CN103914861B - Image processing method and device - Google Patents
Image processing method and device Download PDFInfo
- Publication number
- CN103914861B CN103914861B CN201310006129.9A CN201310006129A CN103914861B CN 103914861 B CN103914861 B CN 103914861B CN 201310006129 A CN201310006129 A CN 201310006129A CN 103914861 B CN103914861 B CN 103914861B
- Authority
- CN
- China
- Prior art keywords
- pixel
- picture
- gradient
- grad
- ownership
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Image Analysis (AREA)
Abstract
The present invention discloses a kind of image processing method and device, and this method includes：Receive pending picture；After detecting picture processing instruction, gradient calculation is carried out according to specific rule to the picture, obtains the gradient information of each pixel；The gradient information for the pixel that each pixel of picture is counted in specific region and is on each preset direction, and belong to gradient direction according to corresponding to each gradient information of statistics finds out each pixel of picture；According to corresponding ownership gradient direction and the Grad of the ownership gradient direction, Fuzzy Processing is carried out to each pixel.Relative to prior art, people can more be embodied by the profile of the picture after the processing of technical solution of the present invention effect is sketched the contours to object in artistic creation, so as to improve the verisimilitude of image effect.
Description
Technical field
The present invention relates to picture processing field, more particularly to a kind of image processing method and device.
Background technology
With the continuous progress of computer technology and increasing substantially for computer graphical processing ability, especially GPU
Rapid development, image processing techniques are also maked rapid progress.Such as rim detection can be carried out by computer processing technology, so as to carve
The profile of picture picture.Prior art is to utilize edge detection algorithm（Such as canny operators, DOG（Difference Of Gauss）
Deng）, the edge of image is detected, is then directly used as lines by the use of edge.Although the edge detection algorithm can depict figure
The profile information of picture, but its caused lines is more robotic, so as to influence computer mould personification in artistic creation to object
The verisimilitude for sketching the contours effect.
The content of the invention
The main purpose of the embodiment of the present invention is to provide a kind of image processing method, it is intended to improves the true to nature of image effect
Property.
To achieve the above object, the embodiment of the present invention proposes a kind of image processing method, comprises the following steps：
Receive pending picture；
After detecting picture processing instruction, gradient calculation is carried out according to specific rule to the picture, obtains each picture
The gradient information of vegetarian refreshments；
The gradient information for the pixel that each pixel of picture is counted in specific region and is on each preset direction,
And belong to gradient direction according to corresponding to each gradient information of statistics finds out each pixel of picture；
According to corresponding ownership gradient direction and the Grad of the ownership gradient direction, fuzzy place is carried out to each pixel
Reason.
The embodiment of the present invention also proposed a kind of picture processing device, including：
Picture gradient calculation module, for receiving pending picture, and after picture processing instruction is detected, to described
Picture carries out gradient calculation according to specific rule, obtains the gradient information of each pixel；
Belong to gradient direction judge module, for counting each pixel of picture in specific region and in each default
The gradient information of pixel on direction, and return according to corresponding to each gradient information of statistics finds out each pixel of picture
Belong to gradient direction；
Fuzzy Processing module, for belonging to gradient direction and the Grad of the ownership gradient direction corresponding to, to every
Individual pixel carries out Fuzzy Processing.
Relative to prior art, the embodiment of the present invention is by calculating the gradient information of each pixel of picture, and according to each
In specific region centered on pixel and the gradient information on the preset direction, it is determined that the ownership gradient side of each pixel
To and its Grad on ownership gradient direction, gradient direction and Grad are then belonged to according to corresponding to pixel, to every
The profile that individual pixel carries out the fuzzy picture formed afterwards can more embody people and sketch the contours effect to object in artistic creation, from
And improve the verisimilitude of image effect.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the embodiment of image processing method one of the present invention；
Fig. 2 is the flow signal that ownership gradient direction corresponding to each pixel is determined in image processing method of the present invention
Figure；
Fig. 3 is the view of pixel in method shown in Fig. 2；
Fig. 4 a are the schematic diagrames of the picture profile after the edge detection algorithm processing using prior art；
Fig. 4 b are the schematic diagrames of the picture profile after the image processing method processing using the present invention；
Fig. 5 is the schematic flow sheet of another embodiment of image processing method of the present invention；
Fig. 6 is the structural representation of picture processing device preferred embodiment of the present invention.
The realization, functional characteristics and advantage of the object of the invention will be described further referring to the drawings in conjunction with the embodiments.
Embodiment
Technical scheme is further illustrated below in conjunction with Figure of description and specific embodiment.It should be appreciated that this
The specific embodiment of place description is not intended to limit the present invention only to explain the present invention.
Reference picture 1, propose the embodiment of image processing method one of the present invention.The image processing method of the embodiment includes following
Step：
Step S110, pending picture is received；
Step S120, after detecting picture processing instruction, gradient calculation is carried out according to specific rule to the picture, obtained
Obtain the gradient information of each pixel；
In the embodiment of the present invention, after picture selection function is triggered, then picture selection menu is ejected, and wait user to select
Select.After the picture selection operation for receiving user, the picture selected by user is obtained, then waits user to send picture processing again
Instruction.Certainly, the picture processing instruction in step S120 can be initiated by user terminal, or receive pending
Automatic triggering produces picture processing instruction during picture.Certainly this is also not limited to, such as in the embodiment of the present invention, can also is：
After picture processing instruction is detected, then picture selection menu is ejected, pending picture is selected for user, receives user's
After picture selection operation, then the picture selected by user is obtained, as pending picture.
Then gradient calculation is carried out to the pending picture, and obtains the gradient information of each pixel.It is first right below
The gradient of picture is simply described.
If the image information of picture is with the discrete function f of a two dimension（X, y）To represent, then the ladder of each pixel of picture
Degree is the two-dimentional discrete function f（X, y）Derivative on each pixel, and the gradient is a vector information, Ke Yitong
Cross equation below expression：
Wherein, Grad [f (x, y)] represents image function f（X, y）Gradient, including（X, y）Pixel gray scale along x
Direction and the rate of change in y directions, point to the maximum direction of grey scale change.Therefore, can be obtained according to above-mentioned gradient（X, y）Pixel
The Grad and gradient direction of point.For example,（X, y）The Grad of pixel is：
（X, y）The gradient direction of pixel is：
The preparation method of gradient information is in the embodiment of the present invention：According to the gradient operator pre-set, by gradient operator
Convolutional calculation is carried out with picture, obtains the vector information of each pixel of picture.Wherein, the gradient operator may include sobel
Operator, Laplace operators, Roberts operators, Prewitt operators etc..Using different gradient operators, itself and picture are entered
Row convolutional calculation, so as to obtain the vector information of each pixel of picture.By taking sobel operators as an example, the operator includes two groups
3*3 matrix, respectively transverse direction and longitudinal direction, it is as follows：
And
Work as calculating（X, y）The gradient of pixel, then centered on the pixel, the gray value of its adjacent pixel is obtained, such as
Under：
| Z1 | Z2 | Z3 |
| Z4 | Z5 | Z6 |
| Z7 | Z8 | Z9 |
Then calculate the pixel respectively in the horizontal direction and vertical direction gradient：
By that analogy, the gradient of each pixel of picture is calculated.
Step S130, the pixel for counting each pixel of picture in specific region and being on each preset direction
Gradient information, and belong to gradient direction according to corresponding to each gradient information of statistics finds out each pixel of picture；
Reference picture 2, the step S130 in the present embodiment may include following steps：
Step S131, each pixel of picture is traveled through, and by pixel centered on the current pixel point of traversal, presets side
The long square gradient information for being used as specific region, obtaining the pixel in the specific region and being on each preset direction, meter
Calculate the Grad for obtaining each pixel；
Specifically, as shown in figure 3, traveling through each pixel of picture, it is then determined that the Grad of each pixel.With picture
Exemplified by vegetarian refreshments a, first the point centered on the pixel, default the square of the length of side are used as specific region, the area surrounded such as dotted line frame
Domain.Then the gradient information for the pixel in the dotted line frame and being on each preset direction is obtained, the preset direction can basis
Particular situation and set, in the embodiment of the present invention, preset direction be 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 °
This 8 directions.Therefore the gradient information of acquired gradient information then pixel as shown in Figure 3, wherein white circle table
Show the pixel for not obtaining gradient information, black circles represent to obtain the pixel of gradient information.Then each pixel is calculated
The Grad of point.
Step S132, compare the Grad of each pixel obtained, central pixel point is pointed into Grad maximum
Ownership gradient direction of the direction of pixel as the central pixel point, and the maximum Grad is as the central pixel point
Belong to the Grad of gradient direction.
Compare the Grad of obtained each pixel, if pixel b Grad is maximum, pixel a is pointed into picture
Ownership gradient directions of the vegetarian refreshments b direction L as pixel a, and pixel b Grad is pixel a terraced in ownership
The Grad spent on direction.
Step S140, according to corresponding ownership gradient direction and the Grad of the ownership gradient direction, to each pixel
Obscured.
Grad on the ownership gradient direction and its ownership gradient direction for determining each pixel, then traversal is each
Pixel, the pixel centered on the current pixel point of traversal, and according to the ownership gradient direction of the central pixel point and should
Belong to the Grad of gradient direction, Gaussian Blur is carried out along ownership gradient direction centered on the central pixel point.Specifically,
Each pixel is obscured by using equation below：
,
Wherein, G（n）For Gaussian parameter, p（n）It is Grad of the central pixel point on each preset direction, due to step
Rapid S130 only records the central pixel point and belongs to the Grad on gradient direction at it, so the Grad of the central pixel point is set
The central pixel point is set in the obtained gradient of its corresponding Gaussian parameter multiplication of Grad that it belongs on gradient direction
Value.It is achieved thereby that the central pixel point is obscured.
As shown in Fig. 4 a and Fig. 4 b, relative to prior art, the embodiment of the present invention is by calculating the ladder of each pixel of picture
Spend information, and according in the specific region centered on each pixel and the gradient information on the preset direction, it is determined that each
The ownership gradient direction of pixel and its Grad on ownership gradient direction, are then belonging to gradient side corresponding to pixel
To and Grad, the profile of the fuzzy picture formed afterwards is carried out to each pixel can more embody people in artistic creation to thing
Body sketches the contours effect, so as to improve the verisimilitude of image effect.
Reference picture 5, propose another embodiment of image processing method of the present invention.On the basis of above method embodiment, this
Image processing method also includes after step S120 in embodiment：
Step S150, the Grad of each pixel is calculated, and filters out the picture that Grad is more than or equal to predetermined threshold value
Vegetarian refreshments.
A threshold value is set, the threshold value is for the pixel where the profile of preliminary screening image.Then more each picture
The Grad of vegetarian refreshments and the threshold value, it is follow-up so as to ensure less than the then deletion of the threshold value more than or equal to the then reservation of the threshold value
Step S130, pixel performed in S140 is the pixel where profile, and no longer needs each pixel to picture
Point is calculated and obscured, so as to further increase the efficiency of the picture processing.
Reference picture 6, propose the embodiment of picture processing device one of the present invention.The picture processing device of the embodiment includes：
Picture gradient calculation module 110 is right for receiving pending picture, and after picture processing instruction is detected
The picture carries out gradient calculation according to specific rule, obtains the gradient information of each pixel；
Belong to gradient direction judge module 120, for counting each pixel of picture in specific region and in each
The gradient information of pixel on preset direction, and each pixel of picture is found out correspondingly according to each gradient information of statistics
Ownership gradient direction；
Fuzzy Processing module 130 is right for belonging to gradient direction and the Grad of the ownership gradient direction corresponding to
Each pixel is obscured.
In the embodiment of the present invention, after the picture selection function on picture processing device is triggered, then picture selection is ejected
Menu, and wait user to select.After the picture selection operation of user is received, picture gradient calculation module 110 then receives use
Picture selected by family, user is then waited to send picture processing instruction again.Certainly, above-mentioned picture processing instruction can be by using
Initiate at family end, or automatic triggering produces picture processing instruction when receiving pending picture.Certainly also it is not limited to
This, such as in the embodiment of the present invention, can also be：Picture gradient calculation module 110 is after picture processing instruction is detected, then bullet
Go out picture selection menu, pending picture is selected for user, user institute is then obtained after receiving the picture selection operation of user
The picture of selection, as pending picture.Then gradient calculation is carried out to the pending picture, and obtains each pixel
Gradient information.The preparation method of gradient information is in the embodiment of the present invention：According to the gradient operator pre-set, gradient is calculated
Son carries out convolutional calculation with picture, obtains the vector information of each pixel of picture.Wherein, the gradient operator may include
Sobel operators, Laplace operators, Roberts operators, Prewitt operators etc..Using different gradient operators, by itself and figure
Piece carries out convolutional calculation, so as to obtain the vector information of each pixel of picture.The each picture of picture is calculated by sobel operators
The process of the gradient of vegetarian refreshments can refer to described in previous methods embodiment, just repeat no more herein.
After the gradient information for obtaining each pixel of picture, ownership gradient direction judge module 120 will travel through the every of picture
Individual pixel, and it is specific to be obtained this as specific region for pixel centered on the current pixel point of traversal, the circle of pre-set radius
In region and the pixel on each preset direction gradient information, calculate the Grad for obtaining each pixel；Compare institute
The Grad of each pixel obtained, central pixel point is pointed into the direction of the maximum pixel of Grad as imago in this
The ownership gradient direction of vegetarian refreshments, and Grad of the maximum Grad as the ownership gradient direction of the central pixel point.Tool
Body, the preset direction can be set as the case may be, in the embodiment of the present invention, preset direction is 0 °, 45 °, 90 °,
135 °, 180 °, 225 °, 270 °, 315 ° of this 8 directions.As shown in figure 3, by taking pixel a as an example, first centered on the pixel
Point, the circle of pre-set radius is as specific region, such as dotted line frame area defined.Then obtain in the dotted line frame and in each pre-
The gradient information of pixel on set direction, i.e., the gradient information of the pixel representated by black circles shown in Fig. 3 are acquired.
Then the Grad of each pixel is calculated.Compare the Grad of obtained each pixel, if pixel b Grad is most
Greatly, then pixel a is pointed to pixel b ownership gradient directions of the direction L as pixel a, and pixel b Grad
For Grad of the pixel a on ownership gradient direction.
Finally, when the Grad on the ownership gradient direction and its ownership gradient direction for determining each pixel, obscure
Processing module 130 then travels through each pixel, the pixel centered on the current pixel point of traversal, using the specific region as mould
Region is pasted, and according to the ownership gradient direction of the central pixel point and the Grad of the ownership gradient direction, with center pixel
Centered on point Gaussian Blur is carried out along ownership gradient direction.Specific blurring process can refer to described in previous methods embodiment,
Just repeat no more herein.
Relative to prior art, the embodiment of the present invention is by calculating the gradient information of each pixel of picture, and according to each
In specific region centered on pixel and the gradient information on the preset direction, it is determined that the ownership gradient side of each pixel
To and its Grad on ownership gradient direction, then belonging to gradient direction and Grad corresponding to pixel, to each
The profile that pixel carries out the fuzzy picture formed afterwards can more embody people and sketch the contours effect to object in artistic creation, so as to
Improve the verisimilitude of image effect.
In another embodiment of picture processing device of the present invention, picture processing device also includes：
Pixel screening module 140, for after the gradient information of each pixel is obtained, calculating the ladder of each pixel
Angle value, and filter out the pixel that Grad is more than or equal to predetermined threshold value.
The pixel that above-mentioned predetermined threshold value is used for where the profile of preliminary screening image.Then the ladder of more each pixel
Angle value and the threshold value, more than or equal to the then reservation of the threshold value, less than the then deletion of the threshold value, so as to ensure subsequently to belong to gradient
Pixel handled by walking direction module 120, Fuzzy Processing module 130 is the pixel where profile, and is no longer needed
Each pixel of picture is calculated and obscured, so as to further increase the efficiency of the picture processing.
The preferred embodiments of the present invention are the foregoing is only, not thereby limit its scope of the claims, it is every to utilize the present invention
The equivalent structure or equivalent flow conversion that specification and accompanying drawing content are made, directly or indirectly it is used in other related technology necks
Domain, it is included within the scope of the present invention.
Claims (8)
1. a kind of image processing method, it is characterised in that comprise the following steps：
Receive pending picture；
After detecting picture processing instruction, gradient calculation is carried out according to specific rule to the picture, obtains each pixel
Gradient information；
The gradient information for the pixel that each pixel of picture is counted in specific region and is on each preset direction, and root
Each gradient information according to statistics finds out ownership gradient direction corresponding to each pixel of picture；
According to corresponding ownership gradient direction and the Grad of the ownership gradient direction, Fuzzy Processing is carried out to each pixel；
Wherein, the gradient of pixel of each pixel of the statistics picture in specific region and on each preset direction
Information, and belong to gradient direction bag according to corresponding to the gradient information of each pixel of statistics finds out each pixel of picture
Include：
Each pixel of picture is traveled through, and by pixel centered on the current pixel point of traversal, presets the square conduct of the length of side
Specific region, the gradient information for the pixel in the specific region and being on each preset direction is obtained, calculates and obtains each picture
The Grad of vegetarian refreshments；
Compare the Grad of each pixel obtained, the direction that central pixel point is pointed to the maximum pixel of Grad is made
For the ownership gradient direction of the central pixel point, and the maximum Grad is as the ownership gradient direction of the central pixel point
Grad.
2. image processing method according to claim 1, it is characterised in that it is described detect picture processing instruction after, it is right
The picture carries out gradient calculation, and obtaining the gradient information of each pixel includes：
After detecting picture processing instruction, according to the gradient operator pre-set, gradient operator and picture are subjected to convolutional calculation,
Obtain the vector information of each pixel of picture.
3. image processing method according to claim 1, it is characterised in that ownership gradient direction corresponding to the basis and
The Grad of the ownership gradient direction, fuzzy include is carried out to each pixel：
Each pixel is traveled through, the pixel centered on the current pixel point of traversal, using the specific region as fuzzy region, and
According to the ownership gradient direction of the central pixel point and the Grad of the ownership gradient direction, the central pixel point is carried out high
This Fuzzy Processing.
4. according to the image processing method described in claim any one of 1-3, it is characterised in that each pixel of acquisition
Also include after gradient information：
The Grad of each pixel is calculated, and filters out the pixel that Grad is more than or equal to predetermined threshold value.
A kind of 5. picture processing device, it is characterised in that including：
Picture gradient calculation module, for receiving pending picture, and after picture processing instruction is detected, to the picture
Gradient calculation is carried out according to specific rule, obtains the gradient information of each pixel；
Belong to gradient direction judge module, for counting each pixel of picture in specific region and being in each preset direction
On pixel gradient information, and according to corresponding to each gradient information of statistics finds out each pixel of picture belong to ladder
Spend direction；
Fuzzy Processing module, for belonging to gradient direction and the Grad of the ownership gradient direction corresponding to, to each picture
Vegetarian refreshments carries out Fuzzy Processing；
Wherein, the ownership gradient calculation module is used for：
Each pixel of picture is traveled through, and by pixel centered on the current pixel point of traversal, presets the square conduct of the length of side
Specific region, the gradient information for the pixel in the specific region and being on each preset direction is obtained, calculates and obtains each picture
The Grad of vegetarian refreshments；
Compare the Grad of each pixel obtained, the direction that central pixel point is pointed to the maximum pixel of Grad is made
For the ownership gradient direction of the central pixel point, and the maximum Grad is as the ownership gradient direction of the central pixel point
Grad.
6. picture processing device according to claim 5, it is characterised in that the picture gradient calculation module is used for：Detect
After measuring picture processing instruction, according to the gradient operator pre-set, gradient operator and picture are subjected to convolutional calculation, schemed
The vector information of each pixel of piece.
7. picture processing device according to claim 5, it is characterised in that the resume module module is used for：
Each pixel is traveled through, the pixel centered on the current pixel point of traversal, using the specific region as fuzzy region, and
According to the ownership gradient direction of the central pixel point and the Grad of the ownership gradient direction, the central pixel point is carried out high
This Fuzzy Processing.
8. according to the picture processing device described in claim any one of 5-7, it is characterised in that also include：
Pixel screening module, for after the gradient information of each pixel is obtained, calculating the Grad of each pixel, and
Filter out the pixel that Grad is more than or equal to predetermined threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310006129.9A CN103914861B (en) | 2013-01-08 | 2013-01-08 | Image processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310006129.9A CN103914861B (en) | 2013-01-08 | 2013-01-08 | Image processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103914861A CN103914861A (en) | 2014-07-09 |
| CN103914861B true CN103914861B (en) | 2018-02-27 |
Family
ID=51040515
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310006129.9A Active CN103914861B (en) | 2013-01-08 | 2013-01-08 | Image processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103914861B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105989575B (en) * | 2015-03-02 | 2019-08-30 | 腾讯科技（深圳）有限公司 | A kind of image fuzzy processing method and device |
| CN105976333A (en) * | 2016-04-29 | 2016-09-28 | 广东小天才科技有限公司 | Picture blur correction method and device and intelligent equipment |
| CN109472848A (en) * | 2018-10-29 | 2019-03-15 | 广东小天才科技有限公司 | Picture processing method for preventing fingerprint information from being leaked and terminal equipment |
| CN111260581B (en) * | 2020-01-17 | 2023-09-26 | 北京达佳互联信息技术有限公司 | Image processing method, device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101465001A (en) * | 2008-12-31 | 2009-06-24 | 昆山锐芯微电子有限公司 | Method for detecting image edge based on Bayer RGB |
| CN101521740A (en) * | 2009-04-01 | 2009-09-02 | 北京航空航天大学 | Real-time athletic estimating method based on multiple dimensioned unchanged characteristic |
| CN102663692A (en) * | 2012-03-28 | 2012-09-12 | 汕头大学 | Adaptive SUSAN diffusion and denoising method of medical ultrasonic image |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1846893B1 (en) * | 2005-02-03 | 2015-09-16 | Koninklijke Philips N.V. | Radial adaptive filter for metal artifact correction |
-
2013
- 2013-01-08 CN CN201310006129.9A patent/CN103914861B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101465001A (en) * | 2008-12-31 | 2009-06-24 | 昆山锐芯微电子有限公司 | Method for detecting image edge based on Bayer RGB |
| CN101521740A (en) * | 2009-04-01 | 2009-09-02 | 北京航空航天大学 | Real-time athletic estimating method based on multiple dimensioned unchanged characteristic |
| CN102663692A (en) * | 2012-03-28 | 2012-09-12 | 汕头大学 | Adaptive SUSAN diffusion and denoising method of medical ultrasonic image |
Non-Patent Citations (2)
| Title |
|---|
| 一种基于梯度的图像滤波算法;李伟;《南通纺织职业技术学院学报(综合版)》;20050925;第5卷(第3期);第18页第2节第1段 * |
| 自动聚焦评价函数的精确度和稳定性研究;高赞;《中国优秀硕士学位论文全文数据库.信息科技辑》;20070915(第03期);第50页5.1.1第4段 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103914861A (en) | 2014-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104916163B (en) | Parking space detection method | |
| CN104408460B (en) | A kind of lane detection and tracking detection method | |
| CN104751146B (en) | A kind of indoor human body detection method based on 3D point cloud image | |
| CN107067382A (en) | A kind of improved method for detecting image edge | |
| CN109284674A (en) | A kind of method and device of determining lane line | |
| CN103914861B (en) | Image processing method and device | |
| CN104331873A (en) | Method for detecting road from single image | |
| CN103455813A (en) | Method for facula center positioning of CCD image measurement system | |
| CN112991374B (en) | Canny algorithm-based edge enhancement method, canny algorithm-based edge enhancement device, canny algorithm-based edge enhancement equipment and storage medium | |
| CN105457908B (en) | The sorting method for rapidly positioning and system of small size glass panel based on monocular CCD | |
| CN104050637A (en) | Quick image defogging method based on two times of guide filtration | |
| CN105139391B (en) | A kind of haze weather traffic image edge detection method | |
| CN114241364A (en) | A rapid calibration method for foreign object targets in overhead transmission lines | |
| CN102821228B (en) | Low-rank video background reconstructing method | |
| CN103955949A (en) | Moving target detection method based on Mean-shift algorithm | |
| Zhang et al. | Edge detection based on improved sobel operator | |
| CN103093223A (en) | Light spot image center quick locating method | |
| CN104077775A (en) | Shape matching method and device combining skeleton feature points and shape context | |
| CN102393964B (en) | Strip gap detection method | |
| CN109833003B (en) | Sweeping robot and method for dividing region thereof | |
| CN102663781A (en) | Sub-pixel level welding center extraction method based on visual sense | |
| CN117372498A (en) | Multi-pose bolt size measurement method based on three-dimensional point cloud | |
| CN116309447A (en) | A method for detecting cracks in dam slopes based on deep learning | |
| CN111739011A (en) | A method and device for detecting the tilt of a utility pole | |
| CN106989672A (en) | A kind of workpiece measuring based on machine vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |