CN109035214A - A kind of industrial robot material shapes recognition methods - Google Patents
A kind of industrial robot material shapes recognition methods Download PDFInfo
- Publication number
- CN109035214A CN109035214A CN201810730332.3A CN201810730332A CN109035214A CN 109035214 A CN109035214 A CN 109035214A CN 201810730332 A CN201810730332 A CN 201810730332A CN 109035214 A CN109035214 A CN 109035214A
- Authority
- CN
- China
- Prior art keywords
- image
- industrial robot
- point
- recognition methods
- material shapes
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/564—Depth or shape recovery from multiple images from contours
-
- 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
-
- 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
- G06T2207/20036—Morphological image processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of industrial robot material shapes recognition methods, it is characterised in that: the following steps are included: step 1: acquisition material image, and by image digitazation；Step 2: by image enhancement, median filtering, frequency domain processing and histogram equalization, image preprocessing is carried out；Step 3: image binaryzation processing is carried out；Step 4: to binary image Morphological scale-space: filtering out the grain noise in image by etching operation；The edge for the target that is corroded moderately is restored using expansive working；Step 5: image outline is extracted；Step 6: go out the shape of object to be identified by contour feature point match cognization.The present invention can find and identify the immediate template of material shapes, and material shapes identification is accurate；Bring shape recognition error in Image Acquisition is avoided simultaneously.
Description
Technical field
The present invention relates to industrial robot field, especially a kind of industrial robot material shapes recognition methods.
Background technique
Industrial robot is multi-joint manipulator or multivariant robot towards industrial circle.Industrial robot is
The automatic mechanical device for executing work is a kind of machine that various functions are realized by self power and control ability.It can connect
It is commanded, can also be run according to the program of preparatory layout, modern industry robot can also be according to artificial intelligence technology by the mankind
The principle program action of formulation.People can be substituted in the industrial production does certain dull, frequent and duplicate long workings, or
Operations dangerous, under adverse circumstances, such as in punching press, compression casting, heat treatment, welding, coating, plastic product forming, machinery
In the processes such as processing and simple assembly.Robot vision mainly simulates the visual performance of people with computer, is not merely people
The simple extension of eye, it is often more important that the part of functions with human brain.Information is extracted from the image of objective things, is located
It manages and is understood, eventually for actually detected, measurement and control.The maximum feature of machine vision is that speed is fast, contain much information,
Function is more.Robot and vision technique are combined, helps robot to be competent at more intelligentized work, has become industry
A kind of trend in robot application field.
Currently, universal robot all uses camera collection image to identify body form, existing robot vision is only
Classified using color, partially focused in the positioning of Small object, the classification of shape often only simply used several
The mode of what template matching is identified, actual conditions cannot be well adapted for；And 3 D workpiece placement angle is different and images
Machine visual field difference bring shape recognition error often causes sorting mistake.
Summary of the invention
Goal of the invention of the invention is: in view of the above problems, providing a kind of industrial robot material shapes knowledge
Other method is realized by strengthening with shape information analysis to the Classification and Identification of material shapes to different shape material
Accurately, stablize, rapid sorting.
The technical solution adopted by the invention is as follows:
A kind of industrial robot material shapes recognition methods of the present invention, comprising the following steps:
Step 1: acquisition material image, and by image digitazation；
Step 2: by image enhancement, median filtering, frequency domain processing and histogram equalization, image preprocessing is carried out；
Step 3: image binaryzation processing is carried out；
Step 4: to binary image Morphological scale-space: filtering out the grain noise in image by etching operation；Using swollen
Swollen operation appropriateness restores the edge for the target that is corroded；
Step 5: image outline is extracted；
Step 6: go out the shape of object to be identified by contour feature point match cognization.
Preferably, the step 3 specifically includes: obtaining the first address of image and the height and width of image；Set initial threshold
Value T；It calculates separately and is less than T and two groups of average gray values u1 and u2 greater than T in image；Threshold value is iterated to calculate, until two thresholds
It is worth equal；According to calculated threshold value, binary conversion treatment is carried out to image.
Preferably, the method for extracting image outline includes: the first address and figure for obtaining image pixel in the step 5
The height and width of picture；8 neighborhood territory pixels of pixel are read in array, if the gray value phase of the sum of the grayscale values central point of neighborhood territory pixel
Difference is less than K, determines that neighborhood territory pixel is identical with central point；If 8 neighborhood territory pixels are all identical with central point, pixel is set it is white, otherwise
It remains unchanged；Each pixel in image is handled from left to bottom right, obtains the profile of image.
Preferably, in the step 6, the matched method of contour feature point: according to the contour feature point a of imagei, with
And the contour feature point b of sample templatej, in conjunction with shape information, the matching value of contour feature point is calculated, when matching value maximum,
The shape of sample template is to identify body form.
Preferably, the matching value calculation formula of the contour feature point:
Wherein, C (ai, bj) indicate be characteristic point matching set, hi(k) it indicates
With characteristic point aiFor the log polar coordinates histogram in the section k of reference, hj(k) it indicates with characteristic point bjFor the log in the section k of reference
Polar coordinates histogram value.
Above method calculates the matching value of contour feature point, finds matching value maximum value, counter sample and object to be identified
Contour feature point matches highest, according to the shape of known sample, it will be able to obtain the shape of object to be identified, realize the essence of material
Quasi- matching.
Preferably, the acquisition methods of the shape information: find the characteristic point of object inside and outside contour, and with it is intensive from
Scatterplot indicates continuous profile sideline；The relative position between characteristic point is indicated by log polar coordinates histogram.
Above method describes shape using relative position information between object inside and outside contour characteristic point；Due to using opposite
Location information is described, thus contour distortion caused by the rotation of object, scaling and video camera it is influenced less,
There is stronger robustness in terms of the identification of shape.
Preferably, the calculation method of log polar coordinates histogram value: a certain characteristic point being made to be located at coordinate system center as ginseng
Examination point can obtain corresponding histogram value by the quantity that counting falls in other characteristic points in each coordinate system section.
Above method, log polar coordinate system make the influence for the characteristic point for closing on reference point be greater than the influence of distant place characteristic point,
To enable fixed reference feature point more sensitive to the change in shape around individual, more appropriate positional relationship is obtained.
In conclusion by adopting the above-described technical solution, the beneficial effects of the present invention are: compared with prior art, this
Invention is reduced by the relative positional relationship of characteristic point and generates influence of the noise to object identification in Image Acquisition operation, obtained
The shape information of material carries out contour feature point matching, finds and identifies the highest template of object matches degree, to improve material
Accuracy of identification.
Detailed description of the invention
Examples of the present invention will be described by way of reference to the accompanying drawings, in which:
Fig. 1 is a kind of industrial robot material shapes recognition methods flow chart of the present invention.
Specific embodiment
All features disclosed in this specification or disclosed all methods or in the process the step of, in addition to mutually exclusive
Feature and/or step other than, can combine in any way.
Any feature disclosed in this specification (including any accessory claim, abstract), unless specifically stated,
It is replaced by other equivalent or with similar purpose alternative features.That is, unless specifically stated, each feature is a series of
An example in equivalent or similar characteristics.
Such as Fig. 1, a kind of industrial robot material shapes recognition methods of the present invention, comprising the following steps:
Step 1: acquisition material image, and by image digitazation；
Step 2: by image enhancement, median filtering, frequency domain processing and histogram equalization, image preprocessing is carried out；
Step 3: image binaryzation processing is carried out；Obtain the first address of image and the height and width of image；Set initial threshold
T；It calculates separately and is less than T and two groups of average gray values u1 and u2 greater than T in image；Threshold value is iterated to calculate, until two threshold values
It is equal；According to calculated threshold value, binary conversion treatment is carried out to image；
Step 4: to binary image Morphological scale-space: filtering out the grain noise in image by etching operation；Using swollen
Swollen operation appropriateness restores the edge for the target that is corroded；
Step 5: image outline is extracted；The method for extracting image outline includes: the first address and image for obtaining image pixel
Height and width；8 neighborhood territory pixels of pixel are read in array, if the gray value of the sum of the grayscale values central point of neighborhood territory pixel differs
Less than K, determine that neighborhood territory pixel is identical with central point；If 8 neighborhood territory pixels are all identical with central point, pixel is set white, otherwise protected
It holds constant；Each pixel in image is handled from left to bottom right, obtains the profile of image；
Step 6: go out the shape of object to be identified by contour feature point match cognization；The matched side of contour feature point
Method: according to the contour feature point a of imageiAnd the contour feature point b of sample templatej, in conjunction with shape information, calculate contour feature
The matching value of point, when matching value maximum, the shape of sample template is to identify body form.
The matching value calculation formula of the contour feature point:
Wherein, C (ai, bj) indicate be characteristic point matching set, hi(k) it indicates
With characteristic point aiFor the log polar coordinates histogram in the section k of reference, hj(k) it indicates with characteristic point bjFor the log in the section k of reference
Polar coordinates histogram value.
The acquisition methods of shape information: the characteristic point of object inside and outside contour is found, and with intensive discrete point come the company of expression
Continuous profile sideline；The relative position between characteristic point is indicated by log polar coordinates histogram；Wherein, log polar coordinates histogram
The calculation method of value: making a certain characteristic point be located at coordinate system center as a reference point, falls in each coordinate system section by counting
In the quantity of other characteristic points can obtain corresponding histogram value.
The invention is not limited to specific embodiments above-mentioned.The present invention, which expands to, any in the present specification to be disclosed
New feature or any new combination, and disclose any new method or process the step of or any new combination.
Claims (7)
1. a kind of industrial robot material shapes recognition methods, it is characterised in that: the following steps are included:
Step 1: acquisition material image, and by image digitazation；
Step 2: by image enhancement, median filtering, frequency domain processing and histogram equalization, image preprocessing is carried out；
Step 3: image binaryzation processing is carried out；
Step 4: to binary image Morphological scale-space: filtering out the grain noise in image by etching operation；It is grasped using expansion
Make the edge that appropriateness restores the target that is corroded；
Step 5: image outline is extracted；
Step 6: go out the shape of object to be identified by contour feature point match cognization.
2. industrial robot material shapes recognition methods as described in claim 1, it is characterised in that: the step 3 is specifically wrapped
It includes: obtaining the first address of image and the height and width of image；Set initial threshold T；It calculates separately and is less than T in image and greater than T's
Two groups of average gray values u1 and u2；Threshold value is iterated to calculate, until two threshold values are equal；According to calculated threshold value, to image into
Row binary conversion treatment.
3. industrial robot material shapes recognition methods as described in claim 1, it is characterised in that: in the step 5, mention
The method for taking image outline includes: the height and width of the first address and image that obtain image pixel；8 neighborhood territory pixels of pixel are read
Enter in array, if the gray value difference of the sum of the grayscale values central point of neighborhood territory pixel is less than K, determines neighborhood territory pixel and central point phase
Together；If 8 neighborhood territory pixels are all identical with central point, pixel is set white, otherwise remained unchanged；From left to bottom right in image
Each pixel is handled, and the profile of image is obtained.
4. industrial robot material shapes recognition methods as described in claim 1, it is characterised in that: in the step 6, wheel
The method of wide Feature Points Matching: according to the contour feature point a of imageiAnd the contour feature point b of sample templatej, in conjunction with shape
Information calculates the matching value of contour feature point, and when matching value maximum, the shape of sample template is to identify body form.
5. industrial robot material shapes recognition methods as claimed in claim 4, it is characterised in that: the contour feature point
Matching value calculation formula:Wherein, C (ai, bj) indicate be characteristic point matching set, hi
(k) it indicates with characteristic point aiFor the log polar coordinates histogram in the section k of reference, hj(k) it indicates with characteristic point bjFor the area k of reference
Between log polar coordinates histogram value.
6. industrial robot material shapes recognition methods as claimed in claim 4, it is characterised in that: the shape information obtains
It takes method: finding the characteristic point of object inside and outside contour, and indicate continuous profile sideline with intensive discrete point；Pass through log
Polar coordinates histogram indicates the relative position between characteristic point.
7. industrial robot material shapes recognition methods as claimed in claim 5, it is characterised in that: log polar coordinates histogram
The calculation method of value: making a certain characteristic point be located at coordinate system center as a reference point, falls in each coordinate system section by counting
In the quantity of other characteristic points can obtain corresponding histogram value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810730332.3A CN109035214A (en) | 2018-07-05 | 2018-07-05 | A kind of industrial robot material shapes recognition methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810730332.3A CN109035214A (en) | 2018-07-05 | 2018-07-05 | A kind of industrial robot material shapes recognition methods |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109035214A true CN109035214A (en) | 2018-12-18 |
Family
ID=64640426
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810730332.3A Pending CN109035214A (en) | 2018-07-05 | 2018-07-05 | A kind of industrial robot material shapes recognition methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109035214A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109724988A (en) * | 2019-02-01 | 2019-05-07 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of pcb board defect positioning method based on multi-template matching |
| CN110070558A (en) * | 2019-04-11 | 2019-07-30 | 西安星舟天启智能装备有限责任公司 | It is a kind of that method and device is sorted based on the statistics of image procossing and Subarea detecting |
| CN111242240A (en) * | 2020-02-13 | 2020-06-05 | 深圳市联合视觉创新科技有限公司 | Material detection method and device and terminal equipment |
| CN111649749A (en) * | 2020-06-24 | 2020-09-11 | 万翼科技有限公司 | Navigation method based on BIM (building information modeling), electronic equipment and related product |
| CN111652270A (en) * | 2020-04-23 | 2020-09-11 | 中南大学 | Coarse-grained soil filler gradation automatic identification method based on image matching and application system |
| CN116723401A (en) * | 2023-08-11 | 2023-09-08 | 深圳金语科技有限公司 | Method and device for compensating image jitter of streaming media rearview mirror |
| CN117299596A (en) * | 2023-08-14 | 2023-12-29 | 江苏秦郡机械科技有限公司 | Material screening system and method for automatic detection |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103559703A (en) * | 2013-10-08 | 2014-02-05 | 中南大学 | Crane barrier monitoring and prewarning method and system based on binocular vision |
| JP2015001750A (en) * | 2013-06-12 | 2015-01-05 | 国立大学法人宇都宮大学 | Shape recognition system, shape recognition method, and shape recognition program |
| EP2884426A1 (en) * | 2013-12-12 | 2015-06-17 | Omron Corporation | Image processing device, image processing method, and image processing program |
| CN106683137A (en) * | 2017-01-11 | 2017-05-17 | 中国矿业大学 | Monocular multi-target identification and positioning method based on artificial mark |
| CN106934813A (en) * | 2015-12-31 | 2017-07-07 | 沈阳高精数控智能技术股份有限公司 | A kind of industrial robot workpiece grabbing implementation method of view-based access control model positioning |
| CN107103323A (en) * | 2017-03-09 | 2017-08-29 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of target identification method based on image outline feature |
| CN107798326A (en) * | 2017-10-20 | 2018-03-13 | 华南理工大学 | A kind of profile visual detection algorithm |
| CN108182704A (en) * | 2016-12-08 | 2018-06-19 | 广州映博智能科技有限公司 | Localization method based on Shape context feature |
-
2018
- 2018-07-05 CN CN201810730332.3A patent/CN109035214A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015001750A (en) * | 2013-06-12 | 2015-01-05 | 国立大学法人宇都宮大学 | Shape recognition system, shape recognition method, and shape recognition program |
| CN103559703A (en) * | 2013-10-08 | 2014-02-05 | 中南大学 | Crane barrier monitoring and prewarning method and system based on binocular vision |
| EP2884426A1 (en) * | 2013-12-12 | 2015-06-17 | Omron Corporation | Image processing device, image processing method, and image processing program |
| CN106934813A (en) * | 2015-12-31 | 2017-07-07 | 沈阳高精数控智能技术股份有限公司 | A kind of industrial robot workpiece grabbing implementation method of view-based access control model positioning |
| CN108182704A (en) * | 2016-12-08 | 2018-06-19 | 广州映博智能科技有限公司 | Localization method based on Shape context feature |
| CN106683137A (en) * | 2017-01-11 | 2017-05-17 | 中国矿业大学 | Monocular multi-target identification and positioning method based on artificial mark |
| CN107103323A (en) * | 2017-03-09 | 2017-08-29 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of target identification method based on image outline feature |
| CN107798326A (en) * | 2017-10-20 | 2018-03-13 | 华南理工大学 | A kind of profile visual detection algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 郎锐: "《数字图像处理学Visual C++实现》", 31 January 2003, 北京希望电子出版社 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109724988A (en) * | 2019-02-01 | 2019-05-07 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of pcb board defect positioning method based on multi-template matching |
| CN109724988B (en) * | 2019-02-01 | 2021-05-18 | 佛山市南海区广工大数控装备协同创新研究院 | PCB defect positioning method based on multi-template matching |
| CN110070558A (en) * | 2019-04-11 | 2019-07-30 | 西安星舟天启智能装备有限责任公司 | It is a kind of that method and device is sorted based on the statistics of image procossing and Subarea detecting |
| CN111242240A (en) * | 2020-02-13 | 2020-06-05 | 深圳市联合视觉创新科技有限公司 | Material detection method and device and terminal equipment |
| CN111242240B (en) * | 2020-02-13 | 2023-04-07 | 深圳市联合视觉创新科技有限公司 | Material detection method and device and terminal equipment |
| CN111652270A (en) * | 2020-04-23 | 2020-09-11 | 中南大学 | Coarse-grained soil filler gradation automatic identification method based on image matching and application system |
| CN111652270B (en) * | 2020-04-23 | 2022-03-29 | 中南大学 | Coarse-grained soil filler gradation automatic identification method based on image matching and application system |
| CN111649749A (en) * | 2020-06-24 | 2020-09-11 | 万翼科技有限公司 | Navigation method based on BIM (building information modeling), electronic equipment and related product |
| CN116723401A (en) * | 2023-08-11 | 2023-09-08 | 深圳金语科技有限公司 | Method and device for compensating image jitter of streaming media rearview mirror |
| CN117299596A (en) * | 2023-08-14 | 2023-12-29 | 江苏秦郡机械科技有限公司 | Material screening system and method for automatic detection |
| CN117299596B (en) * | 2023-08-14 | 2024-05-24 | 江苏秦郡机械科技有限公司 | Material screening system and method for automatic detection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109035214A (en) | A kind of industrial robot material shapes recognition methods | |
| CN107992881B (en) | Robot dynamic grabbing method and system | |
| CN108890692A (en) | A kind of material color identification method for industrial robot vision's sorting | |
| CN108319973B (en) | Detection method for citrus fruits on tree | |
| CN104680519B (en) | Seven-piece puzzle recognition methods based on profile and color | |
| CN110648367A (en) | Geometric object positioning method based on multilayer depth and color visual information | |
| CN110315525A (en) | A kind of robot workpiece grabbing method of view-based access control model guidance | |
| CN104268602A (en) | Shielded workpiece identifying method and device based on binary system feature matching | |
| CN113781561B (en) | Target pose estimation method based on self-adaptive Gaussian weight quick point feature histogram | |
| CN107993224B (en) | Object detection and positioning method based on circular marker | |
| JP6483168B2 (en) | System and method for efficiently scoring a probe in an image with a vision system | |
| CN110110760A (en) | A kind of workpiece positioning and recognition methods based on machine vision | |
| CN111311618A (en) | Circular arc workpiece matching and positioning method based on high-precision geometric primitive extraction | |
| CN108907526A (en) | A kind of weld image characteristic recognition method with high robust | |
| CN113822810A (en) | Method for positioning workpiece in three-dimensional space based on machine vision | |
| CN108109154A (en) | A kind of new positioning of workpiece and data capture method | |
| CN108582075A (en) | A kind of intelligent robot vision automation grasping system | |
| CN112419429A (en) | Large-scale workpiece surface defect detection calibration method based on multiple viewing angles | |
| CN113012098A (en) | Iron tower angle steel punching defect detection method based on BP neural network | |
| CN113469195B (en) | Target identification method based on self-adaptive color quick point feature histogram | |
| CN113034526B (en) | Grabbing method, grabbing device and robot | |
| CN207752527U (en) | A kind of Robotic Dynamic grasping system | |
| CN114257736A (en) | Self-adaptive shooting method for workpieces | |
| CN109493369B (en) | Intelligent robot vision dynamic positioning and tracking method and system | |
| CN114882108A (en) | Method for estimating grabbing pose of automobile engine cover under two-dimensional image |
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: 20181218 |