CN105469401A - Ship groove positioning method based on computer vision - Google Patents
Ship groove positioning method based on computer vision Download PDFInfo
- Publication number
- CN105469401A CN105469401A CN201510826745.8A CN201510826745A CN105469401A CN 105469401 A CN105469401 A CN 105469401A CN 201510826745 A CN201510826745 A CN 201510826745A CN 105469401 A CN105469401 A CN 105469401A
- Authority
- CN
- China
- Prior art keywords
- straight line
- headchute
- image
- line
- edge
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20068—Projection on vertical or horizontal image axis
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
A ship groove positioning method based on a computer vision belongs to the computer vision technology and image processing technology field. The ship groove positioning method comprises the following steps of using a binocular camera to capture an image; carrying out image preprocessing; carrying out mathematical morphology edge detection; carrying out straight line detection; carrying out straight line filtering and clustering; calculating an image depth of each straight line in a plane; projecting a ship groove straight line to a ship groove edge straight line plane; calculating a point of intersection of a ship groove edge line and a ship groove line; and accurately positioning a ship groove. The ship groove positioning method based on the computer vision is used for ship groove positioning. A traditional container loading and unloading working mode which takes a manual guiding as a main point is improved, and a ship container loading and unloading digitization and intelligence level is increased. Production and working environments are improved. The method is good for increasing working efficiency and reducing cost.
Description
Technical field
The invention belongs to computer vision technique and technical field of image processing, be specifically related to a kind of headchute localization method based on computer vision for shipping container handling.
Background technology
Along with developing rapidly of world economic integration and foreign trade economy, because Container Transport has safe and reliable, timely, simple and convenient rapidly feature, the requirement that the haulage time of commodity in foreign trade business is short, damaged less, expense is low can be met, so Container Transport becomes the inevitable development direction of current shipping gradually, thus also become the important symbol of modern trade development.There are corresponding container stacking, transport, loading and unloading operation in the shipping request harbour of container, harbour.Container handling efficiency wherein directly has influence on the market competitiveness of shipping company and harbour handling, and along with the continuous growth of Container Transport amount and the tonnage of container ship increase, the efficiency of loading and unloading of container wharf must be improved.The robotization strengthening container handling is one of effective way improving container handling efficiency, and its core is exactly to shipping container guide rail location (headchute) and casing location, and then guides the subsequent job such as crawl, handling of crane hanger.
In current world's transport field, solid falls apart groceries, liquid fall apart the nearly all cargo type of groceries all start containerzation, in order to realize the robotization of Container Transport, improve the dock container efficiency of loading and unloading and security, the relevant domestic and international researcher in this field has research to this, wherein immediate technical scheme comprises with the present invention: patent of invention (application number: 201410400655.8, title: a kind of container dress, to unload a ship method of operating, shipment operating system) a kind of container handling method cooperated with each other in conjunction with first terminal and the second terminal of middle proposition, the main thought of its aboard ship container handling be by the first terminal at assistant place, the validation of information on ship is clear after, positional information is sent to graphical server, bank bridge operating personnel obtain the positional information in graphical server by the second terminal, carry out the handling of container again, the method and traditional single authoritative method have same shortcoming, namely in the cargo handling process of whole container, too depend on the experience of staff, patent of invention (application number: 201410333783.5, title: a kind of be used for container distribution profile and position intelligent Detection and method) in the distribution profile that proposes to utilize 3 D laser scanning detected set to case and positional information, realize the accurate to case and case of container handling process, the method uses three-dimensional laser scanner to carry out the contour detecting of container, but when there is no container in the container position in storehouse of boats and ships, the profile of container cannot be detected, just cannot carry out the container loading on boats and ships, moreover the cost of three-dimensional laser scanner is too high.
In sum, when the handling realizing shipping container, existing method cannot depart from container itself completely and operate, and too much depend on the experience of operating personnel, particularly when boats and ships are clean ship, how accurately container being placed on the headchute position of specifying, this is a vital job for whole container handling, the present invention is directed to this deficiency and proposes a kind of headchute localization method based on computer vision.
Summary of the invention
The object of the invention is to solve the problem, provide a kind of headchute localization method based on computer vision, greatly facilitate the container handling operation of driver remote manipulation boats and ships, improve shipping container handling digitizing and intelligent level, improve operating efficiency and improve the security of container handling work.
Concrete technical scheme of the present invention is as follows:
Described a kind of headchute localization method based on computer vision, it is characterized in that comprising binocular camera catches image, Image semantic classification, Morphology edge detection, straight-line detection, straight line filters and cluster, the picture depth of calculating each straight line place plane, headchute linear projection to headchute edge line plane, ask the intersection point of headchute edge line and headchute line and headchute accurately to locate.
Described a kind of headchute localization method based on computer vision, is characterized in that specifically comprising the following steps:
Step 1: use the binocular vision system of known internal reference to headchute imaging, obtain binocular image I
1and I
2;
Step 2: to image I
1and I
2carry out Image semantic classification, comprise image gray processing process, image denoising sonication, image sharpening process and image Gaussian smoothing, obtain image f
1and f
2;
Step 3: utilize mathematical morphology to carry out rim detection to the image that step 2 obtains, concrete steps are as follows:
Step 3.1: in order to suppress the negative noise in image, according to formula (1), utilizes morphological dilations and closed operation operation respectively to image f
1with image f
2process, obtain image
, wherein i=1,2;
Step 3.2: in order to suppress the positive noise in image, according to formula (2), utilizes morphological erosion and opening operation operation respectively to image f
1with image f
2process, obtain image
, wherein i=1,2;
Step 3.3: according to formula (3), Multi-structure elements Denoising disposal is carried out to image, finally obtains image
, wherein i=1,2;
Step 3.4: the Image Edge-Detection operator in convolution (1), (2) and (3) obtains last edge-detected image G
i, specifically see formula (4);
Wherein
E in above-mentioned formula (1), (2) and (3)
1, e
2and e
3represent the structural element that morphologic detection is used,
represent etching operation,
represent expansive working,
represent opening operation operation,
represent closed operation operation;
Step 4: utilize Hough transformation line detection method to the headchute edge image G obtained in step 3
icarry out straight-line detection, obtain straight line set
, M is the straight line sum detected; Further to straight line set L
icarry out straight line filtration, obtain headchute anastatic water horizontal line Lh respectively
iwith headchute perpendicular line set Lv
i, specific as follows:
Step 4.1: detect the horizontal line at headchute edge, first by judging straight line
slope, filtering is not the straight line of horizontal edge, namely retains the straight line that straight slope meets-0.087≤K≤0.087, obtains straight line candidates set
, N
hfor filtering the sum of rear straight line; Secondly to straight line candidates set Lsh
icarry out straight line cluster, by slope in straight line candidates set, the identical and straight line that distance d between parallel lines meets d≤4 gathers is a class, obtains straight line polymeric type
,
represent r straight line polymeric type,
for the quantity of straight line polymeric type, recycling least square method carries out matching to the straight line of each straight line polymeric type respectively, and result is straight line set
,
for the straight line sum after fitting a straight line,
represent by straight line polymeric type
the straight line that matching obtains; Finally utilize at image I
istraight line in region
the pixel gray-scale value of process form the variance of collection joint account this set, obtain straight line set Lfh further
icorresponding gray-scale value variance set Var
i, obtain Lfh
ithe minimum straight line of middle gray-scale value variance is headchute horizontal edge straight line
;
Step 4.2: detect headchute perpendicular line, first by judging straight line
slope, filtering is not the straight line of vertical edge, namely retains straight slope and meets
straight line, obtain straight line candidates set further
, N
vfor filtering the sum of rear straight line; Secondly to straight line candidates set Lsv
icarry out straight line cluster, by slope in straight line candidates set, the identical and straight line that distance d between parallel lines meets d≤4 gathers is a class, obtains straight line polymeric type
,
represent r straight line polymeric type,
for the sum of straight line polymeric type, recycling least square method carries out fitting a straight line to each straight line polymeric type respectively, obtains the straight line set after matching further
,
for the straight line sum after fitting a straight line;
Step 5: the headchute horizontal edge straight line Lh utilizing step 4 to obtain
iwith headchute perpendicular line set Lv
i, calculate its picture depth relative to camera coordinates system, the headchute horizontal edge straight line place plane of delineation and the headchute perpendicular line place plane of delineation, specific as follows:
Step 5.1: calculate headchute horizontal edge straight line Lh
ithe picture depth of place plane, first at headchute horizontal edge straight line Lh
ion get a little
, recycling computer vision technique, calculation level
picture depth be
;
Step 5.2: calculate headchute perpendicular line set Lv
ithe picture depth of cathetus place plane, first at headchute perpendicular line
on get a little
, recycling computer vision technique, calculation level
picture depth be
, obtain the picture depth set of headchute perpendicular line further
;
Step 5.3: calculated a little by above-mentioned
picture depth
with perpendicular line set Lv
ipicture depth set
, obtain straight line Lh
iplace plane and straight line set Lv
ipicture depth difference between the plane of place is
;
Step 6: the headchute horizontal edge straight line Lh obtained by step 5
iwith headchute vertical line L v
ibetween picture depth poor
, by straight line
project to straight line Lh
iplace plane, obtains straight line
, recycling two headchute perpendicular line is be that axis of symmetry makes symmetrical feature with picture centre, obtains two headchute perpendicular line
;
Step 7: further, the headchute horizontal edge straight line Lh that calculation procedure 6 obtains
iand with Lh
ibe positioned at conplane headchute perpendicular line
intersection point, obtain intersection point set
;
Step 8: the straight-line intersection coordinate set utilizing step 7 to obtain is demarcated binocular vision system, solves the outer ginseng of video camera, obtains the pixel coordinate in image
coordinate p corresponding in world coordinate system
1, p
2, the headchute image I that other one group of binocular camera is caught
3and I
4, calculate coordinate p by above-mentioned steps 1 to the method for step 7
3, p
4, finally utilize four coordinate points p
1, p
2, p
3, p
4build a rectangle frame, be the particular location of headchute.
Beneficial effect of the present invention is: the headchute localization method based on computer vision that the present invention limits, for the location of headchute, improve tradition based on the container handling mode of operation manually guided, improve shipping container handling digitizing and intelligent level, be one to improve and produce and the multi-win engineering of working environment, to improving operating efficiency, reduce costs highly beneficial.
Accompanying drawing explanation
Fig. 1 is headchute localization method process flow diagram;
The former figure that Fig. 2 (a) is headchute in embodiment;
Fig. 2 (b) is headchute edge detection graph in embodiment;
Fig. 3 (a) is headchute horizontal line fitting result figure in embodiment;
Fig. 3 (b) is headchute horizontal line result figure in embodiment;
Fig. 3 (c) is headchute vertical line fitting result figure in embodiment;
Fig. 3 (d) is headchute perpendicular line result figure in embodiment;
Fig. 4 is headchute anastatic water horizontal line and headchute perpendicular line schematic diagram in embodiment;
Fig. 5 is headchute location schematic diagram in embodiment.
Embodiment
Below in conjunction with Figure of description, the present invention is further detailed.Should be appreciated that specific embodiment described herein only for explaining the present invention, being not intended to limit the present invention.
Fig. 1 is headchute localization method overall flow figure according to an embodiment of the invention.
As shown in Figure 1, according to the process flow diagram of the embodiment of the present invention, headchute localization method comprises that binocular camera catches image, Image semantic classification, Morphology edge detection, straight-line detection, straight line filters and cluster, the picture depth (i.e. the picture depth of each straight line place plane) of calculated line plane, headchute linear projection to headchute edge line plane, ask the intersection point of headchute edge line and headchute line, headchute accurately locates.
A kind of headchute localization method of the present invention, specifically takes following steps:
Step 1: use the binocular vision system of known internal reference to headchute imaging, obtain binocular image I
1and I
2, be only described in detail in image I in the present embodiment
1middle location headchute, image I
2processing procedure is similar, and Fig. 2 (a) is the former figure of boats and ships headchute in the present embodiment;
Step 2: to image I
1carry out Image semantic classification, comprise image gray processing process, image denoising sonication, image sharpening process and image Gaussian smoothing, obtain image f
1;
Step 3: utilize mathematical morphology to carry out rim detection to the image that step 2 obtains, specific as follows:
Step 3.1: in order to suppress the negative noise in image, according to formula (1), utilizes morphological dilations and closed operation operation to image f
1process, obtain image G
1;
Step 3.2: in order to suppress the positive noise in image, according to formula (2), utilizes morphological erosion and opening operation operation to image f
1process, obtain image G
2;
Step 3.3: according to formula (3), Multi-structure elements Denoising disposal is carried out to image, finally obtain image G
3;
Step 3.4: the Image Edge-Detection operator in convolution (1), (2) and (3) obtains last edge-detected image G, specifically sees formula (4);
Wherein
Fig. 2 (b) is the mathematical morphology headchute edge detection graph in the present embodiment.
E in above-mentioned formula (1), (2) and (3)
1, e
2and e
3represent the structural element that morphologic detection is used,
represent etching operation,
represent expansive working,
represent opening operation operation,
represent closed operation operation; In the present embodiment
,
;
Step 4: utilize Hough transformation line detection method to carry out straight-line detection to the headchute edge image G obtained in step 3, obtain straight line set
, M is the straight line sum detected.Then straight line filtration is carried out to straight line set L, obtain ship anastatic water horizontal line Lh and headchute perpendicular line set Lv respectively, specific as follows:
Step 4.1: detect the horizontal line at headchute edge, first by judging straight line L
jslope, filtering is not the straight line of horizontal edge, namely retains the straight line of straight slope in the scope of-0.087≤K≤0.087, obtains straight line candidates set
, N
hfor filtering the sum of rear straight line; Secondly carry out straight line cluster to straight line candidates set Lsh, by slope in straight line candidates set, the identical and straight line that distance d between parallel lines meets d≤4 gathers is a class, obtains straight line polymeric type
, Lch
rrepresent r straight line polymeric type,
for the quantity of straight line polymeric type, recycling least square method carries out matching to the straight line of each straight line polymeric type respectively, and result is straight line set
,
for the straight line sum after fitting a straight line, Lfh
rrepresent by straight line polymeric type Lch
rthe straight line that matching obtains; Finally utilize at image I
1straight line Lfh in region
rthe pixel gray-scale value of process form the variance of collection joint account this set, obtain the gray-scale value variance set Var that straight line set Lfh is corresponding further, obtain the straight line that in Lfh, gray-scale value variance is minimum and be headchute horizontal edge straight line Lh, Fig. 3 (a) is the headchute horizontal line fitted figure in the present embodiment, and Fig. 3 (b) is the headchute horizontal edge straight line result figure in the present embodiment;
Step 4.2: detect headchute perpendicular line, first by judging straight line
slope, filtration is not the straight line of vertical edge, and namely retain straight slope in K≤-11.43, K>=11.43, the straight line in scope, obtains straight line candidates set further
, N
vfor filtering the sum of rear straight line.Secondly carry out straight line cluster to straight line candidates set Lsv, by slope in straight line candidates set, the identical and straight line that distance d between parallel lines meets d≤4 gathers is a class,
, Lcv
rrepresent r straight line polymeric type,
for the sum of straight line polymeric type, again utilize least square method to carry out fitting a straight line to these straight line classes respectively, result is the straight line set after a matching
,
for after fitting a straight line straight line sum, Fig. 3 (c) for headchute perpendicular line fitting result figure, Fig. 3 (d) in the present embodiment be headchute perpendicular line result figure in the present embodiment;
Step 5: the headchute anastatic water flat line utilizing step 4 to obtain and headchute perpendicular line set Lv, calculate it relative to camera coordinates system, the picture depth of place, headchute edge plane of delineation S1 and headchute perpendicular line place plane of delineation S2 is specific as follows:
Step 5.1: the picture depth calculating headchute anastatic water flat line Lh place planar S 1, first gets a p on headchute anastatic water flat line Lh
h, recycling computer vision technique, calculation level p
hpicture depth be z
h;
Step 5.2: the picture depth calculating headchute perpendicular line set Lv cathetus place planar S 2, first at headchute vertical line L v
ron to get be some p
r, recycling computer vision technique, calculation level p
rpicture depth, finally obtain the set of a vertical line picture depth
;
Step 5.3: calculate a p by above-mentioned
hpicture depth z
hwith perpendicular line picture depth set z
v, the picture depth difference obtained between straight line Lh place planar S 1 and straight line set Lv place planar S 2 is
, Fig. 4 is the headchute edge line place plane detected in the drawings in the present embodiment and headchute perpendicular line place floor map;
Step 6: the headchute horizontal edge straight line Lh obtained by step 5 and headchute perpendicular line
between picture depth poor
, by straight line
on coordinate conversion be in conplane coordinate to straight line Lh, obtain straight line Ls
r, recycling two headchute lines is be that axis of symmetry Y makes symmetrical feature with picture centre, the vertical line L s of 2 headchutes that result obtains
1, Ls
2;
Step 7: the headchute edge line Lh utilizing step 6 to obtain and be positioned at conplane headchute vertical line L s with straight line Lh
1, Ls
2intersection point, finally obtain intersection point set
, Fig. 5 is the summit schematic diagram of headchute location in the present embodiment;
Step 8: utilize above step, can obtain binocular image I
2the intersection point of middle ship edge line and headchute perpendicular line
, then the image I that integrating step 7 obtains
1in straight-line intersection set
, demarcate binocular vision system, solve the outer ginseng of video camera, the coordinate obtaining the pixel coordinate correspondence in world coordinate system in image is p
1, p
2, the image I that other one group of binocular camera is caught
3and I
4, calculate coordinate p by above-mentioned method
3, p
4, finally utilize four coordinate points p
1, p
2, p
3, p
4build a rectangle frame, be the particular location of headchute.
After implementing the present invention, for the location of headchute, tradition can be improved based on the container handling mode of operation manually guided, improve shipping container handling digitizing and intelligent level, be a multi-win engineering improving production and working environment, to raising operating efficiency, improve container handling security and reduce costs highly beneficial.
Content described in this instructions embodiment is only enumerating the way of realization of inventive concept; should not being regarded as of protection scope of the present invention is only limitted to the concrete form that embodiment is stated, protection scope of the present invention also and conceive the equivalent technologies means that can expect according to the present invention in those skilled in the art.
Claims (7)
1. based on a headchute localization method for computer vision, it is characterized in that comprising binocular camera catches image, Image semantic classification, Morphology edge detection, straight-line detection, straight line filters and cluster, the picture depth of calculating each straight line place plane, headchute linear projection to headchute edge line plane, ask the intersection point of headchute edge line and headchute line and headchute accurately to locate.
2. a kind of headchute localization method based on computer vision according to claim 1, is characterized in that comprising the following steps:
Step 1: use the binocular vision system of known internal reference to headchute imaging, obtain binocular image I
1and I
2;
Step 2: to image I
1and I
2carry out Image semantic classification, obtain image f
1and f
2;
Step 3: utilize mathematical morphology to carry out rim detection to the image that step 2 obtains, obtain headchute edge image G
i;
Step 4: utilize Hough transformation line detection method to the headchute edge image G obtained in step 3
icarry out straight-line detection, obtain straight line set
, M is the straight line sum detected; Further to straight line set L
icarry out straight line filtration, obtain headchute anastatic water horizontal line Lh respectively
iwith headchute perpendicular line set Lv
i;
Step 5: the headchute horizontal edge straight line Lh utilizing step 4 to obtain
iwith headchute perpendicular line set Lv
i, calculate its picture depth relative to camera coordinates system, the headchute horizontal edge straight line place plane of delineation and the headchute perpendicular line place plane of delineation, obtain headchute horizontal edge straight line Lh
iwith headchute vertical line L v
ibetween picture depth poor
;
Step 6: the headchute horizontal edge straight line Lh obtained by step 5
iwith headchute vertical line L v
ibetween picture depth poor
, by straight line
project to straight line Lh
iplace plane, obtains straight line
, recycling two headchute perpendicular line is be that axis of symmetry makes symmetrical feature with picture centre, obtains two headchute perpendicular line
;
Step 7: the headchute horizontal edge straight line Lh that calculation procedure 6 obtains
iand with Lh
ibe positioned at conplane headchute perpendicular line
intersection point, obtain intersection point set
;
Step 8: the straight-line intersection coordinate set utilizing step 7 to obtain is demarcated binocular vision system, solves the outer ginseng of video camera, obtains the pixel coordinate in image
coordinate points p corresponding in world coordinate system
1, p
2, the headchute image I that other one group of binocular camera is caught
3and I
4, calculate coordinate points p by step 1 to the method for step 7
3, p
4, finally utilize four coordinate points p
1, p
2, p
3, p
4build a rectangle frame, be the particular location of headchute.
3. a kind of headchute localization method based on computer vision according to claim 2, is characterized in that step 2) in Image semantic classification comprise image gray processing process, image denoising sonication, image sharpening process and image Gaussian smoothing.
4. a kind of headchute localization method based on computer vision according to claim 2, it is characterized in that the image in step 3) carries out rim detection, concrete steps are as follows:
Step 3.1: in order to suppress the negative noise in image, according to formula (1), utilizes morphological dilations and closed operation operation respectively to image f
1with image f
2process, obtain image
, wherein i=1,2;
Step 3.2: in order to suppress the positive noise in image, according to formula (2), utilizes morphological erosion and opening operation operation respectively to image f
1with image f
2process, obtain image
, wherein i=1,2;
Step 3.3: according to formula (3), Multi-structure elements Denoising disposal is carried out to image, finally obtains image
, wherein i=1,2;
Step 3.4: the Image Edge-Detection operator in convolution (1), (2) and (3) obtains last edge-detected image G
i, specifically see formula (4);
Wherein
E in above-mentioned formula (1), (2) and (3)
1, e
2and e
3represent the structural element that morphologic detection is used,
represent etching operation,
represent expansive working,
represent opening operation operation,
represent closed operation operation.
5. a kind of headchute localization method based on computer vision according to claim 2, is characterized in that in step 4) as follows to the horizontal line detection method at headchute edge: first by judging straight line
slope, filtering is not the straight line of horizontal edge, namely retains the straight line that straight slope meets-0.087≤K≤0.087, obtains straight line candidates set
, N
hfor filtering the sum of rear straight line; Secondly to straight line candidates set Lsh
icarry out straight line cluster, by slope in straight line candidates set, the identical and straight line that distance d between parallel lines meets d≤4 gathers is a class, obtains straight line polymeric type
,
represent r straight line polymeric type,
for the quantity of straight line polymeric type, recycling least square method carries out matching to the straight line of each straight line polymeric type respectively, and result is straight line set
,
for the straight line sum after fitting a straight line,
represent by straight line polymeric type
the straight line that matching obtains; Finally utilize at image I
istraight line in region
the pixel gray-scale value of process form the variance of collection joint account this set, obtain straight line set Lfh further
icorresponding gray-scale value variance set Var
i, obtain Lfh
ithe minimum straight line of middle gray-scale value variance is headchute horizontal edge straight line
.
6. a kind of headchute localization method based on computer vision according to claim 2, is characterized in that in step 4) as follows to headchute perpendicular line detection method: first by judging straight line
slope, filtering is not the straight line of vertical edge, namely retains straight slope and meets
straight line, obtain straight line candidates set further
, N
vfor filtering the sum of rear straight line; Secondly to straight line candidates set Lsv
icarry out straight line cluster, by slope in straight line candidates set, the identical and straight line that distance d between parallel lines meets d≤4 gathers is a class, obtains straight line polymeric type
,
represent r straight line polymeric type,
for the sum of straight line polymeric type, recycling least square method carries out fitting a straight line to each straight line polymeric type respectively, obtains the straight line set after matching further
,
for the straight line sum after fitting a straight line, be headchute perpendicular line set Lv
i.
7. a kind of headchute localization method based on computer vision according to claim 2, it is characterized in that calculating in step 5 in step 5) its picture depth relative to camera coordinates system, the headchute horizontal edge straight line place plane of delineation and the headchute perpendicular line place plane of delineation, specific as follows:
Step 5.1: calculate headchute horizontal edge straight line Lh
ithe picture depth of place plane, first at headchute horizontal edge straight line Lh
ion get a little
, recycling computer vision technique, calculation level
picture depth be
;
Step 5.2: calculate headchute perpendicular line set Lv
ithe picture depth of cathetus place plane, first at headchute perpendicular line
on get a little
, recycling computer vision technique, calculation level
picture depth be
, obtain the picture depth set of headchute perpendicular line further
;
Step 5.3: calculated a little by above-mentioned
picture depth
with perpendicular line set Lv
ipicture depth set
, obtain straight line Lh
iplace plane and straight line set Lv
ipicture depth difference between the plane of place is
.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510826745.8A CN105469401B (en) | 2015-11-25 | 2015-11-25 | A kind of headchute localization method based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510826745.8A CN105469401B (en) | 2015-11-25 | 2015-11-25 | A kind of headchute localization method based on computer vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105469401A true CN105469401A (en) | 2016-04-06 |
| CN105469401B CN105469401B (en) | 2018-05-29 |
Family
ID=55607060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510826745.8A Active CN105469401B (en) | 2015-11-25 | 2015-11-25 | A kind of headchute localization method based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105469401B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106096606A (en) * | 2016-06-07 | 2016-11-09 | 浙江工业大学 | A kind of container profile localization method based on fitting a straight line |
| CN106931906A (en) * | 2017-03-03 | 2017-07-07 | 浙江理工大学 | A kind of object dimensional size simple measurement method based on binocular stereo vision |
| CN107514994A (en) * | 2017-07-12 | 2017-12-26 | 浙江工业大学 | A kind of headchute localization method based on error compensation |
| CN113450373A (en) * | 2020-08-18 | 2021-09-28 | 中国人民解放军63729部队 | Optical live image-based real-time discrimination method for characteristic events in carrier rocket flight process |
-
2015
- 2015-11-25 CN CN201510826745.8A patent/CN105469401B/en active Active
Non-Patent Citations (4)
| Title |
|---|
| 孔婷等: "《基于双目立体视觉的集装箱卡车定位方法研究》", 《计算机应用与软件》 * |
| 梁晓波等: "《基于双目视觉的集装箱自动识别》", 《机械》 * |
| 梁晓波等: "《集装箱起重机自动装卸***的研究与设计》", 《计算机应用》 * |
| 马蕾: "《基于双目立体视觉的集装箱识别与定位技术研究》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106096606A (en) * | 2016-06-07 | 2016-11-09 | 浙江工业大学 | A kind of container profile localization method based on fitting a straight line |
| CN106096606B (en) * | 2016-06-07 | 2019-11-05 | 浙江工业大学 | A kind of container profile localization method based on straight line fitting |
| CN106931906A (en) * | 2017-03-03 | 2017-07-07 | 浙江理工大学 | A kind of object dimensional size simple measurement method based on binocular stereo vision |
| CN107514994A (en) * | 2017-07-12 | 2017-12-26 | 浙江工业大学 | A kind of headchute localization method based on error compensation |
| CN107514994B (en) * | 2017-07-12 | 2019-11-29 | 浙江工业大学 | A kind of headchute localization method based on error compensation |
| CN113450373A (en) * | 2020-08-18 | 2021-09-28 | 中国人民解放军63729部队 | Optical live image-based real-time discrimination method for characteristic events in carrier rocket flight process |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105469401B (en) | 2018-05-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106067023B (en) | Container number and truck number identification system and method based on image processing | |
| Mi et al. | Research on regional clustering and two-stage SVM method for container truck recognition | |
| CN106097329B (en) | A kind of container profile localization method based on edge detection | |
| CN105719241B (en) | A kind of headchute localization method of view-based access control model scanning | |
| CN105469401A (en) | Ship groove positioning method based on computer vision | |
| Li et al. | Nighttime lane markings recognition based on Canny detection and Hough transform | |
| CN103514448A (en) | Method and system for navicular identification | |
| CN102708351B (en) | Method for fast identifying Data Matrix two-dimensional bar code under complicated working condition background | |
| CN105447892A (en) | Method and device for determining yaw angle of vehicle | |
| CN108460800B (en) | Container image positioning method and system | |
| CN107609510B (en) | Positioning method and device for lower set of quayside container crane | |
| CN109886175B (en) | Method for detecting lane line by combining straight line and circular arc | |
| Huang et al. | Robust lane marking detection under different road conditions | |
| CN108985274A (en) | Water surface method for recognizing impurities | |
| CN113184707A (en) | Method and system for preventing lifting of container truck based on laser vision fusion and deep learning | |
| CN112462373A (en) | Cabin position detection method based on multi-sensor fusion | |
| CN103390259A (en) | Ground image processing method in visual guidance AGV | |
| CN108672988B (en) | A kind of automatic identifying method of weld joint tracking deviation and cross-seam type | |
| Feng et al. | Robust lane detection and tracking for autonomous driving of rubber-tired gantry cranes in a container yard | |
| Zhang et al. | Recognition of pavement surface crack | |
| CN115586552A (en) | Method for accurately secondarily positioning unmanned truck collection under port tyre crane or bridge crane | |
| Jun et al. | A vehicle license plate location and correction method based the characteristics of license plate | |
| CN204726969U (en) | A kind of freight container lead sealing based on planar bar code technology | |
| Jianhong et al. | Adaptive shorelines detection for autonomous surface vessel navigation | |
| CN114419554A (en) | Monitoring and identifying system for violation of regulations on electric power operation site based on image identification |
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 |