CN107657617A - A kind of safflower filament recognition methods - Google Patents
A kind of safflower filament recognition methods Download PDFInfo
- Publication number
- CN107657617A CN107657617A CN201710908265.5A CN201710908265A CN107657617A CN 107657617 A CN107657617 A CN 107657617A CN 201710908265 A CN201710908265 A CN 201710908265A CN 107657617 A CN107657617 A CN 107657617A
- Authority
- CN
- China
- Prior art keywords
- safflower
- image
- filigree
- pixel
- carried out
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 235000003255 Carthamus tinctorius Nutrition 0.000 title claims abstract description 50
- 244000020518 Carthamus tinctorius Species 0.000 title claims abstract description 50
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000000605 extraction Methods 0.000 claims abstract description 9
- 238000012935 Averaging Methods 0.000 claims abstract description 3
- 230000007797 corrosion Effects 0.000 claims abstract description 3
- 238000005260 corrosion Methods 0.000 claims abstract description 3
- 230000001186 cumulative effect Effects 0.000 claims abstract description 3
- 238000010586 diagram Methods 0.000 claims description 10
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 2
- 239000003086 colorant Substances 0.000 claims description 2
- 230000003044 adaptive effect Effects 0.000 claims 1
- 238000003306 harvesting Methods 0.000 abstract description 7
- 239000000284 extract Substances 0.000 abstract description 4
- 238000010835 comparative analysis Methods 0.000 abstract description 2
- 230000004807 localization Effects 0.000 abstract description 2
- 102000003712 Complement factor B Human genes 0.000 abstract 1
- 108090000056 Complement factor B Proteins 0.000 abstract 1
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 238000012545 processing Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 239000000049 pigment Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30204—Marker
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The present invention proposes a kind of method that safflower picking robot quickly identifies safflower filament and marks two-dimensional center point coordinates.According to the color characteristic of filigree, extraction R, G, Factor B, the comparative analysis of color characteristic is carried out with chromatism method；The contours extract of filigree is carried out from Otsu dividing methods, and is converted using corrosion in mathematical morphology and the similar round or regular image of expansion progress profile；The cumulative principle of averaging of pixel is finally used, obtains the two-dimensional center point coordinates of filigree.The beneficial effects of the invention are as follows can fast and effeciently identify safflower filament and carry out the extraction of two-dimensional center point coordinates, three-dimensional localization and harvesting for safflower picking robot provide theoretical foundation.
Description
Technical field
The present invention relates in safflower picking robot field of image recognition, the particularly identification of safflower filament and extraction two dimension
The method of heart point coordinates.
Background technology
Safflower is a kind of very high crops of economic value, integrates the value such as medicinal, edible, pigment extraction, is new
One of major economic crops of boundary.Safflower harvest is divided into receipts flower and receives seed, and the harvesting of safflower filament is with very strong ageing, and one
As choose that the flowers are in blossom period of 2-3 days plucked, now filigree medical value highest.But the harvesting of current safflower is main still
Completed by artificial, harvesting cost is huge, and safflower picking robot can greatly improve picking efficiency, reduces production cost, solution
Labour is put, realizes the development of safflower harvesting intelligent machine.
The quick identification of safflower filament and be accurately positioned be research and develop safflower picking robot key technology and precondition
One of, the filigree for blooming the phase is in thin tube-like, orange red, and countless filigrees form spheroid profile, can use machine vision technique to it
Carry out image recognition and positioning.Have no both at home and abroad at present and carry out safflower filament identification and localization method for safflower picking robot
Research.
The content of the invention
In view of the deficienciess of the prior art, the invention provides a kind of safflower picking robot safflower filament identification side
Method, ensure that safflower picking robot can quickly identify safflower filament, and extract two-dimensional center point coordinates, determine for the three-dimensional of safflower
Position provides theoretical foundation.
The present invention is to realize above-mentioned technical purpose by following technological means.
Safflower picking robot is when carrying out safflower filament harvesting identification, it is characterised in that comprises the steps of：
Step 1：Under conditions of natural lighting, the IMAQ of safflower is carried out；
Step 2：Pretreatment operation is carried out to the image of shooting, and uses RGB color model, extracts the color of safflower image
Information, the color segmentation of image is carried out with R-G chromatism methods；
Step 3：From can adaptively threshold value Otsu split plot designs (maximum variance between clusters) carry out bianry image
Segmentation；
Step 4：The class that filigree profile is carried out with expanding using corroding in mathematical morphology is round or regular figure converts；
Step 5：With the cumulative principle of averaging of pixel, the two-dimensional center point coordinates of filigree is obtained, and be marked.
Further, in the step 1, safflower image is carried out real by the binocular camera in safflower picking machine vision system
When obtain.
Further, the step 2 comprises the following steps that：
A. acquired safflower image is carried out being based on RGB color model, carries out the extraction of R, G, B color characteristic, and paint
Three-dimensional model diagram processed, analyze the difference between its filigree and ambient background color threshold；
B. compare the color characteristic that R-B, R-G, B-G be combined with chromatism method, and draw grey level histogram and three-dimensional mould
Type figure carries out com-parison and analysis, is got colors according to result and is characterized as R-G bianry image, as follow-up pending image.
Further, the step 3 comprises the following steps that：
Bianry image according to acquired in step 2, carry out being based on histogram Two-peak method, maximum variance between clusters and iterative method
Threshold segmentation compare, according to simulation result, the maximum variance between clusters of threshold value can have been automatically determined by, which choosing, carries out image segmentation.
Further, the step 4 comprises the following steps that：
The structural element that respective radius is chosen according to image carries out morphological erosion processing, Ran Houjin to the image after segmentation
Row expansion process, above-mentioned steps are circulated, untill there is similar round or regular image.
Further, the step 5 comprises the following steps that：
Pixel horizontal stroke, the ordinate of the black background part of image after expansion are set to 0, the horizontal, vertical of white pixel point sits
Mark is set to 1.Judgement is identified to the pixel of whole image, if abscissa pixel i is 1, output pixel point information, according to
It is secondary to add up and calculate number, it is otherwise 0；Same ordinate pixel j is judged and added up according to result successively, finally by institute
Obtained pixel point coordinates is averaged, and draws the two-dimensional center coordinate value of filigree.
Brief description of the drawings
Fig. 1 is safflower filament recognition methods flow chart.
Fig. 2 is the safflower image shot under natural lighting.
Fig. 3 is that the three-dimensional model diagram after R color characteristics is extracted under RGB models.
Fig. 4 is that the three-dimensional model diagram after G color characteristics is extracted under RGB models.
Fig. 5 is that the three-dimensional model diagram after B color characteristics is extracted under RGB models.
Fig. 6 is the binary map that R-B is extracted under RGB models.
Fig. 7 is the grey level histogram that R-B is extracted under RGB models.
Fig. 8 is the three-dimensional model diagram that R-B is extracted under RGB models.
Fig. 9 is the binary map that R-G is extracted under RGB models.
Figure 10 is the grey level histogram that R-G is extracted under RGB models.
Figure 11 is the three-dimensional model diagram that R-G is extracted under RGB models.
Figure 12 is the binary map that G-B is extracted under RGB models.
Figure 13 is the grey level histogram that G-B is extracted under RGB models.
Figure 14 is the three-dimensional model diagram that G-B is extracted under RGB models.
Figure 15 is that the image after burn into expansion is carried out to the binary map after extraction R-G color characteristics；Wherein, after a is corrosion
Image；B is the image after expansion.
Figure 16 is the image after extraction safflower filament two-dimensional points coordinate.
Embodiment
Below in conjunction with the accompanying drawings and specific embodiment the present invention is further illustrated, it should be emphasised that, under state
Bright to be merely exemplary, protection scope of the present invention is not limited to this.
For the single safflower that the embodiment of the present invention is collected using safflower picking robot as research object, concrete scheme is as follows：
Step 1：Using the colourful CCD video camera on safflower picking robot, safflower image is carried out under natural lighting
Collection, the image collected are as shown in Figure 2.
Step 2：Limitation that can be because of natural lighting condition because safflower image is shot under natural environment and surrounding ring
Border influences random noise be present, so needing to carry out pretreatment operation to the safflower image of shooting before image procossing, is comparing
After the treatment effect of gaussian filtering, mean filter and medium filtering, from can subtract in the case where not reducing picture contrast
The medium filtering that small exceptional value influences.
Step 3：Extract R component, G components and the B component of image respectively under RGB models, respectively drawing three-dimensional illustraton of model
(such as Fig. 3-Fig. 5), and be analyzed, the safflower image under R component state is found, color threshold distribution is close, size
Reach unanimity (Fig. 3), and filigree and surrounding environment have more obvious difference, but due to the influence of available light, still suffer from and
Region similar in filigree threshold value.And in the figure of G components (Fig. 4) and B component (Fig. 5), color threshold skewness, it is impossible to complete
Represent the contour shape of filigree.
Step 4：Comparative analysis aberration R-B, R-G and G-B binary map, grey level histogram and three-dimensional model diagram (such as Fig. 6-
Figure 14) understand, the three-dimensional model diagram of aberration R-B (Fig. 8) and R-G (Figure 11) factor can be further by safflower filament and surrounding environment
Distinguish, but R-B components grey level histogram (Fig. 7) can be seen that compared with R-G (Figure 10) by the grey level histogram of the two, deposit
In obvious crest and trough, i.e. there is more notable difference with surrounding environment, can thus select in filigree in gamma characteristic
A proper threshold value is taken, determines that each pixel belongs to filigree region or background area in image.
Step 5：Morphological scale-space is carried out to the image that step 4 is chosen, the suitable structural element of radius is chosen and is corroded
(Figure 15 a) and expansion (Figure 15 b) processing, until the image after processing reaches similar round or other regular figures, the figure after processing
As shown in figure 15.
Step 6：Pixel horizontal stroke, the ordinate of the black background part of image after expansion are set to 0, white pixel point
Horizontal, ordinate is set to 1.Judgement is identified to the pixel of whole image, if abscissa pixel i is 1, output pixel point
Information, add up successively and calculate number, be otherwise 0；Same ordinate pixel j judged and added up according to result successively,
Finally resulting pixel point coordinates is averaged, draws the two-dimensional center coordinate value (such as Figure 16) of filigree coordinate.
Above-described embodiment is only the preferable example of the present invention, is not the restriction to embodiment of the present invention, all
All any modification, equivalent and improvement made within the spirit and principles in the present invention etc., it should be included in the power of the present invention
Within the scope of profit is claimed.
Claims (3)
1. safflower picking robot safflower filament recognition methods, it is characterised in that comprise the steps of：
Step 1：Under conditions of natural lighting, the IMAQ of safflower is carried out；
Step 2：Pretreatment operation is carried out to the image of shooting, and uses RGB color model, the color letter of extraction safflower image
Breath, the color segmentation of image is carried out with R-G chromatism methods；
Step 3：From can adaptive threshold value Otsu split plot designs（Maximum variance between clusters）Carry out point of bianry image
Cut；
Step 4：The class circle of filigree profile is carried out with expansion or regular figure converts using corrosion in mathematical morphology；
Step 5：With the cumulative principle of averaging of pixel, the two-dimensional center point coordinates of filigree is obtained, and be marked.
2. safflower picking robot safflower filament recognition methods as claimed in claim 1, it is characterised in that：The step 2
Comprise the following steps that：
A. acquired safflower image is carried out being based on RGB color model, carries out the extraction of R, G, B color characteristic, and draw three
Dimension module figure, analyze the difference between its filigree and ambient background color threshold；
B. compare the color characteristic that R-B, R-G, B-G be combined with chromatism method, and draw grey level histogram and three-dimensional model diagram
Com-parison and analysis is carried out, is got colors according to result and is characterized as R-G bianry image, as follow-up pending image.
3. safflower picking robot safflower filament recognition methods as claimed in claim 1, it is characterised in that：The step 5
Comprise the following steps that：
Pixel horizontal stroke, the ordinate of the black background part of image after expansion are set to 0, horizontal stroke, the ordinate of white pixel point are set
For 1,
Judgement is identified to the pixel of whole image, if abscissa pixelFor 1, then output pixel point information, tires out successively
Adduction calculates number, is otherwise 0；Same ordinate pixelJudged successively and added up according to result, finally by obtained by
Pixel point coordinates averaged, draw the two-dimensional center coordinate value of filigree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710908265.5A CN107657617A (en) | 2017-09-29 | 2017-09-29 | A kind of safflower filament recognition methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710908265.5A CN107657617A (en) | 2017-09-29 | 2017-09-29 | A kind of safflower filament recognition methods |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107657617A true CN107657617A (en) | 2018-02-02 |
Family
ID=61116061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710908265.5A Pending CN107657617A (en) | 2017-09-29 | 2017-09-29 | A kind of safflower filament recognition methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107657617A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108682033A (en) * | 2018-05-29 | 2018-10-19 | 石河子大学 | A kind of phase safflower filament two-dimensional image center in full bloom point extracting method |
| CN108710850A (en) * | 2018-05-17 | 2018-10-26 | 中国科学院合肥物质科学研究院 | A kind of Chinese wolfberry fruit recognition methods of strong applicability and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103714349A (en) * | 2014-01-09 | 2014-04-09 | 成都淞幸科技有限责任公司 | Image recognition method based on color and texture features |
-
2017
- 2017-09-29 CN CN201710908265.5A patent/CN107657617A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103714349A (en) * | 2014-01-09 | 2014-04-09 | 成都淞幸科技有限责任公司 | Image recognition method based on color and texture features |
Non-Patent Citations (1)
| Title |
|---|
| 崔淑娟: "《自然场景下成熟苹果目标的识别及其定位技术研究》", 《陕西科技大学》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108710850A (en) * | 2018-05-17 | 2018-10-26 | 中国科学院合肥物质科学研究院 | A kind of Chinese wolfberry fruit recognition methods of strong applicability and system |
| CN108682033A (en) * | 2018-05-29 | 2018-10-19 | 石河子大学 | A kind of phase safflower filament two-dimensional image center in full bloom point extracting method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108009542B (en) | Weed image segmentation method in rape field environment | |
| Silwal et al. | Apple identification in field environment with over the row machine vision system | |
| CN108319973A (en) | Detection method for citrus fruits on tree | |
| CN111798470B (en) | Crop image entity segmentation method and system applied to intelligent agriculture | |
| CN108537239A (en) | A kind of method of saliency target detection | |
| CN114067206B (en) | Spherical fruit identification positioning method based on depth image | |
| Keller et al. | Soybean leaf coverage estimation with machine learning and thresholding algorithms for field phenotyping | |
| CN108090485A (en) | Display foreground extraction method based on various visual angles fusion | |
| CN102542560B (en) | Method for automatically detecting density of rice after transplantation | |
| Feng et al. | A separating method of adjacent apples based on machine vision and chain code information | |
| CN106683098B (en) | A kind of dividing method of overlapped blades image | |
| CN112131952A (en) | Corn seedling stage plant number information extraction based on unmanned aerial vehicle remote sensing image | |
| CN104992448B (en) | The automatic positioning method of robot antisitic defect grape-picking | |
| CN107657617A (en) | A kind of safflower filament recognition methods | |
| CN105844264B (en) | It is a kind of based on the recognition methods of tree peony fruit image of the oil of stress | |
| Rahman et al. | Identification of mature grape bunches using image processing and computational intelligence methods | |
| CN103729621A (en) | Plant leaf image automatic recognition method based on leaf skeleton model | |
| CN109166127B (en) | Wearable plant phenotype sensing system | |
| CN104408407B (en) | The means of night recognition of apple picking robot | |
| CN107886493A (en) | A kind of wire share split defect inspection method of transmission line of electricity | |
| Wang et al. | A color-texture segmentation method to extract tree image in complex scene | |
| CN110175582B (en) | Intelligent tea tree tender shoot identification method based on pixel distribution | |
| Wongtanawijit et al. | Rubber tapped path detection using k-means color segmentation and distance to boundary feature | |
| CN108682033A (en) | A kind of phase safflower filament two-dimensional image center in full bloom point extracting method | |
| Melnychenko et al. | Apple detection with occlusions using modified YOLOv5-v1 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180202 |