Inspection system

Abstract

The present invention makes it possible to secure sufficient time for inspecting a subject in an inspection system. An inspection system (1) is provided with: a conveyance line (32) that conveys a plurality of subjects (5); an imaging unit (40) that is provided on the conveyance line (32), images the plurality of subjects (5), and acquires image data (V(k)) for each of the subjects (5); an image acquisition unit (110) that stores the image data in a database unit (120); a warehouse system (50) that collects the subjects (5) after being imaged and then transports the subjects (5); and an inspection unit (130) that outputs a determination result (D(k)) indicating whether each of the subjects (5) corresponding to the image data stored in the database unit (120) is a non-defective product (5A) or a defective product (5B).

Description

Inspection system

[0001] The present invention relates to an inspection system.

[0002] As the background art of this technical field, in the summary of Patent Document 1 below, "[Problem] It is desired to identify individual mass-produced products and to properly use the analysis and verification at a later time of the inspection results obtained on the production line. [Solution] An IC chip 2 is attached to each product 1 of mass production. An ID number is stored in this IC chip 2. The ID number is a product-specific number assigned to each product. On the other hand, during or after production, the product is inspected, and this inspection data and inspection results are classified for each product by the ID number and stored as a database. Then, it is retrieved for product verification at a later stage and used for verification and analysis. Further, since the inspection and sorting on the inspection line 3 are performed at different positions (timings), an ID reader 9 is provided for this timing to match the timings, and this record is referred to later, and the inspection results are referred to from the ID added to the subject." is described.

[0003] Japanese Patent Application Laid-Open No. 2004-37259

[0004] By the way, in the above-described technology, since there are significant time constraints for the inspection performed on the production line, it may be difficult to ensure sufficient time for the inspection. This invention has been made in view of the above circumstances, and an object thereof is to provide an inspection system capable of ensuring sufficient time for inspecting a subject.

[0005] To solve the above problems, the inspection system of the present invention includes a transport line for transporting multiple samples, an imaging unit provided on the transport line for photographing the multiple samples and acquiring image data for each, an image acquisition unit for storing the image data in a database unit, a warehouse system for accumulating and then transporting the samples after they have been photographed, an inspection unit for outputting a determination result indicating whether each of the samples corresponding to the image data stored in the database unit is a good product or a defective product, a sorting command unit for outputting a sorting command signal based on the determination result, and sorting the good products from among the multiple samples based on the sorting command signal. The system comprises a sorting unit that supplies defective products to a line and supplies the defective products to a defective product line, a counting unit that outputs a count value which is the result of counting the number of subjects discharged from the warehouse system, an image acquisition unit that stores the image data in a database unit in correspondence with the shooting order, a warehouse system that has a function to discharge the subjects in the order in which they were stored, and a sorting command unit that outputs a sorting command signal corresponding to the count value based on the judgment result corresponding to the count value, and is characterized in that the processing in the warehouse system and the processing in the inspection unit are executed in parallel.

[0006] According to the present invention, sufficient time can be secured for the examination of the subject.

[0007] This is a diagram illustrating the configuration of the inspection system according to the first embodiment. This is a diagram illustrating the operation of the first embodiment. This is a schematic front view of the product in the second embodiment. This is a diagram illustrating the configuration of the inspection system according to the second embodiment. This is a diagram illustrating the configuration of the inspection system according to the third embodiment. This is a schematic front view of the product and the simulated product in the third embodiment. This is a block diagram of the computer.

[0008] [Summary of Embodiments] Applying the contents of Patent Document 1 described above, it is considered possible to perform 100% inspection by in-line inspection at the same speed as the manufacturing speed of the test subject in the test subject manufacturing line, and to sort good products from defective products. However, in recent years, the production capacity of the test subject has been increasing. Therefore, when performing 100% inspection by in-line inspection, the time that can be spent inspecting each test subject becomes limited. In order to realize in-line inspection while addressing this limitation, it is necessary to apply more high-performance and expensive inspection equipment, or to use methods such as parallelizing the inspection equipment.

[0009] Therefore, in the embodiment described later, the subject is photographed inline to acquire image data, and the subject is stored in the warehouse system. Then, inspection based on the acquired image data is performed separately offline. After the inspection is complete, the subject is released from the warehouse system and sorted into good and defective products according to the inspection results. In this way, in order to store the subject in the warehouse system, it is necessary to maintain the identity of the "photographed subject," the "subject stored in the warehouse system," and the "subject released from the warehouse system."

[0010] To maintain the identity of the subjects in this way, it is conceivable to attach an IC chip with an ID number stored in it to each subject, for example, as shown in Patent Document 1. Alternatively, even if an ID number is not assigned to the subjects, identity can be maintained by guaranteeing "first-in, first-out" in the warehouse system, that is, by ensuring that the order in which items are stored in the warehouse matches the order in which they are retrieved from the warehouse. In the embodiment described later, the warehouse system ensures sufficient time for inspecting the subjects while maintaining their identity.

[0011] [First Embodiment] <Configuration of the First Embodiment> Figure 1 is a configuration diagram of the inspection system 1 according to the first embodiment. In Figure 1, the inspection system 1 includes an incoming line 30, a transport line 32, a transport line 34, an imaging unit 40, a warehouse system 50, a transport line 60, a counting unit 62, a sorting unit 64, a good product line 72, a labeler 74, a defective product line 75, a defective product tray 76, a shipping line 78, and a control device 100. The control device 100 also includes an image acquisition unit 110, a database unit 120, an inspection unit 130, and a sorting command unit 140.

[0012] The input line 30 transports the product 5 (subject) manufactured by the manufacturing equipment (not shown) to the transport line 32. The product 5 is, for example, a container such as a vial, ampoule, or syringe filled with liquid. However, the product 5 is not limited to the above examples. The transport line 32 transports the product 5 transported from the input line 30 to the transport line 34. An imaging unit 40 is provided along the transport line 32. The imaging unit 40 photographs the product 5 as it passes through and acquires image data.

[0013] The warehouse system 50 is a warehouse system capable of automatic loading and unloading, storing the products 5 that are brought in and unloading the products 5 in the same order in which they were brought in. In other words, the warehouse system 50 guarantees "first-in, first-out" storage. The warehouse system 50 performs stacking processing on the products 5 that are brought in. Here, "stacking processing" means arranging the products 5 that are brought in in a predetermined order on flat pallets (not shown), and stacking the pallets on which the products 5 are arranged in the vertical direction. By performing this stacking processing, the warehouse system 50 can store a large number of products 5 in a relatively small space.

[0014] Products 5 discharged from the warehouse system 50 are transported to the sorting unit 64 via the transport line 60. The counting unit 62 counts the number of products 5 discharged onto the transport line 60 and supplies the resulting count value m (shooting order) to the control device 100. The control device 100, through a process described later, supplies the sorting unit 64 with a sorting signal SEL(m) indicating whether the product 5 with count value m is a good product 5A or a defective product 5B.

[0015] Based on the sorting signal SEL(m), the sorting unit 64 transports good products 5A to the good product line 72 and defective products 5B to the defective product line 75. The defective products 5B are then collected in the defective product tray 76 via the defective product line 75. The labeler 74 affixes product labels (not shown) to the good products 5A supplied via the good product line 72 and transports them to the shipping equipment (not shown) via the shipping line 78. In this shipping equipment, processes such as assembling boxes to store the good products 5A, inserting the good products 5A and instruction manuals into the assembled boxes, and packaging are performed.

[0016] The image acquisition unit 110 in the control device 100 counts the number of images acquired from the shooting unit 40 and the number of products 5 that have been photographed, and obtains a count value k as a result. The image acquisition unit 110 then stores the acquired image data as image data V(k) corresponding to the count value k in the database unit 120. If the number of products 5 that have been photographed is N, then the count value k will be in the range of "1 ≤ k ≤ N". As a result, the database unit 120 stores N images of image data V(k).

[0017] The inspection unit 130 performs inspection, i.e., a pass / fail judgment, on image data V(k) in the range "1 ≤ k ≤ N", and obtains a judgment result D(k) corresponding to each image data V(k). The judgment result D(k) is information indicating whether the product 5 with count value k is a good product 5A or a defective product 5B. The sorting command unit 140 receives the count value m from the count unit 62 and, based on the judgment result D(m) corresponding to this count value m, supplies the sorting signal SEL(m) described above to the sorting unit 64.

[0018] Here, we will describe various specific examples of inspections performed by the inspection unit 130. In this embodiment, it is sufficient that the judgment result D(k) be determined between the time the imaging unit 40 acquires image data of the product 5 and the time the product 5 is unloaded from the warehouse system 50. Therefore, various methods can be used for the inspection, as described below.

[0019] • Inspection method #1: Image data V(k) is displayed sequentially on a display, and an inspector visually determines whether the product is good or bad, recording the inspection results in the database unit 120. This method allows for more time, enabling inspectors to inspect the product with ample time and reducing their workload. It also allows for the inspection to be adapted to the situation, thus enabling the inspector to make informed decisions. Furthermore, visual inspection does not need to be performed at the factory where the inspection system 1 is installed; it can be performed, for example, via the internet at an overseas factory or a private home.

[0020] • Inspection method #2: A general inspection device is applied as the inspection unit 130, and this inspection device determines whether the product 5 is good or bad based on the image data V(k). According to this embodiment, sufficient time can be secured for inspection, so a slow and inexpensive inspection device can be applied.

[0021] • Inspection method #3: A general-purpose computer is used as the inspection unit 130, and the quality of product 5 is determined based on image data V(k) through software processing. Techniques such as image comparison and machine learning can be applied to the software processing. According to this embodiment, sufficient time can be secured for inspection, so a slow and inexpensive computer can be used. Also, since there are no restrictions on the inspection location, it is possible to execute software processing from a server room or via the internet.

[0022] • Inspection Method #4: Alternatively, inspection methods #1 to #3 described above may be performed in parallel, and the results from each method may be collected to make an overall pass / fail judgment. In this case, methods such as prioritizing the results of visual inspection by inspectors or making a decision by majority vote may be employed.

[0023] <Operation of the First Embodiment> Figure 2 is an explanatory diagram of the operation of the first embodiment. In Figure 2, the white arrows indicate the flow of product 5, and the dashed line indicates the flow of information. In step S2 of Figure 2, the manufacturing apparatus (not shown) manufactures product 5. Next, in step S4, the imaging unit 40 and the image acquisition unit 110 of the control device 100 acquire image data V(k) corresponding to the count value k.

[0024] Next, in step S6, the warehouse system 50 performs an accumulation process on the product 5 from which image data V(k) has been acquired. That is, the product 5 is arranged on flat pallets, and these pallets are stacked vertically. Next, in step S8, the warehouse system 50 performs a storage process on the product 5. That is, the product 5 is kept stationary for a predetermined time (for example, 1 hour or more).

[0025] Next, in step S10, the inspection unit 130 performs inspection processing. That is, it inspects the image data V(k) by image analysis and obtains a judgment result D(k). In this embodiment, since the image data V(k) is stored in the database unit 120, the processing period of steps S2 to S8 and the processing period of step S10 can be separated. Therefore, for example, the processing of steps S2 to S8 may be performed during the daytime when the warehouse system 50 etc. is in operation, and the inspection processing of step S10 may be performed at night when the warehouse system 50 etc. is stopped.

[0026] Next, in step S20, the warehouse system 50 performs an unloading process to unload the products 5 that have finished inspection processing to the transport line 60. Next, in step S22, the sorting command unit 140 and the sorting unit 64 perform a sorting process to sort the products 5 into good products 5A and defective products 5B. That is, the sorting unit 64 unloads the good products 5A to the good product line 72 and the defective products 5B to the defective product line 75.

[0027] Next, in step S24, the labeler 74 applies labels to the good products 5A that have been transported from the good product line 72. Then, in step S26, the shipping equipment (not shown) processes the shipping of the good products 5A with labels attached.

[0028] As described above, the inspection process in step S10 can be performed, for example, at night. In this case, steps S2 to S8 may be performed on products 5 manufactured on the day, and in parallel with this, steps S20 to S26 may be performed on products 5 manufactured the previous day and stored in the warehouse system 50. Thus, according to this embodiment, the processing in the warehouse system 50 (steps S6, S8, S20) and the processing in the inspection unit 130 (step S10) can be performed in parallel. This allows sufficient time to be secured for the inspection process in step S10, enabling precise inspection.

[0029] [Second Embodiment] Next, a second embodiment will be described. In the description of each embodiment, the same reference numerals are used for parts corresponding to parts of the other embodiments described above, and their descriptions may be omitted. Figure 3 is a schematic front view of product 5 in the second embodiment. Product 5 comprises a bottle 82, a cap 84, and a liquid (not shown) which is the contents of the bottle 82. The cap 84 is provided with an identification information display area 86 on which an identification information ID that uniquely identifies product 5 is printed. The identification information display area 86 may be provided in other locations, for example, on the top surface of the cap 84, as long as it does not interfere with the inspection of product 5. Alternatively, instead of providing an identification information display area 86, an IC tag (not shown) may be attached to the cap 84, for example, as shown in Patent Document 1.

[0030] Figure 4 is a diagram showing the configuration of the inspection system 2 according to the second embodiment. The differences from the inspection system 1 of the first embodiment (see Figure 1) will be explained below. In the inspection system 2, instead of the imaging unit 40, warehouse system 50, and counting unit 62 in the inspection system 1, an imaging unit 42, a warehouse system 52, and an ID acquisition unit 68 (identification information acquisition unit) are provided. The imaging unit 42 photographs the product 5 as it passes through the transport line 32, acquires the image data, and reads the identification information ID from the identification information display field 86 (see Figure 3) of the product 5.

[0031] Furthermore, the ID acquisition unit 68 acquires an identification information ID from the identification information display field 86 of the product 5 that has been shipped from the warehouse system 50 to the transport line 60. The identification information ID read by this ID acquisition unit 68 is specifically called the shipping identification information ID m. Also, the warehouse system 52 collects and stores the product 5 in the same way as the warehouse system 50 in the inspection system 1. However, in the warehouse system 52, it is not necessary to guarantee the "first-in, first-out" order of the product 5.

[0032] Furthermore, the image acquisition unit 110 in the inspection system 2 acquires image data and identification information ID from the imaging unit 42. The image acquisition unit 110 then stores the image data in the database unit 120 as image data V (ID) corresponding to the identification information ID.

[0033] The inspection unit 130 acquires a judgment result D (ID) corresponding to the image data V (ID) for all identification information IDs. However, the inspection unit 130 in the inspection system 2 does not include the area of ​​the identification information display field 86 on the cap 84 of the product 5 in its defect judgment. The sorting command unit 140 receives the discharge identification information IDm from the ID acquisition unit 68 and supplies a sorting signal SEL (IDm) to the sorting unit 64 based on the judgment result D (IDm) corresponding to this discharge identification information IDm. As a result, the sorting unit 64 discharges good products 5A to the good product line 72 and defective products 5B to the defective product line 75 based on the sorting signal SEL (IDm). The configuration of the inspection system 2 other than those described above is the same as that of the inspection system 1 in the first embodiment.

[0034] According to this embodiment, based on the identification information ID assigned to product 5, it is possible to associate the captured image data V (ID) with the judgment result D (ID). Therefore, as described above, the warehouse system 52 does not need to guarantee "first-in, first-out" for product 5, and the management of product 5 can be simplified compared to the warehouse system 50 of the first embodiment.

[0035] [Third Embodiment] Next, a third embodiment will be described. Figure 5 is a diagram of the configuration of the inspection system 3 according to the third embodiment. The configuration of the inspection system 3 is the same as that of the inspection system 1 of the first embodiment (see Figure 1), except for the points described below. First, in the inspection system 3, a simulated product mixing section 22 (simulated subject mixing section) is provided instead of the input line 30 of the inspection system 1 (see Figure 1). The simulated product mixing section 22 selects either the product 5 or the simulated product 6 (simulated subject) and delivers it to the transport line 32. Details of the product 5 and the simulated product 6 will be described later.

[0036] In Figure 5, product 5 (which includes both good product 5A and defective product 5B) is represented by a white circle, and the simulated product 6 is represented by a hatched circle. The simulated product mixing unit 22 discharges one simulated product 6 for every M products 5 discharged to the conveyor line 32. Here, the numerical value M is a natural number in the range of "3" to "10000", and the numerical value M does not necessarily have to be constant. In addition, in the inspection system 3, a control device 200 is provided instead of the control device 100.

[0037] Figure 6 is a schematic front view of product 5 and simulated product 6 in the third embodiment. Product 5 in this embodiment includes a bottle 82 and a cap 84, similar to that of the second embodiment (see Figure 3), but the cap 84 does not have an identification information display area 86. Simulated product 6 in this embodiment also includes a bottle 82 and a cap 84, similar to product 5 in the second embodiment. Furthermore, the cap 84 of simulated product 6 is provided with an identification information display area 86 for displaying an identification information ID, similar to product 5 in the second embodiment. This identification information ID uniquely identifies the simulated product 6. Alternatively, instead of providing an identification information display area 86, as in product 5 in the second embodiment, an IC tag (not shown) may be attached to the cap 84.

[0038] Returning to Figure 5, the inspection system 3 is equipped with a simulated product detection unit 66 (simulated subject detection unit) that detects simulated products 6 in the good product line 72 and the defective product line 75. The simulated product detection unit 66 detects simulated products 6 based on whether or not an identification information display field 86 is provided on the cap 84 (see Figure 6), and reads the identification information ID of the simulated product 6 from the identification information display field 86.

[0039] Furthermore, the control device 200 has the same elements as the control device 100 in the first embodiment (see Figure 1). The image acquisition unit 110 stores the image data V(k) corresponding to the cumulative count value k of product 5 and simulated product 6 in the database unit 120. The inspection unit 130 acquires a judgment result D(k) corresponding to each image data V(k), similar to that of the first embodiment.

[0040] In this embodiment, the inspection unit 130 performs a quality determination on the image data V(k), including the portion of the cap 84. Specifically, the inspection unit 130 compares the standard appearance of the cap 84 with the appearance of the cap 84 included in the image data V(k), and if the difference between the two is large, the judgment result D(k) for the image data V(k) is set to "defective product 5B". When the image data V(k) is data for a simulated product 6, its cap 84 is provided with an identification information display area 86 (see Figure 6), so it is clearly different from the standard appearance of the cap 84 in a normal product 5. Therefore, the inspection unit 130 determines that the simulated product 6 is a defective product 5B, and the judgment result D(k) is also "defective product 5B".

[0041] The control device 200 in the inspection system 3 further includes a simulated product recognition unit 210 and a matching unit 250. The simulated product recognition unit 210 determines whether the image data V(k) of the count value k is data from product 5 or simulated product 6 based on whether or not the image data V(k) of the count value k contains the image of the identification information display field 86. If the image data V(k) is data from simulated product 6, it reads the identification information ID from the identification information display field 86 and outputs it as the identification information ID(k) corresponding to the count value k.

[0042] The matching unit 250 stores the identification information ID(k) output from the simulated product recognition unit 210. Then, the matching unit 250 determines whether the "first-in, first-out" in the warehouse system 50 is maintained normally by matching the detection result of the simulated product detection unit 66 and the recognition result of the simulated product recognition unit 210. As described above, in the inspection unit 130, the simulated product 6 is recognized as "defective product 5B". Therefore, if the inspection system 3 is normal, as shown in FIG. 5, all the simulated products 6 should be carried out to the defective product line 75 together with the actual defective products 5B.

[0043] Suppose that the simulated product 6 is conveyed to the non-defective product line 72 and this is detected by the simulated product detection unit 66. This means that some abnormality has occurred in the inspection system 3. For example, it is conceivable that some failure has occurred in the warehouse system 50 and the "first-in, first-out" is no longer maintained. Therefore, in this case, the matching unit 250 generates an alarm indicating that an error has occurred.

[0044] Also, if the inspection system 3 is normal, the order of appearance of the identification information ID(k) recognized by the simulated product recognition unit 210 should match the order of appearance of the identification information ID recognized by the simulated product detection unit 66 in the defective product line 75. If a discrepancy occurs in the order of appearance of the identification information ID, this also means that some error has occurred in the inspection system 3. Therefore, in this case as well, the matching unit 250 generates an alarm indicating that an error has occurred. The configuration of the inspection system 3 other than that described above is the same as that of the inspection system 1 (see FIG. 1) of the first embodiment.

[0045] Thus, in this embodiment, since the simulated product 6 is appropriately mixed into the product 5 conveyed on the conveyance line 32, when a failure occurs in the warehouse system 50 and the "first-in, first-out" is not maintained, an alarm to that effect can be generated.

[0046] [Configuration of Computer] FIG. 7 is a block diagram of a computer 980. Each of the control devices 100, 200, etc. in the above-described first to third embodiments includes one or more computers 980 shown in FIG. 7. In FIG. 7, the computer 980 includes a CPU 981, a storage unit 982, a communication port 983, an input / output port 984, and a media port 985. Here, the storage unit 982 includes a RAM 982a, a ROM 982b, and a SSD (Solid State Drive) 982c. The communication port 983 is connected to a communication circuit 986. The input / output port 984 is connected to an input / output device 987. The media port 985 reads and writes data from / to a recording medium 988.

[0047] The ROM 982b stores an IPL (Initial Program Loader) and the like executed by the CPU. The SSD 982c stores application programs and various data. The CPU 981 realizes various functions by executing the application programs and the like read from the SSD 982c into the RAM 982a. The interiors of the control devices 100, 200 shown in FIGS. 1, 4, and 5 are mainly shown as blocks of functions realized by application programs and the like.

[0048] [Modification Example] The present invention is not limited to the above-described embodiments, and various modifications are possible. The above-described embodiments are exemplified for easy understanding and explanation of the present invention, and are not necessarily limited to those having all the configurations described. Also, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Further, a part of the configuration of each embodiment can be deleted, or other configurations can be added or replaced. Also, the control lines and information lines shown in the figures indicate those considered necessary for explanation, and do not necessarily show all the control lines and information lines required on the product. In fact, it may be considered that almost all the configurations are interconnected. Possible modifications to the above embodiments are, for example, as follows.

[0049] (1) Since the hardware of the control devices 100 and 200 in each of the above embodiments can be implemented using a general-purpose computer, the programs that perform the various processes described above may be stored in a storage medium (a computer-readable recording medium on which the program is recorded) or distributed via a transmission line.

[0050] (2) Although the various processes described above were explained as software processes using a program in the above embodiment, some or all of them may be replaced with hardware processes using an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array), etc.

[0051] (3) The various processes performed in the above embodiment may be performed by a server computer via a network (not shown), and the various data stored in the above embodiment may also be stored on the server computer.

[0052] [Effects of the Embodiments] As described above, the inspection systems 1, 2, and 3 of each embodiment include a transport line 32 for transporting a plurality of subjects (5), imaging units 40 and 42 provided on the transport line 32 for photographing the plurality of subjects (5) and acquiring image data (V(k), V(ID)) for each, an image acquisition unit 110 for storing the image data (V(k), V(ID)) in a database unit 120, warehouse systems 50 and 52 for transporting the subjects (5) after they have been photographed and then collecting them, and a database unit 120 for determining whether each subject (5) corresponding to the image data (V(k)) stored in the database unit 120 is a good product 5A or a defective product 5 The system includes an inspection unit 130 that outputs a determination result (D(k), D(ID)) indicating whether it is B, a sorting command unit 140 that outputs sorting command signals (SEL(m), SEL(IDm)) based on the determination result (D(k), D(ID)), and a sorting unit 64 that supplies good products 5A from among the multiple subjects (5) to the good product line 72 and defective products 5B to the defective product line 75 based on the sorting command signals (SEL(m), SEL(IDm)), and the processing in the inspection unit 130 (step S10) can be executed in parallel. In this way, in each embodiment, the processing in the warehouse systems 50, 52, which collects and then transports the subjects (5) after they have been photographed, and the processing in the inspection unit 130 can be executed in parallel, so that sufficient time can be secured for the inspection in the inspection unit 130.

[0053] Furthermore, as in the inspection systems 1 and 3 of the first and third embodiments, the image acquisition unit 110 stores image data (V(k)) in the database unit 120 in association with the shooting order (k), the warehouse system 50 has a function to unload the subjects (5) in the order in which they were stored, and it is even more preferable that the warehouse system 50 further includes a count unit 62 that outputs a count value m which is the result of counting the number of subjects (5) unloaded from the warehouse system 50, and the sorting command unit 140 outputs a sorting command signal (SEL(m)) corresponding to the count value m based on the judgment result (D(m)) corresponding to the count value m. In this way, by unloading the subjects (5) in the order in which they were stored, the identity between the photographed subjects (5) and the subjects (5) to be sorted can be maintained even if identification information is not assigned to the subjects (5).

[0054] Furthermore, as in the inspection system 2 of the second embodiment, the imaging unit 42 further includes a function to acquire identification information IDs assigned to each subject (5), the image acquisition unit 110 stores image data (V(ID)) in the database unit 120 in association with the identification information IDs, and it is even more preferable that the system further includes an identification information acquisition unit (68) that acquires the removal identification information IDm, which is the identification information ID of the subject (5) removed from the warehouse system 52, and the sorting command unit 140 outputs a sorting command signal (SEL(IDm)) based on the judgment result (D(ID)) corresponding to the removal identification information IDm. As a result, the warehouse system 52 does not need to match the order in which the subject (5) is brought in with the order in which it is removed, thus simplifying the management of the subject (5).

[0055] Furthermore, as in the inspection system 3 of the third embodiment, it is even more preferable to further include a simulated subject mixing unit (22) that mixes multiple simulated subjects (6) which are determined to be defective products 5B by the inspection unit 130 with multiple subject products (5) and supplies them to the transport line 32. This makes it possible to detect abnormalities in the inspection system 3 by checking whether or not the defective product line 75 contains simulated subjects (6).

[0056] Furthermore, it is even more preferable that the inspection system 3 further includes a simulated subject detection unit (66) that detects when a simulated subject (6) is supplied to the good product line 72. This allows for the detection of abnormalities in the inspection system 3 by confirming whether or not a simulated subject (6) is included in the good product line 72.

[0057] Furthermore, the inspection system 3 is more preferably equipped with a matching unit 250 that monitors the order in which the identification information IDs appear, and the simulated subject detection unit (66) has a function to acquire the identification information ID assigned to each simulated subject (6) supplied to the defective product line 75. This makes it possible to detect abnormalities in the inspection system 3 by monitoring the order in which the identification information IDs appear in the defective product line 75.

[0058] 1, 2, 3 Inspection System 5 Product (Subject) 5A Good Product 5B Defective Product 6 Dummy Product (Dummy Subject) 22 Dummy Product Contamination Section (Dummy Subject Contamination Section) 32 Transport Line 40, 42 Imaging Section 50, 52 Warehouse System 62 Counting Section 64 Sorting Section 66 Dummy Product Detection Section (Dummy Subject Detection Section) 68 ID Acquisition Section (Identification Information Acquisition Section) 72 Good Product Line 75 Defective Product Line 110 Image Acquisition Section 120 Database Section 130 Inspection Section 140 Sorting Command Section 250 Verification Section k Count Value (Shooting Order) m Count Value ID Identification Information IDm Dispatch Identification Information

Claims

1. The system comprises: a transport line for transporting multiple subjects; an imaging unit provided on the transport line for photographing multiple subjects and acquiring image data for each; an image acquisition unit for storing the image data in a database unit; a warehouse system for accumulating and then transporting the subjects after they have been photographed; an inspection unit for outputting a determination result indicating whether each subject corresponding to the image data stored in the database unit is a good product or a defective product; a sorting command unit for outputting a sorting command signal based on the determination result; a sorting unit for supplying the good products from among the multiple subjects to a good product line and the defective products to a defective product line based on the sorting command signal; and a counting unit for outputting a count value which is the result of counting the number of subjects transported out of the warehouse system. The image acquisition unit stores the image data in the database unit in correspondence with the shooting order; the warehouse system has a function for transporting the subjects in the order in which they were stored; and the sorting command unit outputs a sorting command signal corresponding to the count value based on the determination result corresponding to the count value. An inspection system characterized by performing the processing in the warehouse system and the processing in the inspection unit in parallel.

2. The inspection system according to claim 1, further comprising a simulated subject mixing unit that mixes multiple simulated subjects, which are determined to be defective by the inspection unit, with multiple subject samples and supplies them to the transport line.

3. The inspection system according to claim 2, further comprising a simulated subject detection unit that detects when a simulated subject is supplied to the good product line.

4. The inspection system according to claim 3, wherein the simulated subject detection unit has a function to acquire identification information assigned to each of the simulated subjects supplied to the defective product line, and further comprises a matching unit that monitors the order in which the identification information appears.

Images