A chip cutting route calculation method, system and readable storage medium

Abstract

The present invention discloses a chip cutting route calculation method, system and readable storage medium, which are based on a visual inspection system, including substrate recognition, angle correction of the substrate, substrate image acquisition and detection, and calculation of the cutting path. The accuracy of the substrate relative to the visual inspection system is improved through angle correction, and the visual inspection system is used to detect the chip in different areas to improve efficiency. At the same time, the actual position coordinates of the chip collected by the visual system are used to accurately calculate the position and angle of the substrate to be cut, so that the size and specifications of the chip after cutting are unified, the operation is simple, the detection speed is fast, and the efficiency is high.

Description

Technical Field

[0001] The present invention relates to the field of image recognition technology, and in particular to a chip cutting route calculation method, system and readable storage medium. Background Art

[0002] As the demand for the chip industry continues to expand, challenges have also been raised to the chip application process. In the production process of chips, in order to adapt to automated production, the chip is usually attached to a larger PCB substrate or ceramic substrate, and then the PCB substrate or ceramic substrate is cut into specified sizes by a cutting machine. However, for large-sized PCB substrates or ceramic substrates, due to the limitations of the manufacturing process, it is inevitable that defects will appear in certain positions of the manufactured products, affecting their use. The visual inspection system can perform quality inspection on the incoming substrates and measure the chip position, but for large-sized substrates, how to quickly collect substrate information and plan the cutting route has become a problem to be solved. Summary of the invention

[0003] In order to overcome the above shortcomings, the purpose of the present invention is to provide a chip cutting route calculation method, which can quickly collect substrate information and accurately and quickly plan the cutting path.

[0004] In order to achieve the above purpose, the technical solution adopted by the present invention is: a chip cutting route calculation method based on a visual inspection system, characterized in that: the method comprises:

[0005] Performing angle correction on the substrate, after the substrate is fed, the visual inspection system calculates the angle deviation of the substrate during feeding, and transmits the value of the angle deviation to a placement mechanism for placing the substrate;

[0006] Substrate image acquisition and detection: the visual inspection system takes pictures of the substrate in different areas and performs quality inspection on the chips on the substrate based on the captured images. For substrates that pass the inspection, the visual inspection calculates the actual position coordinates P of each chip in the coordinate system through the built-in coordinate algorithm model. i ;

[0007] The calculation of the cutting path is based on the actual position coordinates P of each chip i , the cutting path is fitted on the X-axis and Y-axis using the least squares method.

[0008] Furthermore, the visual inspection system calculates the angle deviation of the substrate during feeding, specifically including the following steps:

[0009] S21, the visual inspection system photographs the first correction mark point when the substrate is fed, and the visual inspection system records the visual coordinate VP1 of the first correction mark point;

[0010] S22, the CCD camera of the visual inspection system moves a distance d in the horizontal direction to reach the second correction mark point, the visual inspection system photographs the second correction mark point, and the visual inspection system records the visual coordinates VP2 of the second correction mark point;

[0011] S23, the visual inspection system calculates the angle deviation of the substrate according to the visual coordinates VP1 and the visual coordinates VP2;

[0012] S24, the visual inspection system transmits the angle deviation to the placement mechanism, and the placement mechanism drives the substrate to rotate to correct the angle of the substrate;

[0013] S25, repeating steps S21-S23 to calculate the secondary substrate angle deviation;

[0014] S26, determining whether the secondary substrate angle deviation in step S25 is within a threshold range, if so, performing angle correction, if not, returning to step S24.

[0015] By using the two correction mark points of the substrate, the angle deviation is corrected multiple times until the angle deviation is within the threshold, which can ensure the accuracy of the substrate on the placement mechanism, compensate for the deviation of the substrate when entering the placement mechanism, and improve the accuracy of the chip cutting path calculation.

[0016] Furthermore, in step S23, the angle deviation of the substrate is based on the Pythagorean theorem.

[0017] ΔA=actan(VP2(y)-VP1(y)) / d,

[0018] Among them, VP2(y) is the y coordinate of the visual coordinate VP2, and VP1(y) is the y coordinate of the visual coordinate VP1.

[0019] Use the Pythagorean theorem to quickly calculate the angle deviation.

[0020] Furthermore, the substrate image acquisition and detection specifically includes the following steps:

[0021] S31, dividing a substrate into m*n areas distributed in an array, where m and n are both positive integers, photographing the m*n areas in sequence, and each area includes a*b chips.

[0022] S32, identify and detect the chip in each area, compare the photos taken by the CCD camera with the document information of the substrate, and perform quality inspection on the single chip. Several areas are inspected simultaneously, which greatly reduces the inspection time and improves production efficiency.

[0023] S33, determine whether the substrate is qualified, compare the chip unqualified rate with the set threshold, and for a substrate whose chip detection unqualified rate exceeds the set threshold, directly remove the substrate from the placement mechanism through the robot; for a qualified substrate, the actual position coordinate P of the chip is then measured. i Calculation.

[0024] For unqualified substrates, the calculation of cutting paths is no longer performed, which greatly saves time and avoids unqualified product quality after cutting.

[0025] Furthermore, the actual position coordinates P of the chip i The calculation specifically includes moving the CCD camera of the visual inspection system to the coordinate mark point of the substrate, the coordinate mark point is used as the coordinate origin, and the visual inspection system records the visual coordinate VP (x 0 ,y 0 ) and the mechanism coordinates MP (x 0 ,y 0 ); The CCD camera and the placement mechanism move together. When the CCD camera reaches an area, it takes a picture of the area and records the visual coordinates of a single chip as VP' (x i ,y i ), the mechanism coordinates are MP'(x i ,y i ); Load the coordinate algorithm model VP'-VP+MP'-MP, then the actual position coordinate of this chip is P i (x i ,y i ), x i is the X-axis coordinate value x of the chip i =VP'(x i )-VP(x 0 )+MP'(x i )-MP(x 0 ), y i is the Y-axis coordinate value y of the chip i =VP'(y i )-VP(y 0 )+MP'(y i )-MP(y 0 ), i is a positive integer. Through the coordinate algorithm model, the visual coordinates and the mechanism coordinates are integrated together to obtain the precise coordinates of the actual position of the chip.

[0026] Specifically, in the calculation of the cutting path, the X-axis straight line a is first fitted, and the linear regression equation is used. =b +a, where

[0027] ,

[0028] xi and yi are the actual position coordinates of the chip P i The coordinate values ​​of the X-axis and Y-axis can be fitted to obtain the X-axis straight line a= -b ;

[0029] Then fit the Y-axis straight line and convert the position coordinates of each chip P i Reverse the X coordinate and Y coordinate to get the chip position coordinate. Reverse the coordinate to get the coordinate P. ’ i , through the coordinate P ’ i Use the least squares method to fit the Y axis straight line. Use the least squares method to quickly fit the cutting path.

[0030] Specifically, before the angle of the substrate is corrected, the substrate is first identified. The CCD camera of the visual inspection system captures the identification code on the substrate. After the visual inspection system successfully identifies the identification code, the document information of the product is loaded. However, if the identification code fails to be recognized, an alarm is given.

[0031] The beneficial effect of the present invention is to improve the accuracy of the substrate relative to the visual inspection system through angle correction, and use the actual position coordinates of the chip collected by the visual inspection system to accurately calculate the position and angle at which the substrate needs to be cut, so that the size and specifications of the chips after cutting are unified, the operation is simple, the inspection speed is fast, and the efficiency is high.

[0032] The present invention also provides a chip cutting route calculation system, which can quickly collect substrate information and accurately and quickly plan cutting paths.

[0033] In order to achieve the above purpose, the technical solution adopted by the present invention is: a chip cutting route calculation system, including a visual inspection system and a placement mechanism for substrate placement, the placement mechanism can convey the substrate and drive the substrate to rotate, wherein the visual inspection system includes a collection mechanism and a processing mechanism connected to the collection mechanism in communication, the collection mechanism is used to take pictures of the substrate, and the processing mechanism is used to receive the image taken by the collection mechanism and process the image. The processing mechanism includes

[0034] The identification code detection module is used to identify and detect the identification code on the substrate. When the identification code detection module successfully identifies the identification code, the product document information is loaded; the angle correction module is used to calculate the angle deviation of the substrate and transmit the obtained angle deviation to the placement mechanism; the detection module is used to collect the actual position coordinates P of the chip on the substrate i , and detect the chip quality; a cutting road planning module for planning the cutting road, the cutting road planning module according to the actual position coordinates P of each chip i, the cutting path is fitted on the X-axis and Y-axis using the least squares method.

[0035] Furthermore, the acquisition mechanism is a CCD camera, and the CCD camera uses a variable-magnification telecentric lens, which can meet the detection of chips of different sizes, ensure imaging quality, and reduce measurement errors caused by image distortion.

[0036] In order to achieve the above objectives, the present invention also adopts a technical solution: a readable storage medium having a computer program stored thereon, which implements the above chip cutting route calculation method when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a flow chart of a calculation method in an embodiment of the present invention;

[0038] Figure 2 is a flow chart of substrate angle correction in an embodiment of the present invention;

[0039] Figure 3 4 is a block diagram of a visual inspection system in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0041] Example

[0042] See attached Figure 1 As shown, a chip cutting route calculation method of the present invention adopts a set of visual inspection system and a placement mechanism for substrate placement, the placement mechanism can convey the substrate and drive the substrate to rotate, and the chip cutting route calculation method includes the following steps:

[0043] S1. Perform substrate recognition. The CCD camera of the visual inspection system captures the identification code on the substrate. After the visual inspection system successfully recognizes the identification code, it loads the document information of the product. However, if the identification code recognition fails, an alarm will be issued. The identification code can be a QR code or a barcode, as long as it can be scanned and photographed by a CCD camera.

[0044] S2. Correct the angle of the substrate. Since the substrate will have an angle deviation when feeding, which will cause a deviation in the calculation of the chip coordinates on the substrate, after the substrate is fed, the visual inspection system calculates the angle deviation of the substrate during feeding, and transmits the angle deviation value to the placement mechanism. The placement mechanism drives the substrate to rotate to adjust the angle of the substrate. The steps of angle correction specifically include:

[0045] S21, the visual inspection system photographs the first correction mark point when the substrate is fed, and the visual inspection system records the visual coordinate VP1 of the first correction mark point;

[0046] S22, the CCD camera of the visual inspection system moves a distance d in the horizontal direction to reach the second correction mark point, the visual inspection system photographs the second correction mark point, and the visual inspection system records the visual coordinates VP2 of the second correction mark point;

[0047] S23, the visual inspection system calculates the angle deviation of the substrate according to the visual coordinates VP1 and the visual coordinates VP2;

[0048] S24, the visual inspection system transmits the angle deviation to the placement mechanism, and the placement mechanism drives the substrate to rotate to correct the angle of the substrate;

[0049] S25, repeating steps S21-S23 to calculate the secondary substrate angle deviation;

[0050] S26, judging whether the secondary substrate angle deviation in step S25 is within the threshold range, if so, the angle correction is completed, if not, returning to step S24.

[0051] In step S23, the angle deviation of the substrate is based on the Pythagorean theorem.

[0052] ΔA=actan(VP2(y)-VP1(y)) / d, where VP2(y) is the y coordinate of the visual coordinate VP2, and VP1(y) is the y coordinate of the visual coordinate VP1.

[0053] S3, substrate image acquisition and detection, the visual inspection system takes a picture of the substrate and detects the chip on the substrate based on the captured image.

[0054] Due to the large size of the substrate, the visual inspection system cannot capture the entire substrate in one shot for the sake of accuracy. Therefore, the substrate is divided into multiple areas, and the CCD of the visual inspection system takes pictures of the multiple areas in sequence during the movement. The specific steps include:

[0055] S31, divide a substrate into m*n areas distributed in an array, where m and n are both positive integers, and shoot the m*n areas in sequence, each area including a*b chips. For example, for a 78.5mm*92.6mm substrate, there are a total of 86*71 products, and there are 6*6 products in a single field of view. The CCD can detect 36 products in one shot, which greatly reduces the detection time and improves production efficiency.

[0056] S32, identify and detect the chip in each area, compare the photo taken by CCD with the document information of the product, and detect whether there is a chip where the chip should be; if there is a chip, it means that the place is normal; if there is no chip, it means abnormal. And judge whether each chip has defects, whether the surface is dirty, and whether the position deviation of each chip meets the requirements.

[0057] S33, judging whether the substrate is qualified, comparing the chip unqualified rate with the set threshold, and for a substrate whose chip unqualified rate exceeds the set threshold, directly remove the substrate from the placement mechanism through the robot. For qualified substrates, the cutting path is calculated again.

[0058] After the visual inspection system has taken photos of multiple areas in sequence, it calculates the position coordinates of each chip in the coordinate system through the built-in coordinate algorithm model and integrates each area into an actual substrate map. The calculation of the specific position coordinates of each chip includes the following steps:

[0059] The CCD camera of the visual inspection system moves to the coordinate mark point of the substrate. The coordinate mark point is used as the coordinate origin. The visual inspection system records the visual coordinate VP (x 0 ,y 0 ) (i.e., the coordinates of the mark point in the visual inspection system) and the mechanism coordinates MP (x 0 ,y 0 )(i.e. the coordinates of the coordinate mark point in the placement mechanism). Then the CCD camera and the placement mechanism move together. When the CCD camera reaches an area, it takes a picture of the area and records the visual coordinates of a single chip as VP'(x i ,y i ), the mechanism coordinates are MP'(x i ,y i ), then the actual position coordinates of this chip are P i (x i ,y i ).

[0060] The coordinate algorithm model is, i is the X-axis coordinate value x of the chip i =VP'(x i )-VP(x 0 )+MP'(x i )-MP(x 0 ),y i is the Y-axis coordinate value y of the chip i =VP'(y i )-VP(y 0 )+MP'(y i )-MP(y 0), i is a positive integer. When the CCD camera is moving, the visual inspection system records the actual position coordinates of each chip.

[0061] S4, calculation of the cutting path, based on the actual position coordinates P of each chip i , the cutting path is fitted on the X-axis and Y-axis using the least squares method.

[0062] First, fit the X-axis straight line a, using the linear regression equation =b +a, where

[0063] ,

[0064] xi and yi are the actual position coordinates of the chip P i The coordinate values ​​of the X-axis and Y-axis can be fitted to obtain the X-axis straight line a= -b .

[0065] Then fit the Y axis straight line. Since the cutting machine can only move in the X direction for cutting, first calculate the position coordinates P of each chip. i Reverse the X coordinate and Y coordinate to get the chip position coordinate. Reverse the coordinate to get the coordinate P. ’ i , coordinate P ’ i The Y-axis straight line is fitted by the method of fitting the X-axis straight line.

[0066] Finally, based on the fitted X-axis straight line and Y-axis straight line, the visual inspection system generates the straight line intercept and angle to form a cutting path.

[0067] Using the least squares method, combined with the actual position coordinates P of the chip collected by the visual inspection system i , accurately calculate the position and angle at which the substrate needs to be cut, so that the size and specifications of the chips after cutting are uniform.

[0068] In another embodiment of the present invention, a chip cutting route calculation system is provided, which includes a visual inspection system and a placement mechanism for placing a substrate, wherein the placement mechanism can transport the substrate and drive the substrate to rotate.

[0069] The visual inspection system includes a collection mechanism and a processing mechanism that is connected to the collection mechanism. The collection mechanism is used to take pictures of the substrate. The collection mechanism is a CCD camera that uses a zoom telecentric lens to meet the inspection of different product sizes, ensure imaging quality, and reduce measurement errors caused by image distortion. The processing mechanism is used to receive images taken by the CCD camera and process the images.

[0070] The processing mechanism includes an identification code detection module, which is used for identifying and detecting the identification code on the substrate. When the identification code detection module successfully identifies the product, the product document information is loaded, and an alarm prompt is issued if the identification fails.

[0071] The angle correction module is used to calculate the angle deviation of the substrate and transmit the obtained angle deviation to the placement mechanism, which drives the substrate to rotate and correct the angle of the substrate.

[0072] The detection module is used to collect the actual position coordinates P of the chip on the substrate. i , and according to the actual position coordinates P i Check chip quality, including whether each chip has defects, whether the surface is dirty, and whether the position deviation of each chip meets the requirements.

[0073] The cutting path planning module is used to plan the cutting path according to the actual position coordinates P of each chip. i , the cutting path is fitted on the X-axis and Y-axis using the least squares method.

[0074] In another embodiment of the present invention, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above chip cutting route calculation method are implemented.

[0075] The above implementation modes are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with this technology to understand the content of the present invention and implement it, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be included in the protection scope of the present invention.

Claims

1. A chip cutting route calculation method based on a visual inspection system, characterized in that: The method comprises: Performing angle correction on the substrate, after the substrate is fed, the visual inspection system calculates the angle deviation of the substrate during feeding, and transmits the value of the angle deviation to a placement mechanism for placing the substrate; Substrate image acquisition and detection: the visual inspection system takes pictures of the substrate in different areas and performs quality inspection on the chips on the substrate based on the captured images. For substrates that pass the inspection, the visual inspection calculates the actual position coordinates P of each chip in the coordinate system through the built-in coordinate algorithm model. i ; The calculation of the cutting path is based on the actual position coordinates P of each chip i , the cutting path is fitted on the X-axis and Y-axis using the least square method; The visual inspection system calculates the angle deviation of the substrate during feeding, specifically comprising the following steps: S21, the visual inspection system photographs the first correction mark point when the substrate is fed, and the visual inspection system records the visual coordinate VP1 of the first correction mark point; S22, the CCD camera of the visual inspection system moves a distance d in the horizontal direction to reach the second correction mark point, the visual inspection system photographs the second correction mark point, and the visual inspection system records the visual coordinates VP2 of the second correction mark point; S23, the visual inspection system calculates the angle deviation of the substrate according to the visual coordinates VP1 and VP2. The angle deviation of the substrate is based on the Pythagorean theorem. ΔA=actan(VP2(y)-VP1(y)) / d, Where VP2(y) is the y coordinate of the visual coordinate VP2, and VP1(y) is the y coordinate of the visual coordinate VP1; S24, the visual inspection system transmits the angle deviation to the placement mechanism, and the placement mechanism drives the substrate to rotate to correct the angle of the substrate; S25, repeating steps S21-S23 to calculate the secondary substrate angle deviation; S26, judging whether the secondary substrate angle deviation in step S25 is within the threshold range, if so, the angle correction is completed, if not, returning to step S24.

2. The chip cutting route calculation method according to claim 1, characterized in that: The substrate image acquisition and detection specifically includes the following steps: S31, dividing a substrate into m*n regions distributed in an array, where m and n are both positive integers, photographing the m*n regions in sequence, and each region includes a*b chips; S32, identifying and detecting the chip in each area, comparing the photo taken by the CCD camera with the document information of the substrate, and performing quality inspection of the single chip; S33, determine whether the substrate is qualified, compare the chip unqualified rate with the set threshold, and for a substrate whose chip detection unqualified rate exceeds the set threshold, directly remove the substrate from the placement mechanism through the robot; for a qualified substrate, the actual position coordinate P of the chip is then measured. i Calculation.

3. The chip cutting route calculation method according to claim 2, characterized in that: The actual position coordinates of the chip P i The calculation specifically includes the CCD camera of the visual inspection system moving to the coordinate mark point of the substrate, the coordinate mark point is used as the coordinate origin, and the visual inspection system records the visual coordinates VP (x0, y0) and the mechanism coordinates MP (x0, y0) of the coordinate mark point; the CCD camera and the placement mechanism move together, and when the CCD camera reaches an area, it takes a picture of this area and records the visual coordinates of a single chip as VP' (x i ,y i ), the mechanism coordinates are MP'(x i ,y i ); Load the coordinate algorithm model VP'-VP+MP'-MP, then the actual position coordinate of this chip is P i (x i ,y i ), x i is the X-axis coordinate value x of the chip i =VP'(x i )-VP(x0)+MP'(x i )-MP(x0), y i is the Y-axis coordinate value y of the chip i =VP'(y i )-VP(y0)+MP'(y i )-MP(y0), i is a positive integer.

4. The chip cutting route calculation method according to claim 1, characterized in that: In the calculation of the cutting path, the X-axis straight line a is first fitted, and the linear regression equation is used =b +a, where ， xi and yi are the actual position coordinates of the chip P i The coordinate values ​​of the X-axis and Y-axis can be fitted to obtain the X-axis straight line a= -b ; Then fit the Y-axis straight line and convert the position coordinates of each chip P i Reverse the X coordinate and Y coordinate to get the chip position coordinates. After reversing, get the coordinates P. ’ i , through the coordinate P ’ i The Y-axis straight line is fitted using the least squares method.

5. The chip cutting route calculation method according to any one of claims 1 to 4, characterized in that: Before the angle of the substrate is corrected, the substrate is first identified. The CCD camera of the visual inspection system captures the identification code on the substrate. After the visual inspection system successfully identifies the identification code, the product document information is loaded. However, if the identification code fails to be recognized, an alarm prompts.

6. A chip cutting route calculation system, comprising a visual inspection system and a placement mechanism for placing a substrate, wherein the placement mechanism can convey the substrate and drive the substrate to rotate, wherein the visual inspection system comprises a collection mechanism and a processing mechanism in communication with the collection mechanism, the collection mechanism is used to take pictures of the substrate, and the processing mechanism is used to receive the image taken by the collection mechanism and process the image, characterized in that: Using the method described in any one of claims 1 to 5, the processing mechanism comprises An identification code detection module, used for identifying and detecting the identification code on the substrate, and loading product document information when the identification code detection module successfully identifies the identification code; An angle correction module, used for calculating the angle deviation of the substrate and transmitting the obtained angle deviation to the placement mechanism; The detection module is used to collect the actual position coordinates P of the chip on the substrate. i , and detect chip quality; The cutting path planning module is used to plan the cutting path. The cutting path planning module is based on the actual position coordinates P of each chip. i , the cutting path is fitted on the X-axis and Y-axis using the least squares method.

7. The chip cutting route calculation system according to claim 6, characterized in that: The acquisition mechanism is a CCD camera, and the CCD camera adopts a variable-magnification telecentric lens.

8. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the chip cutting route calculation method according to any one of claims 1 to 5 is implemented.

Images