Inspection device and inspection method

Abstract

To easily confirm propriety of adjustment of an inspection threshold.SOLUTION: An inspection data set 441 indicates position information of a plurality of inspection areas on a substrate and also indicates an inspection kind and inspection threshold with respect to each one of the inspection areas. An inspection data storage section 44 stores the plurality of inspection data sets with respect to a plurality of substrate types. A major defective image storage section 45 stores major defective images 451 each associated with any one of the inspection data sets. A trial inspection control section 41 acquires an adjusted object inspection data set with the adjusted inspection threshold with respect to the substrate type of one production lot when inspecting the one production lot. A confirmation inspection control section 42 applies the adjusted inspection threshold indicated by the adjusted object inspection data set to an inspection threshold corresponding to the other inspection data set, so as to acquire an applied inspection data set, and causes an inspection section to execute inspection processing using the applied inspection data set with respect to the major defective image of the other inspection data set.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a technique for inspecting a substrate.

Background Art

[0002] Conventionally, an inspection apparatus has been used that images a printed circuit board on which a pattern is formed to obtain an image and detects defects using the image. Further, in the inspection apparatus of Patent Document 1, a method for easily adjusting (narrowing down) complex inspection parameters used for defect detection is disclosed. On the other hand, when inspecting printed circuit boards of one production lot, a trial inspection is also performed using some of the printed circuit boards in the production lot, and the inspection threshold is adjusted. In this case, by performing an inspection on all the printed circuit boards in the production lot using the adjusted inspection threshold, it is possible to reduce false alarms specific to the production lot.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in adjusting the inspection threshold, usually, in order to reduce false alarms, the inspection threshold is loosened. In this case, it is necessary to confirm whether a serious defect that cannot be overlooked can be detected using the adjusted inspection threshold, that is, to confirm the appropriateness of the adjustment of the inspection threshold. It is also conceivable to prepare a substrate for confirming serious defects provided with serious defects and inspect the substrate for confirming serious defects using the adjusted inspection threshold. However, preparing a substrate for confirming serious defects requires cost and time, and there is also a problem that the substrate for confirming serious defects deteriorates over time. Therefore, a method for easily confirming the appropriateness of the adjustment of the inspection threshold is required.

[0005] The present invention has been made in view of the above problems, and aims to easily confirm the appropriateness of adjusting the inspection threshold value.

Means for Solving the Problems

[0006] Aspect 1 of the present invention is an inspection apparatus for inspecting a substrate. Using an inspection dataset that indicates the position information of a plurality of inspection regions on the substrate and also indicates the type of inspection and the inspection threshold value for each inspection region, an inspection unit that detects defects in the substrate by executing an inspection process on the captured image of the substrate, an inspection data storage unit that stores a plurality of inspection datasets respectively used for the inspection processes for a plurality of substrate types, a serious defect image storage unit that stores a plurality of serious defect images each associated with any one of the plurality of inspection datasets, an input unit that receives an input for changing the inspection threshold value, when inspecting one production lot, causing the inspection unit to execute the inspection process on the captured images of some substrates in the production lot using the target inspection dataset for the substrate type of the production lot as a trial inspection, and when the input unit receives an input for changing the inspection threshold value based on the result of the trial inspection, causing the inspection unit to execute a new trial inspection using the target inspection dataset with the changed inspection threshold value, thereby obtaining an adjusted target inspection dataset with the adjusted inspection threshold value, and an application inspection control unit that applies the adjusted inspection threshold value indicated by the adjusted target inspection dataset to the inspection threshold value corresponding to the adjusted inspection threshold value in other inspection datasets for other substrate types to obtain an applied inspection dataset, and causes the inspection unit to execute the inspection process using the applied inspection dataset on the serious defect images associated with the other inspection datasets stored in the serious defect image storage unit as a confirmation inspection.

[0007] Aspect 2 of the present invention is the inspection apparatus of Aspect 1, wherein the inspection unit executes the inspection process using the adjusted target inspection dataset on the captured images of a plurality of substrates included in the production lot as a mass production inspection.

[0008] Aspect 3 of the present invention is the inspection apparatus of Aspect 2, wherein when there is a serious defect image not detected by the confirmation inspection, the confirmation inspection control unit does not permit the execution of the mass production inspection by the inspection unit.

[0009] Aspect 4 of the present invention is the inspection apparatus of Aspect 2 (which may be Aspect 2 or 3), wherein when there is a serious defect image not detected by the confirmation inspection, the confirmation inspection control unit gives a notification prompting readjustment of the adjusted inspection threshold in the adjusted target inspection data set.

[0010] Aspect 5 of the present invention is the inspection apparatus of Aspect 2 (which may be any one of Aspects 2 to 4), wherein when the input unit receives an input designating a certain defect detected in the mass production inspection as a serious defect, the image of the defect is stored in the serious defect image storage unit as a serious defect image in a state associated with the adjusted target inspection data set.

[0011] Aspect 6 of the present invention is the inspection apparatus of Aspect 1 (which may be any one of Aspects 1 to 5), wherein when the input unit receives an input designating a certain defect detected in the trial inspection as a serious defect, the image of the defect is stored in the serious defect image storage unit as a serious defect image in a state associated with the target inspection data set.

[0012] Aspect 7 of the present invention is the inspection apparatus of any one of Aspects 1 to 6, wherein the plurality of inspection data sets are classified into a plurality of groups according to the type of surface treatment of the corresponding substrate type and / or the color of the solder resist, and the other inspection data set used for the confirmation inspection is limited to those belonging to the same group as the target inspection data set.

[0013] Aspect 8 of the present invention is an inspection method for inspecting a substrate by an inspection unit, wherein the inspection unit uses an inspection data set that indicates the position information of a plurality of inspection regions on the substrate and also indicates the type of inspection and the inspection threshold for each inspection region to perform an inspection process on the captured image of the substrate, thereby detecting defects of the substrate. The inspection method includes: a) a step of preparing a plurality of inspection data sets respectively used for the inspection processes for a plurality of substrate types; b) a step of preparing a plurality of critical defect images each associated with any one of the plurality of inspection data sets; c) a step of, when inspecting one production lot, causing the inspection unit to perform, as a trial inspection, the inspection process on the captured images of some substrates of the production lot using the target inspection data set for the substrate type of the production lot; d) a step of, when the inspection threshold of the target inspection data set is changed based on the result of the trial inspection, causing the inspection unit to perform a new trial inspection using the target inspection data set with the changed inspection threshold, thereby obtaining an adjusted target inspection data set with the adjusted inspection threshold; e) a step of applying the adjusted inspection threshold indicated by the adjusted target inspection data set to the inspection threshold corresponding to the adjusted inspection threshold in other inspection data sets for other substrate types to obtain an applied inspection data set; and f) a step of causing the inspection unit to perform, as a confirmation inspection, the inspection process using the applied inspection data set on the critical defect images associated with the other inspection data sets.

Advantages of the Invention

[0014] According to the present invention, it is possible to easily confirm the appropriateness of adjusting the inspection threshold in the target inspection data set.

Brief Description of the Drawings

[0015]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 3C

Figure 4

Figure 5A

Figure 5B

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0016] FIG. 1 is a block diagram showing the configuration of an inspection device 1 according to an embodiment of the present invention. The inspection device 1 is a device for inspecting a printed circuit board, and in one example, it is used for the final appearance inspection of a printed circuit board. The inspection device 1 may be used for inspections other than the final appearance inspection. The inspection device 1 includes a device main body 2 and an auxiliary unit 4. The device main body 2 and the auxiliary unit 4 are communicably connected via a network 8 such as a LAN or the Internet.

[0017] The apparatus main body 2 includes an imaging unit 21, a moving mechanism 22, a main body inspection unit 23, and a main body control unit 24. The imaging unit 21 has an imaging element such as a CCD sensor or a CMOS sensor, and images a printed circuit board. In this processing example, the captured image acquired by the imaging unit 21 is a color image. The captured image may be a grayscale image. The moving mechanism 22 has, for example, a motor, a ball screw, etc., and relatively moves the printed circuit board with respect to the imaging unit 21. The main body inspection unit 23 performs an inspection process using the inspection data set 441 described later on the captured image output from the imaging unit 21, and detects a defect from the captured image. The main body control unit 24 is responsible for overall control of the apparatus main body 2. The main body inspection unit 23 and the main body control unit 24 are realized by, for example, a computer and / or an electric circuit.

[0018] The auxiliary unit 4 is realized by a computer. FIG. 2 is a diagram showing the configuration of the computer 3. The computer 3 has the configuration of a general computer system including a CPU 31, a ROM 32, a RAM 33, a storage device 34, a display unit (display) 35, an input unit 36, a reading device 37, a communication unit 38, a GPU 39, and a bus 30. The CPU 31 performs various arithmetic processes. The GPU 39 performs various arithmetic processes related to image processing, etc. The ROM 32 stores a basic program. The RAM 33 and the storage device 34 store various information. The display unit 35 displays various information such as images. The input unit 36 includes a keyboard 36a and a mouse 36b that receive input from an operator. The reading device 37 reads information from a computer-readable recording medium 81 such as an optical disk, a magnetic disk, a magneto-optical disk, a memory card, etc. The communication unit 38 transmits and receives signals to and from the apparatus main body 2, etc. The bus 30 is a signal circuit that connects the CPU 31, the GPU 39, the ROM 32, the RAM 33, the storage device 34, the display unit 35, the input unit 36, the reading device 37, and the communication unit 38. In the computer 3, a touch panel may be provided, and the input unit 36 and the display unit 35 may be realized by the touch panel.

[0019] In computer 3, program 811 is read from recording medium 81 via reading device 37 in advance and stored in storage device 34. Program 811 may be stored in storage device 34 via network 8. CPU 31 and GPU 39 execute arithmetic processing while using RAM 33 and storage device 34 according to program 811. CPU 31 and GPU 39 function as an arithmetic unit in computer 3. Other configurations that function as an arithmetic unit may be adopted in addition to CPU 31 and GPU 39.

[0020] In auxiliary unit 4, when computer 3 executes arithmetic processing and the like according to program 811, the functional configuration shown in FIG. 1 is realized. That is, CPU 31, GPU 39, ROM 32, RAM 33, storage device 34 of computer 3, and their peripheral configurations realize auxiliary unit 4. All or part of the functions of auxiliary unit 4 may be realized by a dedicated electric circuit, or each function may be realized by an individual program. Also, auxiliary unit 4 may be realized by a plurality of computers. In this processing example, auxiliary inspection unit 43 described later includes a dedicated image processing circuit (image processing board).

[0021] Auxiliary unit 4 includes a trial inspection control unit 41, a confirmation inspection control unit 42, an auxiliary inspection unit 43, an inspection data storage unit 44, a critical defect image storage unit 45, a display unit 35, and an input unit 36. Similar to main body inspection unit 23, auxiliary inspection unit 43 executes inspection processing using inspection data set 441 on a captured image showing a printed circuit board, and detects defects from the captured image. Trial inspection control unit 41 controls auxiliary inspection unit 43 to perform a trial inspection described later. Confirmation inspection control unit 42 controls auxiliary inspection unit 43 to perform a confirmation inspection described later. Trial inspection control unit 41 and confirmation inspection control unit 42 are part of unit control unit 40, and unit control unit 40 is also responsible for overall control of auxiliary unit 4. Inspection data storage unit 44 stores a plurality of inspection data sets 441 used in inspection processing. Critical defect image storage unit 45 stores a plurality of critical defect images (data) 451 used in confirmation inspection.

[0022] FIGS. 3A to 3C are diagrams for explaining the inspection data set 441, and show a plurality of inspection regions 92 on the printed circuit board 9. FIG. 3A shows the printed circuit board 9 of substrate type A, FIG. 3B shows the printed circuit board 9 of substrate type B, and FIG. 3C shows the printed circuit board 9 of substrate type C. The substrate type is the type of printed circuit board product, and is distinguished by, for example, the product number. Each inspection data set 441 indicates the position information of a plurality of inspection regions 92 on the printed circuit board 9 (including the size of the inspection region 92), and also indicates the type of inspection and the inspection threshold for each inspection region 92. The plurality of inspection data sets 441 are respectively used for inspection processes for printed circuit boards 9 of different substrate types. As shown in FIGS. 3A to 3C, in the printed circuit board 9, the substrate size, the position and size of the inspection region 92, etc. are different depending on the substrate type.

[0023] Each inspection region 92 belongs to any one of a plurality of region types. Examples of the plurality of region types include a pad region where a metal such as copper is exposed, a solder resist region (hereinafter also referred to as an "SR region") where the lower layer of the solder resist is the base material of the printed circuit board 9, a wiring region where wiring is provided under the solder resist, a silk region such as characters and symbols printed on the solder resist, a through hole region which is the opening of a through hole, and the like.

[0024] In each inspection region 92, an inspection of a type corresponding to the region type is performed. In an inspection region 92 whose region type is a pad region, for example, the minimum distance of the region between two adjacent pads is obtained, and when the minimum distance is less than the inspection threshold, an inspection is performed to detect the region as a defect. In an inspection region 92 whose region type is an SR region, for example, the gradation value at each position is obtained for each color component, and an inspection is performed to detect, as a defect, a set of positions where the gradation value is outside the normal range defined by the inspection threshold (upper limit value and lower limit value). In an inspection region 92 whose region type is a through hole region, for example, the minimum width of the land around the through hole is obtained, and when the minimum width is less than the inspection threshold, an inspection is performed to detect the land as a defect. Actually, various known inspections may be performed according to each region type.

[0025] As described above, the inspection data set 441 indicates the position information of each inspection area 92, as well as the type of inspection and the inspection threshold for the inspection area 92. Values of various parameters in each inspection (for example, the size of the mask area when the mask area or the like is set) may be included in the inspection data set 441. In the following description, it is assumed that the inspection data sets 441 for substrate types A to C are stored in the inspection data storage unit 44. In these substrate types A to C, the design rules and the like are the same. Of course, inspection data sets 441 for two types of substrate types, or four or more types of substrate types, may be stored. Further, in the following description, it is assumed that each inspection area 92 belongs to any one of the first area type, the second area type, and the third area type. The inspection data sets 441 for each substrate type A to C include at least one inspection area 92 of the first area type, at least one inspection area 92 of the second area type, and at least one inspection area 92 of the third area type. The inspection data set 441 may indicate the position information and the like of the inspection areas 92 on both main surfaces of the printed circuit board 9.

[0026] FIG. 4 is a diagram showing a plurality of heavy defect images 451 stored in the heavy defect image storage unit 45. Each heavy defect image 451 is associated with any one of the plurality of inspection data sets 441. In the example of FIG. 4, the heavy defect image 451 associated with the inspection data set 441 of substrate type A is surrounded by a dashed rectangle labeled with symbol A, the heavy defect image 451 associated with the inspection data set 441 of substrate type B is surrounded by a dashed rectangle labeled with symbol B, and the heavy defect image 451 associated with the inspection data set 441 of substrate type C is surrounded by a dashed rectangle labeled with symbol C. Each heavy defect image 451 associated with the inspection data set 441 of substrate type A is detected by an inspection process for the printed circuit board 9 of substrate type A using the inspection data set 441. Actually, the inspection area 92 where the heavy defect image 451 is detected is also specified. The same applies to the heavy defect image 451 associated with the inspection data set 441 of substrate type B and the heavy defect image 451 associated with the inspection data set 441 of substrate type C.

[0027] Incidentally, the printed circuit board 9 is usually manufactured in manufacturing lots. A manufacturing lot is a collection of a plurality of printed circuit boards 9 of the same board type manufactured under the same conditions. One manufacturing lot includes, for example, 1000 printed circuit boards 9. In the manufacture of the printed circuit board 9, due to variations between lots of materials, the influence of temperature and humidity, etc., unique changes may occur for each manufacturing lot. For example, due to variations in solder resist and silk ink, the colors of the SR region and the silk region may change for each manufacturing lot. Also, due to the influence of temperature and humidity during the formation of the solder resist layer, the size of the exposed pads may change for each manufacturing lot. Thus, when unique changes occur for each manufacturing lot, even if the same type of inspection is performed on the same inspection area 92 using the same inspection threshold, the false alarms (detection of false defects) may increase depending on the manufacturing lot. Therefore, it is preferable to adjust the inspection threshold according to the unique changes for each manufacturing lot. Hereinafter, the process of inspecting the printed circuit board 9 while adjusting the inspection threshold for each manufacturing lot will be described.

[0028] FIGS. 5A and 5B are diagrams showing the flow of inspection of the printed circuit board 9 by the inspection apparatus 1. Here, the process of inspecting the printed circuit boards 9 included in the manufacturing lot (hereinafter referred to as the "target manufacturing lot"), which is the first manufacturing lot of the new board type D for which the inspection data set 441 has not yet been created, will be described.

[0029] In the inspection of the printed circuit board 9, a plurality of inspection data sets 441 respectively used for inspection processes for a plurality of board types are stored in advance in the inspection data storage unit 44 and prepared (step S10). Here, the inspection data set 441 for board type A, the inspection data set 441 for board type B, and the inspection data set 441 for board type C are stored in the inspection data storage unit 44.

[0030] Also, in the serious defect image storage unit 45 shown in FIG. 4, serious defect images 451 associated with the inspection data set 441 of substrate type A, serious defect images 451 associated with the inspection data set 441 of substrate type B, and serious defect images 451 associated with the inspection data set 441 of substrate type C are stored and prepared (step S11). The serious defect image 451 corresponding to substrate type A is an image identified as a true defect by an operator in the inspection of the production lot of substrate type A, which was performed before the inspection of the target production lot. Similarly, the serious defect image 451 corresponding to substrate type B is an image identified as a true defect by an operator in the inspection of the production lot of substrate type B, and the serious defect image 451 corresponding to substrate type C is an image identified as a true defect by an operator in the inspection of the production lot of substrate type C.

[0031] In the inspection apparatus 1 of FIG. 1, some printed circuit boards 9 (for example, 20 to 30 printed circuit boards 9) included in the target production lot are carried into the apparatus main body 2, and a plurality of imaging images showing the some printed circuit boards 9 are acquired by the imaging unit 21. Each imaging image in this processing example is an image showing the entire printed circuit board 9 (the entire front and back or one side), but may also be an image showing a part of the printed circuit board 9. As will be described later, since each of the plurality of imaging images is for a trial inspection, hereinafter, it is referred to as a "trial inspection image". The plurality of trial inspection images are transmitted from the apparatus main body 2 to the auxiliary unit 4 via the network 8 and stored in an inspection image storage unit (not shown).

[0032] Subsequently, under the control of the trial inspection control unit 41, in the auxiliary inspection unit 43, inspection processing for a plurality of trial inspection images is executed as a trial inspection (step S12). In the trial inspection, an inspection data set for the substrate type D of the target production lot (hereinafter referred to as the "target inspection data set") is used. In this processing example, the target production lot is the first production lot of the substrate type D, and at the current time, there is no inspection data set for the substrate type D (that is, the target inspection data set). For example, an operator designates a plurality of inspection regions 92 on the printed circuit board 9 via the input unit 36 and specifies the region type of each inspection region 92, thereby creating a target inspection data set. At this time, the type of inspection for each region type is determined in advance. Also, for each region type, an inspection threshold for detecting the heavy defect image 451 as a defect is prepared in advance as an inspection threshold recipe. The created target inspection data set is stored in the inspection data storage unit 44.

[0033] FIG. 6 is a diagram showing a plurality of inspection regions 92 on the printed circuit board 9 represented by the target inspection data set for the substrate type D. The target inspection data set includes at least one inspection region 92 of the first region type, at least one inspection region 92 of the second region type, and at least one inspection region 92 of the third region type. The target inspection data set may be automatically created by the trial inspection control unit 41 specifying the region type of each region of the trial inspection image based on a predetermined algorithm. When creating the target inspection data set, an image obtained by averaging a plurality of captured images may be created as a master image.

[0034] In the trial inspection for a plurality of trial inspection images using the target inspection dataset, for each inspection area 92 (the area corresponding thereto) in the trial inspection image, an inspection of a type corresponding to the area type of the inspection area 92 is performed, and defects are detected based on an inspection threshold. For example, when the inspection area 92 is a pad area, the minimum distance between pads is compared with the inspection threshold. Thus, in the inspection area 92 of the first area type, the inspection for the first area type is performed using the inspection threshold of the first area type. Similarly, in the inspection area 92 of the second area type, the inspection for the second area type is performed using the inspection threshold of the second area type, and in the inspection area 92 of the third area type, the inspection for the third area type is performed using the inspection threshold of the third area type.

[0035] When the trial inspection is completed, the result of the trial inspection is displayed on the display unit 35 (step S13). The result of the trial inspection includes, for example, an image of a defect detected by the trial inspection. An area in the master image indicating the same range as the detected defect may be displayed together with the image of the defect. The image of the defect is preferably displayed in a state where the area type can be specified, and may be displayed separately for each area type. As the result of the trial inspection, the number of defects detected for each area type, etc. may be displayed.

[0036] The results of the trial inspection displayed on the display unit 35 are confirmed by the operator. If there is a defect among the defects displayed on the display unit 35 that the operator determines to be a true defect, an input indicating that the defect is a true defect is made by the operator via the input unit 36 (step S14). For example, the operator selects the defect on the display unit 35 and right-clicks the mouse 36b, and by selecting the registration of a true defect from the displayed menu, the defect can be easily registered as a true defect. In this way, when the input unit 36 receives an input designating a certain defect as a true defect, the image of the defect (true defect) is stored in the multiple-defect image storage unit 45 as a new multiple-defect image 451 by the trial inspection control unit 41 (step S15). The multiple-defect image 451 is associated with the target inspection data set (that is, the inspection data set for the substrate type D) and the detected inspection area 92. If there is no defect that the operator determines to be a true defect, the above input is not made (step S14).

[0037] Also, based on the results of the trial inspection, if the operator determines that it is necessary to change the inspection threshold, the inspection threshold is changed by the operator via the input unit 36 (step S16). For example, if many of the defects displayed on the display unit 35 are present in the inspection area 92 of the first area type and are false alarms, the inspection criteria for the inspection area 92 of the first area type are relaxed (that is, the range determined to be normal becomes wider), and the inspection threshold is changed. In the example where the inspection area 92 is a pad area, the inspection threshold for the minimum distance between pads is decreased. Similarly, for the inspection area 92 of the second area type and the inspection area 92 of the third area type, if there are many false alarms, the inspection threshold is changed. When the input unit 36 receives an input for changing the inspection threshold, inspection processing for a plurality of trial inspection images using the target inspection data set for which the inspection threshold has been changed is executed as a new trial inspection, and the results of the trial inspection are displayed on the display unit 35 (steps S12, S13).

[0038] In the result of the trial inspection, if there are defects that the operator determines to be true defects, in the same manner as above, the images of the defects are stored in the serious defect image storage unit 45 as new serious defect images 451 (steps S14, S15). Also, in the result of the trial inspection, if the operator determines that the false alarm is not sufficiently reduced and the inspection threshold needs to be changed again, the inspection threshold is further changed (step S16). Then, a new trial inspection is further executed using the target inspection data set for which the inspection threshold has been changed, and the result of the trial inspection is displayed on the display unit 35 (steps S12, S13). The change of the inspection threshold and the execution of a new trial inspection are repeated until the operator determines that the change of the inspection threshold is unnecessary (steps S12 to S16). As a result, an adjusted target inspection data set with the inspection threshold adjusted is obtained so that the false alarm can be reduced. Actually, an input indicating that the change of the inspection threshold is unnecessary is received by the input unit 36, and the target inspection data set at that time becomes the adjusted target inspection data set. Hereinafter, the inspection threshold indicated by the adjusted target inspection data set is referred to as the "adjusted inspection threshold".

[0039] When the adjusted target inspection data set for the substrate type D is obtained, in the confirmation inspection control unit 42, the adjusted inspection threshold indicated by the adjusted target inspection data set is applied to the inspection threshold corresponding to the adjusted inspection threshold in the other inspection data sets 441 for the other substrate types A to C. For example, the inspection threshold for the inspection area 92 of the first area type in the inspection data set 441 of the substrate type A is replaced with the adjusted inspection threshold of the first area type in the adjusted target inspection data set. Similarly, the inspection threshold for the inspection area 92 of the second area type is replaced with the adjusted inspection threshold of the second area type, and the inspection threshold for the inspection area 92 of the third area type is replaced with the adjusted inspection threshold of the third area type. As a result, an applied inspection data set for the substrate type A is obtained (step S17). The applied inspection data sets for the substrate type B and the substrate type C are also obtained in the same manner.

[0040] When an applied inspection data set for substrate types A to C is acquired, an inspection process using the applied inspection data set for a defective image 451 associated with an inspection data set 441 other than the target inspection data set is executed as a confirmation inspection (step S18). For example, for each defective image 451 of substrate type A shown in FIG. 4, an inspection process is performed using the applied inspection data set of substrate type A. At this time, when the defective image 451 is associated with, for example, an inspection area 92 of the first area type, an inspection for the first area type is performed on the defective image 451, and it is determined whether it is a defect using the adjusted inspection threshold value of the first area type. For a defective image 451 associated with an inspection area 92 of the second area type, an inspection for the second area type is performed using the adjusted inspection threshold value of the second area type. For a defective image 451 associated with an inspection area 92 of the third area type, an inspection for the third area type is performed using the adjusted inspection threshold value of the third area type. Similarly, for each defective image 451 of substrate type B, an inspection process is performed using the applied inspection data set of substrate type B, and for each defective image 451 of substrate type C, an inspection process is performed using the applied inspection data set of substrate type C.

[0041] When the confirmation inspection is completed, the result of the confirmation inspection is displayed on the display unit 35 (step S19). For example, among the plurality of defective images 451 for which the confirmation inspection has been performed, a defective image 451 that is not detected as a defect (hereinafter referred to as "undetected defective image 451") is displayed on the display unit 35. The undetected defective image 451 is preferably displayed in a state where its area type (that is, the area type of the inspection area 92 with which the defective image 451 is associated) can be specified. While all the defective images 451 are arranged and displayed, for the undetected defective images 451, their presence may be emphasized, for example, by surrounding them with a thick line. When there is an undetected defective image 451 (step S20), a notification prompting readjustment of the adjusted inspection threshold value is issued by the confirmation inspection control unit 42 (step S21). In one example, a message prompting readjustment of the adjusted inspection threshold value is displayed on the display unit 35. The notification prompting readjustment of the adjusted inspection threshold value may also be made by voice or the like.

[0042] When a notification prompting readjustment of the adjusted inspection threshold is given, the adjusted inspection threshold in the adjusted target inspection data set is changed by the operator via the input unit 36 (step S22). That is, an input for changing the adjusted inspection threshold is received by the input unit 36. For example, the adjusted inspection threshold is changed so that the inspection criteria for the inspection area 92 of the area type of the undetected serious defect image 451 become stricter (that is, the range determined to be normal becomes narrower).

[0043] When the adjusted inspection threshold in the adjusted target inspection data set is changed to a new adjusted inspection threshold, the new adjusted inspection threshold is applied to the inspection threshold corresponding to the adjusted inspection threshold in the applied inspection data sets of substrate types A to C. That is, in the applied inspection data sets of substrate types A to C, the inspection threshold for the inspection area 92 of each area type is replaced with the new adjusted inspection threshold of the corresponding area type in the adjusted target inspection data set. Then, an inspection process for the serious defect image 451 using the new applied inspection data set is executed as a new confirmation inspection (step S18), and the result of the confirmation inspection is displayed on the display unit 35 (step S19).

[0044] In the result of the confirmation inspection, the change of the adjusted inspection threshold and the execution of the confirmation inspection using the new applied inspection data set and the like are repeated (steps S18 to S22) until the undetected serious defect image 451 no longer exists. As a result, an adjusted target inspection data set in which all the serious defect images 451 for which the confirmation inspection has been performed can be detected as defects is obtained. The adjusted target inspection data set becomes an inspection data set for mass production inspection for the target production lot described later, and the target inspection data set in the inspection data storage unit 44 is updated to the inspection data set.

[0045] On the one hand, in the result of the initial verification inspection, if there is no undetected serious defect image 451 (step S20), the adjusted inspection threshold is not readjusted, and the adjusted target inspection dataset directly becomes the inspection dataset for mass production inspection for the target manufacturing lot. As described above, the inspection dataset for mass production inspection for the target manufacturing lot is obtained only when there is no undetected serious defect image 451 in the verification inspection. In other words, in the verification inspection, if there is an undetected serious defect image 451, the verification inspection control unit 42 does not permit the execution of mass production inspection, and if there is no undetected serious defect image 451, the verification inspection control unit 42 permits the execution of mass production inspection.

[0046] The inspection dataset for mass production inspection is transmitted to the apparatus main body 2 in FIG. 1 and stored in the main body inspection unit 23. In the apparatus main body 2, a plurality of imaging images indicating a plurality of printed circuit boards 9 included in the target manufacturing lot are sequentially acquired by the imaging unit 21. In the main body inspection unit 23, inspection processing for the plurality of imaging images is executed as mass production inspection (step S23). In the mass production inspection, the inspection dataset for mass production inspection is used. Preferably, the mass production inspection is performed partially in parallel with the acquisition of the plurality of imaging images. For the printed circuit board 9 for which the trial inspection image has been acquired among the plurality of printed circuit boards 9 included in the target manufacturing lot, the trial inspection image may be directly used for the mass production inspection.

[0047] The results of the mass production inspection are sent to the auxiliary unit 4 and displayed on the display unit 35 (step S24). The results of the mass production inspection include, for example, images of the defects detected by the mass production inspection. Regions in the master image indicating the same range as the detected defect may be displayed together with the image of the defect. The results of the mass production inspection displayed on the display unit 35 are confirmed by the operator. If there are defects among those displayed on the display unit 35 that the operator determines to be true defects, an input indicating that the defect is a true defect is made by the operator via the input unit 36 in the same manner as in step S14 (step S25). As a result, the image of the defect is stored in the critical defect image storage unit 45 as a new critical defect image 451 (step S26). The inspection dataset and the detected inspection region 92 are associated with the critical defect image 451. If there are no defects that the operator determines to be true defects, the above input is not made (step S25). When the inspection process for the captured images of all the printed circuit boards 9 included in the target manufacturing lot is completed, the inspection of the printed circuit boards 9 by the inspection apparatus 1 is completed.

[0048] In the inspection apparatus 1, steps S24 to S26 in FIG. 5B are omitted, and the information on the defects detected by the mass production inspection may be output to, for example, an external defect confirmation apparatus. The defect information includes, for example, the image of the defect, the identification information of the printed circuit board 9 including the defect, and the position information on the defective printed circuit board 9. In the defect confirmation apparatus, the region of the defect on the printed circuit board 9 is imaged with reference to the defect information and displayed on the display unit. By the operator confirming the defects included in the displayed image, a determination is made as to whether the defect is a true defect or a false alarm.

[0049] In the above processing example, since the first production lot of the new substrate type D is the inspection target, in the confirmation inspection, there are almost no serious defect images 451 of the substrate type D, and inspection processing is performed on the serious defect images 451 of the other substrate types A to C. On the other hand, if an inspection data set 441 for the substrate type of the target production lot has already been created and there are serious defect images 451 of the substrate type, in the confirmation inspection, inspection processing for the serious defect images 451 of the substrate type may be performed in addition to the inspection processing for the serious defect images 451 of the other substrate types.

[0050] In the inspection apparatus 1 of FIG. 1, the apparatus main body 2 and the auxiliary unit 4 are provided separately. However, for example, the configuration of the auxiliary unit 4 may be realized by a computer included in the apparatus main body 2. In this case, the main body inspection unit 23 and the auxiliary inspection unit 43 are realized by one inspection unit. In other words, in the inspection apparatus 1 of FIG. 1 in which the apparatus main body 2 and the auxiliary unit 4 are provided separately, it can be said that the inspection unit has a first inspection unit (auxiliary inspection unit 43) that executes a trial inspection and a confirmation inspection, and a second inspection unit (main body inspection unit 23) that executes a mass production inspection.

[0051] When the apparatus main body 2 including the imaging unit 21 and the second inspection unit and the auxiliary unit 4 including the trial inspection control unit 41, the confirmation inspection control unit 42, the first inspection unit, the inspection data storage unit 44, and the serious defect image storage unit 45 are provided separately as in the above example, it becomes possible to perform imaging (acquisition of imaging images for mass production inspection) of a plurality of printed circuit boards 9 in the apparatus main body 2 in parallel with the processing of the above steps S12 to S22 in the auxiliary unit 4. As a result, the operation rate of the apparatus main body 2 is improved, and the efficiency related to the inspection of the printed circuit board 9 is improved.

[0052] As described above, the inspection unit of the inspection apparatus 1 in FIG. 1 uses an inspection data set 441 that indicates the position information of a plurality of inspection regions 92 on the printed circuit board 9 and also indicates the type of inspection and the inspection threshold for each inspection region 92, and performs an inspection process on the captured image of the printed circuit board 9 to detect defects in the printed circuit board 9. The inspection data storage unit 44 stores a plurality of inspection data sets 441 respectively used for inspection processes for a plurality of substrate types. When inspecting one production lot, the trial inspection control unit 41 causes the inspection unit (in FIG. 1, the auxiliary inspection unit 43) to perform, as a trial inspection, an inspection process on the captured images of some of the printed circuit boards 9 in the production lot using the target inspection data set for the substrate type of the production lot. Further, when the input unit 36 receives an input for changing the inspection threshold based on the result of the trial inspection, a new trial inspection is performed on the inspection unit using the target inspection data set for which the inspection threshold has been changed, thereby obtaining an adjusted target inspection data set for which the inspection threshold has been adjusted. In this way, in the inspection apparatus 1, by adjusting the inspection threshold based on the result of the trial inspection, it is possible to suppress an increase in false alarms due to specific changes occurring for each production lot.

[0053] Further, the serious defect image storage unit 45 stores a plurality of serious defect images 451 each associated with one of the plurality of inspection data sets 441. The confirmation inspection control unit 42 applies the adjusted inspection threshold indicated by the adjusted target inspection data set to the inspection threshold corresponding to the adjusted inspection threshold in another inspection data set 441 for another substrate type to obtain an applied inspection data set. Then, the inspection unit (in FIG. 1, the auxiliary inspection unit 43) is caused to perform, as a confirmation inspection, an inspection process on the serious defect image 451 associated with the other inspection data set 441 stored in the serious defect image storage unit 45 using the applied inspection data set. Thereby, in the inspection apparatus 1, it is possible to easily confirm whether a serious defect can be detected by the adjusted inspection threshold in the target inspection data set, that is, whether the adjustment of the inspection threshold is appropriate. Also, since there is no need to prepare a substrate for serious defect confirmation, it is possible to reduce the production cost and management cost of the substrate for serious defect confirmation.

[0054] Preferably, the inspection unit (the main body inspection unit 23 in FIG. 1) executes inspection processing using the adjusted target inspection data set on the captured images of a plurality of printed circuit boards 9 included in the manufacturing lot as mass production inspection. By using the adjusted target inspection data set for which detection of serious defects has been confirmed by the confirmation inspection in the mass production inspection, it is possible to reduce false alarms and detect serious defects in the mass production inspection.

[0055] Preferably, when there is a serious defect image not detected by the confirmation inspection, the confirmation inspection control unit 42 does not permit the inspection unit to execute the mass production inspection. Thereby, it is possible to prevent the mass production inspection from being performed using an inappropriate adjusted target inspection data set and overlooking serious defects.

[0056] Preferably, when there is a serious defect image not detected by the confirmation inspection, the confirmation inspection control unit 42 gives a notification prompting readjustment of the adjusted inspection threshold value in the adjusted target inspection data set. Thereby, it is possible to obtain an adjusted target inspection data set capable of reducing false alarms while detecting serious defects, and it is possible to more reliably realize an appropriate mass production inspection.

[0057] Preferably, when the input unit 36 receives an input designating that a certain defect detected in the trial inspection is a true defect, the image of the defect is stored in the serious defect image storage unit 45 as the serious defect image 451 in a state associated with the target inspection data set. Similarly, when the input unit 36 receives an input designating that a certain defect detected in the mass production inspection is a true defect, the image of the defect is stored in the serious defect image storage unit 45 as the serious defect image 451 in a state associated with the adjusted target inspection data set. In either case, it is possible to easily prepare an appropriate serious defect image 451 confirmed by the operator.

[0058] In the above processing example, in a plurality of substrate types corresponding to a plurality of inspection data sets 441, the type of surface treatment of the pad region (copper flux, gold, solder, tin, etc.) and the color of the solder resist (green, black, blue, white, etc.) are the same. Therefore, the heavy defect images 451 of all inspection data sets 441 other than the target inspection data set are used for the confirmation inspection. On the other hand, when there are substrate types in which at least one of the type of surface treatment of the pad region and the color of the solder resist is different from other substrate types among the plurality of substrate types, the heavy defect images 451 used for the confirmation inspection are preferably restricted according to the substrate type of the target inspection data set.

[0059] In the example of FIG. 7, a plurality of inspection data sets 441 for a plurality of substrate types A to D are grouped into a first group 44a and a second group 44b. The first group 44a includes the inspection data sets 441 of substrate types A and C (denoted as "inspection data set A" and "inspection data set C" in FIG. 7). In substrate types A and C, the type of surface treatment and the color of the solder resist are the same as each other. The second group 44b includes the inspection data sets 441 of substrate types B and D (denoted as "inspection data set B" and "inspection data set D" in FIG. 7). In substrate types B and D, the type of surface treatment and the color of the solder resist are the same as each other, but are different from substrate types A and C.

[0060] For example, when the inspection data set 441 of substrate type D is the target inspection data set, in step S17 of FIG. 5A, an applied inspection data set is obtained only from other inspection data sets 441 belonging to the same second group 44b as the target inspection data set, that is, the inspection data set 441 of substrate type B. In other words, the applied inspection data set is not obtained from the inspection data sets 441 of substrate types A and C belonging to the first group 44a. Thereafter, an inspection process using the applied inspection data set of substrate type B for the serious defect image 451 of substrate type B is executed as a confirmation inspection (step S18). In step S18, the confirmation inspection for the serious defect images 451 of substrate types A and C is not performed. The subsequent processing is the same as the above processing.

[0061] As described above, in this processing example, a plurality of inspection data sets 441 are classified into a plurality of groups 44a and 44b regarding the type of surface treatment of the corresponding substrate type and the color of the solder resist. Then, the inspection data sets 441 used for the confirmation inspection are limited to those belonging to the same group as the target inspection data set. Thereby, only the serious defect images 451 of the inspection data sets 441 in the same group as the target inspection data set are used in the confirmation inspection. As a result, it is possible to prevent the differences in the type of surface treatment and the color of the solder resist from affecting the result of the confirmation inspection, and it is possible to appropriately confirm the appropriateness of adjusting the inspection threshold in the target inspection data set. Note that depending on the type of inspection used in the inspection apparatus 1, etc., a plurality of inspection data sets 441 may be classified into a plurality of groups by paying attention to only one of the type of surface treatment of the corresponding substrate type or the color of the solder resist. The number of groups may be three or more.

[0062] Various modifications are possible in the above inspection apparatus 1 and inspection method.

[0063] In the above-described embodiment, all the images specified as true defects by the operator are treated as serious defect images. However, for example, those with a high degree of importance among the true defects may be selected by the operator and stored in the serious defect image storage unit 45 as serious defect images. As described above, when the input unit 36 receives an input designating a certain defect detected in the trial inspection or mass production inspection as a serious defect, the image of the defect may be stored in the serious defect image storage unit 45 as the serious defect image 451. The serious defect image 451 may be acquired other than in the trial inspection and mass production inspection.

[0064] When there is a serious defect image not detected by the confirmation inspection, the operator may determine whether to execute the mass production inspection by the inspection unit. Further, a notification prompting readjustment of the adjusted inspection threshold value may be performed according to other conditions such as when a specific serious defect image is not detected.

[0065] The object to be inspected in the inspection apparatus 1 may be a substrate such as a semiconductor substrate or a glass substrate in addition to the printed circuit board 9.

[0066] The configurations in the above-described embodiment and each modification example may be appropriately combined as long as they do not conflict with each other.

Description of Reference Numerals

[0067] 1 Inspection apparatus 9 Printed circuit board 23 Main body inspection unit 36 Input unit 41 Trial inspection control unit 42 Confirmation inspection control unit 43 Auxiliary inspection unit 44 Inspection data storage unit 44a, 44b Group 45 Serious defect image storage unit 92 Inspection area 441 Inspection data set 451 Serious defect image S10~S26 Steps

Claims

1. An inspection apparatus for inspecting a substrate, using an inspection dataset that indicates the position information of a plurality of inspection regions on the substrate and also indicates the type of inspection and the inspection threshold for each inspection region, and performing an inspection process on the captured image of the substrate to detect defects of the substrate; an inspection unit, an inspection data storage unit that stores a plurality of inspection datasets respectively used for the inspection processes for a plurality of substrate types; a critical defect image storage unit that stores a plurality of critical defect images each associated with any one of the plurality of inspection datasets; an input unit that receives an input for changing the inspection threshold; When inspecting one production lot, causing the inspection unit to perform the inspection process on the captured images of some substrates in the production lot using the target inspection dataset for the substrate type of the production lot as a trial inspection, and when the input unit receives an input for changing the inspection threshold based on the result of the trial inspection, causing the inspection unit to perform a new trial inspection using the target inspection dataset with the changed inspection threshold, thereby obtaining an adjusted target inspection dataset with the adjusted inspection threshold; a trial inspection control unit, applying the adjusted inspection threshold indicated by the adjusted target inspection dataset to the inspection threshold corresponding to the adjusted inspection threshold in other inspection datasets for other substrate types to obtain an applied inspection dataset, and causing the inspection unit to perform the inspection process using the applied inspection dataset on the critical defect images associated with the other inspection datasets stored in the critical defect image storage unit as a confirmation inspection; a confirmation inspection control unit, An inspection apparatus comprising:

2. The inspection apparatus according to claim 1, wherein the inspection unit performs the inspection process using the adjusted target inspection dataset on the captured images of a plurality of substrates included in the production lot as a mass production inspection.

3. The inspection apparatus according to claim 2, wherein when there is a serious defect image not detected by the confirmation inspection, the confirmation inspection control unit does not permit the mass production inspection by the inspection unit.

4. The inspection apparatus according to claim 2, wherein when there is a serious defect image not detected by the confirmation inspection, the confirmation inspection control unit gives a notification prompting readjustment of the adjusted inspection threshold in the adjusted target inspection data set.

5. The inspection apparatus according to claim 2, wherein when the input unit receives an input designating a defect detected in the mass production inspection as a serious defect, the image of the defect is stored in the serious defect image storage unit as a serious defect image in a state associated with the adjusted target inspection data set.

6. The inspection apparatus according to claim 1, wherein when the input unit receives an input designating a defect detected in the trial inspection as a serious defect, the image of the defect is stored in the serious defect image storage unit as a serious defect image in a state associated with the target inspection data set.

7. The inspection apparatus according to any one of claims 1 to 6, wherein the plurality of inspection data sets are classified into a plurality of groups according to the type of surface treatment of the corresponding substrate type and / or the color of the solder resist, and the other inspection data set used for the confirmation inspection is limited to those belonging to the same group as the target inspection data set.

8. An inspection method for inspecting a substrate by an inspection unit, wherein the inspection unit uses an inspection data set indicating position information of a plurality of inspection regions on the substrate and indicating the type of inspection and the inspection threshold for each inspection region, and performs inspection processing on the captured image of the substrate to detect defects of the substrate. the inspection method comprises a) preparing a plurality of inspection data sets respectively used for the inspection processes for a plurality of substrate types; b) preparing a plurality of heavy defect images each associated with any one of the plurality of inspection data sets; c) when inspecting one production lot, causing the inspection unit to perform, as a trial inspection, the inspection process on the captured images of some substrates of the production lot using the target inspection data set for the substrate type of the production lot; d) when the inspection threshold of the target inspection data set is changed based on the result of the trial inspection, obtaining an adjusted target inspection data set with the adjusted inspection threshold by causing the inspection unit to perform a new trial inspection using the target inspection data set with the changed inspection threshold; e) applying the adjusted inspection threshold indicated by the adjusted target inspection data set to the inspection threshold corresponding to the adjusted inspection threshold in other inspection data sets for other substrate types to obtain an applied inspection data set; f) causing the inspection unit to perform, as a confirmation inspection, the inspection process using the applied inspection data set on the heavy defect images associated with the other inspection data sets. An inspection method comprising the above steps.

Images