Detection device, detection method, and method for manufacturing electronic components

Abstract

To provide a detection device, a detection method, and a method for manufacturing an electronic component, capable of detecting a translucent foreign matter and improving a capacity of detecting degeneration on a target surface.SOLUTION: A detection device that detects the presence or absence of abnormalities such as foreign matter or defects on a workpiece, comprises: imaging means that images a plurality of images of the workpiece with different focal points; focus degree calculation means that calculates a focus degree feature value based on a focus degree for each pixel of the image for each imaged pixel; abnormal focus degree setting means that sets a numerical range of the focus degree feature value that is deemed to have abnormalities; and abnormal focus degree area setting means that sets a threshold value for determining that an area that falls within the numerical range of the focus degree feature value set by the abnormal focus degree setting means is deemed to have abnormalities.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a detection device, a detection method, and a method for manufacturing an electronic component. [Background Art]

[0002] In some cases, electronic components (for example, those subjected to wire bonding) are inspected for the presence or absence of abnormalities such as foreign matter and defects. For example, the defect detection method described in Patent Document 1 prepares a reference image (correct pattern image) subjected to out-of-focus correction and an inspection target image by means of a focus calculation unit and a focus adjustment unit. Then, defects in the inspection target pattern are detected based on a comparison result between the correct pattern image and the inspection target image.

[0003] In Patent Document 2, an omnifocal image and a height image are obtained by a focusing method, and foreign matter inspection and other inspections are performed. Specifically, images of the wire and the inspected portion are captured so as to focus stepwise from the inspected portion to the wire, to obtain images having different focal points in the height direction, a focusing degree is calculated for each pixel of the images, and the inspected portion is inspected based on distribution information of the focusing degree in each image. [Prior Art Documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2009-192358 [Patent Document 2] Japanese Unexamined Patent Application Publication No. 2018-98249 [Brief Summary of the Invention] [Problem to be Solved by the Invention]

[0005] The method described in Patent Document 1 is a method of taking the difference between brightness images (grayscale images). Therefore, for example, if a translucent foreign object is attached, it is difficult to create a difference in brightness, and as a result it is difficult to detect foreign objects or defects that do not show a clear difference in brightness images. The method described in Patent Document 2 uses a height image in addition to a brightness image, but it is difficult to detect foreign objects or defects that do not show a clear difference in brightness images. Furthermore, if the foreign object has no height, it is not possible to detect the foreign object or defect using the height image. In other words, conventional visual inspection methods for foreign objects and defects are limited to detecting foreign objects and defects that have changes in brightness, color, or height, and cannot detect translucent foreign objects. Also, on surfaces that do not normally have texture, such as mirror surfaces, it is not possible to detect foreign objects that are difficult to see using brightness.

[0006] Therefore, in view of these circumstances, the present invention aims to provide a detection device, a detection method, and a method for manufacturing electronic components that can detect translucent foreign matter and improve the ability to detect alteration of the surface of an object. [Means for solving the problem]

[0007] The detection device of the present invention is a detection device for detecting the presence or absence of foreign matter or defects on a workpiece, and comprises: an imaging means for capturing multiple images of a workpiece at different focal points; a focus calculation means for performing a focus feature calculation based on the degree of focus for each pixel of the image captured; an abnormal focus setting means for setting a numerical range of the focus feature that indicates an abnormality; and an abnormal focus area setting means for setting a threshold that indicates an abnormality in the area corresponding to the numerical range of the focus feature set by the abnormal focus setting means.

[0008] According to the detection device of the present invention, a focus-based feature quantity is used as a parameter for visual inspection of the workpiece. By using the focus-based feature quantity as a parameter in this way, it is possible to detect abnormalities that cannot be detected by brightness or height information. Furthermore, the user only needs to set the numerical range of the focus-based feature quantity to be considered abnormal and the area threshold for determining that an abnormality exists, making it easy to set up the inspection. [Effects of the Invention]

[0009] The detection device, detection method, and method for manufacturing electronic components of the present invention can detect translucent foreign objects and improve the ability to detect alterations on the surface of an object. [Brief explanation of the drawing]

[0010] [Figure 1] This is a block diagram of the detection device of the present invention. [Figure 2] This is a conceptual diagram illustrating the detection device of the present invention. [Figure 3] This figure shows images created by the detection device of the present invention, where (a) is the full-focus image, (b) is the distance image, and (c) is the focus feature image. [Figure 4] This figure shows the display means for the setting unit of the detection device of the present invention. [Figure 5] This is a flowchart illustrating the procedure for the detection method of the present invention. [Modes for carrying out the invention]

[0011] The embodiments of the present invention will be described below with reference to Figures 1 to 5.

[0012] The detection device of the present invention detects abnormalities such as foreign matter or defects in a workpiece. In this embodiment, as an example of a workpiece, an electronic component (semiconductor device) in which semiconductor chips are connected to each other or to an electrode substrate such as a lead frame by wire bonding will be described, and in particular, the case of detecting the surface other than the wire portion will be described.

[0013] As shown in Figure 1, the detection device 100 of the present invention comprises an image creation unit 1, a setting unit 2, a detection unit 3, and a display means 4.

[0014] The image creation unit 1 is a means for creating image data for displaying a workpiece, and includes an imaging means 5 for capturing an image of the workpiece, a focus calculation means 6 for calculating the degree of focus in the image, a full-focus image creation means 7 for creating a full-focus image from multiple images with different focal points captured by the imaging means 5, a distance image creation means 8 for creating a distance image, and a focus feature image creation means 9 for creating a focus feature image.

[0015] The imaging means 5 can be configured, for example, as a CCD camera or a CMOS camera, and can move the focal plane of an image that is in focus on a part of the object being photographed.

[0016] The focus calculation means 6 calculates the degree of focus based on multiple images with different focal points captured by the imaging means 5, and further calculates a focus feature quantity, which will be described later. The degree of focus is a quantity that represents the degree to which an image is in focus, and is calculated using various known methods, for example, as described in

[0030] of Japanese Patent Application Publication No. 2012-23340. That is, as shown in Figure 2, the focus calculation means 6 calculates the degree of focus for each pixel of each of the multiple images with different focal points captured by the imaging means 5. This makes it possible to derive changes in the degree of focus focusing on a certain pixel position, for example, as shown in A and B of Figure 2, and thereby determine the maximum and minimum values ​​of the degree of focus.

[0017] The focusing degree calculation means 6 further calculates a focusing degree feature quantity based on the obtained calculated value of the focusing degree. The focusing degree feature quantity serves as a parameter for detecting an abnormality, and is, for example, any one of various indicators indicating the focusing degree, such as the maximum value of the focusing degree, the maximum value of the focusing degree divided by the minimum value of the focusing degree, or a Q-value for each pixel, or a combination of a plurality of these indicators. The maximum value of the focusing degree is a value at which the focusing degree becomes maximum at a certain pixel position, as shown in graph A of FIG. 2. The maximum value of the focusing degree divided by the minimum value of the focusing degree represents the reliability of the focusing degree, and is a numerical value indicating how reliable the determination on the presence or absence of an object is. When an object is present, as shown in graph A of FIG. 2, the maximum value of the focusing degree is clear and the difference between the maximum value and the minimum value is large, so the reliability of the focusing degree exhibits a high value. On the other hand, when the presence of an object is ambiguous due to the presence of a foreign substance or the like, as shown in graph B of FIG. 2, the maximum value is not clear and the difference between the maximum value and the minimum value is small, so the reliability of the focusing degree exhibits a value close to 1. Here, the Q-value is an indicator indicating the sharpness of a peak, and generally, maximum value divided by half-maximum width is used. The half-maximum width is obtained by recognizing, as two focal plane positions, one on the side away from the imaging target and the other on the side closer to the imaging target, focal plane positions at which the focusing degree is halved from the focal plane position where the focusing degree is maximum, and is the distance between the two focal plane positions.

[0018] Any of the maximum value of the focusing degree, the maximum value of the focusing degree divided by the minimum value of the focusing degree, and the Q-value may be employed as the focusing degree feature quantity, and in the present embodiment, the focusing degree feature quantity is set as the maximum value of the focusing degree divided by the minimum value of the focusing degree. For this reason, the focusing degree calculation means 6 calculates the maximum value of the focusing degree divided by the minimum value of the focusing degree at a certain pixel position, and performs this calculation at all pixel positions.

[0019] The all-focus image creation means 7 picks up the brightness value in the in-focus frame and creates an all-focus image as shown in Figure 3(a). The all-focus image is created based on multiple image data (brightness images) with different focal points. Here, an all-focus image is created by continuously changing the distance between the imaging system and the object (scanning the camera in the vertical direction, which is the Z direction) and capturing multiple images with different focal points. Then, the focus (brightness value) is evaluated for each local region and the image is reconstructed from a combination of the local images that are in focus. As a result, the all-focus image is in focus at every pixel. In other words, an all-focus image is an image in which the focus is evaluated for each local region and reconstructed from a combination of the local images that are in focus, so that all pixels are in focus. In Figure 3(a), the outlines of the wire portion 10 and the non-wire region 11 are clearly defined. However, translucent foreign objects do not appear in the all-focus image.

[0020] The distance image creation means 8 stores relative distance information at the moment when the focus degree (degree of focus) is maximum for each pixel at the same time as imaging, and creates a distance image from the imaging means 5 to the object being imaged, as shown in Figure 3(b). Here, the distance image is an image represented by brightness (grayscale) according to the distance from the imaging means 5, and the height information of the imaged object can be obtained from the distance image. For example, taller objects (those closer to the imaging means 5) are represented in lighter colors (wire portion 10 in Figure 3(b)), and shorter objects (those farther from the imaging means 5) are represented in darker colors (areas other than the wires 11 in Figure 3(b)). In other words, the distance image is relative distance information at the moment when the focus degree is maximum for each pixel at the same time as imaging by the imaging means 5, and is an image represented by brightness according to the distance from the imaging means 5. In this case, foreign objects that have no height will not appear in the distance image.

[0021] The in-focus degree feature image creating means 9 creates an image as shown in Fig. 3(c) based on the value of the in-focus degree feature quantity calculated by the in-focus degree calculating means 6 (in the present embodiment, the maximum value of in-focus degree ÷ the minimum value of in-focus degree). In this case, when there is a foreign matter 12 in a region 11 other than the wire, the in-focus degree of the foreign matter 12 is different from the in-focus degree of the region 11, so that the region 11 and the foreign matter 12 have different color tones in the in-focus degree feature image.

[0022] The setting unit 2 performs in advance settings required when a user detects an abnormality of a workpiece, and includes a detection area specifying means 15, an abnormal in-focus degree setting means 16, and an abnormal in-focus degree area setting means 17.

[0023] The detection area specifying means 15 is for specifying an area in which a user wants to detect an abnormality in a workpiece. That is, from the image of the workpiece displayed on the image display means 21 of the display unit 4, the user specifies the area where the user wants to detect an abnormality in the displayed image, for example, by dragging a mouse, specifying coordinates, or the like. In addition, an area specifying method of recognizing a wire portion and "setting a portion other than the wire as a detection range" can also be adopted. In this case, the image of the workpiece displayed on the image display means 21 may be an omnifocal image created by the image creating unit 1, a distance image, or the in-focus degree feature image creating means, or may be any of a pre-input image of the workpiece and an image captured by the imaging means 5. In short, any image in which the outline of the area to be detected is clear to a certain extent is acceptable.

[0024] The abnormal in-focus degree setting means 16 is for setting a numerical range of an in-focus degree feature quantity that is regarded as abnormal in the detection area. That is, when a semi-transparent foreign matter exists between the surface of the workpiece and the imaging means 5, the workpiece is blurred, so that the in-focus degree is lowered. In addition, when a foreign matter that is difficult to see with luminance is placed on a surface that does not have texture in a normal state such as a mirror surface, the in-focus degree at that position increases. For this reason, since the numerical range of normal in-focus degree differs depending on the workpiece, the in-focus degree feature quantity that is regarded as having an abnormality is set for each workpiece to be detected, and the in-focus degree feature quantity is used as a parameter for abnormality detection.

[0025] The abnormal focus area setting means 17 sets a threshold for determining whether an area corresponding to the numerical range of the focus feature set by the abnormal focus setting means 16 is abnormal. In other words, the user assumes a boundary between an area that is acceptable (i.e., judged not to be abnormal) and an area that is not acceptable (i.e., judged to be abnormal) for the region corresponding to the numerical range of the focus feature that is considered abnormal, and sets a threshold for that area.

[0026] The detection unit 3 determines whether or not there is an abnormality in the workpiece and includes a determination means 20. Specifically, the determination means 20 determines that there is an abnormality if the area corresponding to the numerical range of the focus characteristic quantity set by the abnormal focus setting means 16 exceeds the threshold of the abnormal focus area setting means 17. Alternatively, the determination means 20 may determine the presence of a foreign object based on the difference between the focus characteristic quantity image and other images (all-focus image, distance image, pre-input image of the workpiece, image captured by the imaging means 5).

[0027] The display unit 4 includes an image display means 21 that displays an overall image and / or images of each processing target according to the user's instructions for each workpiece, and a setting unit display means 22 for the user to set the range and area of ​​a focus characteristic quantity that is to be considered abnormal in advance. Specifically, as shown in Figure 3, the setting unit display means 22 includes a focus setting unit 23 that sets a numerical range (upper and / or lower limit) of the focus characteristic quantity that is to be considered abnormal, and an area threshold setting unit 24 that sets a threshold for the area corresponding to the set numerical range of the focus characteristic quantity that is to be considered abnormal. Furthermore, in Figure 4, the numerical range of the brightness value that is to be considered abnormal and the threshold of the area corresponding to that numerical range, as well as the numerical range of the height that is to be considered abnormal and the threshold of the area corresponding to that numerical range are set, and brightness and / or height information can be selected as parameters. In other words, it is possible to use both anomaly detection using focus characteristic quantities and detection using brightness and / or height information in combination.

[0028] Incidentally, a computer is basically composed of input means with input functions, output means with output functions, storage means with memory functions, arithmetic means with calculation functions, and control means with control functions. The input function is for reading information from the outside into the computer, and the read data and programs are converted into signals in a format suitable for the computer system. The output function displays calculation results and stored data externally. The storage means stores and saves programs, data, processing results, etc. The arithmetic function processes data by calculating and comparing it according to the instructions of the program. The control function decodes the instructions of the program and issues instructions to each means, and this control function oversees all the means of the computer.

[0029] Input methods include keyboards, mice, tablets, microphones, joysticks, scanners, and capture boards. Output methods include monitors, speakers, and printers. Storage methods include memory, hard disks, CDs, CD-Rs, PDs, and MOs. Processing methods include CPUs, and control methods include CPUs and motherboards. Therefore, the imaging means other than the camera can be comprised of a single publicly known and publicly available existing computer.

[0030] The detection method using the detection device described above will be explained using the flowchart in Figure 5. First, the imaging means 1 acquires multiple real images with different focal points in the vertical direction, with the workpiece 20 being within the shooting range (step S1). As a result, multiple image data with different focused pixel positions are stored in memory, and the height information of the workpiece is also stored in memory.

[0031] The focus calculation means 6 calculates the degree of focus at each pixel position based on a plurality of images with different focal points captured by the imaging means. This calculation is performed by a known method, for example, by the method described in paragraph

[0028] of Japanese Patent No. 6704336. Furthermore, the focus calculation means 6 calculates a focus feature quantity at each pixel position (step S2). In this embodiment, the focus feature quantity is the maximum value of the focus divided by the minimum value.

[0032] Based on the calculated value of the focus calculation means 6, the all-focus image creation means 7 creates an all-focus image, the distance image creation means 8 creates a distance image, and the focus feature image creation means 9 creates a focus feature image (step S3).

[0033] The user specifies the area in which they want to detect anomalies in the image displayed on the image display means 21 (one of the created full-focus image, distance image, or focus feature image, one of the actual image, or one of the pre-set work image) by, for example, dragging the mouse or specifying coordinates (step S4).

[0034] Furthermore, the user sets the range of the focus feature that indicates an anomaly in the detection area (step S5). In this case, the range may include setting an upper and lower limit, setting only an upper limit, or setting a lower limit. Focus represents the degree of surface roughness. That is, if the area is in focus and rough, it means the focus is high, and if the area is in focus but not rough, it means the focus is low. For example, if the rough part is normal, as shown in Figure 3, if we define "a focus of 5 or less is a foreign object," then the part with a low focus (part with low roughness) can be considered a foreign object. Also, if the user wants to select brightness values ​​or height information as parameters for anomaly detection, they can select detection using brightness values ​​and / or height information, and it is possible to use anomaly detection using focus features in combination with anomaly detection using brightness values ​​and height information as before. In this case as well, the user inputs the upper and / or lower limits of the brightness value to be considered abnormal, or the upper and / or lower limits of the height information to be considered abnormal.

[0035] Furthermore, the user sets a threshold (step S6) for determining if an area falling within the numerical range of the set focus feature is abnormal. If anomaly detection using brightness values ​​or height information is used in conjunction, separate thresholds for determining abnormal areas are set for each.

[0036] The determination means 20 determines whether or not there is an abnormality in the detection area (step S7). That is, if the area corresponding to the numerical range of the focus feature quantity set by the abnormal focus area setting means 16 exceeds the threshold set by the abnormal focus area setting means 17, then an abnormality is found at that location (step S8). If there is no part that exceeds the threshold, then there is no abnormality in the detection area (step S9). In this case, the determination means 20 may also determine the presence of a foreign object based on the difference between the focus feature quantity image and other images (any of the all-focus image, distance image, pre-inputted work image, or image captured by the imaging means 5).

[0037] If anomaly detection is to continue, return to step S1 when changing the imaging area for a different workpiece or the same workpiece, return to step S4 when performing anomaly detection in the same imaging area, and terminate this detection method when anomaly detection is to be terminated (step S10).

[0038] The inspection apparatus and inspection method of the present invention can detect anomalies that cannot be detected by brightness or height information, such as translucent foreign objects, by using focus characteristics as parameters, and also improve the ability to detect alterations on the surface of the object. Furthermore, the user only needs to set the numerical range of focus characteristics to be considered abnormal and the area threshold for determining that an abnormality exists, making inspection settings easy.

[0039] Although embodiments of the present invention have been described above, the present invention is not limited to the above embodiments and can be modified in various ways. For example, the imaging means is not limited to a CCD camera, but can be a CMOS camera or various other types. In terms of the imaging direction, the imaging means was positioned above the object being imaged, but the imaging means may be positioned horizontally with respect to the object being imaged. Also, the imaging means 25 may be fixed, and the object being imaged may move closer to or further away from the imaging means 25. Furthermore, the workpiece is not limited to one having bonding wires, but can be applied to various electronic components and non-electronic components.

[0040] As part of the procedure, the user may perform settings before imaging. That is, step S3 may be performed before step S1. Furthermore, the creation of various images (all-focus image, distance image, focus feature image) is not necessarily required, and it is possible to detect anomalies by calculation alone without using images. The focus feature can be the maximum value of focus, the Q value, other indicators of focus, or a combination of several of these. In this case, the user may be able to select whether to use confidence, the maximum value of focus, or the Q value as the focus feature for each examination.

[0041] 5. Imaging means 6 Focus degree calculation means 9. Means for creating an image with focus feature 16 Abnormal focus setting means 17 Abnormal Focus Area Setting Means 20 Judgment means

Claims

1. In a detection device that detects the presence or absence of foreign objects or defects on a workpiece, An imaging means for capturing multiple images of a workpiece with different focal points, A focus calculation means that calculates the degree of focus for each of the multiple images with different focal points that have been captured, for each pixel position, and calculates a focus feature quantity, which is a parameter for detecting anomalies, based on the multiple changes in the degree of focus obtained at the same pixel position. An abnormal focus setting means for setting a numerical range for the focus feature that indicates an abnormality, The system includes an abnormal focus area setting means that sets a threshold for determining whether an area corresponding to a numerical range of focus feature quantities set by the abnormal focus setting means is abnormal, A detection device characterized by detecting the presence or absence of an abnormality based on a comparison between an area corresponding to a numerical range of the focus characteristic quantity set by the abnormal focus setting means and a threshold value set by the abnormal focus area setting means.

2. The detection device according to claim 1, characterized in that it detects an abnormality when the area corresponding to the numerical range of the focus characteristic quantity set by the abnormal focus setting means exceeds the threshold set by the abnormal focus area setting means.

3. The detection device according to claim 1, further comprising a determination means for determining that there is an abnormality when the area corresponding to the numerical range of the focus characteristic quantity that is considered abnormal, as set by the abnormal focus setting means, exceeds the threshold of the abnormal focus area setting means.

4. The detection device according to claim 1, further comprising a focus feature image creation means for creating a focus feature image based on the focus feature quantities calculated by the focus calculation means.

5. The detection device according to claim 1, characterized in that the focus feature quantity is the maximum value of focus, the maximum value of focus divided by the minimum value, or the Q value for each pixel.

6. In a detection method for detecting the presence or absence of foreign objects or defects on a workpiece, Multiple images of the workpiece at different focal points are captured, For each of the multiple images with different focal points that have been captured, the degree of focus is calculated for each pixel position, and based on the multiple changes in the degree of focus obtained at the same pixel position, a degree of focus feature quantity, which is a parameter for detecting anomalies, is calculated. Set a numerical range for the focus feature that indicates an abnormality, A threshold is set to determine if an area falling within the specified numerical range of the focus feature is considered abnormal. A detection method characterized by detecting anomalies based on a comparison between the area corresponding to a set numerical range of focus features and a set threshold.

7. After bonding the wires to the designated position on the substrate, Multiple images of the workpiece at different focal points are captured, For each of the multiple images with different focal points that have been captured, the degree of focus is calculated for each pixel position, and based on the multiple changes in the degree of focus obtained at the same pixel position, a degree of focus feature quantity, which is a parameter for detecting anomalies, is calculated. Set a numerical range for the focus feature that indicates an abnormality, A threshold is set to determine if an area falling within the specified numerical range of the focus feature is considered abnormal. A method for manufacturing electronic components, characterized in that an anomaly is detected based on a comparison between the area corresponding to a set numerical range of focus features and a set threshold.

Images