A jigsaw puzzle method for measuring expansion and contraction of flexible circuit boards

Abstract

The invention relates to the field of image processing technology, and discloses a jigsaw puzzle method for measuring the expansion and contraction of a flexible circuit board. In actual use, the surface image of the flexible circuit board is photographed multiple times by using a camera to ensure the clarity of the photographed image, and then two characteristic objects with the same characteristic value are found in the overlapping area of ​​the two images with the overlapping area, and finally the two images with the overlapping area are spliced ​​into one image according to the two characteristic objects with the same characteristic value, so as to ensure the consistency of the images photographed multiple times by the camera with the surface image of the entire flexible circuit board after splicing; in addition, when two characteristic objects with the same characteristic value are not found in the overlapping area of ​​the two images with the overlapping area, the normal jigsaw puzzle of the two images is ensured by manually selecting a reference object in one image and then searching for an object identical to the reference object in another image.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a jigsaw puzzle method for measuring expansion and contraction of a flexible circuit board. Background Art

[0002] In electronic products, chips and components are mostly welded on PCB circuit boards and flexible circuit boards to achieve the connection between chips and components. Compared with PCB circuit boards, flexible circuit boards have excellent characteristics such as light weight, thin thickness, free bending and folding, and are widely used in liquid crystal display and other fields. Steel mesh is used in the production process of flexible circuit boards. However, due to its own material properties and sandwich structure, flexible circuit boards are prone to thermal expansion and contraction in different environments, which in turn causes changes in their own dimensions. Therefore, in the process of generating flexible circuit boards, it is necessary to detect the expansion and contraction values ​​of flexible circuit boards and then select the corresponding type of steel mesh.

[0003] At present, visual inspection is often used to capture the surface image of the flexible circuit board, and the expansion and contraction value of the flexible circuit board is obtained by processing the surface image of the flexible circuit board. However, the images captured by existing visual inspection are all captured once, and there is a problem of unclear images, which affects the calculation of the expansion and contraction value. Summary of the invention

[0004] In view of the shortcomings of the background technology, the present invention provides a mosaic method for measuring the expansion and contraction of a flexible circuit board, which can stitch images of the flexible circuit board taken multiple times to ensure that the images used for calculating the expansion and contraction values ​​are clear.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a jigsaw puzzle method for measuring the expansion and contraction of a flexible circuit board, comprising the following steps:

[0006] S1: Use a camera to capture images of the flexible circuit board at least twice, and the images of the flexible circuit board captured each time have overlapping areas;

[0007] S2: looping through steps S3 to S7 until all images captured by the camera in step S1 are stitched into the same image;

[0008] S3: Select two images with overlapping areas. The selected images cannot be selected again.

[0009] S4: extract all feature objects in the overlapping areas of the two images respectively;

[0010] S5: Calculate the feature value of each feature object;

[0011] S6: traverse all feature objects in the overlapping area of ​​the two images until two feature objects with the same feature values ​​are found in the two images, and the two feature objects with the same feature values ​​are used as reference objects;

[0012] S7: Calculate the center points of two reference objects;

[0013] S8: moving the two selected images with overlapping areas so that the center points of the reference objects on the two images coincide with each other. After the center points of the reference objects on the two images coincide with each other, the two selected images with overlapping areas are stitched together, and the stitched image is used as a new image.

[0014] In a certain embodiment, before using the camera to capture the image of the flexible circuit board in step S1, the camera's shooting parameters are set first. The camera's shooting parameters include a shooting starting point and a shooting number. When the camera captures the image of the flexible circuit board, it starts shooting from the shooting starting point. After the image of the current area of ​​the flexible circuit board is captured, the camera and the surface of the flexible circuit board are moved relative to each other, so that the camera captures the image of the next area to be captured on the flexible circuit board. The overlapping area of ​​the image of the flexible circuit board captured by the camera each time is the same.

[0015] In a certain embodiment, the image of the flexible circuit board taken by the camera is in a rectangular shape. When the width direction of the image of the flexible circuit board taken by the camera is not the same as the reference direction, the image of the flexible circuit board taken by the camera is rotated so that the width direction of the image of the flexible circuit board taken by the camera is the same as the reference direction.

[0016] In some implementations, the feature values ​​in step S5 include the shape, length, width, area, and average pixel RGB contrast of the feature object.

[0017] In a certain embodiment, in step S8, while moving the two selected images with overlapping areas so that the center points of the reference objects on the two images overlap, the distance between the center points of the two reference objects is continuously calculated. When the distance between the center points of the two reference objects is greater than the position threshold, the two images are brought closer to each other at a speed of V1. When the distance between the center points of the two reference objects is less than the position threshold, the two images are brought closer to each other at a speed of V2, and the speed V1 is greater than the speed V2.

[0018] In some implementations, when no feature object is extracted in the overlapping area in step S4, the following steps are performed:

[0019] S9: Manually select a reference object in the overlapping area of ​​an image;

[0020] S10: Calculate the characteristic value of the reference object;

[0021] S11: searching for an object having the same characteristic value as the reference object in the overlapping area of ​​another image;

[0022] S12: Calculate the center point of the reference object and the center point of the object respectively;

[0023] S13: Move the two selected images with overlapping areas so that the center point of the reference object coincides with the center point of the object. After the center point of the reference object coincides with the center point of the object, the two selected images with overlapping areas are stitched together, and the stitched image is used as a new image.

[0024] In some implementations, the characteristic values ​​of the reference object include the area, perimeter, and outline of the reference object.

[0025] Compared with the prior art, the present invention has the following beneficial effects: in actual use, the surface image of the flexible circuit board is captured multiple times by using a camera to ensure the clarity of the captured image, and then two feature objects with the same feature value are found in the overlapping area of ​​the two images with the overlapping area, and finally the two images with the overlapping area are spliced ​​into one image according to the two feature objects with the same feature value, thereby ensuring the consistency of the images captured multiple times by the camera with the entire flexible circuit board surface image after splicing; in addition, when the two feature objects with the same feature value are not found in the overlapping area of ​​the two images with the overlapping area, the normal puzzle of the two images is ensured by manually selecting a reference object in one image and then searching for the object identical to the reference object in the other image. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flow chart of the present invention in an embodiment;

[0027] Figure 2 This is a schematic diagram of the camera photographing the flexible circuit board three times in the embodiment;

[0028] Figure 3 It is a schematic diagram of the camera in the embodiment converting the flexible circuit board image into two images;

[0029] Figure 4 Schematic diagram of an image of a flexible circuit board requiring rotation correction captured by a camera in an embodiment. DETAILED DESCRIPTION

[0030] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0031] like Figure 1 As shown, a jigsaw puzzle method for measuring the expansion and contraction of a flexible circuit board comprises the following steps:

[0032] S1: Use a camera to capture images of the flexible circuit board at least twice, and the images of the flexible circuit board captured each time have overlapping areas;

[0033] S2: looping through steps S3 to S7 until all images captured by the camera in step S1 are stitched into the same image;

[0034] S3: Select two images with overlapping areas. The selected images cannot be selected again.

[0035] S4: extract all feature objects in the overlapping areas of the two images respectively;

[0036] S5: Calculate the feature value of each feature object;

[0037] S6: traverse all feature objects in the overlapping area of ​​the two images until two feature objects with the same feature values ​​are found in the two images, and the two feature objects with the same feature values ​​are used as reference objects;

[0038] S7: Calculate the center points of two reference objects;

[0039] S8: moving the two selected images with overlapping areas so that the center points of the reference objects on the two images coincide with each other. After the center points of the reference objects on the two images coincide with each other, the two selected images with overlapping areas are stitched together, and the stitched image is used as a new image.

[0040] Specifically, refer to Figure 3 , Figure 3 The camera takes two images in total. The overlapping area in one image has circular and elliptical feature objects, and the other image has circular and rectangular feature objects. If the feature values ​​of the circular feature objects in the two images are the same, the circular feature objects in the two images are used as reference objects.

[0041] Specifically, in step S1, before using the camera to capture the image of the flexible circuit board, the camera's shooting parameters are set first. The camera's shooting parameters include a shooting starting point and a shooting number. When the camera captures the image of the flexible circuit board, it starts shooting from the shooting starting point. After the image of the current area of ​​the flexible circuit board is captured, the camera and the surface of the flexible circuit board are moved relative to each other, so that the camera captures the image of the next area to be captured on the flexible circuit board. The area of ​​the overlapping area of ​​the image of the flexible circuit board captured by the camera each time is the same.

[0042] For example, Figure 2 In the example, the camera takes images of the flexible circuit board 1 three times, with the left edge of the flexible circuit board 1 as the starting point, and then moves the camera in the direction of the arrow to take images of the flexible circuit board 1. Figure 3Each dotted line is the width of each captured image. The first captured image and the second captured image have an overlapping area D1, and the third captured image and the second captured image have an overlapping area D2. The overlapping area D1 and the overlapping area D2 are the same in size.

[0043] In actual use, the width of the overlapping area of ​​each image of the flexible circuit board 1 taken by the camera can be automatically calculated according to the set number of camera shooting times, the width of the flexible circuit board, and the width of the image taken by the camera each time, thereby reducing the time required for manual calculation of the overlapping area, thereby improving the stitching speed and making the efficiency more stable.

[0044] In actual use, the camera can be a line scan camera. The images captured by the line scan camera are multiple frame images, and the scanning and transmission speeds are faster than those of existing area array cameras. By capturing the image of the flexible circuit board multiple times, the clarity of the image captured by the camera can be ensured.

[0045] In a certain implementation, the right edge, upper edge or lower edge of the flexible circuit board can be used as the starting point of the camera's shooting, and the number of shots is determined according to the width of each shot of the camera and the length or width of the flexible circuit board. In addition, the camera can be fixed and the flexible circuit board can be moved, or the camera can be moved and the flexible circuit board can be fixed to make the camera and the flexible circuit board move relative to each other.

[0046] like Figure 4 As shown, the width direction of the image of the flexible circuit board 1 taken by the camera may not be parallel to the X direction. When the width directions of multiple images are not in the same direction, it will affect the splicing of the images. Therefore, in step S1 of the present invention, the image of the flexible circuit board 1 taken by the camera is in a rectangular shape. When the width direction of the image of the flexible circuit board 1 taken by the camera is not the same as the reference direction, the image of the flexible circuit board taken by the camera is rotated so that the width direction of the image of the flexible circuit board taken by the camera is the same as the reference direction. In this embodiment, the reference direction is the X direction.

[0047] In this embodiment, the process of image rotation correction is as follows: two reference points 3 are selected on the image of the flexible circuit board 1 taken by the camera, and the straight line formed by the two reference points 3 is perpendicular to the width direction of the image of the flexible circuit board 1 taken by the camera. If the straight line formed by the two reference points is not perpendicular to the X direction, the image of the flexible circuit board taken by the camera is rotated so that the straight line formed by the two reference points is perpendicular to the X direction. In actual use, the width directions of multiple images taken by the camera are made in the same direction by means of rotation correction, which facilitates image stitching. In actual use, the upper left corner position of the image of the flexible circuit board 1 taken by the camera is used as the rotation point, and the image of the flexible circuit board taken by the camera is rotated in a clockwise direction to be a rectangular shape. When the width direction of the image of the flexible circuit board taken by the camera is not the same as the reference direction, the image of the flexible circuit board taken by the camera is rotated so that the width direction of the image of the flexible circuit board taken by the camera is the same as the reference direction.

[0048] Specifically, the feature values ​​in step S5 of this embodiment include the shape, length, width, area and average pixel RGB contrast of the feature object.

[0049] In this embodiment, in order to increase the puzzle speed, in step S8, while moving the two selected images with overlapping areas so that the center points of the reference objects on the two images overlap, the distance between the center points of the two reference objects is continuously calculated. When the distance between the center points of the two reference objects is greater than the position threshold, the two images are moved closer to each other at a speed of V1. When the distance between the center points of the two reference objects is less than the position threshold, the two images are moved closer to each other at a speed of V2, and the speed V1 is greater than the speed V2.

[0050] In actual use, in step S4, no feature object may be extracted in the overlapping area of ​​the two images. In order to ensure the normal puzzle of the two images, when no feature object is extracted in the overlapping area in step S4, the following steps are performed:

[0051] S9: Manually select a reference object in the overlapping area of ​​an image;

[0052] S10: Calculate the characteristic value of the reference object;

[0053] S11: searching for an object having the same characteristic value as the reference object in the overlapping area of ​​another image;

[0054] S12: Calculate the center point of the reference object and the center point of the object respectively;

[0055] S13: Move the two selected images with overlapping areas so that the center point of the reference object coincides with the center point of the object. After the center point of the reference object coincides with the center point of the object, the two selected images with overlapping areas are stitched together, and the stitched image is used as a new image.

[0056] It should be noted that the spliced ​​image described in step S8 and step S13 in this embodiment is used as a new image, and the new image can be re-selected in step S3.

[0057] In actual use, when the camera takes more than four flexible circuit board images, in order to increase the puzzle speed, multiple groups of images can be selected at the same time in step S3, each group of images includes two images with overlapping areas, and the two images in all groups of images are different from each other, and steps S3 to S7 are performed in parallel on all groups of images. For example, when there are only four images, when all steps from step S3 to step S7 are performed on the first image and the second image for puzzle, all steps from step S3 to step S7 are performed on the third image and the fourth image for puzzle, and then the two new images after puzzle are performed once in the order of steps S3-S7.

[0058] In actual use, when the camera captures more than two images of the flexible circuit board, step S3 can be performed on the captured images, so as to realize jigsaw puzzle while shooting, thereby improving jigsaw puzzle efficiency.

[0059] In this embodiment, the characteristic values ​​of the reference object in step S10 include the area, perimeter and outline of the reference object. In actual use, in step S9, a gold surface can be manually selected on the image of the flexible circuit board as a reference object. The gold surface on the image of the flexible circuit board may correspond to a pad or an optical positioning point. Preferably, the gold surface is a regular shape, such as a square or a circle. When the gold surface is a square, the characteristic values ​​calculated in step S10 include width and length. When the gold surface is a circle, the characteristic values ​​are the coordinates of the center of the circle and the radius.

[0060] In summary, the present invention ensures the clarity of the captured images by using a camera to capture the surface images of the flexible circuit board multiple times, and then searches for two feature objects with the same feature values ​​in the overlapping area of ​​the two images with the overlapping area. Finally, the two images with the overlapping area are spliced ​​into one image according to the two feature objects with the same feature values, thereby ensuring the consistency of the images captured by the camera multiple times with the entire flexible circuit board surface image after splicing; in addition, when two feature objects with the same feature values ​​are not found in the overlapping area of ​​the two images with the overlapping area, the normal puzzle of the two images is ensured by manually selecting a reference object in one image and then searching for the object identical to the reference object in the other image.

[0061] The above is based on the present invention as an inspiration. Through the above description, relevant staff can make various changes and modifications without departing from the technical idea of ​​this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A jigsaw puzzle method for measuring the expansion and contraction of a flexible circuit board, characterized in that: The steps include: S1: Use a camera to capture images of the flexible circuit board at least twice, and the images of the flexible circuit board captured each time have overlapping areas; S2: looping through steps S3 to S7 until all images captured by the camera in step S1 are stitched into the same image; S3: Select two images with overlapping areas. The selected images cannot be selected again. S4: extract all feature objects in the overlapping areas of the two images respectively; S5: Calculate the feature value of each feature object; S6: traverse all feature objects in the overlapping area of ​​the two images until two feature objects with the same feature values ​​are found in the two images, and the two feature objects with the same feature values ​​are used as reference objects; S7: Calculate the center points of two reference objects; S8: moving the two selected images with overlapping areas so that the center points of the reference objects on the two images coincide with each other. After the center points of the reference objects on the two images coincide with each other, the two selected images with overlapping areas are stitched together, and the stitched image is used as a new image.

2. A jigsaw puzzle method for measuring expansion and contraction of a flexible circuit board according to claim 1, characterized in that: In step S1, before using the camera to capture an image of the flexible circuit board, the camera's shooting parameters are set first. The camera's shooting parameters include a shooting starting point and a shooting number. When the camera captures the image of the flexible circuit board, it starts shooting from the shooting starting point. After the image of the current area of ​​the flexible circuit board is captured, the camera and the surface of the flexible circuit board are moved relative to each other, so that the camera captures an image of the next area to be captured on the flexible circuit board. The area of ​​the overlapping area of ​​the image of the flexible circuit board captured by the camera each time is the same.

3. A jigsaw puzzle method for measuring expansion and contraction of a flexible circuit board according to claim 1, characterized in that: The image of the flexible circuit board taken by the camera is in a rectangular shape. When the width direction of the image of the flexible circuit board taken by the camera is not the same as the reference direction, the image of the flexible circuit board taken by the camera is rotated so that the width direction of the image of the flexible circuit board taken by the camera is the same as the reference direction.

4. A jigsaw puzzle method for measuring expansion and contraction of a flexible circuit board according to claim 1, characterized in that: The feature values ​​in step S5 include the shape, length, width, area and average pixel RGB contrast of the feature object.

5. A jigsaw puzzle method for measuring expansion and contraction of a flexible circuit board according to claim 1, characterized in that: In step S8, while moving the two selected images with overlapping areas so that the center points of the reference objects on the two images overlap, the distance between the center points of the two reference objects is continuously calculated. When the distance between the center points of the two reference objects is greater than the position threshold, the two images are moved closer to each other at a speed of V1. When the distance between the center points of the two reference objects is less than the position threshold, the two images are moved closer to each other at a speed of V2, where speed V1 is greater than speed V2.

6. A jigsaw puzzle method for measuring expansion and contraction of a flexible circuit board according to claim 1, characterized in that: When no feature object is extracted in the overlapped area in step S4, the following steps are performed: S9: Manually select a reference object in the overlapping area of ​​an image; S10: Calculate the characteristic value of the reference object; S11: searching for an object having the same characteristic value as the reference object in the overlapping area of ​​another image; S12: Calculate the center point of the reference object and the center point of the object respectively; S13: Move the two selected images with overlapping areas so that the center point of the reference object coincides with the center point of the object. After the center point of the reference object coincides with the center point of the object, the two selected images with overlapping areas are stitched together, and the stitched image is used as a new image.

7. A jigsaw puzzle method for measuring expansion and contraction of a flexible circuit board according to claim 6, characterized in that: The characteristic values ​​of the reference object include the area, perimeter and outline of the reference object.

Images