CN116879308A - Industrial machine vision system image processing method - Google Patents
Industrial machine vision system image processing method Download PDFInfo
- Publication number
- CN116879308A CN116879308A CN202311024612.XA CN202311024612A CN116879308A CN 116879308 A CN116879308 A CN 116879308A CN 202311024612 A CN202311024612 A CN 202311024612A CN 116879308 A CN116879308 A CN 116879308A
- Authority
- CN
- China
- Prior art keywords
- image
- light source
- ccd
- machine vision
- industrial
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 8
- 238000000034 method Methods 0.000 claims abstract description 71
- 238000005286 illumination Methods 0.000 claims abstract description 62
- 238000004519 manufacturing process Methods 0.000 claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 33
- 238000006073 displacement reaction Methods 0.000 claims abstract description 16
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000005070 sampling Methods 0.000 claims abstract description 9
- 230000005540 biological transmission Effects 0.000 claims abstract description 8
- 238000009432 framing Methods 0.000 claims abstract description 6
- 238000012994 industrial processing Methods 0.000 claims abstract description 6
- 230000009471 action Effects 0.000 claims abstract description 4
- 238000001514 detection method Methods 0.000 claims description 37
- 238000003384 imaging method Methods 0.000 claims description 33
- 230000033001 locomotion Effects 0.000 claims description 31
- 239000013598 vector Substances 0.000 claims description 30
- 238000009826 distribution Methods 0.000 claims description 17
- 238000005259 measurement Methods 0.000 claims description 15
- 238000012216 screening Methods 0.000 claims description 13
- 238000009776 industrial production Methods 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 238000003702 image correction Methods 0.000 claims description 3
- 230000001678 irradiating effect Effects 0.000 claims description 3
- 238000003825 pressing Methods 0.000 claims description 3
- 229910000938 samarium–cobalt magnet Inorganic materials 0.000 claims description 3
- 241001270131 Agaricus moelleri Species 0.000 claims 1
- 230000007547 defect Effects 0.000 description 27
- 238000010586 diagram Methods 0.000 description 8
- 238000005457 optimization Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 238000007688 edging Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000001105 regulatory effect Effects 0.000 description 3
- 238000005096 rolling process Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011895 specific detection Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Abstract
The application discloses an industrial machine vision system image processing method, which is based on an industrial machine vision system hardware architecture, and comprises a sampling framing system, a displacement device and a feedback linkage system which are mutually connected with data transmission and control transmission, wherein a plurality of light sources are arranged in a high-precision measuring device based on illumination reflection, and the three-dimensional morphology of a flaw product is reconstructed through images in different illumination modes; and embedding a high-precision measuring device based on illumination reflection, and finding whether product flaws exist in the micro-fine pipeline or the small-size characteristic inner hole according to the collected high-precision image feedback. According to the application, through the view finding assembly and the displacement device, the system integrally moves along with the process sequence of the industrial processing production line to be detected, the process precision of the processing production line is monitored, the action of the production line is reversely early-warned or corrected, and the problems generated in the production process are adjusted.
Description
Technical Field
The application discloses an image processing method of an industrial machine vision system, and relates to the technical field of machine vision.
Background
Machine vision is a branch of the rapid development of artificial intelligence. In short, machine vision is to use a machine instead of a human eye to make measurements and decisions. The machine vision system converts the shot target into an image signal through a machine vision product (namely an image shooting device, namely CMOS and CCD), and transmits the image signal to a special image processing system to obtain the form information of the shot target, and converts the form information into a digital signal according to the pixel distribution, brightness, color and other information; the image system performs various operations on these signals to extract characteristics of the object, and further controls the operation of the on-site device according to the result of the discrimination. In the mass repeated industrial production process, the machine vision detection method can greatly improve the production efficiency and the automation degree.
Some methods of industrial inspection using machine vision techniques exist in the prior art, such as:
the application patent of CN104971973B, namely an automobile wheel cover edging system and a using method thereof, refers to the application of a vision system, but is limited to the positioning function of the vision system. In the field of edging, the application of a vision system is stopped on the basis of positioning, detecting the existence or non-existence and the like, and the vision system is applied to track optimization so as to improve the manufacturing quality of edging, and still belongs to the blank.
The application patent of CN201810264297, namely a rolling binding track optimization method and system based on a visual system, mentions that in three processing procedures of rolling binding flanging, pre-binding and final binding, the visual system follows the collected images of binding parts, the forming parameters of a workpiece processed in each procedure are respectively calculated through the images, the quality of binding is judged through deviation analysis, problem diagnosis is carried out by means of a quality database, and the track optimization is carried out by feeding back the rolling binding system. However, the technical scheme is limited to preset track optimization, and cannot solve the problem that the traditional vision detection system based on a single camera cannot cover 360 degrees and the defect detection cannot be realized in a partial area.
The technical scheme disclosed in the patent of CN201810494716, namely a machine vision detection and screening device and a detection system based on a line scanning camera, realizes multi-angle coverage detection, but the technical scheme needs to install a specific rotating mechanism on a production line to detect defects around an object, is inconvenient to use, has low working efficiency, cannot detect the defects on the upper surface of the object, and cannot truly realize omnibearing defect detection.
In addition, the application of the machine vision technology in the industrial system in the prior art has the following problems: with the development of technology, the part structure in the industrial field is increasingly complex, and various defects of micro pipelines or small-size characteristic inner holes are difficult to observe due to detection precision and detection azimuth.
The conventional micro-pipe hole detection is to detect the micro-pipe after the miniaturization of the detection sensor, for example, a micro-endoscope or a CCD camera is used for entering the micro-hole to collect the inner surface image, and then the defects are identified by manual or image processing algorithms, but the methods can only judge whether the defects exist or not, and can not obtain three-dimensional information of the defects. The defect information in the pipe hole can only be qualitatively judged, and quantitative detection of defects in the inner wall of the pipe hole can not be realized.
The application patent of CN201810361225 discloses a device and a method for measuring three-dimensional morphology of a micropore based on an illumination reflection model, wherein the method comprises a motion traction module and a measurement sensor module, the motion traction module is operated to realize movement and positioning of the measurement sensor module in the micropore, and the measurement sensor module is used for completing acquisition of defect images of the inner surface of the micropore under different illumination modes. The method comprises the steps of obtaining an illumination reflection model of the inner surface of a hole based on a reflection model construction method through designing and developing a measurement sensor module, and completing reconstruction of a three-dimensional curved surface of the defect of the inner surface of the hole by utilizing a sequence image under a specific illumination mode adopted by an imaging system, so as to realize measurement of the geometric quantity of the defect of the inner surface of the micro hole, especially the height or depth dimension of the defect based on the illumination reflection model. But the application aims to meet the requirements of the industrial fields of aerospace, automobiles, chemical industry and the like on the measurement of the three-dimensional information of the defects in the inner surfaces of the micropores. The application scene of the method is single, and the use situation of the measurement and calculation method cannot be reasonably utilized.
Disclosure of Invention
The technical problems to be solved by the application are as follows: aiming at the defects of the prior art, an image processing method of an industrial machine vision system is provided, wherein the method is based on the industrial vision system, a top view finding assembly and a side view finding assembly are arranged, a high-precision measuring device based on illumination reflection is embedded in the top view finding assembly or the side view finding assembly, three-dimensional morphology of a micro pipeline or an image of a small-size characteristic inner hole is reconstructed according to the images in different illumination modes, whether product flaws affect the product quality is judged, the flaw generation reasons are deduced reversely, and the problem generated in the production process is regulated.
The application adopts the following technical scheme for solving the technical problems:
an industrial machine vision system image processing method is based on an industrial machine vision system hardware architecture, and comprises a sampling framing system, a displacement device and a feedback linkage system which are mutually connected with data transmission and control transmission, wherein the displacement device comprises a horizontal telescopic cross rod, a vertical telescopic upright post, a rotation adjusting device and a position moving assembly; the feedback linkage system comprises a feedback linkage processing device and a feedback linkage control device; the horizontal telescopic cross rod is connected with a vertical telescopic stand column through a rotary adjusting device, and the vertical telescopic stand column is arranged on the position moving assembly; the feedback linkage control device is arranged in the displacement device, and the feedback linkage processing device is arranged at the control end of the industrial production line; the view finding system comprises a top view finding assembly, a side view finding assembly, a top motion control device, a side motion control device and a high-precision measuring device based on illumination reflection; the front ends of the top and side view finding components are provided with high-precision measuring devices based on illumination reflection, and the top and side view finding components are connected with the horizontal telescopic cross rod through top and side motion control devices respectively; the high-precision measuring device based on illumination reflection comprises 2n light source modules, a view finding imaging module, a CCD spliced gathering reflector and a light source reflector, wherein n is a natural number; the light source modules are uniformly arranged around the view finding imaging module, the view finding imaging module is respectively connected with the CCD spliced gathering reflector and the light source reflector through the first support frame and the second support frame, light rays projected by the light source modules are irradiated into an object to be detected through the light source reflector, and light rays reflected by the object to be detected are gathered and reflected through the CCD spliced gathering reflector and then are projected to the view finding imaging module;
in the method, before measurement, CCD parameter calibration is completed by utilizing laser lattice images formed by a light source module (301) on the inner surface and the outer surface of an object to be measured according to a CCD calibration method;
when in measurement, the light source module (301) is turned on, the light of the light source module (301) irradiates on the light source reflector (303), and irradiates on the appointed position of the object to be measured after reflection; transmitting the illumination reflection intensity distribution in the current illumination mode to a view finding imaging module (302) through a CCD spliced gathering reflector (303) to obtain an aperture surface image in the illumination mode;
sequentially turning on light sources at different positions in a light source module (301) to obtain images of 2n pairs of different illumination modes of the surface position of the same object to be detected, calculating vector field data of the surface to be detected according to the imaging luminosity principle of a camera through the images, and then completing calculation of three-dimensional height field information of the detected area by using a direct corresponding relation between a normal vector and a gradient and using a gradient field to height field reconstruction technology;
and finally, combining CCD calibration parameters to obtain the geometric quantity information of the three-dimensional morphology of the hole surface.
As a further preferable scheme of the application, in the method, three-dimensional morphological parameters of the hole surface are obtained through the illumination reflection distribution state of the hole inner surface in a plurality of multi-angle illumination modes, namely, two-dimensional images of the hole inner surface in different illumination modes, specifically:
a global coordinate system centered on the viewfinder imaging module (302), with 1 coordinate axis aligned with the optic axis of the viewfinder imaging module (302), describes the object surface shape parameters as a function a=b (x, y), whose surface normal vector can be expressed as:
c= (m, n, -1), wherein the direction (m ', n') of the light source is specified with a fixed vector;
the relation expression between the surface reflection light distribution and the surface normal vector and the incidence direction of the light source is established as follows:
r=g (m, n, m ', n', d), where R is the surface reflected light distribution and d is the light source irradiance constant;
the illumination reflection distribution at a certain point on the surface of the object is expressed as R (m, n), and the gray value I (x, y) oc of the image of the surface of the object
R (m, n), and then establishing the relation between the object surface shape parameter and the two-dimensional image gray information;
2 light sources are used for respectively irradiating the surface of an object from a plurality of directions which are not coplanar, 2 images in different illumination directions are obtained, and then the normal vector of the surface is obtained from the 2 images;
let the light source direction of 2 images be ei= [ ei1, ei2], where eij, i, j=1, 2 are the light source and coordinate axis angle parameters, then e= [ E1, E2] constitutes 1 second order square matrix;
setting a vector of gray scale composition of the same point in 2 images, wherein the normal vector of the surface of the point is expressed as H= [ H1, H2], and the reflection coefficient of the point is k, and F= kEH;
under the condition that E-1 exists, H is a normalized vector, and k= |E-1I|; h=e-1I/|e-1 i|, wherein|represents a modulo operation, whereby a surface normal vector of the measured object can be obtained;
when the surface integrality condition of the object is met, calculating to obtain any 2-point height difference:
L AA’ ＝A(m)-A(m 0 )＝∫ S mdx+ndy；
s is any path between 2 points, so that the three-dimensional shape of the object surface is obtained.
When the method is applied to detection and screening of agricultural products, a positioning device and a screening and rejecting device are further arranged in a feedback linkage processing device at the control end of an industrial production line; after the agricultural products to be detected pass through the step of obtaining the three-dimensional shape of the object surface, screening out the agricultural products with the detection result of unqualified states according to comparison, positioning the positions of the agricultural products with the unqualified states in the batch detection products through the positioning device, and removing the unqualified agricultural products from the control end of the industrial production line through the screening and removing device.
As a further preferable aspect of the present application, the method further includes image processing of images acquired under different illumination, including the specific steps of:
step one, sequentially carrying out image acquisition, image correction, edge extraction, target identification, gray comparison, traversing and searching template characteristics and outputting comparison results;
step two, judging whether flaws exist or not by comparing the gray level of the acquired image with a set threshold value;
and thirdly, extracting through the image edge, and positioning the flaw position.
The specific method for acquiring the image comprises the following steps: and acquiring an image of the object to be detected through a sampling view finding system, and storing the image information of the pressing plate in a picture format.
The specific method for correcting the image comprises the following steps: marking the identification position of the qualified industrial object under the standard flow, precision and process on the appointed identification position of the object, wherein the identification position is a round black point, the identification position is positioned as the pixel coordinate of the round black point in the image, the industrial object which does not accord with the standard is identified by taking the pixel coordinate as the reference, and the inclined or deformed image is corrected.
The specific method for extracting the image edge comprises the following steps: the method comprises the steps of setting an image into a plurality of gray value intervals according to gray values of 0-255 in advance, and setting a parameter threshold corresponding to each gray value interval, so that the image is divided into a plurality of sections according to the set gray value intervals, namely, the image with the gray value in the same preset gray value interval is used as one section, the parameter threshold corresponding to each section of image is transmitted to a canny edge extraction function, and the edges of the identification positions are extracted;
the pixel gray value X is processed according to the following formula: x= (R 0 +G 0 +B 0 ) Wherein R is 0 、G 0 、B 0 Respectively isRGB information for each pixel of the picture.
The specific method for gray scale comparison comprises the following steps: and carrying out linear feature comparison state identification on the picture, judging whether the positioning of the identification position accords with the preset position of the qualified product, and further judging whether the object to be detected is in a qualified state.
As a further preferable scheme of the application, in the hardware architecture of the industrial machine vision system based on the method, the top motion control device, the side motion control device, the horizontal telescopic cross rod, the vertical telescopic stand column, the rotation adjusting device and the position moving assembly are all controlled and controlled by a servo motor and a PLC control main board, wherein the top motion control device, the side motion control device, the horizontal telescopic cross rod, the vertical telescopic stand column and the rotation adjusting device are driven by a hydraulic module, and the position moving assembly is driven by a motor and an electric roller.
The private clothing motor is a three-phase permanent magnet alternating current servo motor, and the specific model is Siemens 6SC61 series; the specific model of the PLC control main board is a SmCo permanent magnet alternating current servo motor controller; the specific model of the CCD device is as follows: sony EXVIEW HAD CCD.
The CCD spliced gathering reflector comprises a plurality of CCD devices, the effective pixels of the CCD devices are assembled into a double-row staggered focal plane mode in an end-to-end lap joint mode, namely, gaps formed by the first row of CCD devices are filled by the second row of CCD devices on the same plane, the end-to-end pixels of the adjacent CCD devices are aligned or overlapped for a certain distance, and clear wide-range large-view-field images are generated through integral delay processing.
Compared with the prior art, the technical scheme provided by the application has the following technical effects: according to the technical scheme disclosed by the application, the top and side view finding components are designed, and the number and the positions of the components can be regulated according to actual requirements; setting a displacement device, enabling the whole system to move sequentially along with the process of the industrial processing production line to be detected, and monitoring the process precision of the processing production line through image acquisition and comparison, and carrying out reverse early warning or correcting the action of the production line; embedding a high-precision measuring device based on illumination reflection in the top or side view finding assembly, and finding whether product flaws exist in a micro pipeline or a small-size characteristic inner hole according to the collected high-precision image feedback; and reconstructing three-dimensional morphology of the micro-pipeline or the small-size characteristic inner hole according to images in different illumination modes, judging whether product flaws influence product quality, deducing flaw generation reasons reversely, and adjusting problems generated in the production process.
Drawings
Fig. 1 is a schematic diagram of the system architecture of the present application.
Fig. 2 is a schematic diagram of connection control between functional modules in the present application.
Fig. 3 is a schematic diagram of structural imaging principle of a high-precision measurement device based on illumination reflection in the application.
Fig. 4 is a schematic diagram of an image processing step in the present application.
Wherein: 1. the device comprises a top view finding assembly, a side view finding assembly, a high-precision measuring device based on illumination reflection, a top motion control device, a side motion control device, a horizontal telescopic cross rod, a vertical telescopic upright post, a rotary adjusting device, a position moving assembly, a feedback linkage processing device and a feedback linkage control device, wherein the top view finding assembly, the side view finding assembly, the high-precision measuring device based on illumination reflection, the top motion control device, the side motion control device, the horizontal telescopic cross rod, the vertical telescopic upright post, the rotary adjusting device, the position moving assembly and the feedback linkage processing device are respectively arranged;
301. the device comprises a light source module 302, a view finding imaging module 303, a CCD spliced gathering reflector 304, a light source reflector 305 and an object to be detected.
Detailed Description
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for explaining the present application and are not to be construed as limiting the present application.
The technical scheme of the application is further described in detail below with reference to the accompanying drawings:
the application discloses an industrial machine vision system image processing method, a system hardware structure schematic diagram is shown in figure 1, the industrial machine vision system based on an improved sensing detection device comprises a sampling framing system, a displacement device and a feedback linkage system which are mutually connected with each other by data transmission and control transmission, wherein: the view finding system comprises a top view finding assembly, a side view finding assembly, a top motion control device, a side motion control device and a high-precision measuring device based on illumination reflection; the displacement device comprises a horizontal telescopic cross rod, a vertical telescopic upright post, a rotation adjusting device and a position moving assembly; the feedback linkage system comprises a feedback linkage processing device and a feedback linkage control device; the front ends of the top and side view finding components are provided with high-precision measuring devices based on illumination reflection, and the top and side view finding components are connected with the horizontal telescopic cross rod through top and side motion control devices respectively; the horizontal telescopic cross rod is connected with a vertical telescopic stand column through a rotary adjusting device, and the vertical telescopic stand column is arranged on the position moving assembly; the feedback linkage control device is arranged in the displacement device, and the feedback linkage processing device is arranged at the control end of the industrial production line.
The application designs the top and side view finding components, the number and the positions of the components can be regulated according to actual requirements, and the movement of the view finding components in the actual image acquisition process is automatically controlled by a programmable device.
And the top and side view finding assembly is additionally provided with a high-precision measuring device based on illumination reflection at one or more positions according to actual needs, and a light view finding result is utilized to find flaws or defects which are subtle or hidden in the tubular product. And a plurality of light sources are arranged in the high-precision measuring device based on illumination reflection, and the three-dimensional morphology of the flaw product is reconstructed through images in different illumination modes. And the motion control device of the top and side view finding groups is controlled by presetting a control program according to different product detection production lines, so that the control precision is improved.
The high-precision measuring device based on illumination reflection is provided with a plurality of light sources, three-dimensional morphology of a flaw product is reconstructed through images in different illumination modes, a structural imaging principle schematic diagram of the high-precision measuring device based on illumination reflection is shown in fig. 3, and the high-precision measuring device based on illumination reflection comprises a plurality of light source modules, a view finding imaging module, a CCD spliced gathering reflector and a light source reflector; the light source modules are uniformly arranged around the view finding imaging module, the view finding imaging module is respectively connected with the CCD spliced gathering reflector and the light source reflector through the first support frame and the second support frame, light rays projected by the light source modules irradiate into an object to be detected through the light source reflector, and light rays reflected by the object to be detected are gathered and reflected through the CCD spliced gathering reflector and then are projected to the view finding imaging module.
In order to complete acquisition of the distribution information of the illumination reflection intensity of the inner surface under the multi-illumination mode required by three-dimensional reconstruction of the inner surface of the object to be detected, the optical path structure designed by the application is as follows:
an even number of light source modules 301 which are uniformly distributed at intervals are arranged around the view finding imaging module 302, and can be arranged from 2, preferably 4, the front end of the optical axis of the view finding imaging module 302 is provided with two reflectors which are respectively light source reflectors 303 through a bracket, and the light source reflectors are used for changing the light source route in a narrow space so as to irradiate the inner surface and the outer surface of an object to be measured at different angles; and a CCD spliced gathering reflector 303, which is used for reflecting the image of the observed area on the inner surface of the object to be detected to the framing imaging module 302. Before measurement, a mature CCD calibration method in the prior art is adopted, and CCD parameter calibration is completed by utilizing laser dot matrix images formed by the light source module 301 on the inner surface and the outer surface of an object to be measured.
When in measurement, the light source module 301 is turned on, the light of the light source module 301 irradiates onto the light source reflector 303, and irradiates onto the appointed position (such as the inner surface) of the object to be measured after reflection, meanwhile, the illumination reflection intensity distribution in the current illumination mode is transmitted to the view finding imaging module 302 through the CCD spliced gathering reflector 303, and the hole surface image in the illumination mode is obtained.
Sequentially turning on light sources at different positions in the light source module 301, sequentially cycling to obtain images of 4 pairs of different illumination modes of the surface position of the same object to be detected, calculating vector field data of the surface to be detected through the 4 pairs of images on the premise that the illumination reflection model of the surface of the object to be detected is determined according to the imaging luminosity principle of a camera, and then completing calculation of three-dimensional height field information of a detected region by utilizing a direct corresponding relation between normal vectors and gradients and a reconstruction technology from gradient fields to height fields; and finally, combining CCD calibration parameters to obtain the geometric quantity information of the three-dimensional morphology of the hole surface.
The basic idea of the application is to obtain three-dimensional morphological parameters of the hole surface through the illumination reflection distribution state of the hole inner surface under a plurality of multi-angle illumination modes, namely two-dimensional images of the hole inner surface under different illumination modes.
Assuming that the surface of the object under test itself does not emit light, the object can be observed by the vision system because it is reflected by the light source irradiation. Selecting a global coordinate system centered on the viewfinder imaging module 302 such that 1 coordinate axis is aligned with the optical axis of the viewfinder imaging module 302, the object surface shape parameter can be described by a function a=b (x, y), which represents the change in the vertical distance (a-coordinate) of a point from the lens plane with the point coordinate, so its surface normal vector can be expressed as:
c= (m, n, -1), wherein the direction (m ', n') of the light source is specified with a fixed vector;
given the surface reflectance properties and lighting conditions of an object, a relational expression between the surface reflectance light distribution and the surface normal, the light source incidence direction, can be established as:
r=g (m, n, m ', n', d), where R is the surface reflected light distribution and d is the source irradiance constant.
According to the two-dimensional imaging principle, if the direction of the light source and the irradiance of the light source are fixed, the illumination reflection distribution of a certain point on the surface of the object can be expressed as R (m, n), and since I (x, y) is the image gray value of the surface of the object, the relation between the shape parameter (surface normal vector) of the surface of the object and the gray information of the two-dimensional image can be established. That is, the recovery of shape information of the object surface from the two-dimensional image is achieved using the inverse of the imaging process.
The application utilizes a plurality of gray scale image information to extract three-dimensional information of the object surface. The specific method comprises the following steps: for the same object surface, under the condition that the relative position between the camera and the object is kept unchanged, an even number of light sources are used for respectively irradiating the object surface from a plurality of directions which are not coplanar, 2n images in different illumination directions are obtained, and then the normal vector of the surface is obtained from the 2n images, wherein n is a natural number.
The assumption conditions for this three-dimensional reconstruction method are: the light sources are radiation sources with the same radiance in all directions, scattered light on the surface of the object uniformly irradiates in all directions, and all directions of the light sources are known.
Taking 2 images as an example, the calculation process is as follows:
let the light source directions of 2 images be ei= [ ei1, ei2], where eij, i, j=1, 2 are the light source and coordinate axis angle parameters, and e= [ E1, E2] constitute 1 second-order square matrix.
Let f= [ F1, F2] denote a vector of gray components of the same point in 2 images, and the surface normal vector of the point is denoted as h= [ H1, H2]. Let the reflection coefficient of this point be k, f= kEH. Under the condition that E-1 exists, H is a normalized vector, and k= |E-1I|; h=e-1I/|e-1 i|. Where, || represents a modulo operation, from which a surface normal vector of the measured object can be obtained.
If the object surface integrality condition is met, the integral is irrelevant to the path according to the green theorem, and the height difference of any 2 points can be obtained:
L AA’ ＝A(m)-A(m 0 )＝∫ S mdx+ndy,
s is any path between 2 points, so that the three-dimensional shape of the object surface is obtained. That is, the reconstruction of the normal field to the gradient field and then to the height field is accomplished using the surface normal field data.
In a specific embodiment of the method, when the method is applied to detection and screening of agricultural products, a positioning device and a screening and rejecting device are further arranged in a feedback linkage processing device at the control end of an industrial production line; after the agricultural products to be detected pass through the step of obtaining the three-dimensional shape of the object surface, screening out the agricultural products with the detection result of unqualified states according to comparison, positioning the positions of the agricultural products with the unqualified states in the batch detection products through the positioning device, and removing the unqualified agricultural products from the control end of the industrial production line through the screening and removing device.
In the present application, a schematic diagram of an image processing step is shown in fig. 4, and the high-precision measurement device based on illumination reflection detects according to the collected high-precision image feedback, and the specific detection and determination steps include:
step one, sequentially carrying out image acquisition, image correction, edge extraction, target identification, gray comparison, traversing and searching template characteristics and outputting comparison results;
step two, judging whether flaws exist or not by comparing the gray level of the acquired image with a set threshold value;
and thirdly, extracting through the image edge, and positioning the flaw position.
The method comprises the following specific steps:
(1) Collecting images; acquiring an image of an object to be detected through a sampling view finding system, and storing image information of a pressing plate in a picture format;
(2) Correcting an image; marking the identification position of the qualified industrial object under the standard flow, precision and process on the appointed identification position of the object, wherein the identification position is a round black point, the identification position is positioned as the pixel coordinate of the round black point in the image, the industrial object which does not accord with the standard is identified by taking the pixel coordinate as the reference, and the inclined or deformed image is corrected.
(3) Extracting an image edge; the method comprises the steps of presetting an image as a plurality of gray value intervals according to gray values of 0-255, and setting a parameter threshold corresponding to each gray value interval, so that the image is divided into a plurality of sections according to the set gray value intervals, namely, the image with the gray value in the same preset gray value interval is used as one section, the parameter threshold corresponding to each section of image is transmitted to a canny edge extraction function, and the edges of the identification positions are extracted. The pixel gray value X is processed according to the following formula:
X＝(R 0 +G 0 +B 0 )
wherein R is 0 、G 0 、B 0 RGB information for each pixel of the picture.
(4) Gray scale comparison; carrying out linear feature comparison state identification on the picture, judging whether the positioning of the identification position accords with the preset position of the qualified product, and further judging whether the object to be detected is in a qualified state;
(5) And finding out the to-be-detected object which accords with or does not accord with the set standard by carrying out traversal searching on the to-be-detected object image.
The image recognition method has obvious effect when facing agricultural products such as crops and plant parts to spoil and deteriorate due to collision, the damaged parts of the agricultural products with spoiled and deteriorated surfaces are obviously distinguished from other parts by colors, and the image recognition method can accurately and quickly distinguish the finished agricultural products from unqualified agricultural products with losses.
In a specific embodiment of the application, the top motion control device, the side motion control device, the horizontal telescopic cross rod, the vertical telescopic upright post, the rotation adjusting device and the position moving assembly are controlled and controlled by a servo motor and a PLC control main board, wherein the top motion control device, the side motion control device, the horizontal telescopic cross rod, the vertical telescopic upright post and the rotation adjusting device are driven by a hydraulic module, and the position moving assembly is driven by a motor and an electric roller.
As a preferable scheme of the application, the private clothes motor is a three-phase permanent magnet alternating current servo motor, and the specific model is Siemens 6SC61 series; the specific model of the PLC control main board is a SmCo permanent magnet alternating current servo motor controller; the specific model of the CCD device is as follows: sony EXVIEW HAD CCD.
The above-described device is a preferred embodiment, and any other existing servo motor and controller that satisfies the similar functions, if used in accordance with the disclosed method, would also be considered to be a method employing the disclosed method.
In the application, a connection control schematic diagram among functional modules is shown in fig. 2, an industrial machine vision system integrally moves along with the process sequence of an industrial processing production line to be detected through a displacement device, and image acquisition and comparison are carried out through a sampling view finding system to monitor the process precision of the processing production line and reversely early warn or correct the action of the production line; the displacement device moves according to the process sequence of the industrial processing production line to be detected, and the collected images are fed back to a specific processing and manufacturing flow; judging production process answer when the image detection result is matched with a preset standard image; when the image detection result is not matched with the preset standard image, judging that a product flaw or an industrial process design problem occurs, and sending out an early warning alarm or suspending the production activity of the production line by the system.
As a further preferable scheme of the application, the CCD spliced gathering reflector comprises a plurality of CCD devices, the effective pixels of the CCD devices are assembled into a double-row staggered focal plane mode in an end-to-end lap joint mode, namely, gaps formed by the first row of CCD devices are filled by the second row of CCD devices on the same plane, the end-to-end pixels of the adjacent CCD devices are aligned or overlapped for a certain distance, and clear wide-range large-view-field images are generated through integral delay processing.
The technical scheme disclosed by the application solves the following problems of the existing industrial machine vision system:
1. the traditional vision detection system based on a single camera cannot cover 360 degrees, and partial areas cannot realize defect detection. The detection system based on the line scanning camera is inconvenient to use, low in working efficiency and incapable of detecting defects on the upper surface of an object, and multi-angle omnibearing defect detection cannot be realized because a specific rotating mechanism is required to be installed on a production line to detect the defects on the periphery of the object.
2. In the prior art, the application of the industrial machine vision system is stopped on the basis of positioning, detecting the existence or non-existence and the like, and the detection of the production industry and the manufacturing quality of the detected product can be further realized by applying the machine vision system to track optimization through design.
3. With the development of technology, the part structure in the industrial field is increasingly complex, and various defects of micro pipelines or small-size characteristic inner holes are difficult to observe due to detection precision and detection azimuth.
4. The conventional micro-pipe hole detection is to detect the micro-pipe after the miniaturization of the detection sensor, for example, a micro-endoscope or a CCD camera is used for entering the micro-hole to collect the inner surface image, and then the defects are identified by manual or image processing algorithms, but the methods can only judge whether the defects exist or not, and can not obtain three-dimensional information of the defects. The defect information in the pipe hole can only be qualitatively judged, and quantitative detection of defects in the inner wall of the pipe hole can not be realized.
The embodiments of the present application have been described in detail with reference to the drawings, but the present application is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the spirit of the present application. The present application is not limited to the preferred embodiments, but is capable of modification and variation in detail, and other embodiments, such as those described above, of making various modifications and equivalents will fall within the spirit and scope of the present application.
Claims (7)
1. An industrial machine vision system image processing method is characterized in that: the industrial machine vision system hardware architecture based on the method comprises a sampling framing system, a displacement device and a feedback linkage system which are mutually connected with each other by data transmission and control transmission,
the displacement device comprises a horizontal telescopic cross rod, a vertical telescopic upright post, a rotation adjusting device and a position moving assembly; the feedback linkage system comprises a feedback linkage processing device and a feedback linkage control device; the horizontal telescopic cross rod is connected with a vertical telescopic stand column through a rotary adjusting device, and the vertical telescopic stand column is arranged on the position moving assembly; the feedback linkage control device is arranged in the displacement device, and the feedback linkage processing device is arranged at the control end of the industrial production line;
the view finding system comprises a top view finding assembly, a side view finding assembly, a top motion control device, a side motion control device and a high-precision measuring device based on illumination reflection; the front ends of the top and side view finding components are provided with high-precision measuring devices based on illumination reflection, and the top and side view finding components are connected with the horizontal telescopic cross rod through top and side motion control devices respectively;
the high-precision measuring device based on illumination reflection comprises 2n light source modules, a view finding imaging module, a CCD spliced gathering reflector and a light source reflector, wherein n is a natural number; the light source modules are uniformly arranged around the view finding imaging module, the view finding imaging module is respectively connected with the CCD spliced gathering reflector and the light source reflector through the first support frame and the second support frame, light rays projected by the light source modules are irradiated into an object to be detected through the light source reflector, and light rays reflected by the object to be detected are gathered and reflected through the CCD spliced gathering reflector and then are projected to the view finding imaging module;
in the method, before measurement, CCD parameter calibration is completed by utilizing laser lattice images formed by a light source module (301) on the inner surface and the outer surface of an object to be measured according to a CCD calibration method;
when in measurement, the light source module (301) is turned on, the light of the light source module (301) irradiates on the light source reflector (303), and irradiates on the appointed position of the object to be measured after reflection; transmitting the illumination reflection intensity distribution in the current illumination mode to a view finding imaging module (302) through a CCD spliced gathering reflector (303) to obtain an aperture surface image in the illumination mode;
sequentially turning on light sources at different positions in a light source module (301) to obtain images of 2n pairs of different illumination modes of the surface position of the same object to be detected, calculating vector field data of the surface to be detected according to the imaging luminosity principle of a camera through the images, and then completing calculation of three-dimensional height field information of the detected area by using a direct corresponding relation between a normal vector and a gradient and using a gradient field to height field reconstruction technology;
finally, combining CCD calibration parameters to obtain the three-dimensional shape geometric quantity information of the hole surface;
in the method, three-dimensional morphological parameters of the hole surface are obtained through the illumination reflection distribution state of the hole inner surface in a plurality of multi-angle illumination modes, namely two-dimensional images of the hole inner surface in different illumination modes, wherein the three-dimensional morphological parameters are specifically as follows:
a global coordinate system centered on the viewfinder imaging module (302), with 1 coordinate axis aligned with the optic axis of the viewfinder imaging module (302), describes the object surface shape parameters as a function a=b (x, y), whose surface normal vector can be expressed as:
c= (m, n, -1), wherein the direction (m ', n') of the light source is specified with a fixed vector;
the relation expression between the surface reflection light distribution and the surface normal vector and the incidence direction of the light source is established as follows:
r=g (m, n, m ', n', d), where R is the surface reflected light distribution and d is the light source irradiance constant;
the illumination reflection distribution at a certain point on the surface of the object is expressed as R (m, n), and the gray value I (x, y) oc of the image of the surface of the object
R (m, n), and then establishing the relation between the object surface shape parameter and the two-dimensional image gray information;
2 light sources are used for respectively irradiating the surface of an object from a plurality of directions which are not coplanar, 2 images in different illumination directions are obtained, and then the normal vector of the surface is obtained from the 2 images;
let the light source direction of 2 images be ei= [ ei1, ei2], where eij, i, j=1, 2 are the light source and coordinate axis angle parameters, then e= [ E1, E2] constitutes 1 second order square matrix;
setting a vector of gray scale composition of the same point in 2 images, wherein the normal vector of the surface of the point is expressed as H= [ H1, H2], and the reflection coefficient of the point is k, and F= kEH;
under the condition that E-1 exists, H is a normalized vector, and k= |E-1I|; h=e-1I/|e-1 i|, wherein|represents a modulo operation, whereby a surface normal vector of the measured object can be obtained;
when the surface integrality condition of the object is met, calculating to obtain any 2-point height difference:
L AA’ ＝A(m)-A(m 0 )＝∫ S mdx+ndy；
s is any path between 2 points, so that the three-dimensional shape of the surface of the object is obtained;
when the method is applied to detection and screening of agricultural products, a positioning device and a screening and rejecting device are further arranged in a feedback linkage processing device at the control end of an industrial production line;
after the agricultural products to be detected pass through the step of obtaining the three-dimensional shape of the object surface, screening out the agricultural products with the detection result of an unqualified state according to comparison, positioning the position of the unqualified state agricultural products in the batch detection products by the positioning device, and removing the unqualified agricultural products from the control end of the industrial production line by the screening and removing device;
in the industrial machine vision system hardware architecture based on the method, a top motion control device, a side motion control device, a horizontal telescopic cross rod, a vertical telescopic stand column, a rotation adjusting device and a position moving assembly are controlled and motion-controlled through a servo motor and a PLC control main board, wherein the top motion control device, the side motion control device, the horizontal telescopic cross rod, the vertical telescopic stand column and the rotation adjusting device are driven through a hydraulic module, and the position moving assembly is driven through a motor and an electric roller;
in the hardware architecture of the industrial machine vision system based on the method, the CCD spliced gathering reflector comprises a plurality of CCD devices, the effective pixels of the CCD devices are assembled in a double-row staggered focal plane mode in an end-to-end lap joint mode, namely, gaps formed by the CCD devices in the first row are filled by the CCD devices in the second row on the same plane, the end-to-end pixels of the adjacent CCD devices are aligned or overlapped for a certain distance, and a clear wide-range large-view-field image is generated through integral delay processing;
the industrial machine vision system integrally moves along with the process sequence of the industrial processing production line to be detected through the arranged displacement device, performs image acquisition and comparison through the sampling and framing system, monitors the process precision of the processing production line, and reversely early warns or corrects the action of the production line; the displacement device moves according to the process sequence of the industrial processing production line to be detected, and the collected images are fed back to a specific processing and manufacturing flow; judging production process answer when the image detection result is matched with a preset standard image; when the image detection result is not matched with the preset standard image, judging that a product flaw or an industrial process design problem occurs, and sending out an early warning alarm or suspending the production activity of the production line by the system.
2. The method for processing an image of an industrial machine vision system according to claim 1, further comprising the steps of:
step one, sequentially carrying out image acquisition, image correction, edge extraction, target identification, gray comparison, traversing and searching template characteristics and outputting comparison results;
step two, judging whether flaws exist or not by comparing the gray level of the acquired image with a set threshold value;
and thirdly, extracting through the image edge, and positioning the flaw position.
3. The method for processing the image of the industrial machine vision system according to claim 2, wherein the specific method for acquiring the image is as follows: and acquiring an image of the object to be detected through a sampling view finding system, and storing the image information of the pressing plate in a picture format.
4. The method for processing an image of an industrial machine vision system according to claim 2, wherein the specific method for correcting the image is as follows: marking the identification position of the qualified industrial object under the standard flow, precision and process on the appointed identification position of the object, wherein the identification position is a round black point, the identification position is positioned as the pixel coordinate of the round black point in the image, the industrial object which does not accord with the standard is identified by taking the pixel coordinate as the reference, and the inclined or deformed image is corrected.
5. The method for processing an image of an industrial machine vision system according to claim 2, wherein the specific method for extracting the image edge is as follows: the method comprises the steps of setting an image into a plurality of gray value intervals according to gray values of 0-255 in advance, and setting a parameter threshold corresponding to each gray value interval, so that the image is divided into a plurality of sections according to the set gray value intervals, namely, the image with the gray value in the same preset gray value interval is used as one section, the parameter threshold corresponding to each section of image is transmitted to a canny edge extraction function, and the edges of the identification positions are extracted;
the pixel gray value X is processed according to the following formula: x= (R 0 +G 0 +B 0 ) Wherein R is 0 、G 0 、B 0 RGB information for each pixel of the picture.
6. The method for processing an image of an industrial machine vision system according to claim 2, wherein the specific method for comparing gray scales is as follows: and carrying out linear feature comparison state identification on the picture, judging whether the positioning of the identification position accords with the preset position of the qualified product, and further judging whether the object to be detected is in a qualified state.
7. A method of processing an image of an industrial machine vision system as set forth in claim 1, wherein: in an industrial machine vision system hardware architecture based on the method, the private clothes motor is a three-phase permanent magnet alternating current servo motor, and the specific model is Siemens 6SC61 series; the specific model of the PLC control main board is a SmCo permanent magnet alternating current servo motor controller; the specific model of the CCD device is as follows: sony EXVIEW HAD CCD.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311024612.XA CN116879308A (en) | 2018-11-22 | 2018-11-22 | Industrial machine vision system image processing method |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311024612.XA CN116879308A (en) | 2018-11-22 | 2018-11-22 | Industrial machine vision system image processing method |
| CN201811396457.3A CN109375573A (en) | 2018-11-22 | 2018-11-22 | A kind of industrial machine vision system image processing method |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811396457.3A Division CN109375573A (en) | 2018-11-22 | 2018-11-22 | A kind of industrial machine vision system image processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116879308A true CN116879308A (en) | 2023-10-13 |
Family
ID=65383071
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311024612.XA Pending CN116879308A (en) | 2018-11-22 | 2018-11-22 | Industrial machine vision system image processing method |
| CN201811396457.3A Pending CN109375573A (en) | 2018-11-22 | 2018-11-22 | A kind of industrial machine vision system image processing method |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811396457.3A Pending CN109375573A (en) | 2018-11-22 | 2018-11-22 | A kind of industrial machine vision system image processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN116879308A (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751618B (en) * | 2019-06-05 | 2022-12-30 | 浙江大华技术股份有限公司 | Floater detection method and device and electronic equipment |
| CN111023964B (en) * | 2019-12-10 | 2021-08-17 | 广东省智能制造研究所 | Online detection device for surface profile and surface quality |
| CN111023955B (en) * | 2019-12-10 | 2022-01-18 | 广东省智能制造研究所 | High-dynamic high-precision dimension measurement and defect detection system and method thereof |
| CN111610199A (en) * | 2020-06-30 | 2020-09-01 | 无锡星燎不锈钢有限公司 | Hardware quality inspection system based on visual detection |
| CN115968439A (en) * | 2020-07-29 | 2023-04-14 | 西门子（中国）有限公司 | Method and device for inspecting assembly components of production line |
| CN112461850A (en) * | 2020-09-29 | 2021-03-09 | 江苏南高智能装备创新中心有限公司 | Workpiece surface flaw detection system |
| CN113051992B (en) * | 2020-11-16 | 2022-01-18 | 山东米捷软件有限公司 | Uniform speed identification system applying transparent card slot |
| CN112883842B (en) * | 2021-02-02 | 2022-12-16 | 四川省机械研究设计院(集团)有限公司 | Motorcycle engine assembling method and system based on mutual matching of parts and light source |
| CN113155847A (en) * | 2021-04-23 | 2021-07-23 | 浙江大学 | Oil smoke cover surface defect detecting system based on AI 3D vision |
| CN113607419A (en) * | 2021-08-02 | 2021-11-05 | 广东工业大学 | Engine cylinder block electronic tag detection device and method |
| CN113933300A (en) * | 2021-10-13 | 2022-01-14 | 福州大学 | Garbage misdelivery detection method, device and system based on machine vision |
| CN114170132B (en) * | 2021-10-20 | 2023-05-05 | 中国航发四川燃气涡轮研究院 | Flow tube static pressure hole quality detection method and system based on machine vision |
| CN116087201B (en) * | 2022-12-27 | 2023-09-05 | 广东尚菱视界科技有限公司 | Industrial vision detection system and detection method |
| CN117333480B (en) * | 2023-10-31 | 2024-03-01 | 曲阜冶通铸材科技发展有限公司 | Visual detection method, system, device and medium for flaw edge on surface of casting material |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3300682B2 (en) * | 1999-04-08 | 2002-07-08 | ファナック株式会社 | Robot device with image processing function |
| JP2001042699A (en) * | 1999-08-04 | 2001-02-16 | Minolta Co Ltd | Image forming device |
| JP3859574B2 (en) * | 2002-10-23 | 2006-12-20 | ファナック株式会社 | 3D visual sensor |
| EP2492668B1 (en) * | 2011-02-28 | 2013-08-28 | C.R.F. Società Consortile per Azioni | System and method for monitoring painting quality of components, in particular of motor-vehicle bodies |
| CN105486341B (en) * | 2015-11-25 | 2017-12-08 | 长春乙天科技有限公司 | A kind of large format high-speed, high precision automated optical detection equipment |
| CN206132661U (en) * | 2016-08-29 | 2017-04-26 | 齐鲁工业大学 | Welded pipe detection device |
| CN108447056A (en) * | 2018-03-26 | 2018-08-24 | 广西大学 | Power distribution cabinet circular pressing plate state identification method based on geometric properties clustering |
| CN108458658A (en) * | 2018-04-20 | 2018-08-28 | 南京航空航天大学 | A kind of micropore apparatus for measuring three-dimensional profile and method based on illumination reflection model |
-
2018
- 2018-11-22 CN CN202311024612.XA patent/CN116879308A/en active Pending
- 2018-11-22 CN CN201811396457.3A patent/CN109375573A/en active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| CN109375573A (en) | 2019-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116879308A (en) | Industrial machine vision system image processing method | |
| CN109544679B (en) | Three-dimensional reconstruction method for inner wall of pipeline | |
| CN106392304B (en) | A kind of laser assisted weld seam Intelligent tracing system and method | |
| CN108288288B (en) | Method, device and system for measuring precision shaft dimension based on visual identification | |
| CN100547394C (en) | Fruit quality detection system based on image information fusion technology | |
| CN109297413B (en) | Visual measurement method for large-scale cylinder structure | |
| CN107052086A (en) | Stamping parts surface defect detection apparatus and detection method based on 3D vision | |
| CN106735749B (en) | A kind of laser assisted weld seam Intelligent tracing system | |
| KR102122893B1 (en) | System and method for autonomous crack evaluation of structure based on uav mounted-hybrid image scanning | |
| CN106969706A (en) | Workpiece sensing and three-dimension measuring system and detection method based on binocular stereo vision | |
| CN102589516B (en) | Dynamic distance measuring system based on binocular line scan cameras | |
| CN206981462U (en) | Stamping parts surface defect detection apparatus based on 3D vision | |
| CN110065074A (en) | A kind of the visual servo laser orientation system and method for picking robot | |
| CN110966956A (en) | Binocular vision-based three-dimensional detection device and method | |
| CN102081296A (en) | Device and method for quickly positioning compound-eye vision imitated moving target and synchronously acquiring panoramagram | |
| CN104992446B (en) | The image split-joint method of non-linear illumination adaptive and its realize system | |
| CN106918306A (en) | Industrial products three-dimensional appearance real-time detecting system based on light field one camera | |
| CN114280075A (en) | Online visual inspection system and method for surface defects of pipe parts | |
| CN105023270A (en) | Proactive 3D stereoscopic panorama visual sensor for monitoring underground infrastructure structure | |
| CN109490314B (en) | Industrial machine vision system based on improved sensing detection device | |
| CN105391998B (en) | Automatic detection method and apparatus for resolution of low-light night vision device | |
| CN110097540A (en) | The visible detection method and device of polygon workpeace | |
| CN109631763A (en) | Irregular part detects localization method | |
| CN113189005A (en) | Portable surface defect integrated detection device and surface defect automatic detection method | |
| CN115326835B (en) | Cylinder inner surface detection method, visualization method and detection system |
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 |