CN108416763B - Image edge processing method for relay measurement - Google Patents

Image edge processing method for relay measurement Download PDF

Info

Publication number
CN108416763B
CN108416763B CN201810061024.6A CN201810061024A CN108416763B CN 108416763 B CN108416763 B CN 108416763B CN 201810061024 A CN201810061024 A CN 201810061024A CN 108416763 B CN108416763 B CN 108416763B
Authority
CN
China
Prior art keywords
image
contact
value
relay
judging
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
Application number
CN201810061024.6A
Other languages
Chinese (zh)
Other versions
CN108416763A (en
Inventor
杨将新
孙沛泽
曹彦鹏
曹衍龙
张远松
王帅
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN201810061024.6A priority Critical patent/CN108416763B/en
Publication of CN108416763A publication Critical patent/CN108416763A/en
Application granted granted Critical
Publication of CN108416763B publication Critical patent/CN108416763B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • 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
    • G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformation in the plane of the image
    • G06T3/40—Scaling the whole image or part thereof
    • G06T3/4007—Interpolation-based scaling, e.g. bilinear interpolation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20092—Interactive image processing based on input by user
    • G06T2207/20104—Interactive definition of region of interest [ROI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20172—Image enhancement details
    • G06T2207/20192—Edge enhancement; Edge preservation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection

Abstract

The invention discloses an image edge processing method for relay measurement, which comprises the following specific processing steps: the method comprises the steps of adjusting a device, dividing a recognition area, extracting an upper contact, performing image binarization processing, performing image statistics, performing edge positioning and zooming, expanding an image, acquiring a minimum value of an upper contact gap, judging the upper contact is qualified, extracting a lower contact, performing binarization processing on a lower contact area, acquiring the average width of the lower contact, judging the lower contact is qualified, and judging the relay is qualified; the invention provides the image edge processing method for relay measurement, which has the advantages of high product detection speed, large processing information amount and high product detection accuracy.

Description

Image edge processing method for relay measurement
Technical Field
The invention relates to the field of image processing, in particular to an image edge processing method for relay measurement.
Background
The relay is an automatic circuit control device widely applied to equipment such as daily household appliances, industrial control circuits, electromechanical integration, remote control and remote measurement, aerospace and the like, and has the functions of circuit regulation, circuit protection, circuit conversion and the like. The production of the relay is a complex process involving over ten processes, and the quality of the machining and assembly of each process directly affects the quality of the final finished product. The assembly of the contact of the relay is a difficult point in the production process of the relay, and the assembly quality of the contact is also a main factor causing the fault of the relay, so the assembly quality of the contact needs to be strictly detected in the production process.
The traditional manual detection method is low in detection speed, only suitable for sampling inspection, incapable of meeting the detection requirement of automatic assembly line production, poor in detection repeatability and incapable of guaranteeing the detection quality.
The vision measurement technology is a rapid nondestructive non-contact measurement technology, has high speed and can process larger information quantity, but the error rate of a detection judgment product of the current vision measurement system is larger, the trend of the qualified rate of the product is difficult to judge, the information quantity is difficult to calculate and monitor when being overlarge, which problem causes the qualified rate of the product to be reduced, the information processing speed is improved, the quality inspection time of each product cannot exceed 1s, therefore, effective reasonable inference is difficult to form, and once the problem occurs, the vision measurement technology completely depends on skilled technical workers.
Disclosure of Invention
Aiming at the technical problems, the invention provides the image edge processing method for relay measurement, which has the advantages of high product detection speed, large processing information amount and high product detection accuracy.
In order to solve the technical defects, the technical scheme of the invention is as follows:
the image edge processing method for relay measurement comprises the following specific processing steps:
101) adjusting the device: the light source, the relay and the camera are arranged on the same horizontal line, and a telecentric lens is arranged on the camera through threads which are arranged on the camera and the telecentric lens and matched with each other;
102) identification area division: transmitting an image obtained by shooting through the camera in the step 101) to an image processing terminal, and carrying out framing area identification on two interested areas, namely an ROI (region of interest) of an upper contact and a lower contact of the image;
103) an upper contact extraction step: according to the framing area identification in the step 102), only obtaining the image in the upper contact area frame, and removing the rest parts;
the acquiring of the image in the upper contact area frame specifically includes the following steps:
201) an image binarization processing step: setting the gray value of the pixel point on the acquired image of the upper contact area to be 0 or 255, namely displaying an obvious black and white effect on the whole image;
202) image statistics step: counting the number of black pixel points of the image subjected to the binarization of the image in the step 201), namely acquiring the number of rows and columns of the image, counting the number of pixel points with the gray value of 0 in each column of the image row by row, and storing the pixel points into a corresponding array;
203) and an edge positioning scaling step: comparing the black pixel numbers counted in the step 202), setting a flag count for judging whether to find a left edge, setting the default initial value to be 0, comparing the black pixel numbers counted according to the characteristics of the upper contact area column by column, when three continuous columns of values become smaller gradually, recording the column number j1 by setting the count to be 1, and determining to find an edge column; continuing to compare, when the numerical values of three continuous columns become larger for the first time, recording the column number j2, and confirming that another edge column is found according to the column number j2, thereby determining the range of the extracted image;
104) an image expansion step: expanding each pixel point of the image acquired in the step 103), wherein the expansion adopts the expansion of the pixel points in the horizontal direction and the vertical direction, and the gray value is given to the newly expanded pixel points in a cubic spline curve interpolation mode;
105) acquiring the minimum value of the upper contact clearance: carrying out binarization processing on the expanded image obtained in the step 104), carrying out pixel traversal on the image line by line, recording the number of pixels with the gray value of 255 in each line, storing the number of pixels with the gray value of 255 in a corresponding array, finally obtaining the number of pixels with the gray value of 255, namely the gap value between two contacts, dividing the gap value by 5 to obtain the gap value with the sub-pixel precision of the contacts, and finding the minimum gap value from the gap value, namely the minimum value of the contact gap;
106) and (3) judging the upper contact to be qualified: calibrating the system based on the origin matrix calibration plate and the minimum value of the contact clearance at the upper part of the relay obtained in the step 105) to obtain the actual minimum phase element size of the system, obtaining the numerical value of the actual minimum value of the contact clearance at the upper part, and judging that the upper contact is qualified if the numerical value is within a certain threshold range, otherwise, judging that the upper contact is qualified;
107) a lower contact piece extraction step: according to the framing area identification in the step 102), only obtaining the image in the lower contact connecting ring area frame, and removing the rest parts;
108) and (3) binarization processing of the lower contact wafer region: setting the gray value of the image pixel point obtained in the step 107) as 0 or 255, wherein the gray value of background white is 255, and the gray value of black in the lower contact piece area is 0;
109) acquiring the average width of the lower contact pieces: traversing and counting each line of the binary image obtained in the step 108), respectively calculating the width values of the left and right contact pieces of the lower contact piece, and storing the width values into corresponding arrays; summing the stored width value data, and dividing the summed value by the corresponding image line number to obtain the average width value of the left contact and the right contact of the lower contact;
110) and (3) judging the lower contact sheet to be qualified: judging that the absolute difference value of the average width of the lower contact of the relay obtained in the step 109) and the width of each row of the left contact and the right contact is greater than 12 pixels, judging that the left contact is distorted when the judged mutation point of the left contact exceeds 75, and judging that the right contact is distorted when the judged mutation point of the right contact exceeds 30; when the left contact piece and the right contact piece are not twisted, the lower contact piece is qualified;
111) and (3) judging the qualification of the relay: and judging that the relay is qualified according to the judgment results of the step 106) and the step 110), and judging that the product is qualified.
Further, the expansion in the step 105) is performed by 5 times of the original expansion in the horizontal direction and the vertical direction by using the pixel points.
Further, 106) the threshold value in the upper contact passing determination step is 0.3mm to 0.6 mm.
Furthermore, the hardware equipment adopted by the image edge processing method comprises an observation system, a light source, a workpiece clamping mechanism, a frame supporting system and an image processing terminal; the observation system comprises a telecentric lens, a camera and a fine adjustment sliding table, and the workpiece clamping mechanism comprises a tool chute and a relay; the light source, the relay and the camera are arranged on the same horizontal line.
Due to the adoption of the technical scheme, the invention has the following advantages:
according to the invention, two areas of an upper contact and a lower contact of an MPD-S-112-A type relay contact detection station are precisely positioned and amplified, an original lattice calibration method is fully utilized by a vision measurement method to calibrate a measurement system, internal parameters and external parameters of a camera are determined, and image acquisition and trial shooting are carried out, so that the effect of rapidly positioning the size of an actually measured object is achieved.
The invention utilizes the binarization processing image processing, is convenient for rapid amplification to carry out image expansion and local amplification processing, can greatly improve the resolution recognition degree, greatly reduces the operation speed and improves the execution operation speed. And meanwhile, the cubic spline curve is utilized to more naturally expand image pixels, and high-efficiency operation can be realized. Thereby when realizing real-time measurement to assembly line relay contact, can both make statistics of measured data with data such as range, mean value, maximum value, minimum to carry out fast, and can be with the data real-time recording of statistics, thereby obtain relay contact assembly quality situation, for which problem of product carries out data guidance, avoid judging by the experience of technical worker entirely.
Drawings
FIG. 1 is a schematic diagram of a relay;
FIG. 2 is a schematic view of a three-dimensional model of a measurement platform according to the present invention;
FIG. 3 is a relay image acquired by the measurement platform of the present invention;
FIG. 4 is a detailed flow chart of the present invention;
FIG. 5 is a detailed flow chart of criterion I of FIG. 4;
FIG. 6 is a schematic diagram of the acquisition of an image of the upper contacts of the relay of FIG. 3;
FIG. 7 is a schematic view of the image of FIG. 6 after the criteria of FIG. 5 have been processed;
FIG. 8 is a flow chart of criterion III of FIG. 4;
FIG. 9 is a schematic view of an image processing area according to the present invention;
FIG. 10 is an expanded image of an image pixel according to the present invention;
FIG. 11 is a schematic diagram of a communication process according to the present invention;
FIG. 12 is a front three-dimensional view of the apparatus of the present invention;
FIG. 13 is a rear three-dimensional view of the apparatus of the present invention;
fig. 14 is an exploded view of the area within the box of fig. 13.
Detailed Description
The invention will be further described with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1 to 14, an image edge processing method for relay measurement.
The device comprises an observation system, a light source, a workpiece clamping mechanism, a frame supporting system and an image processing terminal; the observation system comprises a telecentric lens, a camera and a fine adjustment sliding table, and the workpiece clamping mechanism comprises a tool sliding groove and a relay. The relay is arranged on the tool sliding groove, the telecentric lens is arranged on one side of the camera close to the relay, the camera is arranged on the fine adjustment sliding table and is connected with the image processing terminal, and the light source, the relay and the camera are arranged on the same horizontal line; the light source adopts a rectangular block, so that parallel light irradiates on the relay to form a corresponding image, and the image is acquired by the camera.
The specific device structure is shown in fig. 12 to 14:
the observation system further comprises a first sliding table 11, a second sliding table 12, a camera mounting plate 14, a camera 15, a rotating table 13 and a telecentric lens, wherein the first sliding table 11 is protruded in the middle and is mounted in the middle of the right side of the aluminum plate 29. The second sliding table 12 is rectangular, is provided with a groove at the bottom, and is mounted on the upper surface of the first sliding table 11. A first handle is installed on the rear side of the second sliding table 12, a second handle is installed on the front side of the second sliding table and comprises two adjusting blocks, and the adjusting blocks are connected through a guide rod. The first sliding table 11 and the second sliding table can slide along the direction of the formed groove. The rotating table 13 comprises an upper portion and a lower portion, the upper portion is cylindrical, the lower portion is square, and the lower portion of the rotating table 13 is fixed to the second sliding table 12 through bolts. The camera mounting plate 14 is mounted above the rotating table 13. The camera mounting plate 14 is rectangular, four corners of the camera mounting plate are provided with round corners, and the edges of the round corners are respectively provided with a sixth bolt hole; the camera mounting panel 14 is respectively opened at both sides about has a seventh bolt hole, and the first half is the major diameter, and the latter half is the minor diameter, is the step form, major diameter hole be used for with revolving stage 13's cooperation installation adopts the reducing bolt hole, can guarantee firm connection. The front end of the camera 15 is provided with a circular interface. Telecentric lens 8 includes fixing base, mirror neck, changeover portion, camera lens, lens. The fixing base is installed camera 15 front end circular interface, the mirror neck is installed on the fixing base, the camera lens warp the changeover portion with the mirror neck is connected, the lens is installed in the camera lens. It should be noted that the lens neck of the telecentric lens 8 is a telescopic structure, so that the focal length can be flexibly adjusted; in addition, the lens can be disassembled and adjusted as required. It is worth pointing out that the lower and upper parts of the rotary table 13 can be rotated relatively on the R axis.
The light source board 21 is mounted on the fifth cross bar 28, and is square, and four corners of the light source board are provided with chamfers. Two bolt holes are formed below the light source plate 21 and are used for connecting and fixing the fifth cross bar 28. It should be noted that, as shown in fig. 12 and fig. 13, the structure behind the light source plate 21 is not identified, and the structure here includes, but is not limited to, adding a shielding screen as needed, and the structure here is matched with the above eight bolt holes on one hand to further reinforce the light source plate 21, and on the other hand, can enhance the illumination, so that the imaging is clearer, and the detection accuracy is improved.
The workpiece clamping mechanism further comprises a tool fixing block 4, a first angle aluminum 3, a second angle aluminum 5 and a third angle aluminum 7. The first aluminum corner 3 is mounted in the middle of the aluminum plate 29. The first angle aluminum 3 comprises a first horizontal side and a first vertical side. The tool fixing block 4 is in a concave shape and comprises a left side edge, a right side edge and a bottom edge, two bolt holes are formed in the lower portion of the left side edge of the tool fixing block 4, and the two bolt holes are matched with the long circular bolt holes of the first aluminum fillet 3; the lower portion of the right side edge of the tool fixing block 4 is provided with a front bolt hole and a rear bolt hole, the front bolt hole is matched with the second vertical edge long circular bolt hole of the second angle aluminum 5, and the rear bolt hole is matched with the third vertical edge long circular bolt hole of the third angle aluminum 7. Through the oblong bolt holes in the first angle aluminum 3, the second angle aluminum 5 and the third angle aluminum 7, the up-and-down movement and flexible adjustment of the tool fixing block 4 can be realized.
The frame supporting system comprises a first vertical rod 9, a second vertical rod 17, a third vertical rod 20, a fourth vertical rod 1, a first transverse rod 18, a second transverse rod 23, a third transverse rod 27, a fourth transverse rod 30, a fifth transverse rod 28, a first longitudinal rod 24, a second longitudinal rod 19, a third longitudinal rod 26 and a fourth longitudinal rod 6, wherein the cross section of the fourth longitudinal rod is X-shaped, a circular hole is formed in the center of the X-shaped longitudinal rod, four corners of the X-shaped longitudinal rod are arrow-shaped, and a circular corner is formed in the top end of the arrow. By adopting the section in the form, the weight can be reduced on the basis of ensuring the strength, and the device is convenient to move.
In the specific implementation, the equipment is connected with the image processing terminal through the gigabit optical fiber, camera shooting is triggered by the camera equipment after a product to be detected enters a shooting area, shot image information is transmitted to the image processing terminal, high-efficiency image processing is generally carried out on a PC (personal computer) end, the equipment is fed back and forth through (OK) or not through (NG) according to the processing result of the PC end, an instruction is provided for other external mechanical equipment to reject unqualified products, the imaging condition of the product is monitored and recorded in real time in the processing process, data guidance is further carried out on possible problems of the product, judgment is avoided depending on the experience of technical workers, the management cost is greatly reduced, the qualification rate of the product is improved, obstacles are eliminated efficiently, and efficient operation of the equipment can be achieved.
The scheme specifically aims at the measurement and processing of the MPD-S-112-A type relay: the camera transmits image data to an image processing terminal for processing, and the image processing module comprises the following specific processing steps:
101) adjusting the device: the light source, the relay and the camera are arranged on the same horizontal line, and the telecentric lens is arranged on the camera through threads which are arranged on the camera and the telecentric lens and matched with each other.
102) Identification area division: through step 101), an image obtained by shooting by a camera is transmitted to an image processing terminal, and two interesting regions, namely an ROI (region of interest) of an upper contact and a lower contact of the image are subjected to framing region identification. Therefore, whether the product detection equipment normally operates or not can be immediately visually understood by monitoring personnel, and whether the image processing is accurate or not can be achieved.
103) An upper contact extraction step: according to the framed area identification in the step 102), only the image in the upper contact area frame is obtained, and the rest part is removed. The specific data operation processing realizes regional precision processing, eliminates interference items, and achieves the effect as shown in fig. 6.
Specifically, the step of acquiring the image in the upper contact area frame includes the following steps, specifically as shown in fig. 5, so as to achieve the accurate positioning of the areas from fig. 6 to fig. 7.
201) An image binarization processing step: and setting the gray value of the pixel point on the acquired image of the upper contact area to be 0 or 255, namely displaying an obvious black and white effect on the whole image.
202) Image statistics step: counting the number of black pixel points of the image subjected to the image binarization in the step 201), namely acquiring the number of rows and the number of columns of the image, counting the number of pixel points with the gray value of 0 in each column of the image row by row, and storing the pixel points into a corresponding array.
203) And an edge positioning scaling step: comparing the number of black pixels counted in step 202) (the number of black pixels in the number of black pixels is compared with 0.95i as a reference number, i is the number of rows), setting a flag count for judging whether a left edge is found, the default initial value is 0, comparing the number of black pixels counted according to the characteristics of the upper contact area column by column, and when three consecutive columns of values become smaller gradually (that is, Num _ n is used for recording the number of black pixels in the nth column, Num _ (n +1) is smaller than Num _ n, Num _ (n +2) is smaller than Num _ (n +1)), making count 1 record column j1, and determining to find an edge column. Continuing comparison, when the number of consecutive three columns gradually increases for the first time (that is, Num _ n is used as the number of black pixels in the nth column, Num _ (n +1) is greater than Num _ n, Num _ (n +2) is greater than Num _ (n +1)), the number j2 of the columns is recorded, and it is confirmed that another edge column is found, thereby determining the range of the extracted image. This is because the MPD-S-112-A relay inevitably forms a continuously increasing and decreasing area after the image area is reduced, and the edge of the upper contact is determined.
104) An image expansion step: expanding each pixel point of the image obtained in the step 103), wherein the expansion is performed in the horizontal direction and the vertical direction by adopting the pixel points to reach 5 times of the original value (namely reaching as shown in figure 10), and gray values are given to the newly expanded pixel points in a cubic spline curve interpolation mode.
Specifically, as shown in fig. 10, the squares of the large dots are real physical pixel points, and the smaller dots between the physical pixel points are sub-pixel points. The interpolation mode of the cubic spline curve is as follows: set the following nodes
x:a＝x0<x1<x2<…<xn＝b
y:y0y1y2…<yn
Spline curve s (x), s (x) is a piecewise defined equation, and given n +1 data points, there are n intervals, and the cubic spline equation for each interval is as follows:
Si(x)＝ai+bi(x-xi)+ci(x-xi)2+di(x-xi)3formula (1)
Wherein a isi,，bi，ci，diIs four parameters, S (x) Si(x) And S isi(x)＝yiWhere i is 0,1,2 … n, s (x) and its derivative, second derivative are all in [ a, b ]]The intervals are all continuous.
Namely, the following conditions are satisfied:
a. in each segment interval [ xi, xi +1](i＝0,1,2…n-1)，S(x)＝Si(x) Are all cubic polynomials;
b. satisfies Si(x)＝yi(i＝0,1,2…n)；
c.S (x), the derivative S' (x), the second derivative S "(x) are continuous in the [ a, b ] interval, i.e. the S (x) curve is smooth.
Therefore, the gray value of the original pixel point in the acquired image is used as a known data point, a cubic spline curve is arranged between every two nodes, and the intermediate nodes are continuous in second order, so that the gray value of the corresponding sub-pixel point can be obtained, and the image after pixel expansion is obtained.
The method comprises the following specific steps: the precise area of the contact gap is obtained in the original image. Multiplying the horizontal coordinates and the vertical coordinates of all pixel points in the area image by 5 respectively, and correspondingly changing the number of rows and the number of columns of the whole image into 5 times of the original image respectively. In the original image (0,0) corresponds to a new image point (0,0), the original image point (1,1) corresponds to a new image point (5,5), the original image point (2,2) corresponds to a new image point (10,10), and the original image point (n, m) corresponds to a new image point (5n,5 m).
Except for the point (5n,5m), the gray values of the other points (e.g., the point (1,2) (1,3) (2,4) … …) in the newly obtained image are all empty, and the pixel gray values need to be filled by means of cubic spline curve interpolation, where (n, m is 0,1,2,3 … …, and n < the number of rows of the original image and m < the number of columns of the original image). And (3) carrying out cubic spline curve fitting through the known point (5n,5m) gray scale and the image point coordinates to obtain a corresponding cubic spline curve equation, and inputting corresponding coordinate values on the equation to obtain the image gray scale value with the empty defect.
In this project, since the contact gap is the distance between the horizontal pixels, and considering that the program needs to process the output result quickly, only the pixel expansion of the image column is performed by 5 times of the original one, which is most suitable, and the processing efficiency is high and the effect is good.
105) Acquiring the minimum value of the upper contact clearance: and (2) carrying out binarization processing on the expanded image obtained in the step 104), carrying out pixel traversal on the image line by line, recording the number of pixels with the gray value of 255 in each line, storing the number of pixels with the gray value of 255 in a corresponding array, finally obtaining the number of pixels with the gray value of 255, namely the gap value between two contacts, dividing the gap value by 5 to obtain the gap value with the sub-pixel precision of the contacts, and finding the minimum gap value from the gap value, namely the minimum contact gap value.
106) And (3) judging the upper contact to be qualified: and calibrating the system based on the origin array calibration plate and the minimum value of the contact clearance at the upper part of the relay obtained in the step 105) to obtain the actual minimum element size of the system, obtaining the numerical value of the actual minimum value of the contact clearance at the upper part, and judging that the upper contact is qualified if the numerical value is in the range of 0.3mm to 0.6mm, otherwise, judging that the upper contact is unqualified.
107) A lower contact piece extraction step: according to the framing area identification in the step 102), only the image in the lower contact connecting ring area frame is obtained, and the rest part is removed. Interference in other areas is likewise avoided.
108) And (3) binarization processing of the lower contact wafer region: setting the gray value of the image pixel point obtained in the step 107) as 0 or 255, wherein the gray value of background white is 255, and the gray value of black in the lower contact piece area is 0.
109) Acquiring the average width of the lower contact pieces: traversing and counting each line of the binary image obtained in the step 108), respectively calculating the width values of the left and right contact pieces of the lower contact piece, and storing the width values into corresponding arrays; the stored width value data are summed and divided by the corresponding number of image lines to obtain an average width value for the left and right two pads of the lower pad.
The method for respectively calculating the width values of the left contact and the right contact of the lower contact is as follows:
301) a data initialization setting step: acquiring the line number I and the column number J of image pixels, and setting H _ (I, J) to record the pixel gray value of the ith row and the jth column and the boundary number flag quantity count, wherein the boundary number flag quantity is zero by default, Width _ LF (I) to record the Width of a left contact of each row, Width _ RT (I) to record the Width of a right contact, and KeyPoint [4] to store the column number J of the boundary pixel point of each row.
302) Traversing and determining the boundary of the lower contact piece: traversing from the pixel (0,0) in the first row and the first column according to the row number I and the column number J of the image pixel acquired in the step 301), determining that the currently traversed column does not exceed the column number J, and performing line changing if the currently traversed column exceeds the column number J to perform new traversal of the new row; and comparing the front and back traversals if the number J of the columns is not exceeded, when the gray values of the columns obtained by the front and back traversals are different, finding a boundary change point, storing the current column number J into an array KeyPoint [ count ], setting the boundary number flag quantity count to be count +1, continuously traversing and searching as long as the boundary number flag quantity does not exceed 4, and determining the boundary of the lower contact piece if the boundary number flag quantity exceeds 4.
303) The width value of the lower contact piece comprises the following steps: and counting the stored data amount according to KeyPoint [ count ] obtained by traversing each column determined in the step 302), thereby obtaining the width value of the lower contact. Wherein, the left contact is Width _ LF (i) ═ KeyPoint [1] -KeyPoint [0], and the right contact is Width _ RT (i) ═ KeyPoint [3] -KeyPoint [2 ].
110) And (3) judging the lower contact sheet to be qualified: judging that the absolute difference value of the average width of the lower contact of the relay obtained in the step 109) and the width of each row of the left contact and the right contact is greater than 12 pixels, judging that the left contact is distorted when the judged mutation point of the left contact exceeds 75, and judging that the right contact is distorted when the judged mutation point of the right contact exceeds 30; and when the left contact piece and the right contact piece are not twisted, the lower contact piece is qualified.
111) And (3) judging the qualification of the relay: and judging that the relay is qualified according to the judgment results of the step 106) and the step 110), and judging that the product is qualified.
In conclusion, the image processing terminal carries out real-time data statistics and display according to the data, the problems of low traditional manual detection efficiency, poor stability, high labor cost and the like in the aspects of relay contact gap measurement and contact distortion detection are solved, the real-time feedback of the measured data can be counted in modes of range, mean value, maximum value, minimum value and the like, monitoring personnel can conveniently acquire the assembly quality condition of the relay contact in real time, data guidance is carried out on which problem of a product, the situation that technical workers rely on to judge is avoided, the management cost is greatly reduced, the qualification rate of the product is improved, obstacles are eliminated efficiently, and efficient operation of equipment can be achieved.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and decorations can be made without departing from the spirit of the present invention, and these modifications and decorations should also be regarded as being within the scope of the present invention.

Claims (4)

1. The image edge processing method for relay measurement is characterized by comprising the following specific processing steps of:
101) adjusting the device: arranging a light source, a relay and a camera on the same horizontal line, and installing a telecentric lens on the camera through threads matched with the camera and the telecentric lens;
102) identification area division: transmitting an image obtained by shooting through the camera in the step 101) to an image processing terminal, and carrying out framing area identification on two interested areas, namely an ROI (region of interest) of an upper contact and a lower contact of the image;
103) an upper contact extraction step: according to the framing area identification in the step 102), only obtaining the image in the upper contact area frame, and removing the rest parts;
the acquiring of the image in the upper contact area frame specifically includes the following steps:
201) an image binarization processing step: setting the gray value of the pixel point on the acquired image of the upper contact area to be 0 or 255, namely displaying an obvious black and white effect on the whole image;
202) image statistics step: counting the number of black pixel points of the image subjected to the binarization of the image in the step 201), namely acquiring the number of rows and columns of the image, counting the number of pixel points with the gray value of 0 in each column of the image row by row, and storing the pixel points into a corresponding array;
203) and an edge positioning scaling step: comparing the number of black pixels counted in the step 202), setting a flag count for judging whether to find a left edge, setting the default initial value to be 0, comparing the number of black pixels counted according to the characteristics of the upper contact area column by column, when the number of continuous three columns becomes smaller, making count =1 to record the column number j1, and determining to find an edge column; continuing to compare, when the numerical values of three continuous columns become larger for the first time, recording the column number j2, and confirming that another edge column is found according to the column number j2, thereby determining the range of the extracted image;
104) an image expansion step: expanding each pixel point of the image acquired in the step 103), wherein the expansion adopts the expansion of the pixel points in the horizontal direction and the vertical direction, and the gray value is given to the newly expanded pixel points in a cubic spline curve interpolation mode;
105) acquiring the minimum value of the upper contact clearance: carrying out binarization processing on the expanded image obtained in the step 104), carrying out pixel traversal on the image line by line, recording the number of pixels with the gray value of 255 in each line, storing the number of pixels with the gray value of 255 in a corresponding array, finally obtaining the number of pixels with the gray value of 255, namely the gap value between two contacts, dividing the gap value by 5 to obtain the gap value with the sub-pixel precision of the contacts, and finding the minimum gap value from the gap value, namely the minimum value of the contact gap;
106) and (3) judging the upper contact to be qualified: calibrating the system based on the origin matrix calibration plate and the minimum value of the contact clearance at the upper part of the relay obtained in the step 105) to obtain the actual minimum phase element size of the system, and obtaining the numerical value of the actual minimum value of the contact clearance at the upper part, wherein when the numerical value is within a certain threshold range, the upper contact is judged to be qualified, and otherwise, the upper contact is judged to be unqualified;
107) a lower contact piece extraction step: according to the framing area identification in the step 102), only obtaining the image in the lower contact connecting ring area frame, and removing the rest parts;
108) and (3) binarization processing of the lower contact wafer region: setting the gray value of the image pixel point obtained in the step 107) as 0 or 255, wherein the gray value of background white is 255, and the gray value of black in the lower contact piece area is 0;
109) acquiring the average width of the lower contact pieces: traversing and counting each line of the binary image obtained in the step 108), respectively calculating the width values of the left and right contact pieces of the lower contact piece, and storing the width values into corresponding arrays; summing the stored width value data, and dividing the summed value by the corresponding image line number to obtain the average width value of the left contact and the right contact of the lower contact;
110) and (3) judging the lower contact sheet to be qualified: judging that the absolute difference value of the average width of the lower contact of the relay obtained in the step 109) and the width of each row of the left contact and the right contact is greater than 12 pixels, judging that the left contact is distorted when the judged mutation point of the left contact exceeds 75, and judging that the right contact is distorted when the judged mutation point of the right contact exceeds 30; when the left contact piece and the right contact piece are not twisted, the lower contact piece is qualified;
111) and (3) judging the qualification of the relay: and judging that the relay is qualified according to the judgment results of the step 106) and the step 110), and judging that the product is qualified.
2. The method for processing the image edge of the relay measurement according to claim 1, wherein the expansion in the step 105) is performed by 5 times in the horizontal direction and the vertical direction by using pixel points.
3. The image edge processing method for relay measurement according to claim 1, wherein 106) the threshold value in the upper contact qualification determination step is 0.3mm to 0.6 mm.
4. The relay-measured image edge processing method according to claim 1, wherein hardware devices adopted by the image edge processing method comprise an observation system, a light source, a workpiece clamping mechanism, a frame supporting system and an image processing terminal; the observation system comprises a telecentric lens, a camera and a fine adjustment sliding table, and the workpiece clamping mechanism comprises a tool chute and a relay; the light source, the relay and the camera are arranged on the same horizontal line.
CN201810061024.6A 2018-01-22 2018-01-22 Image edge processing method for relay measurement Active CN108416763B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810061024.6A CN108416763B (en) 2018-01-22 2018-01-22 Image edge processing method for relay measurement

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810061024.6A CN108416763B (en) 2018-01-22 2018-01-22 Image edge processing method for relay measurement

Publications (2)

Publication Number Publication Date
CN108416763A CN108416763A (en) 2018-08-17
CN108416763B true CN108416763B (en) 2021-06-29

Family

ID=63126000

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810061024.6A Active CN108416763B (en) 2018-01-22 2018-01-22 Image edge processing method for relay measurement

Country Status (1)

Country Link
CN (1) CN108416763B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111147763B (en) * 2019-12-29 2022-03-01 眸芯科技（上海）有限公司 Image processing method based on gray value and application
CN111678450B (en) * 2020-05-09 2022-01-11 东华大学 Visual detection method and device for precision part

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20070080284A (en) * 2006-02-07 2007-08-10 에스케이 텔레콤주식회사 Voice alarm apparatus of a repeater and its the method
CN106839986B (en) * 2017-03-24 2023-07-21 中国计量大学 Double-station carrier band detection system based on machine vision and machine learning
CN107492109A (en) * 2017-05-19 2017-12-19 温州大学 Based on moment preserving theory MEMS radio-frequency relay contact edges detection method

Also Published As

Publication number Publication date
CN108416763A (en) 2018-08-17

Similar Documents

Publication Publication Date Title
AU589651B2 (en) Surface profile determination
CN204556517U (en) Hot rolled sheet metal surface quality on-line detecting device
CN108416763B (en) Image edge processing method for relay measurement
CN108805935A (en) It is a kind of based on orthogonal pixel equivalent than line-scan digital camera distortion correction method
CN111421001B (en) Online accurate centering system of high-speed wire rod mill and centering method thereof
KR101106045B1 (en) Method for calibrating steel sheet
DE102016120557B4 (en) System for dimensionally measuring an object
CN109974631A (en) A kind of bar Linearity surveying system and detection method
CN211827005U (en) Multi-functional detection device of five-axis numerical control machine tool based on multi-eye vision
CN111322950B (en) Method for positioning cylinder position by line structured light sensor and its use
CN108427949B (en) Image processing and image expansion method for relay measurement
CN108416762B (en) Image extraction and image processing method for relay measurement
CN108759676B (en) Checkerboard-based transmission case end face large-size form and position tolerance detection device and method
CN110084847B (en) Relay measuring device
CN112797900B (en) Multi-camera plate size measuring method
CN114119483A (en) Image processing technology-based quality detection method and device for light wallboard for building
CN207540491U (en) The optical texture that a kind of large scale product quickly detects
CN109470698A (en) Across scale field trash quick analytic instrument device and method based on microphotograph matrix
CN105115434A (en) In-situ measuring device of spacing and depth of V-shaped grooves of conductive rods
Lei et al. Automated vision inspection system for the size measurement of workpieces
CN103905762B (en) Method for automatically detecting projection picture for projection module
CN107478206B (en) Device and method suitable for large scene measurement
CN112102301A (en) Stamping frame cooperative detection system and detection method thereof
CN106331620B (en) Filling production lines medicine bottle method for positioning analyzing
KR20000060731A (en) Calibraion method of high resolution photographing equipment using multiple imaging devices.

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