CN105741308A - Method and system for obtaining multidimensional visual image - Google Patents
Method and system for obtaining multidimensional visual image Download PDFInfo
- Publication number
- CN105741308A CN105741308A CN201610126118.8A CN201610126118A CN105741308A CN 105741308 A CN105741308 A CN 105741308A CN 201610126118 A CN201610126118 A CN 201610126118A CN 105741308 A CN105741308 A CN 105741308A
- Authority
- CN
- China
- Prior art keywords
- image
- shooting
- dimension
- processing system
- image processing
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
Abstract
The invention discloses a method and system for obtaining a multidimensional visual image, which adds a reflection system. The method comprises steps of performing shooting through a visual system of one reflected dimension after reflecting the characteristic area images of all dimensions to one dimension, transmitting the shot image to an image processing system, performing processing on the shot image by the image processing system, wherein the processing comprises decomposition, combination, determination/or compensation, and obtaining the characteristic area images of all dimensions or the processing result. As a result, the visual system is not needed to be set on all the dimensions, which reduces the cost and enables the setting to be simple. The method and system for obtaining the multidimensional visual image are strong in universality, less in cooperation and conversion, and are easy and simple to obtain the characteristic area images of all dimensions.
Description
Technical field
The present invention relates to image processing techniques, particularly to acquisition methods and the system of a kind of multidimensional vision image.
Background technology
In current automated production and/or test system, in order to reach operating position or obtain correct judgement more accurately, generally after the image of erection visual system shooting characteristic area, utilize image processing system that captured feature regional images is detected or/and compensation etc..When being required for doing this monitoring and processing in multiple dimensions, after setting up the shooting that visual system carries out feature regional images in each dimension in a little dimensions to monitor and to process respectively, the feature regional images in each captured dimension is processed by recycling image processing system.
Specifically, it is possible to adopt following three kinds of modes to carry out.
First kind of way
In order to by identifying that label aligns and pastes on the tram of target workpiece, before pasting, it is necessary to judge whether this identification label exists and whether position offsets what identify label to setting up visual system on position；Whether exist and whether its position offsets to judge to be pasted target workpiece simultaneously, it is also required to shooting, judge, compensating and calculating etc. setting up another set of visual system on position in target workpiece, then two groups of offsets after calculating are sent to image processing system and carry out correct compensation, to reach accurately to paste the purpose identifying label, so, comparatively laborious being not easy to realizes.
The second way
For multiaspect object, such as hexahedro cube, when multiple faces of this multiaspect object have characteristic area need monitoring and process, it will usually polyhedral each erection camera system in dimension is shot and image procossing.
The third mode
For being distributed in the multiple target objects in larger space or scope, when needs are automatically monitored and process them, also it is that each target object is shot and processes more by the erection visual system overlapping one-dimensional degree.
Above to the monitoring of the feature regional images in each dimension with when processing, it is necessary to install visual system in each dimension, erection complexity, cost high, and the visual system cooperation conversion installed in each dimension many, incipient fault rate increases, and is unfavorable for safeguarding.
Summary of the invention
In view of this, the embodiment of the present invention provides the acquisition methods of a kind of multidimensional vision image, and the method can simply and readily obtain the feature regional images in each dimension.
The embodiment of the present invention also provides for the acquisition system of a kind of multidimensional vision image, and this system can simply and readily obtain the feature regional images in each dimension.
According to above-mentioned purpose, the present invention is achieved in that
A kind of acquisition methods of multidimensional vision image, described method includes:
Reflex system is set, the feature regional images in each dimension is reflected in a dimension；
Shot by the visual system of described reflection dimension；
Image after shooting is sent to image processing system by described visual system, image processing system the image after shooting is processed, obtain multidimensional vision image.
A kind of acquisition system of multidimensional vision image, including: reflex system, visual system and image processing system, wherein:
Reflex system, for being reflected in a dimension by the feature regional images in each dimension；
Visual system, after being shot by the feature regional images being reflected in a dimension, is sent to image processing system；
Image processing system, for the image after shooting is processed, obtains multidimensional vision image.
Adopt the method and system of the present invention, the embodiment of the present invention increases reflex system, after feature regional images in each dimension being reflected in a dimension, shot by the visual system of described reflection dimension, image after shooting is sent to image processing system, by image processing system, the image after shooting is processed, including decomposition, combination, judgement and/or compensation etc., obtain the feature regional images in each dimension or result.So, avoiding the need for as background technology, set up visual system in each dimension, thus reducing cost, and making erection simple, highly versatile, cooperation conversion is few, simply and readily obtains the feature regional images in each dimension.
Accompanying drawing explanation
The acquisition methods flow chart of the multidimensional vision image that Fig. 1 provides for the embodiment of the present invention；
The acquisition methods object lesson flow chart of the multidimensional vision image that Fig. 2 provides for the embodiment of the present invention
The acquisition system structure schematic diagram of the multidimensional vision image that Fig. 3 provides for the embodiment of the present invention.
Detailed description of the invention
For making the purpose of the present invention, technical scheme and advantage clearly understand, develop simultaneously embodiment referring to accompanying drawing, the present invention is described in further detail.
The present invention is in order to simply and readily obtain the feature regional images in each dimension, reflex system is set, after feature regional images in each dimension being reflected in a dimension, shot by the visual system of described reflection dimension, image after shooting is sent to image processing system, by image processing system, the image after shooting is processed, including decomposition, combination, judgement and/or compensation etc., obtain the feature regional images in each dimension or result.
The acquisition methods flow chart of the multidimensional vision image that Fig. 1 provides for the embodiment of the present invention, it concretely comprises the following steps:
Step 101, reflex system is set, the feature regional images in each dimension is reflected in a dimension；
Step 102, visual system by a dimension of described reflection shoot；
Image after shooting is sent to image processing system by step 103, described visual system, image processing system the image after shooting is processed, obtain multidimensional vision image.
In one embodiment, feature regional images in multiple dimensions is reflected in a dimension by reflex system, namely reflexing in target dimension by the feature regional images in other dimensions, reflex system can adopt concave surface, convex surface, plane, abnormal shape or other combinative structurees.The purpose of reflex system is to focus on the visual system in the real dimension implementing shooting by the characteristic area reflection needing monitoring and/or shooting, the position of reflex system is positioned at the scope that described visual system can shoot, and the setting of angle is biased in the angle of the characteristic area reflected.This angle can regulate as required, regulates this reflex system and enables by till the good observation of visual system acquisition or shooting characteristic area.The structure of reflex system, specifically adopts including size and shape generally according to the characteristic area needing the multiple dimensions changed in reality, adjusts or/and combine.Reflex system both can be simple one side reflecting mirror；Can also be slightly complicated double mirror, such as similar periscope etc.；Can even is that the more complicated combination being permitted polygonal mirror, be such as slr camera view finder.Further, consider the follow-up process to the image after shooting, particularly calculate displacement and/or compensation, consider the field depth that visual system can focus on when shooting, typically require and reflection unit is arranged from visual system slightly closer, in order to the image of multiple dimensions of reflection can all be shot clear by visual system, or the depth of field of visual system is adjusted, it is adjusted to bigger or maximum, in order to visual system can by clear for the image taking of multiple dimensions of reflection.
In one embodiment, visual system is used for shooting the image of the characteristic area in the dimension of place, and this figure is comprehensive or/and composite figure, is namely the image of the characteristic area of one-dimensional degree.
In one embodiment, image processing system is used to analyze and process the system of the image captured by visual system.Image processing system is made up of the image processing software run at computer or server, the shooting image received from visual system is analyzed and processes by it, including with the standard picture pre-saved to when compensate after, obtain visual pattern or the result of multidimensional.
In above-mentioned analysis process, need to use the standard picture pre-saved, the acquisition process of described standard picture is: visual system is arranged on station during normal operation, then characteristic area is arranged on normal station, regulates visual system so as to characteristic area to be carried out good observation and shooting；Then reflex system is set again, carries out position and angular adjustment so that the good of characteristic area that also can be obtained its monitoring corresponding in reflex system by visual system is observed and image.
In image processing system, the described standard picture pre-saved is as standard reference image, when being calculated displacement and/or compensating, also to set up the mapping relations between the coordinate of image captured by the true coordinate of characteristic area and visual system, characteristic area is moved accurate displacement or/and angle, and then obtain image by visual system shooting, can calculate, with reference to the image pixel difference of figure, the displacement obtaining true coordinate compared to primary standard or/and angle corresponds to the displacement of image coordinate or/and the mapping relations of angle.Certainly, considering the factor of error and the precision of images, it usually needs repeat the above steps, displacement that repeatedly moving characteristic region is different is or/and angle, calculate the mapping relations between true coordinate and image coordinate, and adopt error minimum and precision those group data most suitable.Similarly, can also obtaining the mapping relations between true coordinate and described feature regional images coordinate according to the method for the characteristic area part image corresponding to reflex system in shooting image, described part image splits independent process again from entire image.
When specifically processing, if reflex system is one side mirror system, then the area image of non-vision system place dimension is mirror image, so the shooting image-region of non-vision system place dimension is carried out mirror image processing；If reflex system is multiple reflection face or complex reflex system, then the area image of non-vision system place dimension can be normal, left and right on the contrary or contrary up and down, at this moment the shooting image-region of non-vision system place dimension is carried out the opposite direction adjustment of correspondence.
When specifically processing, after the shooting image corresponding region of different dimensions can being split, then alignment processing.This is because the distance of the area image of different dimensions and image model might not be identical, the mapping relations coefficient of the different dimensions that namely reflex system obtains is not necessarily identical.
In one embodiment, obtain multidimensional vision image after step 103, then export, it is possible to by the connection interface output arranged.
The acquisition methods object lesson flow chart of the multidimensional vision image that Fig. 2 provides for the embodiment of the present invention, it concretely comprises the following steps:
Step 201, visual system by a dimension of described reflection shoot；
Step 202, judge shooting whether complete, if it is, perform step 203；Perform if it is not, then return step 201；
Image after shooting is sent to image processing system by step 203, described visual system, image processing system the image after shooting is analyzed；
Step 204, image processing system export final multidimensional vision image or result.
The process in step 203 image after shooting being analyzed is:
According to different test events, the corresponding visual analysis mode arranged is selected to carry out.
Described visual analysis mode includes but not limited to: identify that label bar code, color analysis are or/and character string reading etc..
The acquisition system structure schematic diagram of the multidimensional vision image that Fig. 3 provides for the embodiment of the present invention, its concrete structure is: reflex system, visual system and image processing system, wherein,
Reflex system, for being reflected in a dimension by the feature regional images in each dimension；
Visual system, after being shot by the feature regional images being reflected in a dimension, is sent to image processing system；
Image processing system, for the image after shooting is processed, obtains multidimensional vision image.
In the structure shown here, described image processing system includes: analysis and processing unit and compensation deals unit, wherein,
Analysis and processing unit, for being analyzed comparing with the correspondence image prestored to the image after shooting, obtains analyzing result；Or directly the image after shooting is analyzed；
Compensation deals unit, for image and the correspondence image prestored after shooting to when calculating, obtain displacement and/or offset, the image after shooting can be compensated.
In the structure shown here, the position of described reflex system is positioned at the scope that described visual system can shoot, and angle initialization is partial to the angle of the characteristic area reflected.
In the structure shown here, described included by image processing system: the second analysis and processing unit, wherein,
Second analysis and processing unit, for the image after shooting is split according to the region that different dimensions is corresponding, the principle of reflection of the reflex system according to different dimensions, after carrying out the angle adjustment that image after the shooting of different dimensions is corresponding, then it is analyzed processing by the image after the shooting of the different dimensions after adjusting.
Hereinafter lift several specific embodiment to be described in detail
Specific embodiment one automatic labeling system
Step one gathers image process
First, open camera system, the feature regional images being reflected in a dimension is shot；
Next, it is determined whether shooting terminates, if it is not, then continue shooting；If it is, execution following step；
Finally, image processing system is entered.
The analyzing and processing process of step 2 image processing system
First, selected characteristic region Z, the judgement of operational objective workpiece and compensation calculation function；
Secondly, output obtains the offset compensated when whether characteristic area Z-image target workpiece exists and exist, here, it is necessary to carry out contrast with the correspondence image prestored and obtain；
Again, choosing the characteristic area Y with label, the presence or absence running label judges and compensation calculation function；
Again, output obtains whether label exists and offset when label exists, here, it is necessary to prestore the image that should have label carried out contrast obtain；
Finally, characteristic area Z and characteristic area Y is compensated by image processing system respectively according to calculated offset, to correct characteristic area Z and characteristic area Y.
Specific embodiment two
Step one gathers image process
First, open camera system, the feature regional images being reflected in a dimension is shot；
Next, it is determined whether shooting terminates, if it is not, then continue shooting；If it is, execution following step；
Finally, image processing system is entered.
The analyzing and processing process of step 2 image processing system
First, polyhedron multiaspect pattern restoring function is selected
Secondly, choose captured image A corresponding on the left of polyhedron, carry out pattern on the left of polyhedron and process the image A captured by functional model of reduction；
Again, choose captured image B corresponding on the right side of polyhedron, carry out pattern on the right side of polyhedron and process the image B captured by functional model of reduction；
Again, choose the captured image C that polyhedron front is corresponding, obtain polyhedron front pattern C；
Finally, export the captured image A processed through left side pattern, the image B captured by processing through right side pattern, the image C captured by obtaining is processed through front.
Specific embodiment three
Step one gathers image process
First, open camera system, the feature regional images being reflected in a dimension is shot；
Next, it is determined whether shooting terminates, if it is not, then continue shooting；If it is, execution following step；
Finally, image processing system is entered.
The analyzing and processing process of step 2 image processing system
First, captured image is selected to have the test of Non feature regions；
Secondly, captured characteristic area A corresponding to object a is chosen, it is judged that whether captured characteristic area internal object thing a exists；
Again, captured characteristic area B corresponding to object b is chosen, it is judged that whether captured characteristic area internal object thing b exists；
Again, captured characteristic area C corresponding to object c is chosen, it is judged that whether captured characteristic area internal object thing c exists；
Again, captured characteristic area D corresponding to object d is chosen, it is judged that whether captured characteristic area internal object thing d exists；
Finally, the result that output obtains characteristic area internal object thing a, whether b, c and d exist.
More than lift preferred embodiment; the object, technical solutions and advantages of the present invention have been further described; it is it should be understood that; the foregoing is only presently preferred embodiments of the present invention; not in order to limit the present invention; all any amendment of making, equivalent replace and improvement etc. within the spirit and principles in the present invention, should be included within protection scope of the present invention.
Claims (10)
1. the acquisition methods of a multidimensional vision image, it is characterised in that described method includes:
Reflex system is set, the feature regional images in each dimension is reflected in a dimension；
Shot by the visual system of described reflection dimension；
Image after shooting is sent to image processing system by described visual system, image processing system the image after shooting is processed, obtain multidimensional vision image.
2. the method for claim 1, it is characterised in that the position of described reflex system is positioned at the scope that described visual system can shoot, angle initialization is partial to the angle of the characteristic area reflected.
3. the method for claim 1, it is characterised in that described by image processing system, the image after shooting is carried out process be:
Image after shooting is split according to the region that different dimensions is corresponding, the principle of reflection of the reflex system according to different dimensions, after carrying out the angle adjustment that image after the shooting of different dimensions is corresponding, then it is analyzed processing by the image after the shooting of the different dimensions after adjusting.
4. the method for claim 1, it is characterised in that described by image processing system, the image after shooting is carried out process be:
To the image after shooting according to the test event set, carry out visual analysis；
Described visual analysis includes: identify label bar code, color analysis or/and character string reads.
5. the method for claim 1, it is characterised in that described by image processing system, the image after shooting is carried out process be:
After contrasting with the standard picture pre-saved, obtain shift value or/and offset, the image after shooting is carried out displacement or/and compensate.
6. method as claimed in claim 4, it is characterised in that described method also includes:
The multidimensional vision image output obtained.
7. the acquisition system of a multidimensional vision image, it is characterised in that including: reflex system, visual system and image processing system, wherein:
Reflex system, for being reflected in a dimension by the feature regional images in each dimension；
Visual system, after being shot by the feature regional images being reflected in a dimension, is sent to image processing system；
Image processing system, for the image after shooting is processed, obtains multidimensional vision image.
8. system as claimed in claim 7, it is characterised in that described image processing system includes: the first analysis and processing unit and the first compensation deals unit, wherein,
First analysis and processing unit, for being analyzed comparing with the correspondence image prestored to the image after shooting, obtains analyzing result；Or directly the image after shooting is analyzed；
First compensation deals unit, for carrying out when calculating to the image after shooting with the standard picture prestored, obtains shift value or/and offset, the image after shooting is carried out displacement or/and compensate.
9. system as claimed in claim 7, it is characterised in that the position of described reflex system is positioned at the scope that described visual system can shoot, and angle initialization is partial to the angle of the characteristic area reflected.
10. system as claimed in claim 7, it is characterised in that described included by image processing system: the second analysis and processing unit, wherein,
Second analysis and processing unit, for the image after shooting is split according to the region that different dimensions is corresponding, the principle of reflection of the reflex system according to different dimensions, after carrying out the angle adjustment that image after the shooting of different dimensions is corresponding, then it is analyzed processing by the image after the shooting of the different dimensions after adjusting.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610126118.8A CN105741308A (en) | 2016-03-07 | 2016-03-07 | Method and system for obtaining multidimensional visual image |
| TW105119848A TWI604261B (en) | 2016-03-07 | 2016-06-24 | A method for capturing multi-dimensional visual image and the system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610126118.8A CN105741308A (en) | 2016-03-07 | 2016-03-07 | Method and system for obtaining multidimensional visual image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105741308A true CN105741308A (en) | 2016-07-06 |
Family
ID=56249312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610126118.8A Pending CN105741308A (en) | 2016-03-07 | 2016-03-07 | Method and system for obtaining multidimensional visual image |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105741308A (en) |
| TW (1) | TWI604261B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107093181A (en) * | 2016-12-30 | 2017-08-25 | 西安航天精密机电研究所 | There is the multi-tag detection method of redundancy label in a kind of utilization vision system identification |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101137011A (en) * | 2006-08-29 | 2008-03-05 | 索尼株式会社 | Image processing apparatus, image processing method and computer program |
| CN102243185A (en) * | 2011-04-25 | 2011-11-16 | 苏州大学 | Multi-face imaging measurement device |
| US20140177942A1 (en) * | 2012-12-26 | 2014-06-26 | Industrial Technology Research Institute | Three dimensional sensing method and three dimensional sensing apparatus |
| CN104713527A (en) * | 2015-03-26 | 2015-06-17 | 江苏光导通信科技有限公司 | Single-time multi-angle shooting mechanism |
-
2016
- 2016-03-07 CN CN201610126118.8A patent/CN105741308A/en active Pending
- 2016-06-24 TW TW105119848A patent/TWI604261B/en active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101137011A (en) * | 2006-08-29 | 2008-03-05 | 索尼株式会社 | Image processing apparatus, image processing method and computer program |
| CN102243185A (en) * | 2011-04-25 | 2011-11-16 | 苏州大学 | Multi-face imaging measurement device |
| US20140177942A1 (en) * | 2012-12-26 | 2014-06-26 | Industrial Technology Research Institute | Three dimensional sensing method and three dimensional sensing apparatus |
| CN104713527A (en) * | 2015-03-26 | 2015-06-17 | 江苏光导通信科技有限公司 | Single-time multi-angle shooting mechanism |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107093181A (en) * | 2016-12-30 | 2017-08-25 | 西安航天精密机电研究所 | There is the multi-tag detection method of redundancy label in a kind of utilization vision system identification |
| CN107093181B (en) * | 2016-12-30 | 2020-01-31 | 西安航天精密机电研究所 | multiple label detection method for identifying existence of redundant label by using visual system |
Also Published As
| Publication number | Publication date |
|---|---|
| TWI604261B (en) | 2017-11-01 |
| TW201732412A (en) | 2017-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109949372B (en) | Laser radar and vision combined calibration method | |
| US11003940B2 (en) | System and methods for automatic solar panel recognition and defect detection using infrared imaging | |
| CN109270534B (en) | Intelligent vehicle laser sensor and camera online calibration method | |
| Heng et al. | Leveraging image‐based localization for infrastructure‐based calibration of a multi‐camera rig | |
| EP3033875B1 (en) | Image processing apparatus, image processing system, image processing method, and computer program | |
| CN110910350B (en) | Nut loosening detection method for wind power tower cylinder | |
| EP3496035B1 (en) | Using 3d vision for automated industrial inspection | |
| CN111626139A (en) | Accurate detection method for fault information of IT equipment in machine room | |
| CN110827361B (en) | Camera group calibration method and device based on global calibration frame | |
| CN110136186B (en) | Detection target matching method for mobile robot target ranging | |
| CN108596117B (en) | Scene monitoring method based on two-dimensional laser range finder array | |
| CN114581760B (en) | Equipment fault detection method and system for machine room inspection | |
| CN112605993A (en) | Automatic file grabbing robot control system and method based on binocular vision guidance | |
| CN116310678A (en) | Fire source identification and positioning method by combining solid-state laser radar with thermal imaging vision | |
| CN111738971B (en) | Circuit board stereoscopic scanning detection method based on line laser binocular stereoscopic vision | |
| CN114359865A (en) | Obstacle detection method and related device | |
| Mi et al. | A vision-based displacement measurement system for foundation pit | |
| CN116563391B (en) | Automatic laser structure calibration method based on machine vision | |
| CN105741308A (en) | Method and system for obtaining multidimensional visual image | |
| KR102198028B1 (en) | Position Verification Method for Equipment Layout at 3D Design of Smart Factory | |
| CN104239845A (en) | Double-vision intelligent image recognizing and target positioning method based on infrared thermal imaging technology | |
| CN115546716A (en) | Binocular vision-based method for positioning fire source around power transmission line | |
| CN111815560B (en) | Photovoltaic power station fault detection method and device, portable detection equipment and storage medium | |
| CN113778091A (en) | Method for inspecting equipment of wind power plant booster station | |
| Cheng et al. | 3D Radar and Camera Co-Calibration: A flexible and Accurate Method for Target-based Extrinsic Calibration |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into 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: 20160706 |