Quartz wafer defect detection system and method for identifying gray abrupt change

Abstract

The quartz wafer defect detection system comprises a camera, the camera is fixed to the upper portion of a camera fixing device, a light source objective table is arranged below the camera, a quartz wafer to be detected is placed on the light source objective table, and the camera is connected with an image processor; based on the detection system, the invention further provides a quartz wafer defect detection method for identifying gray level mutation, which comprises the following steps of: acquiring a quartz wafer image under a transmission light imaging condition by utilizing the transparency characteristic of the quartz wafer, extracting a quartz wafer area contour by utilizing the gray level characteristic of a quartz wafer boundary area, and detecting the defect of the quartz wafer. And whether the wafer has defects is judged according to the geometric features of the contour of the quartz wafer area and the gray features of the quartz wafer area, so that the consistency of the detection results of the quartz wafer defects is improved, and the omission ratio and the false detection rate of the quartz wafer defects are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of quartz wafer detection, and in particular to a quartz wafer defect detection system and method for identifying grayscale mutations. Background Art

[0002] Quartz crystal resonator is an electronic component with good frequency stability and is widely used in industrial fields such as communications and aerospace. The quality of the quartz wafers used in quartz crystal resonators directly affects the electrical characteristics and service life of the quartz crystal resonators. Therefore, quartz wafers need to be strictly sorted before they can be used in the production of quartz crystal resonators. In the prior art, manual visual inspection is generally used to detect defects in quartz wafers. Since quartz wafers are transparent and small in size, and the quartz wafers that need to be inspected in actual production have various shapes, this manual visual inspection method has low detection efficiency and poor reliability of detection results, and cannot meet the current requirements of quartz crystal resonator production. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a quartz wafer defect detection system and method for identifying grayscale mutations. Based on machine vision detection technology, it solves the problems of low efficiency and poor reliability of detection results caused by reliance on manual detection in the quartz wafer detection process in the prior art. To achieve the above objectives, the first aspect of the present invention proposes a quartz wafer defect detection system for identifying grayscale mutations, the detection system comprising a camera, the camera is fixed on the upper part of a camera fixing device, a light source stage is arranged below the camera, the quartz wafer to be inspected is placed on the light source stage, and the camera is connected to an image processor. Preferably, the camera is a high-resolution industrial camera. Preferably, the image processor is an industrial computer or an embedded board. Based on the above detection system, the second aspect of the present invention proposes a quartz wafer defect detection method for identifying grayscale mutations, comprising the following steps: Step 1: Obtain the quartz wafer image captured by the camera; Step 2: converting the quartz wafer image into a grayscale image, and obtaining the quartz wafer area contour by identifying the grayscale level difference between the quartz wafer boundary area and the adjacent area under the transmitted light imaging condition; Step 3: judging whether the quartz wafer to be inspected has defects according to the geometric features of the regional contour of the quartz wafer and the typical values ​​of the corresponding geometric features of the quartz wafer to be inspected in a non-defective state; The typical values ​​of the corresponding geometric features of the quartz wafer to be detected in a non-defective state are determined based on a non-defective wafer of the same model as the quartz wafer to be detected, and are pre-set by the system operator during the initialization process. Step 4: Processing the grayscale image of the quartz wafer using an adaptive histogram equalization technique to improve the contrast between the defective region of the quartz wafer and the adjacent region in the grayscale image; Step 5: Binarize the adaptive histogram equalization result map obtained in step 4 to separate the suspected defect area of ​​the quartz wafer from the remaining areas. In the binarized result map of the adaptive histogram equalization result map, the pixel color of the suspected defect area of ​​the quartz wafer is white, and the pixel color of the remaining area is black; Step 6: Using the quartz wafer area contour to screen the suspected defect areas in the binarization result image, the suspected defect areas within the quartz wafer area contour are determined as the real defect areas of the quartz wafer. Preferably, in step 3, the type of geometric features of the quartz wafer area contour and the typical value of the corresponding geometric features of the quartz wafer to be detected in a non-defective state are pre-set by the system operator during the initialization process. The geometric features of the quartz wafer area contour are by default the area of ​​the quartz wafer area contour, and the corresponding process of judging whether the quartz wafer has defects includes calculating the area of ​​the area surrounded by the quartz wafer area contour, and judging whether the quartz wafer has defects in combination with the typical value of the area of ​​the area surrounded by the quartz wafer area contour when the quartz wafer to be detected is in a non-defective state. The judgment formula is: |A-A0|<N, wherein A is the area of ​​the area surrounded by the quartz wafer area contour, A0 is the typical value of the area of ​​the area surrounded by the quartz wafer area contour corresponding to the quartz wafer to be detected when the quartz wafer is in a non-defective state, and N is a positive constant; If the area of ​​the region enclosed by the quartz wafer region outline does not meet the above conditions, it is determined that the quartz wafer to be inspected has a defect. Preferably, in step 3, when the quartz wafer to be inspected is a non-basic geometric figure, a convex hull algorithm is used to fit the regional contour of the quartz wafer to the contour of a basic geometric figure, and it is determined whether the quartz wafer has defects based on the geometric features of the fitted basic geometric figure contour. Preferably, when the detected quartz wafer is a quasi-rectangular wafer, its area contour is fitted into a rectangle, that is, the shape of the minimum area circumscribed contour of the wafer is a rectangle, and whether the quasi-rectangular quartz wafer has defects is determined according to the following formula: Wherein, D1 and D2 are respectively the lengths of the two diagonals of the minimum area circumscribed contour, D0 is the typical value of the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be detected without defects, n is a positive constant, if the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be detected does not meet the above conditions, it is determined that the quartz wafer to be detected has defects; if the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be detected meets the above conditions, it is determined that the quartz wafer to be detected has no defects. Beneficial effects: Compared with the prior art, the present invention can achieve the following technical effects: 1. The detection system of the present invention collects quartz wafer images under transmitted light imaging conditions by setting structures such as a light source stage and utilizing the transparent characteristics of the quartz wafer. The detection method of the present invention replaces manual visual inspection with the technical means of machine vision, and identifies grayscale mutations caused by defects in the above image to detect defects, thereby improving the consistency of quartz wafer defect detection results and reducing the missed detection rate and false detection rate of quartz wafer defects. 2. The detection method of the present invention utilizes the grayscale features of the quartz wafer boundary area to extract the quartz wafer area contour, and utilizes the geometric features of the quartz wafer area contour to locate the quartz wafer area in the image, so that the present invention has excellent adaptability to the shape changes between quartz wafers of different models. 3. The present invention adopts machine vision detection and image processing to replace manual visual inspection, which improves the detection efficiency and the reliability of the detection results, and the hardware structure is simple and easy to maintain. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 A schematic diagram of the structure of a quartz wafer defect detection system provided by an embodiment of the present invention; Figure 2 A three-dimensional schematic diagram of the grayscale mutation phenomenon of a quartz wafer provided by an embodiment of the present invention; (wherein A is an image of the quartz wafer captured by the camera, B is an enlarged view of the portion to which the quartz wafer defect belongs, and C is a three-dimensional schematic diagram of the grayscale mutation phenomenon corresponding to the quartz wafer defect in B; Figure 3 The embodiment of the present invention provides Figure 1 The flowchart of the quartz wafer defect detection method of the quartz wafer defect detection system is shown. Among them: 1. Camera; 2. Light source stage; 3. Image processor; 4. Camera fixing device; 5. Quartz wafer to be tested. DETAILED DESCRIPTION To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. It is understandable that for quartz wafers, low reflectivity leads to poor reflected light imaging effect, and because quartz wafers are also characterized by small size, if reflected light imaging is used, the texture of the surface on which the quartz wafer to be tested is placed will interfere with the defect detection result of the quartz wafer to be tested. Therefore, the present invention chooses to use transmitted light imaging. The vertical downward viewing angle of the camera can make the quartz wafer clearly and completely recorded in the quartz wafer image, reducing the impact of environmental interference on the detection results, and making the quartz wafer defects more prominent in the quartz wafer image. Figure 1 A schematic diagram of the structure of a quartz wafer defect detection system provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the quartz wafer defect detection system of this embodiment includes: a camera 1, a light source stage 2, an image processor 3 connected to the camera, and a camera fixing device 4; wherein, the camera is fixed on the upper part of the camera fixing device 4, a light source stage 2 is arranged below the camera 1, a quartz wafer 5 to be detected is placed on the light source stage 2, and the camera 1 is connected to the image processor 3. In a specific application, the camera 1 described in this embodiment is a high-resolution camera. In a specific application, the camera 1 described in this embodiment may preferably be a high-resolution industrial camera. In a specific application, for example, the image processor 3 in this embodiment may be an industrial computer, an embedded board or other programmable logic processing device. The quartz wafer defect detection method of this embodiment is based on Figure 1 The quartz wafer defect detection system in the embodiment shown in the figure is Figure 1 The quartz wafer defect detection system in the illustrated embodiment is implemented by an image processor and has the characteristics of high detection accuracy, fast detection speed, and high reliability of detection results. Figure 3 The embodiment of the present invention provides Figure 1 The flowchart of the quartz wafer defect detection method of the quartz wafer defect detection system is shown. The quartz wafer defect detection method of this embodiment is described as follows. Step 1: Acquire the quartz wafer image captured by the camera; Step 2: converting the quartz wafer image into a grayscale image, identifying the significant difference in grayscale levels between the boundary area and the adjacent area of ​​the quartz wafer under the condition of transmitted light imaging to obtain the regional contour of the quartz wafer; Step 3: Determine whether the quartz wafer to be detected has defects based on the geometric features of the quartz wafer area contour and the typical values ​​of the corresponding geometric features of the quartz wafer to be detected in a non-defective state; the type of geometric features of the quartz wafer area contour and the typical values ​​of the corresponding geometric features of the quartz wafer to be detected in a non-defective state are pre-set by the system operator during the initialization process; wherein, the type of geometric features of the quartz wafer area contour is the area of ​​the quartz wafer area contour by default; calculate the area of ​​the area surrounded by the quartz wafer area contour, and determine whether the quartz wafer has defects based on the typical values ​​of the area of ​​the area surrounded by the quartz wafer area contour of the quartz wafer to be detected in a non-defective state, and the judgment formula is as follows: |A-A0|<N Wherein, A is the area of ​​the area surrounded by the outline of the quartz wafer area, A0 is the typical value of the area of ​​the area surrounded by the outline of the quartz wafer area corresponding to the quartz wafer to be detected without defects, and N is a positive constant close to 0, which is determined according to the working environment of the system. If the area of ​​the region enclosed by the quartz wafer region outline does not meet the above conditions, it is determined that the quartz wafer to be inspected has a defect; For quartz wafers with special shapes, the convex hull algorithm can be used to fit the contour of the quartz wafer area into a contour of a specific shape, and the geometric features of the fitted contour of the specific shape can be used to determine whether the quartz wafer to be inspected has defects. For example, for a rectangular quartz wafer, the shape of the minimum area circumscribed contour can be set to a rectangle, and the following formula can be used to determine whether the rectangular quartz wafer has defects. Wherein, D1 and D2 are respectively the lengths of the two diagonals of the minimum area circumscribed contour, D0 is the typical value of the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be detected without defects, and n is a positive constant close to 0, which is determined according to the working environment of the system. If the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be detected does not meet the above conditions, it is determined that the quartz wafer to be detected has a defect. If the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be inspected meets the above conditions, it is determined that the quartz wafer to be inspected has no defects. Step 4: Processing the quartz wafer grayscale image using an adaptive histogram equalization technique to improve the contrast between the quartz wafer defect area and the adjacent area in the quartz wafer grayscale image; Step 5: Binarize the adaptive histogram equalization result map obtained in step 4 to separate the suspected defect area of ​​the quartz wafer from the remaining areas. In the binary result map of the adaptive histogram equalization result map, the pixel color of the suspected defect area of ​​the quartz wafer is white, and the pixel color of the remaining area is black; Step 6: Using the quartz wafer area contour to screen the suspected defect areas in the binarization result image, only the suspected defect areas located within the minimum area circumscribed contour corresponding to the quartz wafer to be detected are determined as the real defect areas of the quartz wafer. like Figure 2 As shown in the figure, in the quartz wafer image acquired under the condition of transmitted light imaging, the change in the transmittance of the quartz wafer caused by the quartz wafer defect will cause the pixel grayscale of the corresponding area of ​​the final acquired quartz wafer image to change significantly, and form a closed area with a grayscale level significantly different from the grayscale level of the adjacent area, that is, a grayscale mutation area. By identifying such a grayscale mutation area in the quartz wafer image, the suspected defect area of ​​the quartz wafer can be obtained. Although the specific embodiments of the present invention are described above, those skilled in the art should understand that these are only examples, and various changes or modifications can be made to these embodiments without departing from the principles and essence of the present invention. The scope of the present invention is limited only by the appended claims.

Claims

1. A quartz wafer defect detection system for identifying grayscale mutations, characterized in that: The detection system comprises a camera (1), the camera being fixed on the upper part of a camera fixing device (4), a light source stage (2) being arranged below the camera (1), a quartz wafer (5) to be detected being placed on the light source stage (2), and the camera (1) being connected to an image processor (3).

2. A quartz wafer defect detection system for identifying grayscale mutations according to claim 1, characterized in that: The camera (1) is a high-resolution industrial camera.

3. A quartz wafer defect detection system for identifying grayscale mutations according to claim 1, characterized in that: The image processor (3) is an industrial computer or an embedded board.

4. A quartz wafer defect detection method for identifying grayscale mutations using the system as claimed in claim 1, characterized in that: The following steps are involved: Step 1: Obtain the quartz wafer image captured by the camera; Step 2: converting the quartz wafer image into a grayscale image, and obtaining the quartz wafer area contour by identifying the grayscale level difference between the quartz wafer boundary area and the adjacent area under the transmitted light imaging condition; Step 3: judging whether the quartz wafer to be inspected has defects according to the geometric features of the regional contour of the quartz wafer and the typical values ​​of the corresponding geometric features of the quartz wafer to be inspected in a non-defective state; Step 4: Processing the grayscale image of the quartz wafer using an adaptive histogram equalization technique to improve the contrast between the defective region of the quartz wafer and the adjacent region in the grayscale image; Step 5: Binarize the adaptive histogram equalization result map obtained in step 4 to separate the suspected defect area of ​​the quartz wafer from the remaining areas. In the binarized result map of the adaptive histogram equalization result map, the pixel color of the suspected defect area of ​​the quartz wafer is white, and the pixel color of the remaining area is black; Step 6: Using the quartz wafer area contour to screen the suspected defect areas in the binarization result image, the suspected defect areas within the quartz wafer area contour are determined as the real defect areas of the quartz wafer.

5. A quartz wafer defect detection method for identifying grayscale mutations according to claim 4, characterized in that: The method of judging whether the quartz wafer to be detected has defects is based on the geometric features of the quartz wafer area contour and the typical values ​​of the corresponding geometric features of the quartz wafer to be detected in a non-defective state, wherein the type of the geometric features of the quartz wafer area contour and the typical values ​​of the corresponding geometric features of the quartz wafer to be detected in a non-defective state are both preset during the system initialization process; wherein the type of the geometric features of the quartz wafer area contour is the area of ​​the quartz wafer area contour; calculating the area of ​​the area surrounded by the quartz wafer area contour, and judging whether the quartz wafer has defects in combination with the typical value of the area of ​​the area surrounded by the quartz wafer area contour of the quartz wafer to be detected in a non-defective state, wherein the judgment formula is: |A-A0|<N, wherein, A is the area of ​​the area surrounded by the outline of the quartz wafer area, A0 is the typical value of the area of ​​the area surrounded by the outline of the quartz wafer area corresponding to the quartz wafer to be tested when there is no defect, and N is a positive constant; If the area of ​​the region enclosed by the quartz wafer region outline does not meet the above conditions, it is determined that the quartz wafer to be inspected has a defect.

6. The quartz wafer defect detection method for identifying grayscale mutations according to claim 4, characterized in that: In step 3, when the quartz wafer to be inspected is a non-basic geometric figure, the convex hull algorithm is used to fit the regional contour of the quartz wafer to the contour of the basic geometric figure, and the quartz wafer is judged whether there is a defect according to the geometric features of the fitted basic geometric figure contour.

7. The quartz wafer defect detection method for identifying grayscale mutations according to claim 6, characterized in that: When the quartz wafer to be detected is a quasi-rectangular wafer, its area contour is fitted into a rectangle, that is, the shape of the minimum area circumscribed contour of the wafer is a rectangle, and the following formula is used to determine whether the quasi-rectangular quartz wafer has defects: Wherein, D1 and D2 are respectively the lengths of the two diagonals of the minimum area circumscribed contour, D0 is the typical value of the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be detected without defects, n is a positive constant, if the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be detected does not meet the above conditions, it is determined that the quartz wafer to be detected has defects; if the diagonal length of the minimum area circumscribed contour corresponding to the quartz wafer to be detected meets the above conditions, it is determined that the quartz wafer to be detected has no defects.

Images