CN109520706A - Automobile fuse box assembly detection system, image-recognizing method and screw hole positioning mode - Google Patents
Automobile fuse box assembly detection system, image-recognizing method and screw hole positioning mode Download PDFInfo
- Publication number
- CN109520706A CN109520706A CN201811389468.9A CN201811389468A CN109520706A CN 109520706 A CN109520706 A CN 109520706A CN 201811389468 A CN201811389468 A CN 201811389468A CN 109520706 A CN109520706 A CN 109520706A
- Authority
- CN
- China
- Prior art keywords
- image
- fuse box
- measured
- character
- detection
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 66
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000002156 mixing Methods 0.000 claims description 2
- 238000005286 illumination Methods 0.000 description 10
- 238000004519 manufacturing process Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 241000238097 Callinectes sapidus Species 0.000 description 1
- 239000002775 capsule Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000002518 distortionless enhancement with polarization transfer Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M11/00—Testing of optical apparatus; Testing structures by optical methods not otherwise provided for
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23P—METAL-WORKING NOT OTHERWISE PROVIDED FOR; COMBINED OPERATIONS; UNIVERSAL MACHINE TOOLS
- B23P19/00—Machines for simply fitting together or separating metal parts or objects, or metal and non-metal parts, whether or not involving some deformation; Tools or devices therefor so far as not provided for in other classes
- B23P19/04—Machines for simply fitting together or separating metal parts or objects, or metal and non-metal parts, whether or not involving some deformation; Tools or devices therefor so far as not provided for in other classes for assembling or disassembling parts
- B23P19/06—Screw or nut setting or loosening machines
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23P—METAL-WORKING NOT OTHERWISE PROVIDED FOR; COMBINED OPERATIONS; UNIVERSAL MACHINE TOOLS
- B23P19/00—Machines for simply fitting together or separating metal parts or objects, or metal and non-metal parts, whether or not involving some deformation; Tools or devices therefor so far as not provided for in other classes
- B23P19/10—Aligning parts to be fitted together
- B23P19/102—Aligning parts to be fitted together using remote centre compliance devices
- B23P19/105—Aligning parts to be fitted together using remote centre compliance devices using sensing means
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The present invention discloses a kind of automobile fuse box assembly detection system, image-recognizing method and screw hole positioning mode.It include: detection platform, multi-cam vision collecting system, program-controlled torque torque spanner and computer, fuse box to be measured is located in the detection platform, and multi-cam vision collecting system is used to obtain the fuse box image to be measured from three angles；Program-controlled torque torque spanner is located at the top of the fuse box to be measured, and program-controlled torque torque spanner is by judging that nut model selects suitable dynamics by the screw tightening on fuse box to be measured；Computer is connect with multi-cam vision collecting system and program-controlled torque torque spanner, computer is used to receive the image information of multi-cam vision collecting system acquisition, and identifies whether fuse assembles the correct and program-controlled torque torque spanner of adjustment and tighten the nut of fuse box to be measured according to image information.It can be improved the detection accuracy and detection efficiency of automobile fuse box using the present invention.
Description
Technical field
The present invention relates to automobile fuse box detection fields, more particularly to a kind of automobile fuse capsule detection system
System, image-recognizing method and screw hole positioning mode.
Background technique
Requirement of the automobile fuse box to detection and assembly is very stringent, and the type detected needed for such product is more, dress
With process complexity, so that needing to put into the quality that a large amount of staff carries out product under the current technological process of production
Detection.Meanwhile there are the connectors such as screw and nut in such product, reach required for the different connectors on different location
The degree of coupling (i.e. the torque of screw) arrived is not also identical, causes in assembling process manual labor's intensity big, at high cost and can not
Guarantee higher reliability.On the other hand, it depends on manual labor unduly and increases enterprise's production cost, reduce production efficiency,
The space of Improving The Quality of Products is limited.
In recent years, machine vision has become the important component of industrial automation system, is widely suitable for all kinds of
Product line.Its internal structure is complex with assembly system for existing automobile fuse box detection system, so that one
It is directly mutually independent two systems, it is low to result in fitness between existing system, and inspection is separated with assemble flow, occupies treasured
Expensive production space also increases input cost of the factory in equipment.And detection system currently used in the market is due to skill
Art is limited to the more demanding of environment, it is desirable to provide the additional inspection environment without ambient light interference reduces detection efficiency.Separately
On the one hand, although gradually offer uploading detection data are genuinely convinced in for external high-end fuse box Automatic Production System at present
Be engaged in device function, but still cannot achieve and the entire production line interconnect.At the same time, although Manufacturing
Execution System and Enterprise Resource Planning system have largely used in the factory, but domestic
Outer detection and assembly system, with MES and ERP system without good interoperability, so that there is still a need for by being accomplished manually for factory
Data between production and management system are transmitted.
Although automobile fuse box mechanized production system is developed rapidly in recent years, due to its technical limitation
Bring equipment cost is higher, harsh to environmental requirement and the shortcomings that be difficult to existing information system combination, causes its fresh
Have and is disposed in factory's large area.
Summary of the invention
The object of the present invention is to provide the positioning of a kind of automobile fuse box assembly detection system, image-recognizing method and screw hole
Method can be improved the detection accuracy and detection efficiency of automobile fuse box.
To achieve the above object, the present invention provides following schemes:
A kind of automobile fuse box assembling detection device, comprising: detection platform, multi-cam vision collecting system, program-controlled
Torque torque spanner and computer, fuse box to be measured are located in the detection platform, the multi-cam vision collecting system
Including X-axis camera, Y-axis camera and Z axis camera, the multi-cam vision collecting system is used to obtain from three angles
The image of the fuse box to be measured；The program-controlled torque torque spanner is located at the top of the fuse box to be measured, the journey
Control-torque torque spanner is used to adjust the stubborn dynamics of the nut on the fuse box to be measured；The computer and more camera shootings
Head vision collecting system is connected with the program-controlled torque torque spanner, and the computer is adopted for receiving the multi-cam vision
The image information of collecting system acquisition, and identify whether fuse assembles the correct and adjustment program-controlled power according to described image information
The torsion and torque of square torque spanner tighten the nut of the fuse box to be measured.
Optionally, further includes:
Light even compensation system, positioned at the top of the fuse box to be measured, for being adopted for the multi-cam vision
Collecting system provides stable test environment.
Optionally, further includes:
Pneuma-lock system, between the fuse box to be measured and the detection platform, for fixing guarantor to be measured
Dangerous silk box.
Optionally, the orientation of the X-axis camera and the Y-axis camera is chosen not in same level mutually
Vertically, the Y-axis camera keeps 30 ° of angles with vertical direction to the detection platform outside.
A kind of detection recognition method based on color, comprising:
Obtain image to be detected of multi-cam vision collecting system acquisition；
Described image to be detected is transformed into HSV color space, by the different HSV color spaces H component and S points
Amount is mixed, the pixel after being mixed；The H component is the tone of tested pixel, and the S component is tested picture
The saturation degree of vegetarian refreshments；
Threshold segmentation is carried out to the pixel, obtains the color-ratio of fuse in image；
Choose average color of the highest color as fuse in the color-ratio.
A kind of detection and recognition methods based on character, comprising:
Obtain the character picture of multi-cam vision collecting system acquisition；
Printed page analysis is carried out to described image, the character picture after being analyzed；
Character picture after the analysis is split using row, obtains discrete character；
The discrete character is identified by classifier, obtains single character；
The single character is corrected using Markov model, obtains correct complete character.
Optionally, described to carry out printed page analysis to described image, the character picture after being analyzed specifically includes:
The pixel of the non-person's handwriting in described image is filtered out by stroke wide algorithm, obtains filtering image；
Branch is carried out using projection histogram to the filtering image, the character picture after being analyzed.
Optionally, described to described image progress printed page analysis, before the character picture after being analyzed, further includes:
Histogram equalization processing is carried out to described image.
A kind of screw hole coordinate extraction method, comprising:
Obtain the image of multi-cam vision collecting system acquisition；
Described image is traversed using gray level thresholding, obtains bianry image；
Extract the shadow region in the bianry image；
According to the shadow region, calculate X-axis camera direction and Y-axis camera direction in the shadow region in
Heart position；
Calculate the distance of each center and described image edge；
According to the distance, the coordinate of torque spanner is determined；
Judge whether the coordinate falls in the critical region of calibration；
If so, exporting the coordinate；
If it is not, being then back to the image step for obtaining multi-cam vision collecting system acquisition.
The specific embodiment provided according to the present invention, the invention discloses following technical effects: the present invention provides a kind of vapour
Vehicle fuse box assembling detection device, comprising: detection platform, multi-cam vision collecting system, program-controlled torque torque spanner and
Computer, fuse box to be measured are located in the detection platform, and the multi-cam vision collecting system includes X-axis camera, Y
Axis camera and Z axis camera, the multi-cam vision collecting system are used to obtain the fuse to be measured from three angles
The image of box；The program-controlled torque torque spanner is located at the top of the fuse box to be measured, the program-controlled torque torque spanner
For adjusting the stubborn dynamics of the nut on the fuse box to be measured；The computer and the multi-cam vision collecting system
It is connected with the program-controlled torque torque spanner, the computer is used to receive the figure of the multi-cam vision collecting system acquisition
Identify whether fuse assembles torsion that is correct and adjusting the program-controlled torque torque spanner as information, and according to described image information
Power and torque tighten the nut of the fuse box to be measured.Automobile fuse box can be improved using above-mentioned apparatus of the invention
Detection accuracy and detection efficiency.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is automobile fuse box of embodiment of the present invention assembling detection device structural schematic diagram；
Fig. 2 is detection recognition method flow chart of the embodiment of the present invention based on color；
Fig. 3 is detection of the embodiment of the present invention based on character and recognition methods flow chart；
Fig. 4 is screw hole of embodiment of the present invention coordinate extraction method flow chart.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide the positioning of a kind of automobile fuse box assembly detection system, image-recognizing method and screw hole
Method can be improved the detection accuracy and detection efficiency of automobile fuse box.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
Fig. 1 is automobile fuse box of embodiment of the present invention assembling detection device structural schematic diagram.As shown in Figure 1, a kind of vapour
Vehicle fuse box assembling detection device, comprising: detection platform 1, multi-cam vision collecting system, program-controlled torque torque spanner 3
And computer, fuse box 4 to be measured are located in the detection platform 1, the multi-cam vision collecting system includes X-axis camera shooting
First 21, Y-axis camera 22 and Z axis camera 23, the multi-cam vision collecting system are used for from described in three angles acquisitions
The RGB24 format video image of fuse box to be measured；The program-controlled torque torque spanner 3 is located at the fuse box 4 to be measured
Top, the program-controlled torque torque spanner 3 are used for the dynamics that the nut on the fuse box to be measured is stubborn；The computer with
The multi-cam vision collecting system is connected with the program-controlled torque torque spanner, and the computer for receiving described take the photograph more
Identify whether fuse assembles correct and adjustment as the image information of head vision collecting system acquisition, and according to described image information
The torsion and torque of the program-controlled torque torque spanner tighten the nut of the fuse box to be measured.
Fuse box assembling detection device, further includes:
Light even compensation system 5, positioned at the top of the fuse box 4 to be measured, for being the multi-cam vision
Acquisition system provides stable test environment.
Pneuma-lock system 6, it is to be measured for fixing between the fuse box 4 to be measured and the detection platform 1
Fuse box 4.
The orientation of the X-axis camera 21 and the Y-axis camera 22 is chosen mutually hangs down not in same level
Directly, the Y-axis camera 22 keeps 30 ° of angles with vertical direction to the detection platform outside.
Fig. 2 is detection recognition method flow chart of the embodiment of the present invention based on color.As shown in Fig. 2, a kind of be based on color
Detection recognition method, comprising:
Step 101: obtaining image to be detected of multi-cam vision collecting system acquisition；
Step 102: described image to be detected being transformed into HSV color space, by the H in the different HSV color spaces
Component and S component are mixed, the pixel after being mixed；The H component is the tone of tested pixel, and the S divides
Amount is the saturation degree of tested pixel；
Step 103: Threshold segmentation being carried out to the pixel, obtains the color-ratio of fuse in image；
Step 104: choosing average color of the highest color as fuse in the color-ratio.
The fuse of different model specification has different colors, while surface is also sprayed with fuse type, therefore can pass through
The fuse color in certain region is detected to judge whether the fuse assembles correctly.
Due to RGB (Red, Green, Blue) color model, its color change need to calculate each color maximum of three reference axis point
The line for measuring vertex and YMC (Yellow, Magenta, Cyan) color vertex, after being directly unfavorable for using original RGB data
Continuous color region segmentation.And HSV (Hue Saturation Value) three components of color model respectively correspond color
Three kinds of type, the depth and light and shade information, are transformed into HSV color space for image to be detected, by by the H of different colored regions with
Threshold value delimited after the mixing of S component, can simply be divided, the precision of color extraction and classification can be effectively improved.
Firstly, set variable max as the maximum value in r, g and b component of measured point, min be in the three-component of same point most
Small value.(h, s, v) value in its corresponding HSV space are as follows:
V=max
Wherein h is the tone of measured point, and value is between 0 to 360 °；S is the saturation degree of measured point, and value arrives for 0
Between 100%；V is the lightness of measured point, and value is 0 between max.
After carrying out above-mentioned processing by each pixel in altimetric image, Threshold segmentation is carried out to color by following formula,
To obtain the color proportion of certain fuse in specified image-region.
Dst (I)=lowerb (I)0≤src(I)0≤upperb(I)0
Λlowerb(I)1≤sro(I)1≤upperb(I)1
Wherein lower and upper is respectively the minimum value and most in the section HSV of various different model fuse surface colors
Big value；Dst is the value of the bianry image target point after over-segmentation.
It can get specified image district by calculating the ratio of brightened dot and all points in the bianry image obtained after segmentation
The color proportion of certain fuse in domain finally takes average face of the highest color of accounting as the region in the region
Color, thus the fuse model where obtaining the region with the corresponding table of model according to fuse color.
Fig. 3 is detection of the embodiment of the present invention based on character and recognition methods flow chart.As shown in figure 3, a kind of be based on word
The detection and recognition methods of symbol, comprising:
Step 201: obtaining the character picture of multi-cam vision collecting system acquisition；
Step 202: printed page analysis being carried out to described image, the character picture after being analyzed；
Step 203: the character picture after the analysis being split using row, obtains discrete character；
Step 204: the discrete character being identified by classifier, obtains single character；
Step 205: the single character being corrected using Markov model, obtains correct complete character.
Step 202, it specifically includes:
The pixel of the non-person's handwriting in described image is filtered out by stroke wide algorithm, obtains filtering image；
Branch is carried out using projection histogram to the filtering image, the character picture after being analyzed.
The method also includes:
Histogram equalization processing is carried out to described image.
Since the illumination condition variation that may occur in the white balance problem and detection process of camera will lead to upper one
It walks the average color got and deviation occurs, so that the identification of the more similar fuse model of minority HSV value occurs centainly
Erroneous judgement in degree.In order to improve fuse detection accuracy, detection error is reduced, using the dangerous wire type number indicated on fuse
(being generally made of number) further judges whether fuse assembles correctly.
But it is more demanding to picture quality since character extracts for color differentiation, therefore need to by altimetric image
It is pre-processed, the interference of external noise bring is reduced, to improve the efficiency and precision of digital extraction.
Firstly, histogram equalization processing is used, so that image overall color is evenly distributed.Extract RGB image
In one-component for example R component constitute a discrete single channel gray level image { x }, enable niIndicate that gray value i goes out in image { x }
Existing number, then gray value is that the probability of occurrence of the pixel of i can be indicated by following formula in image.
L is grey all in the gray level image, and n is the sum of all pixels in image, Px(i) constituting pixel value is i
Image histogram.Histogram is normalized, P will be corresponded toxCumulative distribution function cdfxIs defined as:
Wherein, cdfxIt (i) is the accumulative normalization histogram of image.Mapping function y=T (x) of the creation one about x,
One y is generated for each point in the gray level image { x } that extracts, then the cumulative probability function cdf of yyIt can pass through
Following formula is linearized in value range.
adfy(i)=iK
Wherein K is linear scale constant.According to Inverse distribution function it is found that cdfyIt can do such as down conversion:
cdfy(y ')=cdfy(T (x))=cdfx(x)
Wherein y' is the gray value after equalization.In order to by equalization after gray value y' be emitted back towards initial domain, need as
Down conversion:
Y '=y ' (max { x }-min { x })+min { x }
Wherein max { x }, min { x } are respectively maximum gradation value, the minimum gradation value in image { x }, and y " is to complete histogram
New gray level image after figure equalization.
Processing method on above-mentioned single channel image is respectively used to red, the green and blue point of image RGB color value
Amount, can be completed the processing to color image.
By carrying out histogram equalization to original image, can be obtained overall color be evenly distributed, the figure without color lump, few noise
Picture.
Then, using gamma transformation to overexposure or excessively dark picture is modified:
S=crγ
Wherein γ is Gamma factor, according to the different value Selective long-range DEPTs of γ or the contrast of reduction gray areas；C is
Normalize proportionality constant.γ=3 are obtained by experiment is repeated several times, and obtained picture contrast is higher, discrimination is preferable, place
Reason comparison is as shown in figure Fig. 4.
Character in image is identified.The first step is printed page analysis, by the wide algorithm of stroke the picture of non-person's handwriting
Element filters out, and reuses projection histogram and carries out branch, convenient for below by row processing.Second step is that row is split, and passes through connected domain
Analysis carries out row to be split as discrete character.Third portion is that identification character is converted discrete character by training SVM classifier
For single character.4th step is correction, and by Markov model, the character that mistake is identified in preceding step is corrected, from
And obtain correct, complete character content.
When specifically judging whether fuse assembly is correct, it is strong that illumination is adaptively adjusted according to ambient light first
Degree, after ambient light meets threshold value set by system, Z axis camera obtains fuse box orthography, then uses and mentions
Two kinds of identifications out obtain fuse rigging position and model with detection method, to judge whether fuse assembly is correct.
Fig. 4 is screw hole of embodiment of the present invention coordinate extraction method flow chart.As shown in figure 4, a kind of screw hole coordinate extraction side
Method, comprising:
Step 301: obtaining the image of multi-cam vision collecting system acquisition；
Step 302: described image being traversed using gray level thresholding, obtains bianry image；
Step 303: extracting the shadow region in the bianry image；
Step 304: according to the shadow region, calculating X-axis camera direction and Y-axis camera direction in the shadow region
Center in domain；
Step 305: calculating the distance of each center and described image edge；
Step 306: according to the distance, determining the coordinate of torque spanner；
Step 307: judging whether the coordinate falls in the critical region of calibration；
Step 308: if so, exporting the coordinate；
If it is not, being then back to the image step for obtaining multi-cam vision collecting system acquisition.
Vision collecting system follows pin-hole imaging model, by the way that two cameras are individually positioned in X and Y-direction, and protects
The intersection of its optic centre line is held, physically common O-xy coordinate system is constituted；Meanwhile the perspective relation up and down based on camera
Z axis is established, and merges with above-mentioned XY axis and establishes O-xyz three-dimensional coordinate system.It need to be convenient for worker operation in view of physical device, in Y-axis
Without vertical in same level with X-axis in the selection in the orientation of camera, but protected to testboard outside with vertical direction
30 degree of angle is held, so that will not influence the normal operation of whole system in the positive routine operation of testboard.
Measured piece image obtains:
The original image resolution as accessed by system is 1280*720, and image resolution ratio is excessively high, is contained a large amount of
Garbage, influence processing speed, it is therefore desirable to be cut out to original image.Since object under test is located at the two of X-axis and Y-axis
A camera center line point of intersection, so the strip image of the horizontal centre region of interception image totally 30~100px.
Since illumination condition is complicated and changeable under industrial production environment, image irradiation intensity distribution is caused to occur non-uniform existing
As needing to carry out illumination compensation to acquired image to eliminate the interference of ambient noise.By hardware illumination compensation method and soft
Part illumination compensation method combines, and efficiently solves the problems, such as external ambient light bring noise jamming.
On the one hand, multiple optical sensors are installed in test zone, pass through detection environment illumination intensity dynamic adjustment LED light
Source intensity of illumination, so that tested region illumination keeps stable state.
On the other hand, linear smoothing filtering is carried out to the data that camera is passed back using Gaussian filter algorithm, to eliminate
Gaussian noise.
Wherein σ is standard deviation, and x and y are respectively point transverse and longitudinal coordinate, and h is to pass through filtered gray value.
Screw hole coordinate extracts:
Since the background of image-region is homogeneous bands, can will be carried on the back when the contact head of torque spanner is placed in identification region
Light source in scape blocks, so that occurring biggish rectangular shadow region in image.
First using all pixels in gray level thresholding traversal image, original image f is divided into binary map according to threshold value T
As g, to extract the shadow region in image.Its thresholding formula is as follows:
Wherein i and j is the coordinate at correspondence image midpoint.Since the contact head of torque spanner can also result in X-axis and Y-axis side
There is rectangular shade to the different zones of image, so dashed horizontal center and image by calculating separately X-axis and Y direction
Coordinate (x, y) where the distance at edge can determine torque spanner.
Whether coordinates regional determined by finally judging, which falls in the critical region that system is demarcated, can determine whether that torsion is pulled
Screw hole corresponding to hand realizes the assembly of nut so that control system exports corresponding torque.
Torque spanner positioning experiment:
The staking-out work for manually completing coordinate where fuse screw in O-xyz coordinate system first, then in different illumination
Under the conditions of torque spanner is placed on four different screw T1, T2, T3, T4, according to the screw parameter of table 1, detection system
Whether the coordinate point number identified is correct, so that can test macro correctly identify screw corresponding to torque spanner.
1 test point result of table
The test point coordinate information being calculated in upper table is analyzed, coordinate spacing is obvious between each point, distinguishes
Degree is high, by combining X and Y axis coordinate that can correctly distinguish contact point information, and does not easily lead to erroneous judgement, can satisfy in engineer application
Needs.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other
The difference of embodiment, the same or similar parts in each embodiment may refer to each other.For system disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is said referring to method part
It is bright.
Used herein a specific example illustrates the principle and implementation of the invention, and above embodiments are said
It is bright to be merely used to help understand method and its core concept of the invention；At the same time, for those skilled in the art, foundation
Thought of the invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (9)
1. a kind of automobile fuse box assembling detection device characterized by comprising detection platform, multi-cam vision collecting
System, program-controlled torque torque spanner and computer, fuse box to be measured are located in the detection platform, the multi-cam vision
Acquisition system includes X-axis camera, Y-axis camera and Z axis camera, and the multi-cam vision collecting system is used for from three
Angle obtains the image of the fuse box to be measured；The program-controlled torque torque spanner is located at the upper of the fuse box to be measured
Side, the program-controlled torque torque spanner are used to adjust the stubborn dynamics of the nut on the fuse box to be measured；The computer with
The multi-cam vision collecting system is connected with the program-controlled torque torque spanner, and the computer for receiving described take the photograph more
Identify whether fuse assembles correct and adjustment as the image information of head vision collecting system acquisition, and according to described image information
The torsion and torque of the program-controlled torque torque spanner tighten the nut of the fuse box to be measured.
2. automobile fuse box assembling detection device according to claim 1, which is characterized in that further include:
Light even compensation system, positioned at the top of the fuse box to be measured, for being multi-cam vision collecting system
System provides stable test environment.
3. automobile fuse box assembling detection device according to claim 1, which is characterized in that further include:
Pneuma-lock system, between the fuse box to be measured and the detection platform, for fixing fuse to be measured
Box.
4. automobile fuse box assembling detection device according to claim 1, which is characterized in that the X-axis camera and
The orientation of the Y-axis camera is chosen not to be mutually perpendicular in same level, and the Y-axis camera is flat to the detection
30 ° of angles are kept with vertical direction on the outside of platform.
5. a kind of detection recognition method based on color characterized by comprising
Obtain image to be detected of multi-cam vision collecting system acquisition；
Described image to be detected is transformed into HSV color space, by the different HSV color spaces H component and S component into
Row mixing, the pixel after being mixed；The H component is the tone of tested pixel, and the S component is tested pixel
Saturation degree；
Threshold segmentation is carried out to the pixel, obtains the color-ratio of fuse in image；
Choose average color of the highest color as fuse in the color-ratio.
6. a kind of detection and recognition methods based on character characterized by comprising
Obtain the character picture of multi-cam vision collecting system acquisition；
Printed page analysis is carried out to described image, the character picture after being analyzed；
Character picture after the analysis is split using row, obtains discrete character；
The discrete character is identified by classifier, obtains single character；
The single character is corrected using Markov model, obtains correct complete character.
7. the detection and recognition methods according to claim 6 based on character, which is characterized in that it is described to described image into
Row printed page analysis, the character picture after being analyzed, specifically includes:
The pixel of the non-person's handwriting in described image is filtered out by stroke wide algorithm, obtains filtering image；
Branch is carried out using projection histogram to the filtering image, the character picture after being analyzed.
8. the detection and recognition methods according to claim 6 based on character, which is characterized in that it is described to described image into
Row printed page analysis, before the character picture after being analyzed, further includes:
Histogram equalization processing is carried out to described image.
9. a kind of screw hole coordinate extraction method characterized by comprising
Obtain the image of multi-cam vision collecting system acquisition；
Described image is traversed using gray level thresholding, obtains bianry image；
Extract the shadow region in the bianry image；
According to the shadow region, the centre bit of X-axis camera direction and Y-axis camera direction in the shadow region is calculated
It sets；
Calculate the distance of each center and described image edge；
According to the distance, the coordinate of torque spanner is determined；
Judge whether the coordinate falls in the critical region of calibration；
If so, exporting the coordinate；
If it is not, being then back to the image step for obtaining multi-cam vision collecting system acquisition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811389468.9A CN109520706B (en) | 2018-11-21 | 2018-11-21 | Screw hole coordinate extraction method of automobile fuse box |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811389468.9A CN109520706B (en) | 2018-11-21 | 2018-11-21 | Screw hole coordinate extraction method of automobile fuse box |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109520706A true CN109520706A (en) | 2019-03-26 |
| CN109520706B CN109520706B (en) | 2020-10-09 |
Family
ID=65777669
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811389468.9A Active CN109520706B (en) | 2018-11-21 | 2018-11-21 | Screw hole coordinate extraction method of automobile fuse box |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109520706B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110329760A (en) * | 2019-07-05 | 2019-10-15 | 四川凯达化工有限公司 | A kind of emulsion medicinal strip detection device and detection method |
| CN111445466A (en) * | 2020-04-01 | 2020-07-24 | 济南浪潮高新科技投资发展有限公司 | Bolt anti-screwing-leakage detection method and device and medium |
| CN111562267A (en) * | 2020-05-29 | 2020-08-21 | 重庆施鲁逊智能科技有限公司 | Visual detection system for assembling automobile fuse box |
| CN111571190A (en) * | 2020-05-22 | 2020-08-25 | 中国科学院自动化研究所 | Three-dimensional visual automatic assembly system and method |
| CN112643324A (en) * | 2020-12-28 | 2021-04-13 | 深圳硅纳智慧科技有限公司 | Automatic screw driving equipment and automatic screw driving method adopting same |
| CN113624145A (en) * | 2021-08-09 | 2021-11-09 | 吉林大学 | Visual realization method for measuring assembly height and inclination degree of device in automobile fuse box |
| CN113798834A (en) * | 2021-09-23 | 2021-12-17 | 鼎捷自动化科技有限公司 | Automatic assembling machine for bolt and nut of automobile fuse box |
| CN117415603A (en) * | 2023-12-18 | 2024-01-19 | 北京李尔现代坦迪斯汽车系统有限公司 | Bolt screwing robot for automobile seat production |
| CN117557561A (en) * | 2024-01-11 | 2024-02-13 | 凌源日兴矿业有限公司 | Underground roadway wall gap rapid detection method based on artificial intelligence |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1674047A (en) * | 2004-03-25 | 2005-09-28 | 上海大学 | Six freedom visual tracking method and system based on micro machine parallel processing structure |
| CN101751565A (en) * | 2008-12-10 | 2010-06-23 | 中国科学院自动化研究所 | Method for character identification through fusing binary image and gray level image |
| WO2012037343A1 (en) * | 2010-09-15 | 2012-03-22 | Ascentia Imaging, Inc. | Imaging, fabrication, and measurement systems and methods |
| CN105138999A (en) * | 2015-09-16 | 2015-12-09 | 三峡大学 | Shadow-based night object single-camera locating device and method |
| CN105631893A (en) * | 2016-03-09 | 2016-06-01 | 中国矿业大学 | Method and device for detecting whether capacitor is correctly mounted through photographing |
| CN106296744A (en) * | 2016-11-07 | 2017-01-04 | 湖南源信光电科技有限公司 | A kind of combining adaptive model and the moving target detecting method of many shading attributes |
| CN206037947U (en) * | 2016-08-31 | 2017-03-22 | 昆山泰丰自动化技术有限公司 | Car insurance silk box automated inspection machine |
-
2018
- 2018-11-21 CN CN201811389468.9A patent/CN109520706B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1674047A (en) * | 2004-03-25 | 2005-09-28 | 上海大学 | Six freedom visual tracking method and system based on micro machine parallel processing structure |
| CN101751565A (en) * | 2008-12-10 | 2010-06-23 | 中国科学院自动化研究所 | Method for character identification through fusing binary image and gray level image |
| WO2012037343A1 (en) * | 2010-09-15 | 2012-03-22 | Ascentia Imaging, Inc. | Imaging, fabrication, and measurement systems and methods |
| CN105138999A (en) * | 2015-09-16 | 2015-12-09 | 三峡大学 | Shadow-based night object single-camera locating device and method |
| CN105631893A (en) * | 2016-03-09 | 2016-06-01 | 中国矿业大学 | Method and device for detecting whether capacitor is correctly mounted through photographing |
| CN206037947U (en) * | 2016-08-31 | 2017-03-22 | 昆山泰丰自动化技术有限公司 | Car insurance silk box automated inspection machine |
| CN106296744A (en) * | 2016-11-07 | 2017-01-04 | 湖南源信光电科技有限公司 | A kind of combining adaptive model and the moving target detecting method of many shading attributes |
Non-Patent Citations (3)
| Title |
|---|
| 李睿等: "基于机器视觉的螺母定位系统", 《机床与液压》 * |
| 郭康德等: "基于视觉技术的三维指尖跟踪算法", 《计算机研究与发展》 * |
| 陈阳光: "工业机器人视觉定位与工件装配研究", 《万方》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110329760A (en) * | 2019-07-05 | 2019-10-15 | 四川凯达化工有限公司 | A kind of emulsion medicinal strip detection device and detection method |
| CN110329760B (en) * | 2019-07-05 | 2024-04-12 | 四川凯达化工有限公司 | Emulsion explosive strip detection device and detection method |
| CN111445466A (en) * | 2020-04-01 | 2020-07-24 | 济南浪潮高新科技投资发展有限公司 | Bolt anti-screwing-leakage detection method and device and medium |
| CN111445466B (en) * | 2020-04-01 | 2023-05-05 | 山东浪潮科学研究院有限公司 | Bolt anti-leakage screwing detection method, equipment and medium |
| CN111571190A (en) * | 2020-05-22 | 2020-08-25 | 中国科学院自动化研究所 | Three-dimensional visual automatic assembly system and method |
| CN111571190B (en) * | 2020-05-22 | 2022-01-25 | 中国科学院自动化研究所 | Three-dimensional visual automatic assembly system and method |
| CN111562267A (en) * | 2020-05-29 | 2020-08-21 | 重庆施鲁逊智能科技有限公司 | Visual detection system for assembling automobile fuse box |
| CN112643324A (en) * | 2020-12-28 | 2021-04-13 | 深圳硅纳智慧科技有限公司 | Automatic screw driving equipment and automatic screw driving method adopting same |
| CN113624145B (en) * | 2021-08-09 | 2022-04-26 | 吉林大学 | Visual realization method for measuring assembly height and inclination degree of device in automobile fuse box |
| CN113624145A (en) * | 2021-08-09 | 2021-11-09 | 吉林大学 | Visual realization method for measuring assembly height and inclination degree of device in automobile fuse box |
| CN113798834B (en) * | 2021-09-23 | 2022-09-23 | 鼎捷自动化科技有限公司 | Automatic assembling machine for bolt and nut of automobile fuse box |
| CN113798834A (en) * | 2021-09-23 | 2021-12-17 | 鼎捷自动化科技有限公司 | Automatic assembling machine for bolt and nut of automobile fuse box |
| CN117415603A (en) * | 2023-12-18 | 2024-01-19 | 北京李尔现代坦迪斯汽车系统有限公司 | Bolt screwing robot for automobile seat production |
| CN117415603B (en) * | 2023-12-18 | 2024-03-12 | 北京李尔现代坦迪斯汽车系统有限公司 | Bolt screwing robot for automobile seat production |
| CN117557561A (en) * | 2024-01-11 | 2024-02-13 | 凌源日兴矿业有限公司 | Underground roadway wall gap rapid detection method based on artificial intelligence |
| CN117557561B (en) * | 2024-01-11 | 2024-03-22 | 凌源日兴矿业有限公司 | Underground roadway wall gap rapid detection method based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109520706B (en) | 2020-10-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109520706A (en) | Automobile fuse box assembly detection system, image-recognizing method and screw hole positioning mode | |
| US7680320B2 (en) | Image processing method, substrate inspection method, substrate inspection apparatus and method of generating substrate inspection data | |
| DE102015210440A1 (en) | Image inspection device, image inspection method, image inspection program, computer readable recording medium and recording device | |
| CN107945184A (en) | A kind of mount components detection method positioned based on color images and gradient projection | |
| CN109993154B (en) | Intelligent identification method for single-pointer sulfur hexafluoride instrument of transformer substation | |
| CN103065334A (en) | Color cast detection and correction method and device based on HSV (Hue, Saturation, Value) color space | |
| CN110059701B (en) | Unmanned aerial vehicle landmark image processing method based on poor illumination | |
| CN117152648B (en) | Auxiliary teaching picture recognition device based on augmented reality | |
| KR20060048746A (en) | Area segmentation of an image | |
| CN208254424U (en) | A kind of laser blind hole depth detection system | |
| CN108805144A (en) | Shell hole recognition methods based on morphology correction and system, indication of shots equipment | |
| CN114926661B (en) | Textile surface color data processing and identifying method and system | |
| CN107094219B (en) | Image processing apparatus and image processing method | |
| CN114241438B (en) | Traffic signal lamp rapid and accurate identification method based on priori information | |
| CN101609508A (en) | Sign structure and recognition methods to object identification and orientation information calculating | |
| CN114067222A (en) | Urban water body remote sensing classification method and device | |
| CN103200349A (en) | Scanned image color cast automatic detection method | |
| CN116485801B (en) | Rubber tube quality online detection method and system based on computer vision | |
| CN113313638A (en) | Appearance defect detection method | |
| CN109784317B (en) | Traffic signal lamp identification method and device | |
| CN101815190A (en) | Projection display system and characteristic analysis method thereof | |
| CN115484712A (en) | Control method and device for tunnel entrance lighting system and storage medium thereof | |
| CN110288662B (en) | Display detection method and system | |
| CN111798442B (en) | Whiteness measuring method and parameter calibration method in whiteness measurement | |
| CN113834488B (en) | Robot space attitude calculation method based on remote identification of structured light array |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| OL01 | Intention to license declared | ||
| OL01 | Intention to license declared |