CN105654097B - The detection method of quadrangle marker in image - Google Patents
The detection method of quadrangle marker in image Download PDFInfo
- Publication number
- CN105654097B CN105654097B CN201511008855.XA CN201511008855A CN105654097B CN 105654097 B CN105654097 B CN 105654097B CN 201511008855 A CN201511008855 A CN 201511008855A CN 105654097 B CN105654097 B CN 105654097B
- Authority
- CN
- China
- Prior art keywords
- line segment
- image
- edge
- detection
- straight line
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of detection methods of quadrangle marker in image, comprising: makees back projection processing to image to be detected as template using the distribution of color in image to be detected in quadrangle label object area, obtains back projection result figure；Edge detection is made to back projection result figure, obtains edge detection results figure；Using preset threshold as condition, the four edges edge line segment for the condition that meets is extracted from edge detection results figure；Calculate the intersection point of straight line where any two adjacent segments in four edges edge line segment；Four vertex of the quadrangle marker are obtained according to four obtained intersection points.Detection method of the present invention screens the edge line segment of extraction by default accumulative threshold value, reduces the straight line number for participating in subsequent calculating, detection speed and detection accuracy can be improved；And back projection processing is made to image to be detected before performing edge detection, can reduce background image to the interference of detection zone and influence, detection accuracy can be improved.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of detection side of quadrangle marker in image
Method.
Background technique
Extraction and detection in machine vision and field of image processing, using the marker of advantages of simple to image information
It is of great significance to, helps to improve the accuracy and speed to image procossing.Rectangle marked object is due to its own shape
Simple structure is easy to the characteristics of being extracted characteristic point and is widely used, in practical applications, rectangle marked object institute in the picture
The shape of display is irregular quadrilateral, and therefore, the identification to rectangle marked object area is exactly in image this irregular four
The detection of side shape.
In recent years, have a few studies personnel to start to expand the detection of quadrilateral area in image research, it is public
The application publication number opened is the patent document of CN103198444A, entitled " detection method of image processing apparatus and rectangle "
In, multiple candidate rectangles are constituted by extracting four straight lines of pairwise orthogonal from a plurality of straight line detected, further according to each
Edge pixel distribution near each angle of a candidate rectangle acquire each angle like angular travel degree, detect rectangle finally to determine.
But above-mentioned detection method have the following disadvantages: participated in since it extracts a plurality of straight line calculate straight line number and
Intersection point number is more, therefore haves the shortcomings that detection speed is slow, precision is low.
Summary of the invention
Detection can be improved compared with the prior art in the detection method that the present invention provides quadrangle marker in a kind of image
Speed and detection accuracy.
To achieve the above object, the invention provides the following technical scheme:
The detection method of quadrangle marker in a kind of image, comprising:
Described image to be detected is made using the distribution of color in image to be detected in quadrangle label object area as template anti-
To projection process, back projection result figure is obtained；
Edge detection is made to the back projection result figure, obtains edge detection results figure；
Using preset threshold as condition, the four edges edge line segment for the condition that meets is extracted from the edge detection results figure；
Calculate the intersection point of straight line where any two adjacent segments in the four edges edge line segment；
Four vertex of the quadrangle marker are obtained according to four obtained intersection points.
Optionally, the distribution of color using in image to be detected in quadrangle label object area is template to described to be checked
Altimetric image makees back projection processing, before further include:
Noise reduction process is made to image to be detected application gaussian filtering, and carries out gradation conversion processing.
Optionally, the distribution of color using in image to be detected in quadrangle label object area is template to described to be checked
Altimetric image makees back projection processing, comprising:
Back projection behaviour is made to described image to be detected as template using the distribution of color in quadrangle label object area
Make, and carries out binary conversion treatment and multiple closed operation processing.
Optionally, described using preset threshold as condition, that the condition that meets is extracted from the edge detection results figure four
Edge line segment, comprising:
Using default accumulative threshold value as condition, the edge detection results figure is made at accumulated probability type straight line Hough transformation
Reason extracts four edge line segments for obtaining meeting condition.
Optionally, the intersection point for calculating straight line where any two adjacent segments in the four edges edge line segment, comprising:
When judging that four edge line segments all have slope, the slope of every edge line segment is calculated；
Four edge line segments are successively sorted according to the size of slope absolute value, are L1, L2, L3, L4；
Straight line and line segment where the intersection point and line segment L2 of straight line where calculating line segment L1 and the place line segment L3, L4 straight line
The intersection point of straight line where L3, L4, obtains four intersection points.
Optionally, the intersection point for calculating straight line where any two adjacent segments in the four edges edge line segment, is also wrapped
It includes:
When judge one in four edge line segments or two lines section there is no when slope, calculate its excess-three item or
There are the slopes of the edge line segment of slope by person two；
By three or two, there are the edge line segments of slope successively to sort according to the size of slope absolute value, is
L1, L2, L3 or L1, L2；
When slope is not present in a line segment, the intersection point of straight line where calculating this line segment and the place line segment L2, L3 straight line,
And the intersection point of straight line where straight line where line segment L1 and line segment L2, L3, obtain four intersection points；
When slope is not present in two lines section, straight line where calculating two lines section respectively with straight line where line segment L1, L2
Intersection point, obtain four intersection points.
Optionally, four intersection points that the basis obtains obtain four vertex of the quadrangle marker, comprising:
The distance that any two intersection point is fastened in image coordinate in four intersection points is calculated, lowest distance value is obtained；
Respectively centered on four intersection points, four pros are set by side length of the half of the lowest distance value
Shape detection zone；
Detection has the harris angle point of maximum eigenvalue respectively in four square detection zones, is determined as institute
State four vertex of quadrangle marker.
It is optionally, described that detection has the angle harris of maximum eigenvalue respectively in four square detection zones
Point is determined as four vertex of the quadrangle marker, comprising:
Sub-pixization processing is made to the four harris angle points detected, using processing result as the quadrangle marker
Four vertex.
Optionally, described to calculate the distance that any two intersection point is fastened in image coordinate in four intersection points, it obtains most
Small distance value, comprising:
Calculated distance value is formed into distance set；
The distance value in the distance set is traversed, lowest distance value is obtained.
Optionally, the lenth ratio of the longest edge of quadrangle marker and most short side described in described image to be detected is little
In 2.
As shown from the above technical solution, in image provided by the present invention quadrangle marker detection method, comprising: with
Distribution of color in image to be detected in quadrangle label object area is that template makees back projection processing to described image to be detected,
Obtain back projection result figure；Edge detection is made to the back projection result figure, obtains edge detection results figure；To preset threshold
Value is condition, and the four edges edge line segment for the condition that meets is extracted from the edge detection results figure；Calculate the four edges edge line
The intersection point of straight line where any two adjacent segments in section；The four of the quadrangle marker is obtained according to four obtained intersection points
A vertex.
The detection method of quadrangle marker in image of the present invention is extracting edge line from edge detection results figure
Duan Shi extracts the four edges edge line segment for the condition that meets using preset threshold as condition by the way that preset threshold is arranged, to be based on extracting
Four line segments arrived calculate and determine four vertex of quadrangle marker, compared with prior art, make the straight line for participating in calculating
Number is few, so that detection speed and detection accuracy can be improved；Also, back projection is made to figure to be detected before performing edge detection
Processing, can reduce background image to the interference of detection zone and influence, detection accuracy can be improved.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow chart of the detection method of quadrangle marker in a kind of image provided in an embodiment of the present invention；
Fig. 2 is the flow chart of step S103 in the embodiment of the present invention；
Fig. 3 is the flow chart of step S104 in the embodiment of the present invention.
Specific embodiment
Technical solution in order to enable those skilled in the art to better understand the present invention, below in conjunction with of the invention real
The attached drawing in example is applied, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described implementation
Example is only a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, this field is common
Technical staff's every other embodiment obtained without making creative work, all should belong to protection of the present invention
Range.
Referring to FIG. 1, Fig. 1 is the stream of the detection method of quadrangle marker in a kind of image provided in an embodiment of the present invention
Cheng Tu.
The detection method of quadrangle marker in image provided by the present embodiment need to meet the following conditions in application:
(1) color in quadrangle label object area and the color of background image have obvious differentiation；(2) quadrangle marks object area
Interior distribution of color will more uniformly；(3) lenth ratio of the longest edge of quadrangle marker and most short side is not more than 2；(4) four
Area ratio/occupancy ratio of the side shape marker in entire image to be detected cannot be too small.
The detection method comprises the following steps for the present embodiment:
S100: being template to described image to be detected using the distribution of color in image to be detected in quadrangle label object area
Make back projection processing, obtains back projection result figure.
Before carrying out the step, first image to be detected of acquisition is pre-processed, is specifically included: using gaussian filtering pair
Image to be detected makees noise reduction process, then carries out gradation conversion processing to it and be converted to gray level image.Work can be used in the present embodiment
Industry camera acquires image, and acquisition camera is required to have a certain distance away from rectangle marked object when acquiring image.
It is template to be checked using the distribution of color in quadrangle label object area after having carried out pretreatment to image to be detected
Altimetric image makees backprojection operations, and result is carried out binary conversion treatment and multiple closed operation processing, obtains back projection result
Figure.
S101: edge detection is made to the back projection result figure, obtains edge detection results figure.
S102: using preset threshold as condition, the four edges edge line for the condition that meets is extracted from the edge detection results figure
Section.
Specifically, making the processing of accumulated probability type straight line Hough transformation to obtained edge detection results figure, wherein pass through conjunction
The default accumulative threshold value of reason setting extracts the four edges edge line segment for obtaining meeting threshold condition using default accumulative threshold value as condition, right
The each edge of marker is answered only to extract the edge line segment for meeting condition.
Therefore compared with the prior art in detection method, the present embodiment the method is by default accumulative threshold value, to mentioning
The edge line segment taken is screened, and is only extracted four edge line segments for meeting threshold condition, is made the line segment for participating in subsequent calculating
Number is reduced, and can reduce calculation amount, improves detection speed.
S103: the intersection point of straight line where any two adjacent segments in the four edges edge line segment is calculated.
In the step, referring to FIG. 2, first determining whether to detect before the intersection point of straight line where calculating acquisition edge line segment
Four edges edge line segment out whether there is slope, and judging result includes following two situation: (1) four edges edge line segment all exists
Slope；(2) having one or two in four edges edge line segment, there is no slopes.
When being judged as the first situation, i.e., when four edges edge line segment all has slope, calculate the detailed process of intersection point such as
Under:
S200: the slope of every edge line segment is calculated；
Specifically, the slope k of i-th line sectioniBy the starting point coordinate (x of the line segmenti、yi) and terminal point coordinate (xi′、yi'), root
According to calculation formulaBe calculated, wherein starting point coordinate and terminal point coordinate by previous step to edge detection knot
Fruit figure makees accumulated probability type straight line Hough transformation and handles to obtain.
S201: four edge line segments are successively sorted according to the size of slope absolute value, are L1, L2, L3, L4；
Straight line where S202: the intersection point of straight line where straight line and line segment L3, L4 where calculating line segment L1 and line segment L2 with
The intersection point of straight line where line segment L3, L4, obtains described four intersection points A1, A2, A3, A4.
When being judged as second situation, i.e., there are one or two lines section in four edges edge line segment there is no when slope,
Calculating process is specific as follows:
S300: calculating its excess-three item or two, there are the slopes of the edge line segment of slope；
The calculation method of slope is same as above, specifically: the slope k of i-th line sectioniBy the starting point coordinate (x of the line segmenti、yi) and
Terminal point coordinate (xi′、yi'), according to calculation formulaIt is calculated, wherein starting point coordinate and terminal point coordinate are by upper one
Make accumulated probability type straight line Hough transformation to edge testing result figure in step to handle to obtain.
S301: by three or two, there are the edge line segments of slope successively to arrange according to the size of slope absolute value
Sequence is L1, L2, L3 or L1, L2；
S302: when slope is not present in a line segment, straight line where calculating this line segment and straight line where line segment L2, L3
The intersection point of intersection point and line segment L1 place straight line and the place line segment L2, L3 straight line, obtains described four intersection points A1, A2, A3, A4；
When slope is not present in two lines section, two lines section place straight line intersection point with straight line where line segment L1, L2 respectively is calculated, is obtained
Obtain described four intersection points A1, A2, A3, A4.
When slope is not present in a line segment, the intersection point of straight line where calculating this line segment and the place line segment L2, L3 straight line,
And the intersection point of straight line where straight line where line segment L1 and line segment L2, L3, obtain four intersection points；When slope is not present in two lines section
When, two lines section place straight line intersection point with straight line where line segment L1, L2 respectively is calculated, four intersection points are obtained
Therefore detection method described in the present embodiment, where calculating it in the edge line segment according to extraction when the intersection point of straight line,
By comparing the slope size between each edge line segment, to prejudge the positional relationship between each line segment, (opposite side faces frontier juncture
System), it then only calculates the intersection point faced between the section of sideline and obtains four intersection points, so that the intersection point number for participating in subsequent calculating be made to subtract
It is few.And detection method in the prior art needs to calculate the intersection point between all line segments extracted, participates in the friendship of subsequent calculating
Point number is more, will lead to heavy workload, speed is slow, therefore compared with prior art, and detection method described in the present embodiment can make to count
Operator workload substantially reduces, and detection speed can be improved.
S104: four vertex of the quadrangle marker are obtained according to four obtained intersection points.
In the previous step, four intersection points are obtained by the intersection point of straight line where calculating any two edges line segment, is obtained
Four intersecting point coordinates obtained are the rough estimate values to four vertex of quadrangle marker, are obtained in this step based on upper step
To four intersection points four vertex of quadrangle marker are accurately measured.
Referring to FIG. 3, specifically includes the following steps:
S400: calculating the distance that any two intersection point is fastened in image coordinate in four intersection points, obtains minimum range
Value.
It is specifically included in this step: calculating any two intersection point in four intersection points A1, A2, A3, A4 and fastened in image coordinate
Distance, by calculated distance value form distance set D；The distance value in the distance set D is traversed, most narrow spacing is obtained
From value, it is denoted as min { D }.
S401: respectively centered on four intersection points, using the half of the lowest distance value as side length setting four
A square detection zone.It is that side length is being arranged four just with min { D }/2 i.e. respectively centered on four intersection points A1, A2, A3, A4
Rectangular detection zone.
S402: detection has the harris angle point of maximum eigenvalue respectively in four square detection zones, really
It is set to four vertex of the quadrangle marker.
Specifically include in this step: detection has the harris of maximum eigenvalue respectively in four square detection zones
Angle point makees sub-pixization processing to the four harris angle points detected, using processing result as the quadrangle marker
Four vertex, obtain the fine estimation on four vertex of quadrangle marker, to complete to quadrangle marker
Detection.
Therefore detection method described in the present embodiment, when detecting harris angle point, can be obtained according to rough estimate four
Harris Corner Detection region is reasonably arranged in a intersection point, to reduce detection range, exclusive PCR harris angle point.Compared to existing
The some methods for needing manual setting detection zone having, have certain intelligence and convenience.Simultaneously in detection zone
Detection harris angle point can guarantee detection accuracy.
Therefore, in a kind of image provided in this embodiment quadrangle marker detection method, by marker four
The detection on vertex can preferably obtain the shape feature of marker, vertex be detected as harris angle point, energy
Enough guarantee detection accuracy.
When extracting edge line segment, by the way that accumulative threshold value is rationally arranged, the line segment of extraction is screened, correspondence markings
The each edge of object only extracts the edge line segment for meeting condition, keeps the line segment number for participating in calculating less, reduces calculating
Detection speed can be improved in workload.
At four vertex of rough estimate marker, (opposite side faces the positional relationship that can judge in advance between line segment
Frontier juncture system), the intersection point between straight line where facing sideline section is only calculated, keeps the intersection point number for participating in subsequent calculating few, helps to improve
Detect speed.
At accurately four vertex of estimation marker, can be rationally arranged according to the rough estimate position on four vertex
Detection zone reduces the detection range of harris angle point, further increases detection accuracy.
In addition, detection method described in the present embodiment, handles image to be detected with back projection by ingenious, it will be with label
Object area color has the background obviously distinguished to filter, therefore makes interference very little of the detection process by complex background.
The detection method of quadrangle marker in a kind of image provided by the present invention is described in detail above.This
Apply that a specific example illustrates the principle and implementation of the invention in text, the explanation of above example is only intended to
It facilitates the understanding of the method and its core concept of the invention.It should be pointed out that for those skilled in the art,
Without departing from the principles of the invention, can be with several improvements and modifications are made to the present invention, these improvement and modification are also fallen
Enter in the protection scope of the claims in the present invention.
Claims (8)
1. the detection method of quadrangle marker in a kind of image characterized by comprising
Reversed throw is made to described image to be detected using the distribution of color in image to be detected in quadrangle label object area as template
Shadow processing, obtains back projection result figure；
Edge detection is made to the back projection result figure, obtains edge detection results figure；
Using preset threshold as condition, the four edges edge line segment for the condition that meets is extracted from the edge detection results figure；
Calculate the intersection point of straight line where any two adjacent segments in the four edges edge line segment；
Four vertex of the quadrangle marker are obtained according to four obtained intersection points；
The intersection point for calculating straight line where any two adjacent segments in the four edges edge line segment, comprising:
When judging that four edge line segments all have slope, the slope of every edge line segment is calculated；
Four edge line segments are successively sorted according to the size of slope absolute value, are L1, L2, L3, L4；
Straight line and line segment L3, L4 where the intersection point and line segment L2 of straight line where calculating line segment L1 and the place line segment L3, L4 straight line
The intersection point of place straight line obtains four intersection points；
The intersection point for calculating straight line where any two adjacent segments in the four edges edge line segment, further includes:
When judging one in four edge line segments, perhaps there is no calculate its excess-three item or two when slope for two lines section
There are the slopes of the edge line segment of slope for item；
By three or two, there are the edge line segments of slope successively to sort according to the size of slope absolute value, be L1, L2,
L3 or L1, L2；
When slope is not present in a line segment, the intersection point of straight line where calculating this line segment and the place line segment L2, L3 straight line, and
The intersection point of straight line where line segment L1 and the place line segment L2, L3 straight line, obtains four intersection points；
When slope is not present in two lines section, the friendship with straight line where line segment L1, L2 respectively of straight line where calculating the two lines section
Point obtains four intersection points.
2. detection method as described in claim 1, which is characterized in that described to mark object area with quadrangle in image to be detected
Interior distribution of color is that template makees back projection processing to described image to be detected, before further include:
Noise reduction process is made to image to be detected application gaussian filtering, and carries out gradation conversion processing.
3. detection method as described in claim 1, which is characterized in that described to mark object area with quadrangle in image to be detected
Interior distribution of color is that template makees back projection processing to described image to be detected, comprising:
Backprojection operations are made to described image to be detected as template using the distribution of color in quadrangle label object area, and
Carry out binary conversion treatment and multiple closed operation processing.
4. detection method as described in claim 1, which is characterized in that it is described using preset threshold as condition, it is examined from the edge
It surveys in result figure and extracts the four edges edge line segment for the condition that meets, comprising:
Using default accumulative threshold value as condition, the processing of accumulated probability type straight line Hough transformation is made to the edge detection results figure, is mentioned
Obtain four edge line segments of the condition of satisfaction.
5. detection method as described in claim 1, which is characterized in that four intersection points that the basis obtains obtain four side
Four vertex of shape marker, comprising:
The distance that any two intersection point is fastened in image coordinate in four intersection points is calculated, lowest distance value is obtained；
Respectively centered on four intersection points, four square inspections are set by side length of the half of the lowest distance value
Survey region；
Detection has the harris angle point of maximum eigenvalue respectively in four square detection zones, is determined as described four
Four vertex of side shape marker.
6. detection method as claimed in claim 5, which is characterized in that described to distinguish in four square detection zones
The harris angle point with maximum eigenvalue is detected, four vertex of the quadrangle marker are determined as, comprising:
Sub-pixization processing is made to the four harris angle points detected, using processing result as the institute of the quadrangle marker
State four vertex.
7. detection method as claimed in claim 5, which is characterized in that described to calculate any two intersection point in four intersection points
In the distance that image coordinate is fastened, lowest distance value is obtained, comprising:
Calculated distance value is formed into distance set；
The distance value in the distance set is traversed, lowest distance value is obtained.
8. such as the described in any item detection methods of claim 1-7, which is characterized in that quadrangle described in described image to be detected
The lenth ratio of the longest edge of marker and most short side is not more than 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511008855.XA CN105654097B (en) | 2015-12-29 | 2015-12-29 | The detection method of quadrangle marker in image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511008855.XA CN105654097B (en) | 2015-12-29 | 2015-12-29 | The detection method of quadrangle marker in image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105654097A CN105654097A (en) | 2016-06-08 |
| CN105654097B true CN105654097B (en) | 2019-04-16 |
Family
ID=56477280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511008855.XA Expired - Fee Related CN105654097B (en) | 2015-12-29 | 2015-12-29 | The detection method of quadrangle marker in image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105654097B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110097065A (en) * | 2019-05-07 | 2019-08-06 | 厦门商集网络科技有限责任公司 | A kind of line detection method and terminal based on FreeMan chain code |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127778B (en) * | 2016-06-27 | 2019-01-04 | 安徽慧视金瞳科技有限公司 | It is a kind of for projecting the line detection method of interactive system |
| CN108665495B (en) * | 2017-03-30 | 2021-11-26 | 展讯通信（上海）有限公司 | Image processing method and device and mobile terminal |
| CN108304840B (en) * | 2017-08-31 | 2022-11-11 | 腾讯科技（深圳）有限公司 | Image data processing method and device |
| CN108109169B (en) * | 2017-12-12 | 2021-12-14 | 深圳市神州云海智能科技有限公司 | Pose estimation method and device based on rectangular identifier and robot |
| CN110136156B (en) * | 2018-02-02 | 2020-12-01 | 北京三快在线科技有限公司 | Polygonal area detection method and device |
| CN108734743A (en) * | 2018-04-13 | 2018-11-02 | 深圳市商汤科技有限公司 | Method, apparatus, medium and electronic equipment for demarcating photographic device |
| CN108805124B (en) * | 2018-04-18 | 2019-10-18 | 北京嘀嘀无限科技发展有限公司 | Image processing method and device, computer readable storage medium |
| CN109214396A (en) * | 2018-08-24 | 2019-01-15 | 国网安徽省电力有限公司阜阳供电公司 | A kind of industrial equipment image characteristic extracting method and equipment |
| CN109949211A (en) * | 2019-03-07 | 2019-06-28 | 北京麦哲科技有限公司 | A kind of rectangle file and picture cutting method and device |
| CN110018174B (en) * | 2019-05-22 | 2021-10-22 | 合肥联宝信息技术有限公司 | Method and device for detecting object appearance |
| CN110751631B (en) * | 2019-10-10 | 2023-04-07 | 郑州大学 | Quick and high-precision rectangle detection method |
| CN112132163B (en) * | 2020-09-21 | 2024-04-02 | 杭州睿琪软件有限公司 | Method, system and computer readable storage medium for identifying object edges |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101727581A (en) * | 2009-12-10 | 2010-06-09 | 上海名图软件有限公司 | Plate number tilt correcting method based on character pre-cut |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4801551B2 (en) * | 2006-09-27 | 2011-10-26 | 富士通株式会社 | Image area detection method, program, and apparatus |
-
2015
- 2015-12-29 CN CN201511008855.XA patent/CN105654097B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101727581A (en) * | 2009-12-10 | 2010-06-09 | 上海名图软件有限公司 | Plate number tilt correcting method based on character pre-cut |
Non-Patent Citations (3)
| Title |
|---|
| 四边形分类识别算法;黄柳;《信息技术》;20100630(第06期);第58-60页 |
| 基于Hough变换的四边形分类识别算法研究;刘智;《广西计算机学会2010年学术年会论文集》;20100930;第194页第1段-198页倒数第1段 |
| 直方图反向投影多目标检测优化算法;张辰 等;《计算机系统应用》;20140215;第23卷(第2期);第178页第1段-183页倒数第2段 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110097065A (en) * | 2019-05-07 | 2019-08-06 | 厦门商集网络科技有限责任公司 | A kind of line detection method and terminal based on FreeMan chain code |
| CN110097065B (en) * | 2019-05-07 | 2021-09-10 | 厦门商集网络科技有限责任公司 | Freeman chain code-based line detection method and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105654097A (en) | 2016-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105654097B (en) | The detection method of quadrangle marker in image | |
| CN107871124B (en) | A kind of Remote Sensing Target detection method based on deep neural network | |
| CN106485275B (en) | A method of realizing that cover-plate glass is bonded with liquid crystal display positioning | |
| CN105046245B (en) | Video human face method of determination and evaluation | |
| CN104408460B (en) | A kind of lane detection and tracking detection method | |
| CN102654902B (en) | Contour vector feature-based embedded real-time image matching method | |
| CN102704215B (en) | Automatic cutting method of embroidery cloth based on combination of DST file parsing and machine vision | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN108319949A (en) | Mostly towards Ship Target Detection and recognition methods in a kind of high-resolution remote sensing image | |
| CN108364280A (en) | Structural cracks automation describes and width accurately measures method and apparatus | |
| CN115564771B (en) | Concrete crack identification method based on building foundation column | |
| CN109993099A (en) | A kind of lane line drawing recognition methods based on machine vision | |
| CN108898047A (en) | The pedestrian detection method and system of perception are blocked based on piecemeal | |
| CN107392929B (en) | Intelligent target detection and size measurement method based on human eye vision model | |
| CN108537787A (en) | A kind of quality judging method of facial image | |
| CN106504262A (en) | A kind of small tiles intelligent locating method of multiple features fusion | |
| CN106067031A (en) | Cooperate with the degree of depth learning network Machine Vision Recognition system based on artificial mechanism for correcting errors | |
| CN113096121A (en) | Pavement crack detection method and system based on cross fracture mechanics and image processing | |
| CN104966295B (en) | A kind of naval vessel extracting method based on wire-frame model | |
| CN107292238A (en) | Contour of building recognition methods with end point is constrained based on spatial topotaxy | |
| CN108492306A (en) | A kind of X-type Angular Point Extracting Method based on image outline | |
| CN103699876A (en) | Method and device for identifying vehicle number based on linear array CCD (Charge Coupled Device) images | |
| CN103679724B (en) | Slope approximant straight line detection method | |
| CN108197540A (en) | A kind of fire image Feature extraction and recognition method based on SURF | |
| CN115876786B (en) | Wedge-shaped welding spot detection method and motion control device |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190416 Termination date: 20191229 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |