A method for detecting the offset between the contour of a manufactured polygonal object and the design contour of the polygonal object based on its design.

Abstract

A method for adjusting the orientation of a polygonal object's contour image, obtained by detecting the first contour of the polygonal object, is disclosed. [Solution] This method includes (a) determining a plurality of feature points from the first contour of a polygonal object; (b) identifying a plurality of position parameters for the plurality of feature points; (c) obtaining a virtual contour of the first contour by associating the plurality of position parameters; (d) determining a center of rotation based on the virtual contour; and (e) rotating the virtual contour along the center of rotation to align it with the direction to generate a directional contour.

Description

Technical Field

[0001] Cross - reference to related applications and claims of priority This application claims the benefit of Taiwan Patent Application No. 113139050, filed with the Taiwan Intellectual Property Office on October 14, 2024, the disclosure of which is hereby incorporated by reference in its entirety.

[0002] The present invention relates to a method for inspecting a product by detecting a real - time image of the product. In particular, the present invention relates to a method for adjusting the orientation of a design contour based on a polygon object by detecting an offset between the contour of a manufactured polygon object and a design contour based on the design of the polygon object, based on a real - time image of the manufactured polygon object.

Background Art

[0003] When mass-producing IC carrier boards and circuit boards, they are usually manufactured on large motherboards designed for mass production. IC carrier boards and circuit boards have a polygonal shape. For example, each circuit board has various electronic components mounted on it, such as ICs, their carrier boards, resistors, capacitors, semiconductors, and LEDs, either directly or indirectly. The IC carrier board is a crucial component in the packaging process, acting as a carrier for transporting ICs, protecting and securing the circuits placed on it, and dissipating heat generated by the ICs and circuits. Therefore, extremely strict precision requirements are demanded in the manufacturing of IC carrier boards. On the other hand, when processing polygonal objects like IC carrier boards and incorporating them into the internal space of equipment, devices, structures, or enclosures, it is necessary to make the object compact to fully utilize the volume of the limited space. The tolerance for incorporating an object into a space is not large. Conversely, if the tolerance for incorporating an object is increased, the space for other electronic components becomes smaller, and it becomes impossible to accommodate even more electronic components in the limited space. Conventional techniques, for example, to determine whether the contour of a polygonal object differs from its design contour, use an image sensor such as a CCD (charge-coupled device) to detect the contour of an IC carrier board, treating the IC carrier board as the object. By moving the image sensor along the four sides of the object / IC carrier, features such as the edges, interior angles, and boundaries between circuit and non-circuit areas (also called non-effective areas) on the object can be optically detected. Therefore, by comparing the data obtained from the edges and interior angles with data corresponding to the design outline, an offset / deviation between the object's outline and the design outline can be obtained. When detecting multiple objects such as circuit boards and IC carrier boards, they are arranged in a matrix on a tray, but in many cases, they are not necessarily arranged neatly on the tray. That is, each edge of each object is not oriented or aligned with a desired direction (also called the reference direction or design direction) that has been pre-set for translational / rotational offset detection. Therefore, it is necessary to compare each object with a translational and / or rotational offset in the tray with the design outline of the object.Therefore, before comparing the data obtained from the detected object with the design contour data, it is extremely important to align the image of the object detected by the optical sensor (an image converted into detection data such as edge length and corner coordinates) with the direction of the design contour and compare them to determine whether the object's contour is within manufacturing tolerances. Furthermore, using conventional technology, an image of the entire object can be obtained by capturing an image with a single camera device and extracting the contour and corners from that image. However, in a panoramic image of the object, the proportion of each corner of the object in the image is very small. Taking circuit boards and IC carrier boards as examples, the objects usually have a rectangular surface, and in practice, the contour of the entire object is detected by detecting images of the four corners of the object and sequentially connecting virtual lines that connect all adjacent corners to calculate and obtain their position values ​​or coordinates. Therefore, this can be achieved by moving one camera over a tray on which at least one object to be detected is placed. This allows one camera to be oriented in different directions over the upper sides of each corner of at least one object. However, this technology introduces movement errors when the moving device moves the tray and at least one object placed on it, potentially resulting in errors in each image of the object captured by the camera's optical lens. These errors also lead to errors in the measurements. Therefore, to improve measurement accuracy, it is crucial to reduce the number of camera movements and, after acquiring data for each corner of the object, to properly adjust and align the acquired images so that they can be more accurately compared to the object's design contours. These are urgent issues that need to be addressed. [Disclosure of the Invention] [Problems that the invention aims to solve]

[0004] First, to facilitate comparison between the object's contour and the design contour, the present invention provides a method for capturing an image of the object, obtaining data from each corner of the object from the image, forming a virtual contour of the object by connecting virtual lines between two adjacent corners, and adjusting the orientation of the virtual contour of the object to a predetermined orientation that matches the orientation of the design contour. Next, to avoid errors caused by multiple movements of a single camera moving above or around the object's interior angles, or to avoid the lack of resolution of the object caused by a panoramic image captured by a single camera at all of the object's interior angles, the object can be captured using multiple cameras positioned above each of the object's interior angles (typically a number of cameras corresponding to the number of interior angles of the object), thereby eliminating errors caused by the movement of a single camera or improving the resolution of images captured at all interior angles. Thus, the accuracy of comparing the object's virtual contour with the design contour is improved. [Means for solving the problem]

[0005] According to one aspect of the present disclosure, a method is disclosed for detecting an offset between the contour of a manufactured polygonal object and a design contour based on the design of the polygonal object. The polygonal object is provided with a dead region and a circuit region enclosed by the dead region, the circuit region having a plurality of interior angles and a plurality of solder joints. The method includes the steps of: capturing a contour image of the contour of the polygonal object to obtain a virtual contour; capturing (1) a plurality of interior angle images for the plurality of interior angles, and (2) a plurality of solder joint images for the plurality of solder joints closest to the plurality of interior angles; determining a plurality of feature points based on either (1) the plurality of interior angle images and (2) the plurality of solder joint images; forming a first virtual line connecting two feature points that are not adjacent to each other in the virtual contour; forming a second virtual line connecting two other feature points; determining an intersection point based on the first virtual line and the second virtual line; using the intersection point as the center of rotation; rotating the virtual contour in a specific direction along the center of rotation so that the virtual contour coincides with the design contour to generate a directionally aligned contour; and comparing the directionally aligned contour with the design contour to obtain an offset.

[0006] Another aspect of the present disclosure discloses a method for detecting an offset of the contour of a polygonal object generated from a design contour. The method includes the steps of: determining at least four feature points from the contour of the polygonal object; obtaining a virtual polygonal contour by connecting the at least four feature points in either a clockwise or counterclockwise direction; forming a first virtual line connecting two of a plurality of non-adjacent feature points in the virtual polygonal contour; forming a second virtual line connecting the other two feature points; using the intersection of the first virtual line and the second virtual line as the center of rotation; rotating the virtual polygonal contour in a specific direction along the center of rotation to generate an orientation-aligned contour corresponding to the design contour; and comparing the orientation-aligned contour with the design contour to obtain an offset.

[0007] A further aspect of the present disclosure discloses a method for adjusting the orientation of an image of a polygonal object obtained by detecting a first contour of the polygonal object. The method includes (a) determining a plurality of feature points from the first contour of the polygonal object; (b) identifying a plurality of position parameters for the plurality of feature points; (c) obtaining a virtual contour of the polygonal object by associating the plurality of position parameters; (d) determining a center of rotation based on the virtual contour; and (e) rotating the virtual contour along the center of rotation to align it with the direction to generate an orientation-aligned contour. [Brief explanation of the drawing]

[0008] The above-mentioned objectives and advantages of the present invention will become more readily apparent to those skilled in the art by examining the following detailed description and accompanying drawings.

[0009] [Figure 1] This is a schematic diagram showing the upper mark or portion of a polygonal object that is the object to be measured according to one embodiment of the present invention. [Figure 2] This is a schematic diagram showing a mark or portion on the back side of a polygonal object that is the object to be measured according to one embodiment of the present invention. [Figure 3] This is a schematic diagram showing a plurality of polygonal objects arranged on a tray according to one embodiment of the present invention. [Figure 4] This is a schematic diagram showing a device having multiple cameras according to one embodiment of the present invention. [Figures 5A-5D] This is a schematic diagram showing the interior corner of a polygonal object, which is the target of detection according to one embodiment of the present invention, as viewed from above. [Figures 6A-6D] This is a schematic diagram showing the interior corner of a polygonal object, which is the target of detection according to one embodiment of the present invention, as viewed from the back. [Figure 7] This is a schematic diagram showing the state in which the virtual contour of the polygonal object to be detected according to the present invention is being aligned with the design contour of the polygonal object. [Figure 8] This is a schematic diagram showing a comparison between the virtual contour and the designed contour of a polygonal object according to the present invention. [Figure 9] This is a schematic diagram showing a comparison of the upper and lower sides of the virtual contour of a polygonal object according to another embodiment of the present invention. [Figure 10] Figure 9 is a schematic diagram showing a front view. [Figure 11] This is a schematic diagram showing one interior angle on the upper side of the virtual contour of a polygonal object according to the present invention, and a solder joint on the back side of that interior angle. [Modes for carrying out the invention]

[0010] When reading the following detailed description, please refer to all the drawings of the present invention. All the drawings of the present invention illustrate various embodiments of the invention through illustration and will help those skilled in the art to understand how to carry out the invention. The examples herein provide sufficient embodiments to illustrate the spirit of the invention. Each embodiment is non-contradictory to another, and new embodiments can be carried out in any combination. Therefore, the present invention is not limited to the embodiments disclosed herein.

[0011] Unless otherwise defined in a specific example, the following definitions apply to terms used throughout this specification:

[0012] This invention can be used to detect polygonal objects such as substrates (e.g., IC carrier boards) and PCBs, which require precise measurement to guarantee that they are good products before shipment. The measurement data can be further analyzed, and based on its variations, points that should be considered in the preceding process, such as the degree of wear of the tools used to cut the product and the setting of the precision required for cutting, can be determined. To perform more accurate comparison and correction, this invention uses a single camera or a multi-camera system to take photographs of each of the various feature points of the object to be measured (e.g., multiple interior angle images of the interior angles), acquires a contour image of each, and then uses the data obtained from these photographs, i.e., the contour images of the multiple feature points, to further form a virtual contour of the object to be measured (i.e., a polygon result of the polygonal relationship), and then rotates the virtual contour of the object to align it with the design contour based on the design of the polygonal object and compare them. The number of multiple feature points is at least three, and feature points can be selected from the same number of interior angles among the multiple interior angles of the polygonal object. Alternatively, provided that multiple feature points are not selected from multiple interior angles, a predetermined number of solder joints can be selected as feature points from among multiple solder joints placed on an object. The solder joint images captured from each solder joint are processed appropriately by an image processing unit. The present invention further provides a multi-camera device that functions as an image sensing unit, detecting each angle of a polygonal object using multiple camera devices, and eliminating errors caused by hardware movement (i.e., in the case of a single camera) by sequentially or simultaneously capturing and detecting angle images of each angle with camera devices placed at predetermined positions. Capturing the detection target (its interior angles) at once improves shooting efficiency, and improving the stability of detection accuracy because the distance between multiple camera devices does not change when the camera devices are moved to detect the next object.

[0013] Figure 1 is a schematic diagram showing a mark or portion formed on the upper side of a polygonal object that is the object to be measured according to one embodiment of the present invention. Referring to Figure 1, for example, an IC carrier board functions as a polygonal object detected by an image processing unit connected to a camera device and acquiring an image from it. When viewed from the upper side (also called the top surface) 31F of the carrier board 31, it is normally a quadrilateral (also called a quadrilateral), and it can be seen that the upper side 31F represents the polygonal result of the aforementioned polygonal relationship. On the upper side 31F of the carrier board 31, the circuit region A1 has four angles: the first interior angle 31F1, the second interior angle 31F2, the third interior angle 31F3, and the fourth interior angle 31F4. Similarly, the first exterior angle 31F1', the second exterior angle 31F2', the third exterior angle 31F3', and the fourth exterior angle 31F4' of the upper side 31F can also be seen in the non-circuit region A2 surrounding the circuit region A1. The interior angle image of each interior angle is captured by a camera device and identified as multiple feature points on the polygonal object. The image processing unit can virtually draw a first interior angle line 3101 connecting two adjacent interior angles (i.e., interior angle parts) 31F1 and 31F4. The image processing unit can also virtually draw three virtual lines in sequence: a second interior angle line 3102, a third interior angle line 3103, and a fourth interior angle line 3104 connecting interior angles 31F2, 31F3, and 31F4. Similarly, the image processing unit can virtually draw four virtual lines: a first outer edge line 3101', a second outer edge line 3102', a third outer edge line 3103', and a fourth outer edge line 3104' passing through the four exterior angles (i.e., exterior angle parts) 31F1', 31F2', 31F3', and 31F4'. Furthermore, the image processing unit virtually draws the first upper diagonal DG13F as the first virtual line connecting the first interior angle 31F1 and the third interior angle 31F3, and the second upper diagonal DG24F as the second virtual line connecting the second interior angle 31F2 and the fourth interior angle 31F4. The two diagonals DG13F and DG24F intersect by the image processing unit, generating the intersection node 1. Also, for the convenience of direction correction, an upper triangle mark TF is usually created near the interior angle of the upper side 31F of the carrier board 31. In Figure 1, the mark TF is located in the upper left of the upper side 31F, for example, adjacent to or closest to the first interior angle 31F1, but is not limited to this position.Generally speaking, the present invention uses intersection node1 as a reference point for comparing the virtual contour and the design contour in the design of an object. Intersection node1 is superimposed on the intersection of two corresponding diagonals on the design contour. Subsequently, the upper side 31F of the virtual contour is rotated by an image processing unit to the same direction as the design contour, or to coincide with the design contour, and compared to generate a oriented contour. An oriented contour aligned with the design contour can be achieved by selecting one side of the design contour of the object as a reference edge (or reference border) and aligning the corresponding side of the virtual contour with the reference edge of the design contour. The reference edge of the oriented contour is parallel to the corresponding reference edge of the design contour and is compared by an image sensing unit.

[0014] In an alternative embodiment of the present invention, the polygonal object is further provided with a pair of alignment marks, and the step of rotating the virtual contour of the polygonal object in a specific direction is performed based on the detection of the positions of two of the pair of alignment marks.

[0015] Figure 2 is a schematic diagram showing a mark or portion on the back side of a polygonal object to be measured according to one embodiment of the present invention. Referring to Figure 2, an IC carrier board is used as the polygonal object to be detected. When viewed from the back side (also called the back surface) 31B of the carrier board 31, it is usually a quadrilateral (also called a quadrilateral), and it can be seen that the back side 31B represents the polygonal result on the back side of the polygonal relationship. On the back surface 31B of the carrier board 31, the circuit region A1 has a plurality of internal solder joints P and four external angles (also called external angles) 31B1', 31B2', 31B3', and 31B4', and either the image of the plurality of internal solder joints P or the four external angles 31B1', 31B2', 31B3', and 31B4' can be determined as multiple feature points of the back surface polygon result. The multiple internal angle solder joints P include a first internal angle solder joint 31B1, a second internal angle solder joint 31B2, a third internal angle solder joint 31B3, and a fourth internal angle solder joint 31B4. Similarly, the first external angle 31B1', second external angle 31B2', third external angle 31B3', and fourth external angle 31B4' of the back side 31B can be observed in the non-circuit region A2. By directly photographing with a camera device, the first internal angle line 3101, the second internal angle line 3102, the third internal angle line 3103, and the fourth internal angle line 3104 can be obtained. Similarly, each external angle (or external angle section) can be virtually drawn to form the first outer edge line 3101', the second outer edge line 3102', the third outer edge line 3103', and the fourth outer edge line 3104'. Furthermore, the first interior angle solder joint 31B1 and the third interior angle solder joint 31B3 are virtually connected to form the first back diagonal DG13B, and the fourth interior angle solder joint 31B2 and the fourth interior angle solder joint 31B4 are virtually connected to form the second back diagonal DG24B. The two diagonals DG13B and DG24B intersect, generating opposing intersection point node2. In addition, for the convenience of orientation / correction, a back triangle mark TB is usually created at the corner of the back side 31B of the carrier board 31. In Figure 2, the mark TB is located adjacent to or closest to the first interior angle solder joint 31B1, for example, at the upper right corner of the back side 31B of the carrier board 31 (corresponding to the upper left side of the first upper side 31F), but is not limited to this position. However, the back triangle mark TB is usually mirror-symmetric to the upper triangle mark TF.Generally, the present invention uses the opposite intersection point node2 as a reference point for comparing the virtual contour with the design contour of the object. The opposite intersection point node is superimposed on the intersection of two corresponding diagonals on the design contour, and then the back side 31B of the virtual contour is rotated to the same orientation as the design contour or to coincide with the design contour for comparison. Since the upper side 31F and the back side 31B are mirror images of each other, the back side 31B shown in Figure 2 is actually a polygon obtained by flipping the carrier board 31 horizontally, and the points and edges on the back side 31B are named sequentially to correspond to the mirror image of the upper side 31F.

[0016] Figure 3 is a schematic diagram showing a plurality of polygonal objects arranged on a tray according to one embodiment of the present invention. Referring to Figure 3, a plurality of polygonal objects to be detected, for example, carrier boards 31, are arranged on the tray 30. Here, we will explain using an example of a carrier board 31. Each carrier board 31 has a first interior angle 321, a second interior angle 322, a third interior angle 323, and a fourth interior angle 324 corresponding to the tray 30. In order to capture images of the above interior angles, the present invention includes a plurality of camera devices arranged corresponding to each interior angle of each carrier board 31, for capturing images of each interior angle. Further explanation of this will be given below with reference to Figure 4. The X and Y directions shown in Figure 3 coincide with the corresponding directions shown in Figure 4.

[0017] FIG. 4 is a schematic diagram showing an apparatus having a plurality of cameras according to an embodiment of the present invention. Referring to FIG. 4, a light detection module 10 serving as an image sensing unit includes a first camera device 11 located at a first position 21, a second camera device 12 located at a second position 22, a third camera device 13 located at a third position 23, and a fourth camera device 14 located at a fourth position 24. The X direction and the Y direction of the tray 30 shown in FIG. 3 correspond to the X direction and the Y direction of the light detection module 10 shown in FIG. 4. Therefore, the first inner angle 321, the second inner angle 322, the third inner angle 323, and the fourth inner angle 324 of the carrier board 31 shown in FIG. 3 respectively correspond to the first position 21, the second position 22, the third position 23, and the fourth position 24 of the light detection module 10. When the light detection module 10 is moved onto the carrier board 31, each camera device on the light detection module 10 faces each position, captures each inner angle 321-324 of the carrier board 31 on the tray 30, and obtains each image of the inner angle portions shown in FIGS. 1 and 2.

[0018] Figures 5A to 5D are schematic diagrams of the interior corners of a polygonal object to be detected according to one embodiment of the present invention, viewed from above. The arrangement order of Figures 5A to 5D corresponds to the order of the first position 21, second position 22, third position 23, and fourth position 24 shown in Figure 4. That is, the X and Y directions shown in Figures 5A to 5D are the same as the X and Y directions shown in Figure 4. Figures 5A to 5D are also the same as those shown in Figure 4. Figure 5A shows an image captured by the first camera device 11 when the interior corner of the upper side 31F of the carrier board 31 is located at the first position 21 shown in Figure 4. The captured image includes, as described above, the first interior angle line 3101, the second interior angle line 3102, the first outer edge line 3101', and the second outer edge line 3102'. The captured image also includes the upper triangle mark TF, the first interior angle 31F1, and the first outer angle 31F1'. Furthermore, the internal-external distance DA at position A can be determined from the distance between the first internal angle line 3101 and the first outer edge line 3101'. Similarly, the internal-external distance DE at position E can be determined from the distance between the second internal angle line 3102 and the second outer edge line 3102'. These two internal-external distances DA and DE are extremely important data for determining whether the position of the first internal angle 31F1 at the corner of the circuit region A1 (i.e., the effective region) is within the acceptable range for the design outline of the object to be detected, relative to the position of the first external angle 31F1' at the corner of the non-circuit region A2 (i.e., the non-effective region).

[0019] Figures 5A to 5D are schematic diagrams of the interior corners of a polygonal object to be detected according to one embodiment of the present invention, viewed from above. Figure 5B shows an image captured by the second camera device 12 when the corner of the upper 31F of the carrier board 31 is located at the second position 22 shown in Figure 4. As described above, the captured image includes the third interior angle line 3103, the second interior angle line 3102, the third outer edge line 3103', and the second outer edge line 3102'. The captured image also includes the second interior angle 31F2 and the second outer angle 31F2'. Furthermore, the interior-exterior distance DB at position B can be determined from the distance between the third interior angle line 3103 and the third outer edge line 3103'. Similarly, the interior-exterior distance DF at position F can be determined from the distance between the second interior angle line 3102 and the second outer edge line 3102'. These two internal and external distances, DB and DF, are crucial data for determining whether the position of the second internal angle 31F2 at the corner of the circuit region A1 (i.e., the effective region) is within the acceptable range in the design shape of the object being detected, relative to the position of the second external angle 31F2' at the corner of the non-circuit region A2 (i.e., the ineffective region).

[0020] Figures 5A to 5D are schematic diagrams of the interior corners of a polygonal object to be detected according to one embodiment of the present invention, viewed from above. Figure 5C shows an image captured by the third camera device 13 when the corner of the upper 31F of the carrier board 31 is at the third position 23 shown in Figure 4. As mentioned above, the captured image includes the third interior angle line 3103, the fourth interior angle line 3104, the third outer edge line 3103', and the fourth outer edge line 3104'. Furthermore, the captured image also includes the third interior angle 31F3 and the third outer angle 31F3'. In addition, the interior-exterior distance DC at position C can be determined from the distance between the third interior angle line 3103 and the third outer edge line 3103'. Similarly, the interior-exterior distance DG at position G can be determined from the distance between the fourth interior angle line 3104 and the fourth outer edge line 3104. These two internal and external distances, DC and DG, are crucial data for determining whether the position of the third internal angle 31F3 at the corner of the circuit region A1 (i.e., the effective region) is within the acceptable range in the design outline of the object being detected, relative to the position of the third external angle 31F3' at the corner of the non-circuit region A2 (i.e., the ineffective region).

[0021] Figs. 5A to 5D are schematic views of the inner corner portions of a polygonal object to be detected according to an embodiment of the present invention, as viewed from above. Fig. 5D shows an image captured by the fourth camera device 14 when the corner portion of the upper side 31F of the carrier board 31 is located at the fourth position 24 shown in Fig. 4. As described above, the captured image includes the first inner corner line 3101, the fourth inner corner line 3104, the first outer edge line 3101', and the fourth outer edge line 3104'. The captured image also includes the fourth inner corner 31F4 and the fourth outer corner 31F4'. Further, from the distance between the first inner corner line 3101 and the first outer edge line 3101', the inner-outer distance D-D at the position D can be obtained. Similarly, from the distance between the fourth inner corner line 3104 and the fourth outer edge line 3104', the inner-outer distance D-H at the position H can be obtained. These two inner-outer distances D-D and D-H are very important data for determining whether the position of the fourth inner corner 31F4 at the corner of the circuit region A1 (i.e., the effective region) is within the allowable range with respect to the position of the fourth outer corner 31F4' at the corner of the non-circuit region A2 (i.e., the non-effective region) in the designed outer shape of the object to be detected. Taking the inner-outer distance D-A at the position A in Fig. 5A as an example, the inner-outer distance D-A at the position A is the actual vertical distance between a position close to the first inner corner 31F1 on the first inner corner line 3101 and the position on the first outer edge line 3101'. All the other inner-outer distances at the positions B to H are also defined in the same concept.

[0022] Figures 6A to 6D are schematic diagrams of the corners of a polygonal object to be detected according to one embodiment of the present invention, viewed from the back. Since the images in Figures 6A to 6D were taken from the back of the tray 30, they show the corners when the tray 30 is rotated left and right. Therefore, the order of Figures 6A to 6D corresponds to the order in which the tray 30 is flipped left and right, as shown in Figures 5A to 5D, and multiple solder joints Ps located on the back of the object can be confirmed. Figure 6A shows an image taken by the fourth camera device 14 when the corner of the back side 31B (see Figure 2) of the carrier board 31 is located at the fourth position 24 shown in Figure 4. The captured image includes, as described above, the first interior corner line 3101, the second interior corner line 3102, the first outer edge line 3101', and the second outer edge line 3102'. The captured image also includes the back triangle mark TB, the first interior corner solder joint 31B1, and the first outer corner solder joint 31B1'. Furthermore, the distance between the first internal corner solder joint 31B1 (e.g., its center) and the first outer edge line 3101' gives the internal-to-external distance DAb at position A of the object's bottom. Similarly, the distance between the first internal corner solder joint 31B1 (e.g., its center) and the second outer edge line 3102' gives the internal-to-external distance DEb at position E of the object's bottom. These two internal-to-external distances, DAb and DEb, are crucial data for determining whether the vertical displacement or tilt between the upper surface 31F and the back side 31B (or side edge) of the corner of the carrier board 31, caused by machining errors, is within the tolerances allowed for the design outline of the object being detected. The displacement can be obtained by calculating the offset between the two internal-to-external distances DA and DAb, and between the two internal-to-external distances DE and DEb.

[0023] Figures 6A to 6D are schematic diagrams of the corners of a polygonal object to be detected according to one embodiment of the present invention, viewed from the back. Since Figures 6A to 6D are images taken from the back of the tray 30, they show the corners when the tray 30 is rotated left and right. Therefore, the order of Figures 6A to 6D corresponds to the order in which the tray 30 is flipped left and right, as shown in Figures 5A to 5D, and multiple solder joints Ps located on the back of the object can be confirmed. Figure 6B shows an image taken by the third camera device 13 when the corner of the back side 31B of the carrier board 31 (see Figure 2) is at the third position 23 shown in Figure 4. As mentioned above, the captured image includes the third inner corner line 3103, the second inner corner line 3102, the third outer edge line 3103', and the second outer edge line 3102''. The captured image further includes the second inner corner solder joint 31B2 and the second outer corner solder joint 31B2'. Furthermore, the internal-to-external distance DBb at position B on the bottom surface of the object can be determined from the distance between the second internal angle solder joint 31B2 (for example, its center) and the third outer edge line 3103'. Similarly, the internal-to-external distance DFb at position F on the bottom surface of the object can be determined from the distance between the second internal angle solder joint 31B2 (for example, its center) and the second outer edge line 3102'. These two internal-to-external distances DBb and DFb are very important data for determining whether the vertical displacement or tilt between the upper side 31F and the back side 31B (or side edge) of the corner of the carrier board 31 caused by machining errors is within the acceptable range for the design outline of the object to be detected. These displacements can be determined by calculating the offset between the two internal-to-external distances DB and DBb, and the offset between the two internal-to-external distances DF and DFb.

[0024] Figures 6A to 6D are schematic diagrams of the corners of a polygonal object to be detected according to one embodiment of the present invention, viewed from the back. Since the images in Figures 6A to 6D were taken from the back of the tray 30, Figures 6A to 6D show the corners when the tray 30 is rotated left and right. Therefore, the order of Figures 6A to 6D corresponds to the order in which the tray 30 is flipped left and right as shown in Figures 5A to 5D, and multiple solder joints Ps located on the back of the object can be confirmed. Figure 6C shows an image taken by the second camera device 12 when the corner of the back side 31B of the carrier board 31 (see Figure 2) is in the second position 22 shown in Figure 4. As mentioned above, the captured image includes the third inner corner line 3103, the fourth inner corner line 3104, the third outer edge line 3103', and the fourth outer edge line 3104'. The captured image further includes the third inner corner solder joint 31B3 and the third outer corner solder joint 31B3'. Furthermore, the internal-to-external distance DCb at position C on the bottom surface of the object can be determined from the distance between the third internal angle solder joint 31B3 (for example, its center) and the third outer edge line 3103'. Similarly, the internal-to-external distance DGb at position F on the bottom surface of the object can be determined from the distance between the third internal angle solder joint 31B3 (for example, its center) and the fourth outer edge line 3104'. These two internal-to-external distances DCb and DGb are very important data for determining whether the vertical displacement or tilt between the upper side 31F and the back side 31B (or side edge) of the corner of the carrier board 31 caused by machining errors is within the permissible range in the design outline of the object to be detected. These displacements can be determined by calculating the offset between the two internal-to-external distances DC and DCb, and the offset between the two internal-to-external distances DG and DGb.

[0025] Figures 6A to 6D are schematic diagrams of the corners of a polygonal object to be detected according to one embodiment of the present invention, viewed from the back. Since the images in Figures 6A to 6D were taken from the back of the tray 30, Figures 6A to 6D show the corners when the tray 30 is rotated left and right, respectively. Therefore, the order of Figures 6A to 6D corresponds to the order in which the tray 30 is flipped left and right, as shown in Figures 5A to 5D, and multiple solder joints Ps located on the back of the object can be confirmed. Figure 6D shows an image taken by the first camera device 11 when the corner of the back side 31B (see Figure 2) of the carrier board 31 is at the first position 21 shown in Figure 4. As mentioned above, the captured image includes the first inner corner line 3101, the fourth inner corner line 3104, the first outer edge line 3101', and the fourth outer edge line 3104'. The captured image further includes the fourth internal angle solder joint 31B4 and the fourth external angle solder joint 31B4'. Furthermore, the internal-to-external distance DDb at position D on the object's bottom surface can be determined from the distance between the fourth internal angle solder joint 31B4 (for example, its center) and the first outer edge line 3101'. Similarly, the internal-to-external distance DHb at position H on the object's bottom surface can be determined from the distance between the fourth internal angle solder joint 31B4 (for example, its center) and the fourth outer edge line 3104'. These two internal-to-external distances, DDb and DHb, are very important data for determining whether the vertical displacement or tilt between the upper side 31F and the back side 31B (or side edge) of the corner of the carrier board 31 caused by machining errors is within the permissible range in the design outline of the object to be detected. The displacement can be obtained by calculating the offset between the two internal-to-external distances DD and DDb, and the offset between the two internal-to-external distances DH and DHb.

[0026] Figure 7 is a schematic diagram showing the state in which the virtual contour of the polygonal object to be detected according to the present invention is being aligned with the design contour of the polygonal object. In order to clearly show how the acquired polygon result Re of the virtual contour is aligned with the design contour according to the present invention, Figure 7 shows that the upper side 31F of the acquired polygon result Re has a relatively exaggerated rotation angle with respect to the design contour. Those skilled in the art will understand that such an exaggerated misalignment does not actually occur. Based on the description of the images taken as Figures 1 and 2, in Figures 5A to 5D and 6A to 6B, the first interior angle line 3101, the second interior angle line 3102, the third interior angle line 3103 and the fourth interior angle line 3104, as well as the first outer edge line 3101', the second outer edge line 3102', the third outer edge line 3103' and the fourth outer edge line 3104' and their positions are clearly identified, and all the necessary distance values ​​have been calculated. A comparison platform 100 equipped with an image processing unit is provided to detect the discrepancy between the virtual contour of an object and its design contour. The virtual field of view of the image processing unit on the comparison platform 100 can be made to match, for example, the effective field of view of each camera device (such as a camera device equipped with a CCD sensor or a CMOS sensor). As shown in Figure 7, the virtual field of view of the comparison platform 100 includes the upper edge 100U, the lower edge 100D, the left edge 100L, and the right edge 100R. When an upper triangle mark TF is placed on the carrier board 31 to be detected, and an upper alignment mark LU is placed within the field of view of the screen, the direction that coincides with the upper alignment mark LU is set as the predetermined direction. Referring to Figure 7, here, as explained in Figure 1, the intersection point node1 is determined as a reference point for comparing the virtual contour with the design contour of the object. It can be seen that the polygon result Re of the virtual contour is shifted counterclockwise from the screen field of view of the comparison platform 100. By rotating the polygon result Re of the virtual contour clockwise around intersection node1 as the rotation center, and aligning the upper triangle mark TF of the virtual contour with the upper alignment mark LU, the polygon result Re of the virtual contour with the upper triangle mark TF will be perpendicular to the screen field of view of the comparison platform 100 with the upper alignment mark LU. At this point, the preliminary alignment operation before comparison with the design contour is complete.Similarly, as explained in Figure 2, the opposite intersection node 2 is determined as a reference point for comparing the virtual contour with the design contour of the object. The back side 31B of the virtual contour can be rotated around the opposite intersection node 2 as the rotation center until the back triangle mark TB within the virtual contour faces the same direction as the back alignment mark RU. This means that the polygon result Re of the virtual contour with the back triangle mark TB is aligned perpendicularly to the field of view of the comparison platform 100 screen with the back alignment mark RU. The obtained data is processed by the comparison platform 100 using software for image analysis and comparison. For ease of explanation, in this invention the comparison platform 100 is visualized and presented as a platform and can be considered as a planar coordinate system. The direction and orientation of the comparison platform 100 are exactly the same as those of the design contour, and the polygon result Re of the virtual contour is aligned with the design contour for comparison. For example, after alignment, the horizontal direction of the virtual contour acquired by the multiple camera devices shown in Figure 4 is parallel to the upper edge 100U and lower edge 100D of the comparison platform 100.

[0027] See Figure 7 for further reference. If neither the upper triangle mark TF nor the lower triangle mark TB exists on the carrier board 31, one of the first interior angle line 3101, the second interior angle line 3102, the third interior angle line 3103, and the fourth interior angle line 3104 can be selected as a marking line (or reference line), and one of the upper edge 100U, lower edge 100D, left edge 100L, and right edge 100R of the comparison platform 100 can be set as a reference line to align with the inner edge selected as the marking line. For example, by setting the intersection node 1 as the rotation center, the first interior angle line 3101 as the marking line, and the upper edge 100U as the reference line, the polygon result Re rotates with respect to the intersection node 1 until the first interior angle line 3101 is adjacent to and parallel to the upper edge 100U. Similarly, the polygon result of the back side 31B shown in Figure 2 (i.e., the polygon result obtained and calculated from the acquired data of the entire back side 31B shown in Figure 2) can also be rotated with the opposite intersection node2 as the rotation center until the selected indicator line is adjacent to and parallel to the selected reference line.

[0028] Figure 8 is a schematic diagram showing a comparison between the virtual contour and the design contour of a polygonal object according to the present invention. Figure 8 should be referred to as an example of a view from the direction facing the first interior corner line 3101 on the upper side 31F of the carrier board 31. The numbers on the design contour 31CAD of the carrier board are the same as those shown in the previous figure, and redundant explanations are omitted. The measured first outer edge line 3101'x represents the polygon result obtained by calculation after photographing the carrier board 31 with a camera device. The in-out-out distance DA at position A (i.e., the distance from the measured first outer edge line 3101'x to the first interior angle line 3101 adjacent to the first interior angle 31F1) is 315 units, and the in-out-out distance DD at position D is 303 units (the distance from the measured first outer edge line 3101'x to the first interior angle line 3101 adjacent to the fourth interior angle 31F4). By comparing the measured first outer edge line 3101'x with the first outer edge line 3101' of the carrier board design contour 31CAD, the actual offset between the measured first outer edge line 3101'x and the first outer edge line 3101' of the carrier board design contour 31CAD can be obtained, and based on a predetermined tolerance, it can be determined whether the carrier board 31 has been processed well, should be reprocessed / modified, or should be discarded. The above uses the in-out-out distance between the first interior angle line 3101 and the measured first outer edge line 3101'x as an example, and other similar examples are also achievable.

[0029] Figure 9 is a schematic diagram showing a comparison of the upper and lower sides of the virtual contour of a polygonal object according to another embodiment of the present invention. Figure 10 is a schematic diagram showing the front view of Figure 9. The reference numerals in Figure 9 follow those of the previously mentioned figures, so a detailed explanation is omitted. Please refer to Figures 9 and 10 together. Figure 9 shows the polygonal result obtained by combining the upper side 31F (see Figure 1) and the lower side 31B (see Figure 2) of the carrier board 31. The intersection node 1 (see Figure 1) is superimposed with the opposite intersection node 2 (see Figure 2), and the corresponding outer edge lines of the upper side 31F and the lower side 31B are compared. Before comparison, the polygonal results shown in Figures 5A to 5D are rotated left and right so that the directions of the second interior angle line 3102 and the second outer edge line 3102' shown in Figures 5A and 5B coincide with the directions of the second interior angle line 3102 and the second outer edge line 3102' shown in Figures 6A and 6B. Figure 10 shows the second outer edge line 3102'B and the upper second outer edge line 3102'F of the carrier board 31, viewed from a direction parallel to the upper side 31F and the back side 31B. Figure 10 can also be considered a side view of Figure 9. Since Figure 9 shows a view from the direction facing the back side 31B of the carrier board 31, Figure 9 can also be considered a rear view of the carrier board 31. Comparing them, we can obtain the second outer edge line 3102'B of the back side 31B and the upper second outer edge line 3102'F of the upper side 31F. It can be seen that the upper second outer edge line 3102'F protrudes more. Alternatively, due to the difference in viewpoint, it can be seen that the second outer edge line 3102'B on the back side is recessed. Furthermore, referring to Figure 10, it can be seen that the second side edge 31S2 of the carrier board 31 is inclined at this point. Therefore, from Figures 9 and 10, it can be seen that the second side edge 31S2 of the carrier board 31 undergoes some deviation during machining. This means that the second side edge 31S2 is inclined and not perpendicular to either the upper side 31F or the lower side 31B of the carrier board 31. "Processing" here refers to cutting the carrier board 31 from the motherboard and / or trimming any burrs present on the side edge of the carrier board 31 after cutting.Furthermore, referring again to Figure 10, by introducing the thickness 31t shown in the design outline of the carrier board 31 to the second side edge 31S2, the inclination or slope of the second side edge 31S2 can be obtained, and by determining whether the inclination or slope exceeds the allowable value, it is possible to determine whether the carrier board 31 needs to be reprocessed, modified, or discarded.

[0030] More specifically, the polygonal object has a first surface, the first surface includes a first circuit region having a plurality of interior angles, and the plurality of feature points are determined by the steps of capturing a plurality of interior angle images for the plurality of interior angles and determining the plurality of feature points from the plurality of interior angle images. The polygonal object further has a second surface opposite to the first surface, the polygonal object has a second contour that is a mirror image of the first contour, the polygonal object includes a second circuit region opposite to the first circuit region, the second circuit region has a plurality of opposing interior angles opposite to the plurality of interior angles, the second circuit region has a plurality of solder joints, the method includes the steps of capturing a plurality of solder joint images for four solder joints included in the plurality of solder joints, each closest to the plurality of opposing interior angles, determining four opposing feature points from the plurality of solder joint images, and the four opposing feature points in a clockwise direction The method further includes the steps of forming opposing virtual contours by connecting them in either a counterclockwise or counterclockwise direction; forming a third virtual line connecting two opposing feature points that are not adjacent to each other along the opposing virtual contours; forming a fourth virtual line connecting two other opposing feature points; determining opposing intersection points based on the third and fourth virtual lines; setting the opposing intersection points as opposing rotation centers; rotating the opposing virtual contours along the opposing rotation centers to generate opposing direction-aligned contours; and comparing the direction-aligned contours with the opposing direction-aligned contours to obtain top-back offsets.

[0031] The plurality of feature points on the first surface are four feature points, the four feature points and the four opposing feature points are captured by a camera device for comparing the orientation-aligned contour and the opposing orientation-aligned contour, the polygonal object has thickness, and the method further includes the step of calculating the slope of the side having the thickness of the polygonal object based on the offset of the top back.

[0032] Figure 11 is a schematic diagram showing one interior angle on the upper side of the virtual contour of a polygonal object according to the present invention, and a solder joint on the back side of that interior angle. By comparing the upper side 31F and the back side 31B of the carrier board 31 shown in Figures 9 and 10, the relative position between the upper first interior angle 31F1 and the first interior angle solder joint 31B1 can be measured, and it is possible to determine whether the relative position to the interior angle compared to the designed outer shape, i.e., the distance, satisfies the design value.

[0033] From the attached drawings and description, it can be seen that the present invention greatly contributes to the smooth inspection of the size, shape, and contour of a carrier board by rotating the polygonal result of the virtual contour around a rotation center to match the design contour of the carrier board to be detected. After rotating the virtual contour in a direction that matches the design contour, the deviation of the size of the manufactured carrier board can be determined by comparing the obtained image with the design contour. Furthermore, the rotation center does not necessarily have to be based on the intersection of two lines connecting each feature point. The centroid, center of mass, and centroid obtained from the polygonal result can also be used as the rotation center. This improves the flexibility of the present invention, as the method can be used for inspecting targets with irregular edge shapes (e.g., arc shapes). The user no longer needs to precisely fix the carrier board on the tray; they only need to confirm that the carrier board is fixed on the tray and does not fall off the tray as usual. This greatly saves operating time and greatly contributes to the calibration and comparison of products such as carrier boards and circuit boards with polygonal planar structures.

[0034] Embodiment 1: The present invention provides a method for detecting an offset between the contour of a manufactured polygonal object and a design contour based on the design of the polygonal object. The polygonal object is provided with an invalid region and a circuit region surrounded by the invalid region, the circuit region having a plurality of interior angles and a plurality of solder joints. The method includes the steps of: capturing a contour image of the contour of the polygonal object to obtain a virtual contour; capturing (1) a plurality of interior angle images for the plurality of interior angles, and (2) a plurality of solder joint images for the plurality of solder joints closest to the plurality of interior angles; determining a plurality of feature points based on either (1) the plurality of interior angle images and (2) the plurality of solder joint images; forming a first virtual line connecting two feature points that are not adjacent to each other in the virtual contour; forming a second virtual line connecting two other feature points; determining an intersection point based on the first virtual line and the second virtual line; using the intersection point as the center of rotation; rotating the virtual contour in a specific direction along the center of rotation so that the virtual contour coincides with the design contour to generate a directionally aligned contour; and comparing the directionally aligned contour with the design contour to obtain an offset.

[0035] Embodiment 2: The present invention provides a method according to Embodiment 1, wherein the polygonal object has a pair of alignment marks, and the step of rotating the virtual contour of the polygonal object in the specific direction is based on detecting the positions of two of the pair of alignment marks.

[0036] Embodiment 3: The present invention provides a method according to Embodiment 1 or 2, wherein, after the step of rotating the virtual contour in the particular direction, the method further includes the step of selecting the sides of the polygonal object to coincide with corresponding sides in the design in order to compare the orientation-aligned contour with the design contour.

[0037] Embodiment 4: The present invention provides a method according to any one of Embodiments 1 to 3, wherein the steps of capturing the contour image and capturing the interior angle image are performed by a camera device for sensing the contour image to be compared with the design contour, and the base edge for defining the specific direction is parallel to one side of the design contour.

[0038] Embodiment 5: The present invention provides a method for detecting an offset of the contour of a polygonal object generated from a design contour. The method includes the steps of: determining at least four feature points from the contour of the polygonal object; obtaining a virtual polygonal contour by connecting the at least four feature points in either a clockwise or counterclockwise direction; forming a first virtual line connecting two of a plurality of feature points in the virtual polygonal contour that are not adjacent to each other; forming a second virtual line connecting the other two feature points; using the intersection of the first virtual line and the second virtual line as the center of rotation; rotating the virtual polygonal contour in a specific direction along the center of rotation to generate an orientation-aligned contour corresponding to the design contour; and comparing the orientation-aligned contour with the design contour to obtain an offset.

[0039] Embodiment 6: The present invention provides the method according to Embodiment 5, wherein at least four feature points are multiple feature points, and the first virtual line and the second virtual line are the two virtual diagonals of the virtual contour, respectively.

[0040] Embodiment 7: The present invention provides a method according to Embodiment 5 or 6, wherein the polygonal object has a pair of alignment marks, and the step of rotating the virtual contour of the polygonal object in the specific direction is based on detecting the positions of two of the pair of alignment marks.

[0041] Embodiment 8: The present invention provides a method according to any one of Embodiments 5 to 7, wherein, after the step of rotating the virtual contour in the particular direction, the method further includes the step of selecting the sides of the polygonal object to coincide with corresponding sides in the design in order to compare the orientation-aligned contour with the design contour.

[0042] Embodiment 9: The present invention provides a method according to any one of Embodiments 5 to 8, wherein at least four feature points are captured by a camera device for comparing the orientation-aligned contour with a design contour, and a base edge for defining a specific direction of the orientation-aligned contour is parallel to one edge of the design contour.

[0043] Embodiment 10: The present invention provides a method for adjusting the orientation of an image of a polygonal object obtained by detecting a first contour of the polygonal object. The method includes (a) determining a plurality of feature points from the first contour of the polygonal object; (b) identifying a plurality of position parameters for the plurality of feature points; (c) obtaining a virtual contour of the polygonal object by associating the plurality of position parameters; (d) determining a rotation center based on the virtual contour; and (e) rotating the virtual contour along the rotation center to align it with the direction to generate an orientation-aligned contour.

[0044] Embodiment 11: The present invention provides a method according to Embodiment 10, the method further comprising the step of obtaining an offset by comparing the orientation-aligned contour with the design contour.

[0045] Embodiment 12: The present invention provides the method according to Embodiment 10 or 11, wherein the offset includes at least one of a displacement offset and an angular offset.

[0046] Embodiment 13: The present invention provides a method according to any one of Embodiments 10 to 12, wherein the step (d) of determining the rotation center based on the virtual contour includes (d1) forming a first virtual line along the virtual contour connecting two feature points of the plurality of feature points that are not adjacent to each other, (d2) forming a second virtual line connecting two other feature points of the plurality of feature points, and (d3) determining the intersection of the first virtual line and the second virtual line as the rotation center.

[0047] Embodiment 14: The present invention provides a method according to any one of Embodiments 10 to 13, wherein the plurality of feature points are four feature points, and the first virtual line and the second virtual line are the two virtual diagonals of the virtual contour, respectively.

[0048] Embodiment 15: The present invention provides a method according to any one of Embodiments 10 to 14, wherein the polygonal object has a first surface, the first surface includes a first circuit region having a plurality of interior angles, and the plurality of feature points are determined by the steps of capturing a plurality of interior angle images for the plurality of interior angles and determining the plurality of feature points from the plurality of interior angle images.

[0049] Embodiment 16: The present invention provides a method according to any one of Embodiments 10 to 15, wherein the polygonal object further has a second surface opposite to the first surface, the polygonal object has a second contour that is a mirror image of the first contour, the polygonal object includes a second circuit region opposite to the first circuit region, the second circuit region has a plurality of opposing interior angles opposite to the plurality of interior angles, the second circuit region has a plurality of solder joints, the method comprises the steps of capturing a plurality of solder joint images for four solder joints included in the plurality of solder joints and closest to each of the plurality of opposing interior angles, and determining four opposing feature points from the plurality of solder joint images. The method further includes the steps of: forming a virtual opposing contour by connecting the four opposing feature points in either a clockwise or counterclockwise direction; forming a third virtual line connecting two non-adjacent opposing feature points along the virtual opposing contour; forming a fourth virtual line connecting two other opposing feature points; determining an opposing intersection point based on the third and fourth virtual lines; setting the opposing intersection point as the opposing rotation center; rotating the virtual opposing contour along the opposing rotation center to generate an opposing direction-aligned contour; and comparing the direction-aligned contour with the opposing direction-aligned contour to obtain a top-back offset.

[0050] Embodiment 17: The present invention provides a method according to any one of Embodiments 10 to 16, wherein a contour image, an interior angle image and a plurality of solder joint images are captured by a camera device equipped with an image processing unit for comparing the contour image with the design contour, the polygonal object having thickness, and the method further includes the steps of introducing thickness as a parameter to the image processing unit and calculating the slope of the sides having thickness of the polygonal object based on the top back offset.

[0051] Embodiment 18: The present invention provides the method according to Embodiments 10 to 17, wherein the contour image and the interior angle image are captured by a camera device equipped with an image processing unit for comparing the contour image with the design contour, and the base edge for defining the direction is parallel to one side of the design contour compared by the image sensing unit.

[0052] Embodiment 19: The present invention provides a method according to any one of Embodiments 10 to 18, wherein the center of rotation is determined as one of the center of gravity, center of mass, and center of shape based on the virtual contour of the polygonal object.

[0053] Embodiment 20: The present invention provides a method according to any of Embodiments 10 to 19, wherein the polygonal object has a pair of alignment marks, and the step of rotating the virtual contour of the polygonal object is based on detecting the position of the pair of alignment marks.

[0054] Furthermore, based on the description in Figures 1 and 2, the present invention provides a method for detecting an offset of the contour of a polygonal object generated from a design contour. The method includes the steps of: determining at least four feature points from the contour of the polygonal object; obtaining a virtual polygonal contour by connecting the at least four feature points in either a clockwise or counterclockwise direction; forming a first virtual line connecting two of a plurality of feature points that are not adjacent to each other in the virtual polygonal contour; forming a second virtual line connecting the other two feature points; using the intersection of the first virtual line and the second virtual line as the center of rotation; rotating the virtual polygonal contour in a specific direction along the center of rotation to generate a directional contour corresponding to the design contour; and comparing the directional contour with the design contour to obtain an offset.

[0055] Furthermore, based on the description in Figures 1 and 2, the present invention provides a method for adjusting the orientation of an image of a polygonal object obtained by detecting the first contour of the polygonal object. The method includes (a) determining a plurality of feature points from the first contour of the polygonal object; (b) identifying a plurality of position parameters for the plurality of feature points; (c) obtaining a virtual contour of the polygonal object by associating the plurality of position parameters; (d) determining a rotation center based on the virtual contour; and (e) rotating the virtual contour along the rotation center to align it with the direction to generate an orientation-aligned contour.

[0056] While the present invention has been described based on embodiments considered most practical and preferred at present, it should be understood that the invention is not limited to the disclosed embodiments. Rather, the invention is intended to encompass a variety of modifications and similar configurations included in the spirit and scope of the appended claims, which should be interpreted most broadly to include all such modifications and similar configurations. [Explanation of symbols]

[0057] 10: Light detection module 11: First camera device 12: Second camera device 13: Third camera device 14: Fourth camera device 21: First position 22:Second position 23: Third position 24:Fourth position 30: Tray 31: Carrier board 31B: Reverse side 31B1': First outside corner 31B2': Second outside corner 31B3': Third outside angle 31B4': Fourth exterior angle 31B1: First internal corner solder joint 31B2: Second interior corner solder joint 31B3: Third interior angle solder joint 31B4: Fourth interior angle solder joint 31B1': First outer corner solder joint 31B2': Second outer corner solder joint 31B3': Third outer corner solder joint 31B4': Fourth outer corner solder joint 31CAD: Design Contour 31F: Upper side 31F1, 321: First interior corner 31F2, 322: Second interior angle 31F3, 323: Third interior angle 31F4, 324: Fourth interior angle 31F1': First outside corner 31F2': Second outside corner 31F3': Third outside angle 31F4': Fourth outside angle 31S2:Second side edge 31t: Thickness 100: Comparison Platform 100U: Upper edge 100D: Lower edge 100L: Left edge 100R: Right edge 3101: First interior angle line 3102: Second interior angle line 3103: Third interior angle line 3104: Fourth interior angle 3101': First outer edge 3102': Second outer edge 3103': Third outer margin 3104': Fourth outer margin 3101'x: Measured first outer edge line 3102'B: Second outer edge 3102'F: Upper second outer edge line A, B, C, D, E, F, G, H: Position A1: Circuit area A2: Non-circuit area DA, DB, DC, DD, DE, DF, DG, DH, DAb, DBb, DCb, DDb, DEb, DFb, DHb: Inner / outer distance DG13F: First upper diagonal DG13B: First back diagonal DG24F: Second upper diagonal DG24B: Second back diagonal LU: Upper alignment mark P: Internal solder joint Ps: Solder joint Re: polygon result RU: Rear alignment mark TB: Reverse triangle mark TF: Upward triangle mark node1, node2: intersection n

Claims

1. A method for detecting the offset between the contour of a manufactured polygonal object and the design contour of the polygonal object based on its design, The aforementioned polygonal object is provided with an invalid region and a circuit region enclosed by the invalid region. The circuit region has multiple interior angles and multiple solder joints, The aforementioned method, A step of capturing a contour image of the contour of the polygonal object and obtaining a virtual contour, (1) capturing multiple interior angle images for the multiple interior angles, and (2) capturing one of the multiple solder joint images for the multiple solder joints closest to the multiple interior angles. (1) A step of determining a plurality of feature points based on either the plurality of interior angle images or (2) the plurality of solder joint images, The steps include forming a first virtual line connecting two feature points that are not adjacent to each other in the virtual contour, The step of forming a second virtual line connecting two other feature points, A step of determining the intersection point based on the first virtual line and the second virtual line, A step with the aforementioned intersection as the center of rotation, The steps include: rotating the virtual contour along the rotation center in a specific direction so that the virtual contour coincides with the design contour to generate a directionally aligned contour; A method characterized by comprising the step of obtaining an offset by comparing the oriented contour with the design contour.

2. The aforementioned polygonal object has a pair of alignment marks placed on it. The step of rotating the virtual contour of the polygonal object in the specific direction is based on the detection of the positions of two marks of the pair of alignment marks, After the step of rotating the virtual contour in the specific direction, the method further includes the step of selecting the sides of the polygonal object to coincide with the corresponding sides in the design in order to compare the orientation-aligned contour with the design contour, The method according to claim 1, characterized in that the step of capturing the contour image and the step of capturing the interior angle image are each performed by a camera device for sensing the contour image to be compared with the design contour, and the base edge for defining the specific direction is parallel to one side of the design contour.

3. A method for detecting the offset of the contour of a polygonal object generated from a design contour, The steps include determining at least four feature points from the contour of the polygonal object, The steps include obtaining a virtual polygonal contour by connecting the at least four feature points in either a clockwise or counterclockwise direction, The steps include forming a first virtual line connecting two of the multiple feature points that are not adjacent to each other in the virtual polygonal contour, The step of forming a second virtual line connecting the other two feature points, A step in which the intersection of the first virtual line and the second virtual line is used as the center of rotation, The steps include: rotating the virtual polygonal contour along the rotation center in a specific direction to generate a directionally aligned contour corresponding to the design contour; A method characterized by comprising the step of obtaining an offset by comparing the oriented contour with the design contour.

4. A method for adjusting the orientation of an image of a polygonal object, The polygonal object is obtained by detecting the first contour of the polygonal object. The aforementioned method, (a) A step of determining a plurality of feature points from the first contour of the polygonal object, (b) The step of identifying a plurality of positional parameters for the plurality of feature points, (c) The step of obtaining the virtual contour of the polygonal object by relating the plurality of position parameters, (d) The step of determining the center of rotation based on the virtual contour, (e) A method comprising the step of rotating the virtual contour along the rotation center in the direction to generate a direction-aligned contour.

5. The method according to 4, further comprising the step of obtaining an offset by comparing the orientation-aligned contour with the design contour, wherein the offset includes at least one of a displacement offset and an angular offset.

6. Step (d) of determining the rotation center based on the virtual contour is: (d1) A step of forming a first virtual line along the virtual contour, connecting two feature points from the plurality of feature points that are not adjacent to each other, (d2) The step of forming a second virtual line connecting two other feature points among the plurality of feature points, The method according to 4, characterized by comprising the step of (d3) determining the intersection point of the first virtual line and the second virtual line as the center of rotation.

7. The method according to 6, characterized in that the plurality of feature points are four feature points, and the first virtual line and the second virtual line are each two virtual diagonals of the virtual contour.

8. The aforementioned polygonal object has a first surface, The first surface includes a first circuit region having a plurality of interior angles, The aforementioned multiple feature points are, The method according to 6, characterized in that it is determined by the steps of capturing a plurality of interior angle images for the plurality of interior angles and determining the plurality of feature points from the plurality of interior angle images.

9. The polygonal object further has a second surface opposite to the first surface, The polygonal object has a second contour that is a mirror image of the first contour, and the polygonal object includes a second circuit region opposite to the first circuit region. The second circuit region has a plurality of opposing interior angles opposite to the plurality of interior angles, The second circuit region has the plurality of solder joints, The aforementioned method, The steps include capturing multiple solder joint images for four solder joints that are included in the plurality of solder joints and are closest to the plurality of opposing interior angles, The steps include determining four opposing feature points from the aforementioned multiple solder joint images, The steps include forming opposing virtual contours by connecting the four opposing feature points in either a clockwise or counterclockwise direction, The steps include forming a third virtual line connecting two opposing feature points that are not adjacent to each other along the opposing virtual contour, The step of forming a fourth virtual line connecting two other opposing feature points, The steps include determining opposing intersections based on the third virtual line and the fourth virtual line, The step of making the aforementioned opposing intersections the opposing centers of rotation, The steps include: rotating the opposing virtual contours along the opposing rotation center to generate opposing direction-aligned contours; The method of 8, further comprising the step of comparing the oriented alignment contour with the opposing oriented alignment contour to obtain the top back offset.

10. The aforementioned multiple feature points are four feature points, The four feature points and the four opposing feature points are captured by a camera device for comparing the orientation-aligned contour and the opposing orientation-aligned contour. The aforementioned polygonal object has thickness, The method according to 9, further comprising the step of calculating the slope of the side having the thickness of the polygonal object based on the offset of the top back.

Images