatus, and described image processing unit includes:
Image block determining module, for reference image block Ni of the determination centered on reference image vegetarian refreshments i, and described in determination
Target pixel points j in the contiguous range of reference image vegetarian refreshments i, and the target figure is determined centered on the target pixel points j
As block Nj；
Similitude determining module, it is similar between the reference image block Ni and the target image block Nj for determining
Property；
Weighted value determining module, for determining the target pixel points j to the reference image vegetarian refreshments i according to the similitude
The weighted value influenced when denoising；
Denoising module denoises the reference image vegetarian refreshments i for the weighted value according to the reference image vegetarian refreshments i.
Third aspect present invention provides a kind of terminal device, including memory, processor and is stored in the memory
In and the computer program that can run on the processor, the processor realize such as this hair when executing the computer program
The step of bright first aspect the method.
Fourth aspect present invention provides a kind of computer readable storage medium, and the computer-readable recording medium storage has
Computer program, when the computer program is executed by processor realize as described in the first aspect of the invention method the step of.
Image processing method, device, equipment and storage medium provided by the invention, by calculating reference image block and target
Similitude between image block, and then target pixel points are obtained to the weighing factor of reference image vegetarian refreshments, and then to reference image vegetarian refreshments
Denoising is carried out, since noise changes the distance between image block in an identical manner, is based on image block similarity measurements
Amount can more effectively remove picture noise.
Detailed description of the invention
Fig. 1 is the schematic flow chart of the method for the application one embodiment；
Fig. 2 shows the schematic block diagrams of the application one embodiment；
Fig. 3 shows the schematic flow chart of the method for the application one embodiment；
Fig. 4 shows the schematic diagram of the application one embodiment；
Fig. 5 shows the schematic block diagram of the device of the application one embodiment；
Fig. 6 shows the schematic block diagram of the device of another embodiment of the application.
Specific embodiment
Below in conjunction with attached drawing, the technical solution in the application is described.
There is the correlation of height firstly the need of the information being appreciated that in image, each pixel is not isolated presence
, they not only have the similitude of gray scale, but also the similitude with geometry.In addition, there is phase with object pixel
The regional area that may be not necessarily limited to image like the pixel of attribute, the pixel of any position may all be shown in image
Very strong correlation, such as cyclic pattern, elongated edge and the texture with repetitive structure etc..That is, scheming naturally
Often there is information largely with repetitive structure as in.Therefore, if we are using the figure that can describe image spatial feature
The similitude between two pixels is measured as block, it should can be more accurate than the measurement between single pixel point, so that it is more preferable
Some structural informations of ground holding image itself.
Based on this, the embodiment of the present application provides a kind of image processing method, the executing subject of this method can be figure
As processing unit, image processing method includes:
Step 110, the reference image block Ni centered on reference image vegetarian refreshments i is determined.
Step 120, it determines the target pixel points j in the contiguous range of reference image vegetarian refreshments i, and with target pixel points j is
The heart determines target image block Nj.
Step 130, the similitude between reference image block Ni and target image block Nj is determined.
Step 140, the weighted value influenced when determining that target pixel points j denoises reference image vegetarian refreshments i according to similitude.
Step 150, according to the weighted value of reference image vegetarian refreshments i, reference image vegetarian refreshments i is denoised.
Specifically, as an implementation, it can realize that the image processing method, Fig. 2 show by image processing apparatus
The schematic block diagram of image processing apparatus one embodiment in the application is gone out.Above-mentioned image processing apparatus can be such as Fig. 2
Shown in several units: COMS image sensor pixel array unit 210, analog circuitry processes unit 220, AD conversion unit
230, digital circuit processing unit 240 and image data output control unit 250.
The movement that said units specifically execute includes, prior to step 110, COMS image sensor pixel array unit
210 will acquire the original signal of image to be processed, and original image signal passes through analog circuit unit 220 and analog-digital converter unit
230, it converts analog signals into digital information transmission to digital circuit processing unit 240 and carries out processing of circuit.
Further, in step 110, digital circuit processing unit 240 will obtain the pixel currently to be denoised,
It is exactly that reference image block Ni is obtained centered on reference image vegetarian refreshments i.
It should be understood that the size of reference image block Ni and target image block Nj can also be other pixel sizes, the application is not
It limits, this is because when doing denoising to some pixel in image one can be taken centered on reference image vegetarian refreshments Ni
The image block that a size is 3 × 3 or 5 × 5, and using it as reference image block, the selection of reference picture block size will affect figure
The denoising effect of current pixel point as in, while also will affect the computation complexity of algorithm, the size of practical reference block needs simultaneous
Both care for.
Next, digital circuit processing unit 240 can be handled image to be processed, specifically, work as reference image block
After Ni is selected, the searching image block similar with reference block Ni in image to be processed, i.e., the target image block Nj in step 120,
As target image block Nj with the size of reference image block Ni is, two image blocks are more similar, then the current figure for participating in calculating
As the weight that the center pixel of block is contributed during denoising is also bigger, vice versa.
It is assumed that current pixel to be processed is i, j is the neighborhood territory pixel point of i.Ni indicates the figure centered on pixel i
The gray value vectors of shape block, the pixel in the Ni of field are expressed as u (Ni), then the similitude between pixel i and pixel j according to
Rely the similarity measurement between u (Ni) and u (Nj), u (Ni) and u (Nj) are more similar, then between pixel j and pixel i
It is more similar, so the contribution for calculating mean value when pixel j denoises pixel i is also bigger.
In order to measure the similitude between pixel i and pixel j, usual way is the Gauss weighting calculated between image block
Euclidean distance.In fact this measurement can be used for any additive white noise, this is because noise changes image in an identical manner
The distance between block.Image block similarity measurement based on Euclidean distance be it is reliable, effective, can robust reflection pixel it
Between similitude.
Optionally, it as the application one embodiment, determines similar between reference image block Ni and target image block Nj
Property, including, the Euclidean distance between reference image block Ni and target image block Nj is calculated using following equation:
Wherein, u (Ni) is the gray value vectors of the pixel in reference image block Ni, and u (Nj) is in target reference block Nj
The gray value vectors of pixel, Dist (i, j) indicate the Euclidean distance between image block Ni and image block Nj.
It optionally, can be by corresponding in two image blocks when calculating similarity between image block Ni and image block Nj
The absolute value of the difference of pixel is made and replaces Euclidean distance, can reduce the complexity of calculating in this way.
Optionally, as the application one embodiment, determine that target pixel points j removes reference image vegetarian refreshments i according to similitude
Weighing factor value when making an uproar, including, weighted value is determined using following formula:
Wherein, w (i, j) indicates that the weighted value that pixel j is contributed when denoising to pixel i, h are similarity weight power
Weight parameter.
Specifically, h is similarity weight parameter, it determines the smoothness of image.
Optionally, when calculating weighting function, the piecewise function that monotone decreasing can be used carrys out approximate fits exponential function, this
Sample also can be reduced the complexity of calculating.
Optionally, reference image vegetarian refreshments i is carried out according to the weighted value of reference image vegetarian refreshments i as the application one embodiment
Denoising, comprising:
Use the estimated value after the denoising of following formula estimation target pixel points i:
Wherein De (i) indicates the estimated value after pixel i denoising, and Ω (i) indicates the field of search centered on pixel i
Domain, u (j) indicate the value at pixel j, and C (i) is the normalization factor of weight, and expression formula is as follows:
C (i)=∑j∈Ω(i)w(i,j)。
Above-mentioned denoising process is carried out to pixel each in image to be processed, finally, image data-outputting unit 250 will be defeated
Image after denoising out.
Therefore, method provided by the embodiments of the present application, it is similar between reference image block and target image block by calculating
Property, and then target pixel points are obtained to the weighing factor of reference image vegetarian refreshments, and then denoising is carried out to reference pixel point.Due to
Noise changes the distance between image block in an identical manner, therefore can more effectively be removed based on image block similarity measurement
Picture noise.
Since the influence meeting of noise directly judges the similar inaccuracy between pixel when so that denoising, so that denoising effect
Fruit is bad, and we estimated using the image block centered on the pixel to be compared its similitude can more acurrate and robust,
It is allowed during denoising in this way and contributes bigger weight with the more like pixel of pixel to be processed, so denoising effect
Also more preferable.
There can be many similar structures in natural image, pixel to be processed is no longer confined to during denoising
The neighborhood of one very little can find the similar pixel of pixel more and to be processed in this way, this can also improve denoising
Effect.
Fig. 3 shows the schematic flow chart of the method for the application one embodiment.As shown in figure 3, this method comprises: step
Rapid 301, COMS image sensor pixel array unit 210 incudes the original signal of image to be processed；Step 302, image is original
Signal passes through analog circuit 220 and analog-digital converter 230, by digital data transmission to digital processing circuit processing 240 after conversion；
Step 303, it is 3 × 3 or 5 × 5 that digital processing circuit processing 240 takes a size centered on the pixel currently to be denoised
Image is as reference block；Step 304, digital processing circuit processing 240 calculate in the picture cut centered on some pixel and
Similarity between the identical image block of reference block image size and reference block；Step 305, image data output control unit 250
Pixel weight for sharing in denoising is determined according to the similarity being calculated, and then image after being denoised.
Specifically, Fig. 4 shows the schematic diagram of the application one embodiment.As shown in figure 4, showing one reality of the application
Apply reference image vegetarian refreshments i and target pixel points j1, the target pixel points j2 and target pixel points j3 in example.By calculating separately target
Pixel j1 the weighted value w (i, j1) to target pixel points i, weighted value w (i, j2) of the target pixel points j2 to reference image vegetarian refreshments i
And target pixel points j3 can denoise the weighted value w (i, j3) of reference image vegetarian refreshments i to target pixel points i, thus
To preferable denoising result.
Fig. 5 shows the schematic block diagram of the device of the application one embodiment.As shown in figure 5, image processing apparatus 500
Include:
Image block determining module 510 for the determining reference image block Ni centered on reference image vegetarian refreshments i, and determines
Target pixel points j in the contiguous range of reference image vegetarian refreshments i, and target image block Nj is determined centered on target pixel points j；
Similitude determining module 520, for determining the similitude between reference image block Ni and target image block Nj；
Weighted value determining module 530, shadow when for determining that target pixel points j denoises reference image vegetarian refreshments i according to similitude
Loud weighted value；
Denoising module 540 denoises reference image vegetarian refreshments i for the weighted value according to reference image vegetarian refreshments i.
Optionally, similitude determining module 520 is specifically used for:
Reference image block Ni and target figure are determined according to the Euclidean distance between reference image block Ni and target image block Nj
As the similitude between block Nj, and using following equation calculate between reference image block Ni and target image block Nj it is European away from
From:
Wherein, u (Ni) is the gray value vectors of the pixel in reference image block Ni, and u (Nj) is in target reference block Nj
The gray value vectors of pixel, Dist (i, j) indicate the Euclidean distance between reference image block Ni and target image block Nj.
Optionally, weighted value determining module 530 is specifically used for:
Weighted value is determined using following equation:
Wherein, w (i, j) indicates that the weighted value that pixel j is contributed when denoising to pixel i, h are similarity weight power
Weight parameter；
Optionally, denoising module 540 is specifically used for:
Use the estimated value after the denoising of following formula estimation target pixel points i:
Wherein De (i) indicates the estimated value after pixel i denoising, and Ω (i) indicates the field of search centered on pixel i
Domain, u (j) indicate the value at pixel j, and C (i) is the normalization factor of weight, and expression formula is as follows:
C (i)=∑j∈Ω(i)w(i,j)。
Another kind embodiment of the invention provides a computer readable storage medium, stores on the computer readable storage medium
There is computer program, which realizes the image processing method in above-described embodiment when being executed by processor, to avoid
It repeats, which is not described herein again.Alternatively, the computer program realizes image processing apparatus in above-described embodiment when being executed by processor
In each module/unit function, to avoid repeating, which is not described herein again.
Fig. 6 shows the schematic block diagram of the device of another embodiment of the application.The device is able to carry out implementation of the present invention
The image processing method that example provides.Wherein, which includes: processor 601 and memory 602.Wherein, memory 602
It can be used for storing the program code and data of the network equipment.Therefore, which can be inside processor 601
Storage unit, be also possible to the independent external memory unit of processor 601, can also be depositing inside including processor 601
Storage unit and component with the independent external memory unit of processor 601.
Optionally, above-mentioned apparatus can also include bus 603.Wherein, memory 602 can pass through bus 603 and processing
Device 601 connects；Bus 603 can be Peripheral Component Interconnect standard (Peripheral Component Interconnect,
PCI) bus or expanding the industrial standard structure (Extended Industry Standard Architecture, EISA) bus
Deng.Bus 603 can be divided into address bus, data/address bus, control bus etc..For convenient for indicating, only with a thick line table in Fig. 6
Show, it is not intended that an only bus or a type of bus.
Processor 601 for example can be central processing unit (Central Processing Unit, CPU), general procedure
Device, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit (Application-
Specific Integrated Circuit, ASIC), field programmable gate array (Field Programmable Gate
Array, FPGA) either other programmable logic device, transistor logic, hardware component or any combination thereof.It can
To realize or execute various illustrative logic blocks, module and circuit in conjunction with described in the disclosure of invention.The place
Reason device is also possible to realize the combination of computing function, such as combines comprising one or more microprocessors, DSP and microprocessor
Combination etc..
It should be understood that the method that processor 601 can execute above-described embodiment description, can be realized above-described embodiment description
Beneficial effect, for brevity, details are not described herein.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or the second equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of image processing method, which is characterized in that described image processing method includes:
Determine the reference image block Ni centered on reference image vegetarian refreshments i；
Determine the target pixel points j in the contiguous range of the reference image vegetarian refreshments i, and centered on the target pixel points j really
The fixed target image block Nj；
Determine the similitude between the reference image block Ni and the target image block Nj；
The weighted value influenced when determining that the target pixel points j denoises the reference image vegetarian refreshments i according to the similitude；
According to the weighted value of the reference image vegetarian refreshments i, the reference image vegetarian refreshments i is denoised.
2. image processing method according to claim 1, which is characterized in that the determination reference image block Ni and institute
State the similitude between target image block Nj, comprising:
The reference image block Ni is determined according to the Euclidean distance between the reference image block Ni and the target image block Nj
With the similitude between the target image block Nj, and the reference image block Ni and the target figure are calculated using following equation
As the Euclidean distance between block Nj:
Wherein, u (Ni) is the gray value vectors of the pixel in the reference image block Ni, and u (Nj) is the target reference block Nj
The gray value vectors of interior pixel, Dist (i, j) indicate European between the reference image block Ni and target image block Nj
Distance.
3. image processing method according to claim 2, which is characterized in that described to determine the mesh according to the similitude
The weighted value that mark pixel j influences when denoising on the reference image vegetarian refreshments i, comprising:
The weighted value is determined using following equation:
Wherein, w (i, j) indicates that the weighted value that pixel j is contributed when denoising to pixel i, h are similarity weight ginsengs
Number.
4. image processing method according to claim 3, which is characterized in that the power according to the reference image vegetarian refreshments i
Weight values denoise the reference image vegetarian refreshments i, comprising:
Estimated value after estimating the denoising of the target pixel points i using following equation:
Wherein De (i) indicates the estimated value after pixel i denoising, and Ω (i) indicates the region of search centered on pixel i, u
(j) value at pixel j is indicated, C (i) is the normalization factor of weight, and expression formula is as follows:
C (i)=∑j∈Ω(i)w(i,j)。
5. image processing method according to any one of claim 1 to 4, which is characterized in that the reference image block Ni
It is identical with the pixel size of the target image Nj.
6. a kind of image processing apparatus, which is characterized in that described image processing unit includes:
Image block determining module, for determining the reference image block Ni centered on reference image vegetarian refreshments i, and the determining reference
Target pixel points j in the contiguous range of pixel i, and the target image block is determined centered on the target pixel points j
Nj；
Similitude determining module, for determining the similitude between the reference image block Ni and the target image block Nj；
Weighted value determining module, for determining that the target pixel points j denoises the reference image vegetarian refreshments i according to the similitude
When the weighted value that influences；
Denoising module denoises the reference image vegetarian refreshments i for the weighted value according to the reference image vegetarian refreshments i.
7. image processing apparatus according to claim 6, which is characterized in that the similitude determining module is specifically used for:
The reference image block Ni is determined according to the Euclidean distance between the reference image block Ni and the target image block Nj
With the similitude between the target image block Nj, and the reference image block Ni and the target figure are calculated using following equation
As the Euclidean distance between block Nj:
Wherein, u (Ni) is the gray value vectors of the pixel in the reference image block Ni, and u (Nj) is the target reference block Nj
The gray value vectors of interior pixel, Dist (i, j) indicate European between the reference image block Ni and target image block Nj
Distance.
8. image processing apparatus according to claim 7, which is characterized in that the weighted value determining module is specifically used for:
The weighted value is determined using following equation:
Wherein, w (i, j) indicates that the weighted value that pixel j is contributed when denoising to pixel i, h are similarity weight weight ginsengs
Number；
The denoising module is specifically used for:
Estimated value after estimating the denoising of the target pixel points i using following formula:
Wherein De (i) indicates the estimated value after pixel i denoising, and Ω (i) indicates the region of search centered on pixel i, u
(j) value at pixel j is indicated, C (i) is the normalization factor of weight, and expression formula is as follows:
C (i)=∑j∈Ω(i)w(i,j)。
9. a kind of terminal device, including memory, processor and storage are in the memory and can be on the processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1 to 5 when executing the computer program
The step of any one the method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In when the computer program is executed by processor the step of any one of such as claim 1 to 5 of realization the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711002231.6A CN109697699A (en) | 2017-10-24 | 2017-10-24 | Image processing method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711002231.6A CN109697699A (en) | 2017-10-24 | 2017-10-24 | Image processing method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109697699A true CN109697699A (en) | 2019-04-30 |
Family
ID=66227843
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711002231.6A Pending CN109697699A (en) | 2017-10-24 | 2017-10-24 | Image processing method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109697699A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111462155A (en) * | 2020-03-26 | 2020-07-28 | 深圳万拓科技创新有限公司 | Movement detection method and device, computer 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 |
| CN111860836A (en) * | 2020-07-17 | 2020-10-30 | 深圳高性能医疗器械国家研究院有限公司 | Self-supervision learning method and application |
| CN117115261A (en) * | 2023-10-17 | 2023-11-24 | 深圳市青虹激光科技有限公司 | Knife wheel cutting positioning method and system based on thin wafer |
| WO2024078179A1 (en) * | 2022-10-11 | 2024-04-18 | 腾讯科技（深圳）有限公司 | Lighting map noise reduction method and apparatus, and device and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103020918A (en) * | 2013-01-09 | 2013-04-03 | 西安电子科技大学 | Shape-adaptive neighborhood mean value based non-local mean value denoising method |
| CN103150712A (en) * | 2013-01-18 | 2013-06-12 | 清华大学 | Image denoising method based on projection sequential data similarity |
-
2017
- 2017-10-24 CN CN201711002231.6A patent/CN109697699A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103020918A (en) * | 2013-01-09 | 2013-04-03 | 西安电子科技大学 | Shape-adaptive neighborhood mean value based non-local mean value denoising method |
| CN103150712A (en) * | 2013-01-18 | 2013-06-12 | 清华大学 | Image denoising method based on projection sequential data similarity |
Non-Patent Citations (1)
| Title |
|---|
| 周立云，刘航军编著: "《数码相机修理从入门到精通[M]》", 31 May 2011 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111462155A (en) * | 2020-03-26 | 2020-07-28 | 深圳万拓科技创新有限公司 | Movement detection method and device, computer equipment and storage medium |
| CN111462155B (en) * | 2020-03-26 | 2023-10-24 | 深圳万拓科技创新有限公司 | Motion detection method, device, computer 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 |
| CN111709894B (en) * | 2020-06-16 | 2022-09-02 | 展讯通信（上海）有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN111860836A (en) * | 2020-07-17 | 2020-10-30 | 深圳高性能医疗器械国家研究院有限公司 | Self-supervision learning method and application |
| WO2024078179A1 (en) * | 2022-10-11 | 2024-04-18 | 腾讯科技（深圳）有限公司 | Lighting map noise reduction method and apparatus, and device and medium |
| CN117115261A (en) * | 2023-10-17 | 2023-11-24 | 深圳市青虹激光科技有限公司 | Knife wheel cutting positioning method and system based on thin wafer |
| CN117115261B (en) * | 2023-10-17 | 2024-03-19 | 深圳市青虹激光科技有限公司 | Knife wheel cutting positioning method and system based on thin wafer |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109697699A (en) | Image processing method, device, equipment and storage medium | |
| CN108230292B (en) | Object detection method, neural network training method, device and electronic equipment | |
| CN109741356B (en) | Sub-pixel edge detection method and system | |
| JP5846517B2 (en) | Quality evaluation of image composition change | |
| CN108416789A (en) | Method for detecting image edge and system | |
| CN107403134B (en) | Local gradient trilateral-based image domain multi-scale infrared dim target detection method | |
| CN107301661A (en) | High-resolution remote sensing image method for registering based on edge point feature | |
| US11841421B2 (en) | Synthetic aperture radar image analysis system, synthetic aperture radar image analysis method, and synthetic aperture radar image analysis program | |
| CN108140291A (en) | Mist detecting device, method and image processing equipment | |
| CN104732509B (en) | Self-adaptive image segmentation method, device and equipment | |
| CN107221005B (en) | Object detection method and device | |
| JP2019194821A (en) | Target recognition device, target recognition method, and program | |
| WO2015178148A1 (en) | State estimation device, program, and integrated circuit | |
| WO2017070923A1 (en) | Human face recognition method and apparatus | |
| CN110837079B (en) | Target detection method and device based on radar | |
| CN108830828B (en) | Method and device for detecting change of remote sensing image | |
| CN109978855A (en) | A kind of method for detecting change of remote sensing image and device | |
| CN112530003B (en) | Three-dimensional human hand reconstruction method and device and electronic equipment | |
| JP7484492B2 (en) | Radar-based attitude recognition device, method and electronic device | |
| CN105225219B (en) | Information processing method and electronic equipment | |
| CN117809044A (en) | Method, device and equipment for denoising three-dimensional point cloud of array interference synthetic aperture radar | |
| CN116977671A (en) | Target tracking method, device, equipment and storage medium based on image space positioning | |
| CN117351350A (en) | CNN neural network InSAR phase filtering method for weighted MAE training | |
| CN115984246B (en) | Machine vision-based defect rapid detection method and device, equipment and storage medium | |
| CN113508395A (en) | Method for detecting an object |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210204 Address after: 518119 No.1 Yan'an Road, Kuiyong street, Dapeng New District, Shenzhen City, Guangdong Province Applicant after: BYD Semiconductor Co.,Ltd. Address before: 518118 BYD Road, Pingshan New District, Shenzhen, Guangdong 3009 Applicant before: BYD Co.,Ltd. |
|
| TA01 | Transfer of patent application right |