CN108106554B - Machine vision-based large ring outer diameter detection system and method - Google Patents
Machine vision-based large ring outer diameter detection system and method Download PDFInfo
- Publication number
- CN108106554B CN108106554B CN201810184799.2A CN201810184799A CN108106554B CN 108106554 B CN108106554 B CN 108106554B CN 201810184799 A CN201810184799 A CN 201810184799A CN 108106554 B CN108106554 B CN 108106554B
- Authority
- CN
- China
- Prior art keywords
- laser
- image
- ring
- ring piece
- outer diameter
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/08—Measuring arrangements characterised by the use of optical techniques for measuring diameters
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a machine vision-based large ring outer diameter detection system and a machine vision-based large ring outer diameter detection method, wherein the system comprises an image acquisition unit, a motion control unit, an image processing unit and a feedback control unit; firstly, positioning a detection system, measuring the distance L between the center of a laser and the rolling center of the ring piece, then controlling the laser to rotate through a rotary table, enabling a green laser line to move left and right on the surface of the ring piece, obtaining whether the edge searching is successful or not according to an image processing algorithm, and if the edge searching is unsuccessful, controlling the rotary table to continue to move until the edge searching is successful; if the 'edge finding success' is displayed, the rotary table is controlled to rotate reversely to find the other edge of the ring piece, and finally the outer diameter size of the ring piece is confirmed through the distance L and the rotation angle relation of the laser. The method can accurately acquire the outer diameter size parameters of the ring piece, and simultaneously provides accurate data reference for the ring rolling machine rolling and expanding system, and provides a closed loop system for the rolling and expanding of the whole ring piece.
Description
Technical Field
The invention relates to a machine vision-based large ring outer diameter detection system and method, and belongs to the technical field of machine vision detection.
Background
The large seamless ring is widely applied to various industrial fields of heavy machinery, rail transit, wind power generation, aerospace, war industry and the like, such as wheels and wheel hoops of trains, bearing rings and gear ring blanks of wind power equipment, cabin bodies of carrier rockets, pressure vessels, reinforcing rings of nuclear reactors and the like. The manufacturing industry of the large ring piece has wide economic range and strong economic mobility, plays an important role in an economic industry chain, and is a fundamental industry for national economic development. The manufacturing technology of the large ring piece in China is relatively lagged behind, the material utilization rate is only 50% -55%, the production efficiency is low, the cost is high, and the resource and energy consumption is overlarge. The factors causing the phenomenon are many, and one important reason is that the size measurement of the forged piece is inaccurate in the forging process, the measurement time is long, and the production efficiency and the quality of the forged piece are greatly influenced. For large-size ring pieces, if the temperature drop is too large, the size requirement of the process cannot be met before the final forging temperature, and waste products are caused because the ring pieces cannot be heated in a furnace again. The diameter machining allowance of the ring piece has to be increased in a 'fat head big ear' mode, and data show that the material loss of a main large ring piece machining enterprise in China is up to 25% only due to the forging allowance.
Although forging enterprises in China have been developed rapidly, certain achievements are achieved; but still can not compare with developed countries abroad, besides the forging process is behind, the technical problem of detection of the sizes of the blank and the forging in the forging process is solved. The high temperature measurement technology has different temperature measurement requirements and modes in different fields, and can be divided into contact type and non-contact type according to the measurement mode. The traditional manual caliper method measurement commonly used by enterprises in China belongs to point-to-point measurement, and is low in precision and large in harm to human bodies; the ultrasonic measurement method is easily influenced by sound velocity, environmental media and the like and can not perform uninterrupted scanning; the electromagnetic measurement method is not suitable for measuring the metal material; optical measurement is an important solution to the problem in non-contact measurement, and image processing is an important approach to solve the problem.
The invention provides a novel detection system and a novel detection method for outer diameter parameters of large-sized ring pieces, which are characterized in that characteristic information is obtained by carrying out image processing and analysis on high-temperature images, the outer diameter and height dimension parameters of the ring pieces can be obtained at high precision by adopting a computer vision detection theory, the size machining efficiency and precision requirements of the ring pieces are accurately measured, ring rolling technological parameter adjustment is guided, the machining precision and machining efficiency of the large-sized ring pieces are improved, and the rejection rate is reduced.
Disclosure of Invention
The purpose of the invention is as follows: in order to overcome the defects in the prior art, the invention provides a machine vision-based large-sized ring outer diameter detection system and method, which can accurately obtain the outer diameter size parameters of the ring, provide accurate data reference for a ring rolling machine rolling and expanding system and provide a closed-loop system for the rolling and expanding of the whole ring.
The technical scheme is as follows: in order to achieve the purpose, the invention adopts the technical scheme that:
a large ring outer diameter detection system based on machine vision comprises:
the image acquisition unit acquires images of the side of the ring piece in real time through a CCD industrial camera, and projects a high-brightness laser line parallel to the axis of the ring piece on the surface of the ring piece through a laser to serve as a detection reference;
the motion control unit comprises a linear motion mechanism, a frame plate erected on the linear motion mechanism and two adjacent rotary tables arranged on the frame plate, the two rotary tables are used for respectively controlling the autorotation motion of the laser and the CCD camera, and the linear motion of the laser and the CCD camera on the frame plate parallel to the axis of the ring piece is controlled by the linear motion mechanism;
the image processing unit is used for carrying out image processing on the image acquired by the CCD camera through an opencv (computer vision library), and utilizing an image color space change theory and designing an algorithm to extract target image characteristics;
and the feedback control unit judges whether the edge is searched by comparing and analyzing the contact ratio of the image features extracted by the image processing unit and the highlight laser lines, controls the rotary table to drive the laser to rotate in the reverse direction if the edge is searched, and controls the rotary table to drive the laser to rotate continuously in the direction until the edge is judged to be searched successfully if the edge is not searched.
Furthermore, because the ring piece is in a high-temperature red hot state during rolling and expanding, in order to improve color contrast and make reference lines clear, a green linear laser is adopted as the laser, and the power is 1000 mw.
Further, an angular displacement encoder is arranged on a frame plate on one side of the rotary table where the laser is located.
A large ring outer diameter detection method based on machine vision is characterized in that a visual detection mode in a non-contact measurement mode is selected for detection according to analysis of the surface temperature of a ring up to thousands of degrees centigrade; firstly, the position of a detection system is positioned, the distance L between the center of a laser and the rolling and expanding center of the ring piece is measured, and the outer diameter size of the ring piece is confirmed according to the distance L and the rotation angle relation of the laser. The method specifically comprises the following steps:
step 1: the detection system is placed on one side of the ring piece to be detected, and the distance L between the center of the laser and the center of the ring piece is measured by using the laser range finder (by detecting the position relationship between the control system equipment and the ring piece to be detected, on one hand, the distance between the instrument and the high-temperature ring piece can be kept, so that the equipment such as a CCD camera and the like can be ensured to be normally used without being influenced by high-temperature heat radiation, the working environment of the equipment is stabilized, the service life of the equipment is prolonged, and on the other hand, the outer diameter size of the;
step 2: acquiring images of the side part of the ring piece in real time through a CCD camera, and projecting a high-brightness laser line parallel to the axis of the ring piece on the surface of the ring piece through a laser to serve as a detection reference; in the image acquisition process, the laser and the CCD camera on the frame plate are controlled to linearly move parallel to the axis of the ring piece through the linear motion mechanism so as to ensure the complete imaging of the CCD camera, meanwhile, the laser is controlled to rotate left and right by the rotary table, the other rotary table controls the rotation of the CCD camera, and the high-brightness laser line in the image is tracked;
and step 3: analyzing the position of the side edge of the thermal ring by an image processing unit, and controlling a rotary table to drive a laser to rotate towards the edge position of the ring by a feedback control unit;
and 4, step 4: an equation of the high-brightness laser line under an image pixel coordinate system is analyzed through an image processing unit, and the equation is set to be u-u because the laser line projected by a laser is parallel to the axis of the ring piece0(where u refers to the image equation expression in the image pixel coordinate system, and since the green stripe of light is projected parallel to the ring axis, the equation form is u-u0)；
Step 6, by comparing u0,u1,u2Whether the edge of the ring piece is tracked by the highlight laser line at the moment is obtained according to the size of the ring piece, namely: if it is | u1-u0Delta u or | u is less than or equal to |)2-u0If | ≦ Δ u, u can be determined0And u1/u2Equal, the laser line has sought "one side", whereDelta u is a pixel error (the specific value of delta u is determined according to the diameter of the ring piece and the distance between the detection system and the ring piece), otherwise, the turntable continues to rotate towards the direction until the system judges to be equal;
and 7, if the system judges that one side is found, the system controls the laser to rotate reversely through the rotary table, finds another boundary, records the reverse rotation angle theta of the rotary table where the laser is located, and obtains the outer diameter size of the ring piece as
Further, the step 4 specifically includes:
step 401, performing noise reduction processing on an image acquired by a CCD camera in real time through opencv;
step 402, analyzing image characteristics, performing color space conversion on the image characteristics, and converting the image characteristics from an original RGB color space to an HSV color space;
step 403, performing selective binarization processing on the image by using an vary function in opencv to obtain the position of the laser line in the image;
step 404, thinning lines in the binary image by adopting a Zhang fast parallel thinning algorithm;
step 405, performing straight line fitting on the thinned lines by using a least square method to obtain a straight line equation u-u of the highlight laser lines under the image pixel coordinate system0。
Further, the step 5 specifically includes:
step 501, performing noise reduction processing on an image acquired by a CCD camera in real time through opencv;
502, analyzing image characteristics, performing color space conversion on the image characteristics, and converting the image characteristics from an original RGB color space to an HSV color space;
step 503, carrying out selective binarization processing on the image by using an vary function in opencv to obtain the position of the thermal ring piece in the image;
step 504, performing morphological operation on the binary image, and lubricating the image contour by using open operation;
step 505, performing rectangle detection on the processed image through opencv, specifically including: firstly, extracting a target contour in a binary image by using a library function findContours in opencv, obtaining a rectangle containing a point set with the minimum area by using a minAreaRect function, detecting the contour of the rectangle by using a set constraint condition (the condition is determined to be suitable for analysis of an actual image), obtaining four vertex coordinates of the rectangle, and setting the four vertex coordinates as A (u)a,va)，B(ub,vb)，C(uc,vc)，D(ud,vd) The term "u, v" as used herein refers to a point coordinate system in the image pixel coordinate system, and u is determined by the judgmenta，ub，uc，udThe two-side perpendicular line equation U-U of the rectangular detection is solved according to the size (two-two are equal in the error range of delta U)1And u ═ u2I.e. the ring edge equation.
Has the advantages that: compared with the prior art, the large ring outer diameter detection system and method based on machine vision provided by the invention have the following effects: 1. the structure is simple, the operation is safe, the use is convenient, the characteristic information is obtained by carrying out image processing analysis on the high-temperature image, and the outer diameter and height dimension parameters of the ring piece with high precision are obtained by adopting a computer vision detection theory; 2. the method can accurately measure the size machining efficiency and precision requirements of the ring piece, guide the adjustment of ring rolling technological parameters, contribute to improving the machining precision and machining efficiency of the large-size ring piece and reduce the rejection rate.
Drawings
FIG. 1 is a frame diagram of a ring outer diameter detection system according to the present invention;
FIG. 2 is a schematic diagram of the outer diameter measurement of the ring member according to the present invention;
the figure includes: 1. ring piece, 2, rotary table, 3, CCD camera, 4, laser, 5, angular displacement encoder, M-laser center, O-ring piece center, P1、P2-the point of tangency of the laser ray with the edge of the ring, the distance between the center of the L-laser and the center of the ring, the theta-laser ray sweeping from one side edge of the ring to the other side edgeThe angular size of the rim.
Detailed Description
The present invention will be further described with reference to the accompanying drawings.
Fig. 1 shows a large ring outer diameter detection system based on machine vision, which includes:
the image acquisition unit acquires images of the side of the ring piece in real time through a CCD camera, and projects a high-brightness laser line parallel to the axis of the ring piece on the surface of the ring piece through a laser to serve as a detection reference;
the motion control unit comprises a linear motion mechanism (such as a ball screw, a linear slide rail and the like), a frame plate erected on the linear motion mechanism, and two adjacent rotary tables arranged on the frame plate, wherein the two rotary tables are used for respectively controlling the autorotation motion of the laser and the CCD camera, and the linear motion of the laser and the CCD camera on the frame plate parallel to the axis of the ring piece is controlled by the linear motion mechanism;
the image processing unit is used for carrying out image processing on the image acquired by the CCD camera through an opencv (computer vision library), and utilizing an image color space change theory and designing an algorithm to extract target image characteristics;
and the feedback control unit judges whether the edge is searched by comparing and analyzing the contact ratio of the image features extracted by the image processing unit and the highlight laser lines, controls the rotary table to drive the laser to rotate in the reverse direction if the edge is searched, and controls the rotary table to drive the laser to rotate continuously in the direction until the edge is judged to be searched successfully if the edge is not searched.
In the embodiment, because the ring piece is in a high-temperature red-hot state during rolling and expanding, in order to improve color contrast and make reference lines clear, a green linear laser is adopted as the laser, and the power is 1000 mw; and an angular displacement encoder is arranged on the frame plate at one side of the rotary table where the laser is positioned.
A large ring outer diameter detection method based on machine vision is characterized in that a visual detection mode in a non-contact measurement mode is selected for detection according to analysis of the surface temperature of a ring up to thousands of degrees centigrade; firstly, the position of a detection system is positioned, the distance L between the center of a laser and the rolling and expanding center of the ring piece is measured, and the outer diameter size of the ring piece is confirmed according to the distance L and the rotation angle relation of the laser. The method specifically comprises the following steps:
1. motion control
The PC end controls the laser to rotate left and right through the turntable, so that the green laser light bar can be projected on the surface of the ring piece in parallel to the axis of the ring piece. In addition, the PC end controls the ball screw to do lifting motion, so that the CCD camera installed on the other turntable can clearly image.
The PC end sends an instruction to control the rotary table to rotate, so that the green laser line moves leftwards or rightwards on the surface of the ring piece, whether the edge searching is successful or not is obtained according to an image processing algorithm, and if the edge searching is unsuccessful, the PC end controls the rotary table to continue to move until the edge searching is successful; if the 'edge finding success' is displayed, the PC end controls the rotary table to rotate reversely to find the other edge of the ring piece.
2. Image processing algorithm analysis
2.1 Gauss filtering noise reduction
As the ring piece runs in the rolling and expanding process, huge noise generated by friction among the core rod, the conical rod and the ring piece can cause certain interference on image acquisition, and the influence of the noise on image imaging is eliminated by adopting a Gaussian filtering mode.
The gaussian filtering is a process of weighted average of the whole image, and the value of each pixel point is obtained by weighted average of itself and other pixel values in the neighborhood. The specific operation of gaussian filtering is: each pixel in the image is scanned using a template (or convolution, mask), and the weighted average gray value of the pixels in the neighborhood determined by the template is used to replace the value of the pixel in the center of the template.
2.2 color space conversion
First, the matrix type of the picture captured by the CCD camera is CV _8UC3, which is an 8-bit unsigned character type matrix, where 3 is 3 channels, i.e. R, G, B channels. The RGB color space is based on three basic colors of R (red), G (green), and B (blue), and different levels of superposition are performed, thereby producing abundant and wide colors.
The HSV color space is a color space created according to the intuitive characteristics of colors, H-Hue, S-Saturation, and V-Value.
The hue H is measured by angle, and has a value ranging from 0 to 360 °, starting from red and counting in the counterclockwise direction, with red being 0 °, green being 120 °, and blue being 240 °. Their complementary colors are: yellow is 60 °, cyan is 180 °, and magenta is 300 °.
The saturation S represents the degree to which the color approaches the spectral color. A color can be seen as the result of a mixture of a certain spectral color and white. The greater the proportion of spectral colors, the higher the degree of color approaching spectral colors and the higher the saturation of colors. High saturation and dark and bright color. The white light component of the spectral color is 0, and the saturation reaches the highest. Usually the value ranges from 0% to 100%, the larger the value, the more saturated the color.
Lightness V represents the degree of brightness of the color, for a light source color, the lightness value is related to the lightness of the illuminant; for object colors, this value is related to the transmittance or reflectance of the object. Values typically range from 0% (black) to 100% (white).
The algorithm for RGB to HSV conversion is as follows:
max＝max(R,G,B)；
min＝min(R,G,B)；
V＝max(R,G,B)；
S＝(max-min)/max；
if(R＝max)H＝(G-B)/(max-min)*60；
if(G＝max)H＝120+(B-R)/(max-min)*60；
if(B＝max)H＝240+(R-G)/(max-min)*60；
if(H<0)H＝H+360；
the specific functions of the opencv built-in function library can be realized only by calling a cvtColor function, and at the moment, the influence of the heat radiation of the high-temperature ring in an HSV color space is small, so that the contour edge of the ring can be clearly found.
2.3 binarization
Through color space conversion, the ring image boundary can be observed more intuitively, but further contrast is still needed, so that the ring image boundary is converted into a binary image most intuitively.
The following is the fuzzy range of the HSV component calculated by experiment.
H：0-180；
S：0-255；
V：0-255；
Here red is assigned to the purple range:
the binarization function is realized by inRange function in opencv. I.e. by checking whether the array element is between the other two array element values. If so, it is 255, otherwise it is 0.
2.4 morphological operations-on operations
The opening operation is a process of corrosion first and then expansion, and can remove isolated points, burrs and small bridges (small points connecting two areas), eliminate the boundaries of small objects and smooth large objects, and do not change the area obviously.
2.5 rectangle detection
Since the high-temperature ring appears rectangular in the image, the detection is performed by using the principle of rectangular detection. Because the outer diameter of the ring piece is measured, the maximum external rectangle of the picture is detected. Firstly, extracting a target contour in a binary image by using a library function findContours in opencv, solving a rectangle containing a minimum area of a point set by using a minAreaRect function, and determining a proper constraint condition through analysis on an actual image to detect the rectangle.
This method can obtain coordinates of four vertices of a rectangle, and the coordinates are denoted as A (u)a,va)，B(ub,vb)，C(uc,vc)， D(ud,vd). The form (u, v) refers to a point coordinate form in an image pixel coordinate system. By judging ua，ub，uc， udThe two-side perpendicular line equation U-U of the rectangular detection is solved according to the size (two-two are equal in the error range of delta U)1And u ═ u2。
2.6 refinement
And thinning the extracted green stripline binary image by adopting a Zhang parallel rapid thinning algorithm.
2.7 fitting
And performing linear fitting on the thinned lines by using a least square method to obtain a linear equation.
Let y be F (x), and some undetermined coefficients a are contained in the equationnGive the true value { (x)i，yi) I 1, 2.. n, substituting these x, y values into the equation and then making a difference, can describe the error: y isi-F(xi) To take the overall error into account, the sum of squares can be taken, and the square is taken to account that the error can be added positively or negatively and can be cancelled, so the error is recorded as:
ei＝∑(yi-F(xi))2
it is a multivariate function having anA total of n unknowns, the minimum value is now required. It must be satisfied that the partial derivatives for the variables are equal to 0, thus yielding n equations:
...
it is theoretically possible to solve the n equations to determine the n unknowns as constants. The method of performing a regression equation by using such an error analysis method is a least square method.
Linear regression:
if the empirical equation is linear, in the form of y ═ ax + b, it is a linear regression. According to the above analysis, the error function is:
e＝∑(yi-axi-b)2
the partial derivatives are:
de/da＝2∑(yi-axi-b)xi＝0
de/db＝-2∑(yi-axi-b)＝0
a system of linear equations for a, b is then obtained:
(∑xi 2)a+(∑xi)b＝∑yixi
(∑xi)a+nb＝∑yi
let A ═ Σ xi 2，B＝∑xi，C＝∑yixi，D＝∑yiThen, it is formulated as:
Aa+Bb＝C
Ba+nb＝D
and (3) solving a and b to obtain:
a＝(Cn-BD)/(An-BB)
b＝(AD-CB)/(An-BB)
here, the y ═ ax + b equation also refers to the equation in the image pixel coordinate system: y represents v, referring to the ordinate; x denotes u, referring to the abscissa, i.e. the equation for the fitted curve is v au + b; dissolving to obtain u ═ v-b)/a, namely u1 ═ v1-b1)/a 1; u2 ═ v2-b2)/a 2.
2.8 error analysis
If it is | u1-u0Delta u or | u is less than or equal to |)2-u0If | ≦ Δ u, u can be determined0And u1/u2Equally, the laser line has sought "one side". The value of au is determined based on the diameter of the ring and the distance from the ring to the detection system. Where u refers to the image equation expression in the image pixel coordinate system, and since the green stripe of light is projected parallel to the ring axis, the equation form is u-u0。
3 ring outside diameter detection theory
As shown in fig. 2, firstly, the distance L between the detection device and the ring rolling center is measured when the detection system is fixed; suppose P1Position sum P2The position is the edge of the ring piece, when the system prompts the success of edge finding, the system records the angular displacement at the momentAnd the angle of the encoder is recorded again by the system when the system prompts that the edge searching is successful again, the difference between the angles of the encoder and the angular displacement encoder is the angle theta, and the outer diameter d of the ring piece can be obtained by the following formula:
the above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.
Claims (6)
1. A large ring outer diameter detection system based on machine vision is characterized by comprising:
the image acquisition unit acquires images of the side of the ring piece in real time through a CCD camera, and projects a high-brightness laser line parallel to the axis of the ring piece on the surface of the ring piece through a laser to serve as a detection reference;
the motion control unit comprises a linear motion mechanism, a frame plate erected on the linear motion mechanism and two adjacent rotary tables arranged on the frame plate, the two rotary tables are used for respectively controlling the autorotation motion of the laser and the CCD camera, and the linear motion of the laser and the CCD camera on the frame plate parallel to the axis of the ring piece is controlled by the linear motion mechanism;
the image processing unit is used for carrying out image processing on the image acquired by the CCD camera through opencv and designing an algorithm to extract the characteristics of the target image by utilizing an image color space change theory;
and the feedback control unit judges whether the edge is searched by comparing and analyzing the contact ratio of the image features extracted by the image processing unit and the highlight laser lines, controls the rotary table to drive the laser to rotate in the reverse direction if the edge is searched, and controls the rotary table to drive the laser to rotate continuously in the direction until the edge is judged to be searched successfully if the edge is not searched.
2. The machine vision-based large ring outer diameter detection system is characterized in that a green linear laser is adopted as the laser, and the power is 1000 mw.
3. The machine vision-based large ring outer diameter detection system is characterized in that an angular displacement encoder is arranged on a frame plate on one side of a turntable where a laser is located.
4. A detection method based on the large ring outer diameter detection system of claim 1 is characterized by comprising the following steps:
step 1: placing the detection system to one side of the ring piece to be detected, and measuring the distance L between the center of the laser and the center of the ring piece by using a laser range finder;
step 2: acquiring images of the side part of the ring piece in real time through a CCD camera, and projecting a high-brightness laser line parallel to the axis of the ring piece on the surface of the ring piece through a laser to serve as a detection reference; in the image acquisition process, the laser and the CCD camera on the frame plate are controlled to linearly move parallel to the axis of the ring piece through the linear motion mechanism so as to ensure the complete imaging of the CCD camera, meanwhile, the laser is controlled to rotate left and right by the rotary table, the other rotary table controls the rotation of the CCD camera, and the high-brightness laser line in the image is tracked;
and step 3: analyzing the position of the side edge of the thermal ring by an image processing unit, and controlling a rotary table to drive a laser to rotate towards the edge position of the ring by a feedback control unit;
and 4, step 4: an equation of the high-brightness laser line under an image pixel coordinate system is analyzed through an image processing unit, and the equation is set to be u-u because the laser line projected by a laser is parallel to the axis of the ring piece0；
Step 5, further analyzing an equation of two side boundaries of the thermal ring in the image under the image pixel coordinate system through the image processing unit, and setting the equation as u-u1，u＝u2；
Step 6, by comparing u0,u1,u2Whether the highlight laser line is traced to the ring or not is obtained according to the size of the highlight laser linePiece edge, namely: if it is | u1-u0Delta u or | u is less than or equal to |)2-u0If | ≦ Δ u, where Δ u is the set pixel error, then u can be determined0And u1Or u2Equal, the laser line has sought "one side"; otherwise, the rotary table continues to rotate towards the direction until the system judges that the rotation directions are equal;
and 7, if the system judges that one side is found, the system controls the laser to rotate reversely through the rotary table, finds another boundary, records the reverse rotation angle theta of the rotary table where the laser is located, and obtains the outer diameter size of the ring piece as
5. The detection method according to claim 4, wherein the step 4 specifically comprises:
step 401, performing noise reduction processing on an image acquired by a CCD camera in real time through opencv;
step 402, analyzing image characteristics, performing color space conversion on the image characteristics, and converting the image characteristics from an original RGB color space to an HSV color space;
step 403, performing selective binarization processing on the image by using an vary function in opencv to obtain the position of the laser line in the image;
step 404, thinning lines in the binary image by adopting a Zhang fast parallel thinning algorithm;
step 405, performing straight line fitting on the thinned lines by using a least square method to obtain a straight line equation u-u of the highlight laser lines under the image pixel coordinate system0。
6. The detection method according to claim 4, wherein the step 5 specifically comprises:
step 501, performing noise reduction processing on an image acquired by a CCD camera in real time through opencv;
502, analyzing image characteristics, performing color space conversion on the image characteristics, and converting the image characteristics from an original RGB color space to an HSV color space;
step 503, carrying out selective binarization processing on the image by using an vary function in opencv to obtain the position of the thermal ring piece in the image;
step 504, performing morphological operation on the binary image, and lubricating the image contour by using open operation;
step 505, performing rectangle detection on the processed image through opencv, specifically including: firstly, extracting a target contour in a binary image by using a library function findContours in opencv, obtaining a rectangle containing a point set with a minimum area by using a minAreaRect function, detecting the contour of the rectangle by using a set constraint condition, obtaining four vertex coordinates of the rectangle, and setting the four vertex coordinates as A (u) for the four vertex coordinatesa,va)，B(ub,vb)，C(uc,vc)，D(ud,vd) The term "u, v" as used herein refers to a point coordinate system in the image pixel coordinate system, and u is determined by the judgmenta，ub，uc，udTo find the equation u-u of the two perpendicular lines of the rectangular detection1And u ═ u2I.e. the ring edge equation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810184799.2A CN108106554B (en) | 2018-03-07 | 2018-03-07 | Machine vision-based large ring outer diameter detection system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810184799.2A CN108106554B (en) | 2018-03-07 | 2018-03-07 | Machine vision-based large ring outer diameter detection system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108106554A CN108106554A (en) | 2018-06-01 |
| CN108106554B true CN108106554B (en) | 2020-05-05 |
Family
ID=62205973
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810184799.2A Active CN108106554B (en) | 2018-03-07 | 2018-03-07 | Machine vision-based large ring outer diameter detection system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108106554B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110400321A (en) * | 2019-07-26 | 2019-11-01 | 广东工业大学 | The extracting method of leather material profile based on machine vision and actual size |
| CN110595369B (en) * | 2019-08-14 | 2021-04-02 | 太原理工大学 | Pipe diameter measuring device and method based on machine vision |
| CN111735384B (en) * | 2020-04-28 | 2021-11-30 | 南京工程学院 | Engine installation measurement and installation method and device based on dynamic interference analysis |
| CN111891379B (en) * | 2020-06-28 | 2022-10-14 | 南京工程学院 | Stable attitude adjusting and mounting method for aero-engine based on interference pre-analysis |
| CN112697219A (en) * | 2021-01-29 | 2021-04-23 | 惠州拓邦电气技术有限公司 | Cup capacity detection method and device for beverage equipment, beverage equipment and medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB0306467D0 (en) * | 2003-03-20 | 2003-04-23 | Molins Plc | Improvements in or relating to machine vision equipment |
| CN101059340B (en) * | 2007-05-18 | 2010-05-19 | 东南大学 | Vehicle tread measurement method based on stereo vision and laser |
| CN201107639Y (en) * | 2007-11-30 | 2008-08-27 | 华南理工大学 | Large-sized producing workpiece geometric measuring apparatus using machine vision |
| CN202101655U (en) * | 2011-05-11 | 2012-01-04 | 瓦房店光阳轴承集团有限公司 | Large ball bearing measuring instrument |
| CN102519400B (en) * | 2011-12-15 | 2014-04-09 | 东南大学 | Large slenderness ratio shaft part straightness error detection method based on machine vision |
| CN203274695U (en) * | 2013-05-24 | 2013-11-06 | 武汉铁路局武汉大型养路机械运用检修段 | Steel-rail wear measurement device based on machine vision |
-
2018
- 2018-03-07 CN CN201810184799.2A patent/CN108106554B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108106554A (en) | 2018-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108106554B (en) | Machine vision-based large ring outer diameter detection system and method | |
| CN109612390B (en) | Large-size workpiece automatic measuring system based on machine vision | |
| CN109141232B (en) | Online detection method for disc castings based on machine vision | |
| CN110766684B (en) | Stator surface defect detection system and detection method based on machine vision | |
| CN102636490B (en) | Method for detecting surface defects of dustproof cover of bearing based on machine vision | |
| CN110530885A (en) | A kind of Systems for optical inspection suitable for industrial production line | |
| CN102441581A (en) | Machine vision-based device and method for online detection of structural steel section size | |
| CN104680519A (en) | Seven-piece puzzle identification method based on contours and colors | |
| CN113793337B (en) | Locomotive accessory surface abnormal degree evaluation method based on artificial intelligence | |
| CN107014291B (en) | A kind of vision positioning method of material precision transfer platform | |
| CN113012098B (en) | Iron tower angle steel punching defect detection method based on BP neural network | |
| Chu et al. | A study on welding quality inspection system for shell-tube heat exchanger based on machine vision | |
| CN114037703B (en) | Subway valve state detection method based on two-dimensional positioning and three-dimensional attitude calculation | |
| CN111968100B (en) | Machine vision detection method and system | |
| CN111308987A (en) | Automatic uncoiling control system of uncoiler based on image processing and detection method | |
| CN111210419B (en) | Micro magnetic shoe surface defect detection method based on human visual characteristics | |
| Ali et al. | Camera based precision measurement in improving measurement accuracy | |
| CN111539951B (en) | Visual detection method for outline size of ceramic grinding wheel head | |
| CN111815575A (en) | Bearing steel ball part detection method based on machine vision | |
| CN110899147A (en) | Laser scanning-based online stone sorting method for conveyor belt | |
| CN110969357A (en) | Visual detection method for holes of aluminum alloy machined part | |
| Haitz et al. | Corrosion detection for industrial objects: from multi-sensor system to 5D feature space | |
| Sun et al. | Defects detecting of gloves based on machine vision | |
| KR20240089263A (en) | Defect detection method, defect detection system, and defect detection program | |
| CN111080623B (en) | Visual detection device and method for circular part missing |
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 |