Block cutting control method, device and block cutting system

Abstract

The block cutting control method, device and block cutting system provided by the embodiment of the present invention include: obtaining a block image to be cut taken by a camera, and a mark to be cut is set on the block to be cut; using a preset standard mark image as a convolution kernel, calculating the area in the block image to be cut with the highest similarity to the standard mark image as the target area; calculating the image position of the center of the target area in the block image to be cut; based on the conversion relationship between the preset camera coordinate system and the world coordinate system, calculating the actual position of the center according to the image position, so that the machine can cut the block to be cut based on the actual position. The present invention can determine the area where the mark to be cut is located in the collected image of the block to be cut according to the standard mark image, and use the center of the area as the reference for block cutting so that the machine can accurately cut the block to be cut, thereby improving production efficiency and reducing work hazards.

Description

Technical Field

[0001] The present invention relates to the field of electronic component manufacturing, and in particular to a block cutting control method, device and block cutting system. Background Art

[0002] Chip-type multilayer electronic components are a new type of components that are indispensable in the electronics, communications and information industries. They are widely used in surface mounting of consumer electronic devices such as communications, computers, and household appliances. In the manufacturing process of such electronic components, multiple diaphragms are usually stacked to form a green substrate, and then the green substrate is compacted by lamination to prepare a large bar block, which then needs to be cut into 4 small bar blocks for subsequent production processes.

[0003] The existing method of cutting a large bar block into four small bars mainly relies on manual operation. The worker puts the bar block on a simple rack with a visual camera, and manually moves the bar block cutting point to the cutting position while watching the camera image. After the adjustment is completed, press the cutter button to trigger the cutter to cut. This production technology relies on manual loading and unloading, and the production efficiency is low. The method of distinguishing the cutting point by the naked eye cannot guarantee the cutting accuracy of the bar block. In addition, when manually moving and adjusting the bar block position, the hand is close to the cutter, which is dangerous. Summary of the invention

[0004] The embodiments of the present invention provide a block cutting control method, device and block cutting system, which can determine the area where the to-be-cut mark in the collected block to-be-cut image is located according to the standard mark image, and use the center of the area as the reference for block cutting so that the machine can accurately cut the block to be cut, thereby improving production efficiency and reducing work risks.

[0005] In order to achieve the above object, an embodiment of the present invention provides a block cutting control method, comprising:

[0006] Acquire an image of the to-be-cut block captured by a camera; wherein a to-be-cut mark is set on the to-be-cut block;

[0007] Using a preset standard marker image as a convolution kernel, calculating the area in the image to be cut that has the highest similarity with the standard marker image as the target area;

[0008] Calculate the image position of the center of the target area in the image of the block to be cut;

[0009] Based on the conversion relationship between the preset camera coordinate system and the world coordinate system, the actual position of the center is calculated according to the image position, so that the machine can cut the block to be cut based on the actual position.

[0010] As an improvement of the above solution, before using the preset standard marker image as the convolution kernel and calculating the area in the image to be cut that has the highest similarity with the standard marker image as the target area, the method further includes:

[0011] Based on a preset threshold, identify the area where the to-be-cut mark in the to-be-cut block image is located, and use the identified area as a region of interest;

[0012] An opening operation is performed on the region of interest to highlight the mark to be cut.

[0013] As an improvement of the above solution, the opening operation is performed on the region of interest to highlight the mark to be cut, specifically including:

[0014] Binarizing the region of interest based on a preset binarization threshold;

[0015] An erosion-and-dilation operation is performed on the region of interest.

[0016] As an improvement of the above solution, the structured element of the open operation is a "Z"-shaped structure.

[0017] In order to achieve the above object, an embodiment of the present invention further provides a block cutting control device, comprising:

[0018] An image acquisition module is used to acquire an image of the to-be-cut block taken by a camera; wherein a to-be-cut mark is set on the to-be-cut block;

[0019] A target region determination module is used to use a preset standard marker image as a convolution kernel, and calculate the region in the image to be cut that has the highest similarity with the standard marker image as the target region;

[0020] An image position determination module, used for calculating the image position of the center of the target area in the image of the block to be cut;

[0021] The actual position determination module is used to calculate the actual position of the center according to the image position based on the conversion relationship between the preset camera coordinate system and the world coordinate system, so that the machine can cut the block to be cut based on the actual position.

[0022] As an improvement of the above scheme, it also includes:

[0023] An area of ​​interest determination module is used to identify the area where the to-be-cut mark in the to-be-cut block image is located based on a preset threshold, and use the identified area as the area of ​​interest;

[0024] The mark highlighting module is used to perform an opening operation on the region of interest to highlight the mark to be cut.

[0025] As an improvement of the above solution, the structured element of the open operation is a "Z"-shaped structure.

[0026] In order to achieve the above-mentioned purpose, an embodiment of the present invention further provides a bar block cutting system, including a camera, a machine platform and a controller, wherein the machine platform includes a servo motor, a bar block cutting platform and a bar block cutting actuator; the servo motor is used to drive the bar block cutting platform to move; the bar block cutting platform is used to place a bar block to be cut, and a to-be-cut mark is set on the bar block to be cut; the bar block cutting actuator is used to cut the bar block to be cut;

[0027] The controller is used to:

[0028] Controlling the servo motor to drive the block cutting platform to move until the block to be cut moves into the shooting field of view of the camera;

[0029] Controlling the camera to shoot the block to be cut, and obtaining an image of the block to be cut;

[0030] Using a preset standard marker image as a convolution kernel, calculating the area in the image to be cut that has the highest similarity with the standard marker image as the target area;

[0031] Calculate the image position of the center of the target area in the image of the block to be cut;

[0032] Based on a preset conversion relationship between the camera coordinate system and the world coordinate system, the actual position of the center is calculated according to the image position;

[0033] Based on the actual position, the block cutting actuator is controlled to cut the block to be cut.

[0034] As an improvement of the above solution, the controller is also used for:

[0035] Before using a preset standard marker image as a convolution kernel and calculating the area in the to-be-cut block image with the highest similarity to the standard marker image as the target area, identifying the area where the to-be-cut marker is located in the to-be-cut block image based on a preset threshold, and using the identified area as the region of interest;

[0036] An opening operation is performed on the region of interest to highlight the mark to be cut.

[0037] As an improvement of the above solution, the structured element of the open operation is a "Z"-shaped structure.

[0038] Compared with the prior art, the block cutting control method, device and block cutting system provided by the embodiments of the present invention firstly use the preset standard mark image as the convolution kernel, and perform convolution operation with the block image to be cut obtained by the camera, calculate the similarity between each area in the block image to be cut and the standard mark image to select the area with the highest similarity as the target area, then calculate the image position of the center of the target area; finally, perform coordinate transformation on the image position of the center to obtain the actual position so that the control machine can cut the block to be cut. The embodiments of the present invention determine the area where the mark to be cut in the collected block image to be cut is located according to the standard mark image, and use the center of the area as the reference for block cutting so that the machine can accurately cut the block to be cut, thereby improving production efficiency and reducing work hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of a block cutting control method provided by an embodiment of the present invention;

[0040] Figure 2 Schematic diagram of the principle of image convolution operation provided by one embodiment of the present invention;

[0041] Figure 3 is a simplified schematic diagram of a template image provided by an embodiment of the present invention;

[0042] Figure 4 It is a structural block diagram of a block cutting control device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 , is a flow chart of a block cutting control method provided by an embodiment of the present invention. The method comprises steps S11 to S14:

[0045] S11, obtaining an image of a to-be-cut block captured by a camera, wherein a to-be-cut mark is set on the to-be-cut block;

[0046] S12, using a preset standard marked image as a convolution kernel, calculating and obtaining the region in the image to be cut with the highest similarity to the standard marked image as the target region;

[0047] S13, calculating the image position of the center of the target area in the image of the block to be cut;

[0048] S14. Based on the conversion relationship between the preset camera coordinate system and the world coordinate system, the actual position of the center is calculated according to the image position, so that the machine cuts the block to be cut based on the actual position.

[0049] In step S1, the camera is a 1.30W pixel industrial camera and is equipped with a customized light source. It is worth noting that the type of camera is not limited to the above specific settings.

[0050] Specifically, in order to more clearly explain the process of determining the target area, the following introduces the similarity extraction of the convolution kernel, see Figure 2 , Figure 2 The number 2 in the figure is the image to be tested, and the slash is the 3*3 convolution kernel. Figure 2 It can be seen that after the convolution operation, the value calculated for the area with the same convolution kernel in the image to be tested is 3, which is larger than the calculation results of other areas. Therefore, the maximum value obtained by convolution calculation is used as a reference for similarity.

[0051] In step S12, for example, see Figure 3 The simplified schematic diagram of the template image shown in FIG. 1 is to use the template image (standard marker image) as the convolution kernel, and to perform convolution operation with the photo captured by the camera (the image of the block to be cut) to obtain the convolution result, and to use the convolution result as the image similarity, and to divide the area in the photo with the highest similarity to the template image as the target area. Among them, the template image is an image about the standard marker, and each layer of the stacked block is set with a specific mark. In the template image, the marks of each layer are on the same straight line and the straight line is perpendicular to the upper surface of the block.

[0052] In steps S13 and S14, after obtaining the area in the photo that has the highest matching degree with the template image, the center of the area is calculated, and the cutting position is calculated using the information of the center of the area and fed back to the cutting actuator of the machine. The actual position of the center of the area can be obtained by using the conversion relationship between the camera coordinate system and the world coordinate system to perform coordinate transformation on the center of the area to obtain its actual position on the machine. For example, after calculating the actual position of the center of the area, the actual position is used as the standard marking position of the block, and the machine is controlled to cut the block.

[0053] It can be understood that each layer of the laminated block is set with a mark to be cut. In the actual manufacturing process of the block, it is inevitable that the block stacking will be misaligned, resulting in poor uniformity of the mark to be cut and inaccurate cutting position judgment by the human eye. By identifying the area where the mark is located and calculating the center of the area, and using the center of the area as the standard mark position, the cutting accuracy of each layer can be balanced as much as possible during cutting, thereby achieving accurate cutting of the block.

[0054] The embodiment of the present invention determines the area where the to-be-cut mark is located in the collected image of the block to be cut according to the standard mark image, and uses the center of the area as the reference for the block cutting so that the machine can accurately cut the block to be cut, thereby improving production efficiency and reducing work risks.

[0055] In one embodiment, before using the preset standard marker image as the convolution kernel and calculating the area in the image to be cut that has the highest similarity with the standard marker image as the target area, the method further includes:

[0056] Based on a preset threshold, identify the area where the to-be-cut mark in the to-be-cut block image is located, and use the identified area as a region of interest;

[0057] An opening operation is performed on the region of interest to highlight the mark to be cut.

[0058] Specifically, first, threshold segmentation is performed to highlight the subject, that is, the image is expanded to connect the discrete pixels generated by interference, and the area where the mark to be cut is located in the block image to be cut is detected based on the preset threshold by setting a preset threshold, and the identified area is framed in the ROI domain as the region of interest, and the region of interest is opened to highlight the Mark mark feature. It is worth noting that the preset threshold is set according to the characteristics of the mark in the actual application.

[0059] In one embodiment, performing an opening operation on the region of interest to highlight the mark to be cut specifically includes:

[0060] Binarizing the region of interest based on a preset binarization threshold;

[0061] An erosion-and-dilation operation is performed on the region of interest.

[0062] Specifically, the Mark mark (the mark to be cut) is corroded first to shrink its boundary inward, and then the Mark mark is expanded. All the background points it contacts will be merged into the Mark mark, and the boundary will expand outward. This process can eliminate the burrs around the Mark mark.

[0063] In one implementation, the structured element of the open operation is a "Z"-shaped structure.

[0064] Specifically, the following is a detailed introduction to the open operation in the implementation of the present invention:

[0065] 1. The specific process of corroding the Mark mark is as follows: (1) binarizing each pixel of the Mark mark within the field of view, that is, setting a binarization threshold (threshold between 100 and 150) for the grayscale value of the pixel point (grayscale value range 0-255). If the grayscale value is greater than the threshold (binarization threshold), it is distinguished as 1, otherwise it is 0, so that each pixel becomes a binary image (the optimal threshold in this process is 120); (2) Selection of structuring elements: Scan the Mark mark within the field of view, and accurately save the position of the surrounding area of ​​the mark based on the mark image. Determine the filtered image based on the surrounding position image. The filtered image determined in the embodiment of the present invention is "Z" shaped and has geometric similarity with the misaligned part of the stacked Mark mark in the inspected image. Binarize it and then reduce the pixel count to form a new heteromorphic structuring element. Use the structuring element to scan and capture each pixel of the Mark mark in the inspected image, and perform an "AND" operation with the binary image covered by it. If both are 1, the pixel of the result image is 1, otherwise it is 0. Operation effect: The Mark in the inspected image shrinks inward.

[0066] 2. The specific process of the Mark expansion operation is as follows: ① Binarize each pixel of the Mark within the field of view (the same as the specific process of the Mark corrosion); ② Use the structural element in the corrosion process to scan each pixel of the Mark, and use the structural element to perform an "or" operation with the binary image covered by it. If both are 0, the pixels of the resulting image are all 0, otherwise they are 1. Operation effect: The binary image is expanded by one circle, that is, the entire Mark is expanded outward by one circle.

[0067] 3. In actual production, most Mark points are rectangular. In an ideal stacking situation, the Mark image is a rectangle. However, in the actual process, due to stacking misalignment, the Mark image presents a rectangular middle part and burrs on both sides with many rectangular protrusions. In this case, the commonly used 3*3 square structuring element is selected as the filtering unit. Although it can also achieve the effect of removing burrs, the actual deviation of the cutting line from the theory is still large; using a rectangle consistent with the Mark image formed by the ideal stacking situation, compared with the 3*3 square structuring element, although the effect is improved, because the burrs are also rectangular in the actual misaligned image, many burrs will be missed during the filtering process, resulting in position offset when calculating the cutting line, affecting the cutting effect. The embodiment of the present invention selects the "Z" shape as the structuring element, combined with the actual situation, to remove burrs as little as possible. Preferably, the size is in the range of 3*3-5*5 matrix.

[0068] Compared with manual search for cutting points, the visual method of the embodiment of the present invention is faster and more accurate, and can realize unmanned operation, which is safer.

[0069] See also Figure 4 , is a structural block diagram of a bar block cutting control device provided by an embodiment of the present invention, the bar block cutting control device comprises:

[0070] The image acquisition module 11 is used to acquire an image of the to-be-cut block taken by a camera; wherein a to-be-cut mark is set on the to-be-cut block;

[0071] The target region determination module 12 is used to use a preset standard marker image as a convolution kernel, and calculate the region in the image to be cut that has the highest similarity with the standard marker image as the target region;

[0072] An image position determination module 13, used for calculating the image position of the center of the target area in the image of the block to be cut;

[0073] The actual position determination module 14 is used to calculate the actual position of the center according to the image position based on the conversion relationship between the preset camera coordinate system and the world coordinate system, so that the machine can cut the block to be cut based on the actual position.

[0074] In one embodiment, it further includes:

[0075] An area of ​​interest determination module is used to identify the area where the to-be-cut mark in the to-be-cut block image is located based on a preset threshold, and use the identified area as the area of ​​interest;

[0076] The mark highlighting module is used to perform an opening operation on the region of interest to highlight the mark to be cut.

[0077] In one implementation, the structured element of the open operation is a "Z"-shaped structure.

[0078] It is worth noting that the specific working process of the bar cutting control device can refer to the working process of the bar cutting control method described in the above embodiment, which will not be repeated here.

[0079] The embodiment of the present invention further provides a bar block cutting system, comprising a camera, a machine platform and a controller, wherein the machine platform comprises a servo motor, a bar block cutting platform and a bar block cutting actuator; the servo motor is used to drive the bar block cutting platform to move; the bar block cutting platform is used to place a bar block to be cut, and a to-be-cut mark is set on the bar block to be cut; the bar block cutting actuator is used to cut the bar block to be cut;

[0080] The controller is used to:

[0081] Controlling the servo motor to drive the block cutting platform to move until the block to be cut moves into the shooting field of view of the camera;

[0082] Controlling the camera to shoot the block to be cut, and obtaining an image of the block to be cut;

[0083] Using a preset standard marker image as a convolution kernel, calculating the area in the image to be cut that has the highest similarity with the standard marker image as the target area;

[0084] Calculate the image position of the center of the target area in the image of the block to be cut;

[0085] Based on a preset conversion relationship between the camera coordinate system and the world coordinate system, the actual position of the center is calculated according to the image position;

[0086] Based on the actual position, the block cutting actuator is controlled to cut the block to be cut.

[0087] In one embodiment, the controller is further configured to:

[0088] Before using a preset standard marker image as a convolution kernel and calculating the area in the to-be-cut block image with the highest similarity to the standard marker image as the target area, identifying the area where the to-be-cut marker is located in the to-be-cut block image based on a preset threshold, and using the identified area as the region of interest;

[0089] An opening operation is performed on the region of interest to highlight the mark to be cut.

[0090] In one implementation, the structured element of the open operation is a "Z"-shaped structure.

[0091] It is worth noting that the specific working process of the controller in the bar cutting system can refer to the working process of the bar cutting control method described in the above embodiment, which will not be repeated here.

[0092] In order to more clearly explain the working process and advantages of the bar block cutting system, the following is an explanation of the specific test method and specific cutting control process of the ceramic capacitor bar block:

[0093] 1. Test method:

[0094] 1. The test method for the scrap rate is: take 1,000 pieces of cut products, define the products whose absolute value of the difference with the theoretical cutting line is greater than 5mm as scraps, and calculate the scrap rate.

[0095] 2. Cutting line position deviation: Take 100 cut products and calculate the average difference of the theoretically calculated cutting line position deviation.

[0096] 2. Specific cutting control process:

[0097] 1. Specific cutting control process of the embodiment of the present invention:

[0098] S1: Acquire images: calibrate the camera and determine the transformation relationship between the camera coordinate system and the world coordinate system.

[0099] S2: Get the image: Use the servo motor to drive the platform to move in the X and θ directions, send the block mark to the camera's field of view, capture the picture, and get the original image.

[0100] S3: Threshold segmentation, highlighting the subject. Dilate the image to connect the discrete pixels generated by interference.

[0101] S4: Determine whether there is a Mark mark, otherwise return to S2; if there is a Mark mark, enter S5.

[0102] S5: Identify the Mark marker and divide the image into a region of interest (ROI).

[0103] S6: Open the region. First, erode the Mark mark, and binarize each pixel of the Mark mark within the field of view, that is, the grayscale value of the pixel point is set to a threshold of 120, and the grayscale value greater than the threshold is distinguished as 1, otherwise it is 0, so that each pixel becomes a binary image, and the "Z" shape of the 3*3 matrix is ​​selected as the structural element. The structural element is used to scan and capture each pixel of the Mark mark in the inspected image, and the "AND" operation is performed with the binary image covered by it. If all are 1, the pixel of the result image is 1, otherwise it is 0; then the Mark mark is expanded, and all the background points it contacts will be merged into the Mark mark, so that the boundary expands outward. This process can eliminate the burrs around the Mark mark and highlight the Mark mark features.

[0104] S7: performing a convolution operation on the template image and the photo captured by the camera, and performing a correlation coefficient calculation on the grayscale image obtained after the convolution operation.

[0105] ;

[0106] ;

[0107] Among them, D(i,j) represents the similarity; M represents the number of vertical pixels of the convolution kernel; N represents the number of horizontal pixels of the convolution kernel; S(i,j) represents the grayscale value of the image point (i,j) captured by the camera; T(s,t) represents the grayscale value of the convolution kernel image point (s,t), m represents the number of vertical pixels of the image captured by the camera, and n represents the number of horizontal pixels of the image captured by the camera.

[0108] S8: Get the area with the highest matching degree in the original image and calculate the center of the area. Use the information of the center of the area to calculate the cutting position and feed it back to the actuator.

[0109] In this embodiment, the position deviation of the cutting line after cutting calculated by calculation is within ±0.1mm from the cutting line calculated by theoretical calculation, the scrap rate of the slitting machine is 0.1%, and the time for dividing the large Bar block into 4 small Bar blocks is 1s. The structural element (the "Z" shape of the 3*3 matrix) is replaced by the "Z" shape of the 4*4 matrix, and the others remain unchanged. The position deviation of the cutting line after cutting calculated by calculation is within ±0.3mm from the cutting line calculated by theoretical calculation, the scrap rate of the slitting machine is 0.4%, and the time for dividing the large Bar block into 4 small Bar blocks is 1s. The structural element (the "Z" shape of the 3*3 matrix) is replaced by the "Z" shape of the 5*5 matrix, and the others remain unchanged. The position deviation of the cutting line after cutting calculated by calculation is within ±0.44mm from the cutting line calculated by theoretical calculation, the scrap rate of the slitting machine is 0.3%, and the time for dividing the large Bar block into 4 small Bar blocks is 1s.

[0110] 2. A 3*3 square is used as the structural element. The position deviation between the calculated cutting line and the theoretically calculated cutting line is ±1mm. The scrap rate of the slitting machine is 5%. The time to divide the large Bar block into 4 small Bar blocks is 1s.

[0111] 3. Select the "Z" shape of the 6*6 matrix as the structuring element. The position deviation between the cutting line after calculation and the cutting line calculated theoretically is within ±1mm. The scrap rate of the slitting machine is 3%. The time to divide the large Bar block into 4 small Bar blocks is 1s.

[0112] It can be seen that choosing a “Z” shape with a size ranging from 3*3 to 5*5 matrices as the structuring element makes the cutting position closest to the theoretical cutting position, which greatly improves the production quality.

[0113] In summary, the block cutting control method, device and block cutting system provided by the embodiments of the present invention firstly use the preset standard mark image as the convolution kernel, and perform convolution operation with the image of the block to be cut obtained by the camera, calculate the similarity between each area in the image of the block to be cut and the standard mark image to select the area with the highest similarity as the target area, and then calculate the image position of the center of the target area; finally, perform coordinate transformation on the image position of the center to obtain the actual position so that the control machine can cut the block to be cut. The embodiment of the present invention determines the area where the mark to be cut in the collected image of the block to be cut is located according to the standard mark image, and uses the center of the area as the reference for block cutting so that the machine can accurately cut the block to be cut, which saves the time of repeated manual adjustments, saves the time of loading and unloading, improves production efficiency, reduces the risk of work, ensures the consistency of cutting accuracy, accurately selects the position with the most concentrated distribution of mark points for cutting, and greatly reduces the scrap rate.

[0114] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A block cutting control method, characterized in that: include: Acquire an image of the to-be-cut block captured by a camera; wherein the to-be-cut block is a stacked block and each layer is provided with a to-be-cut mark; Using a preset standard marker image as a convolution kernel, calculating the area in the image to be cut that has the highest similarity with the standard marker image as the target area; Calculate the image position of the center of the target area in the image of the block to be cut; Based on the conversion relationship between the preset camera coordinate system and the world coordinate system, the actual position of the center is calculated according to the image position, so that the machine can cut the block to be cut based on the actual position; Before using the preset standard marked image as the convolution kernel and calculating the area in the image to be cut that has the highest similarity with the standard marked image as the target area, the method further includes: Based on a preset threshold, identify the area where the to-be-cut mark in the to-be-cut block image is located, and use the identified area as a region of interest; Performing an opening operation on the region of interest to highlight the mark to be cut; The performing an opening operation on the region of interest to highlight the mark to be cut specifically includes: Binarizing the region of interest based on a preset binarization threshold; Performing an erosion and then dilation operation on the region of interest; The structured element of the open operation is a "Z" shaped structure.

2. A block cutting control device, characterized in that: include: An image acquisition module is used to acquire an image of the to-be-cut block taken by a camera; wherein the to-be-cut block is a stacked block and each layer is provided with a to-be-cut mark; A target region determination module is used to use a preset standard marker image as a convolution kernel, and calculate the region in the image to be cut that has the highest similarity with the standard marker image as the target region; An image position determination module, used for calculating the image position of the center of the target area in the image of the block to be cut; An actual position determination module, used to calculate the actual position of the center according to the image position based on a preset conversion relationship between the camera coordinate system and the world coordinate system, so that the machine can cut the block to be cut based on the actual position; An area of ​​interest determination module is used to identify the area where the to-be-cut mark in the to-be-cut block image is located based on a preset threshold, and use the identified area as the area of ​​interest; The mark highlighting module is used to perform an opening operation on the region of interest to highlight the mark to be cut; the mark highlighting module is specifically used to: binarize the region of interest based on a preset binarization threshold; perform an erosion and then dilation operation on the region of interest; The structured element of the open operation is a "Z" shaped structure.

3. A block cutting system, characterized in that: The invention comprises a camera, a machine platform and a controller, wherein the machine platform comprises a servo motor, a block cutting platform and a block cutting actuator; the servo motor is used to drive the block cutting platform to move; the block cutting platform is used to place blocks to be cut, wherein the blocks to be cut are stacked blocks and each layer is provided with a mark to be cut; the block cutting actuator is used to cut the blocks to be cut; The controller is used to: Controlling the servo motor to drive the block cutting platform to move until the block to be cut moves into the shooting field of view of the camera; Controlling the camera to shoot the block to be cut to obtain an image of the block to be cut; Using a preset standard marker image as a convolution kernel, calculating the area in the image to be cut that has the highest similarity with the standard marker image as the target area; Calculate the image position of the center of the target area in the image of the block to be cut; Based on a preset conversion relationship between the camera coordinate system and the world coordinate system, the actual position of the center is calculated according to the image position; Based on the actual position, controlling the block cutting actuator to cut the block to be cut; The controller is also used for: Before using a preset standard marker image as a convolution kernel and calculating the area in the to-be-cut block image with the highest similarity to the standard marker image as the target area, identifying the area where the to-be-cut marker is located in the to-be-cut block image based on a preset threshold, and using the identified area as the region of interest; Performing an opening operation on the region of interest to highlight the mark to be cut; The opening operation of the region of interest to highlight the mark to be cut specifically includes: binarizing the region of interest based on a preset binarization threshold; corroding and then dilating the region of interest; The structured element of the open operation is a "Z" shaped structure.

Images