JP6052448B2 - Inspection processing apparatus, inspection system, inspection processing method and program - Google Patents

Description

  The present invention relates to an inspection processing apparatus, an inspection system, an inspection processing method, and a program.

  Whether or not the product to be shipped is the product described in the purchase order compared to the purchase order before shipping the product in the process of taking out and shipping the distribution product handled by the logistics company such as warehousing or wholesale. There is work to confirm. This work is called inspection work. Among the inspection operations, there are also operations for inspecting a plurality of items such as “100 parts of the same catalog” and “60 parts of the same box”. When inspecting such products, in order to ship the correct quantity of the products on the purchase order, the products are checked only on the top or back of the target product bundle, and the remaining products are checked. It is common to count the number and confirm that it is the correct quantity.

  Specifically, a method is conceivable in which the product placed on top is visually checked and the number of products is counted by hand. Also, a method of automating the matching operation of products by taking an image to be inspected with a camera and collating with an image of a product registered in advance and automatically confirming a match or mismatch can be considered. Patent Document 1 describes an inspection processing apparatus to which such a method is applied. According to this method, first, a data table that defines the quantity of merchandise for each merchandise and the method of stacking the merchandise is prepared. Next, an image in which the product is stacked is captured by the camera, and the product and the stacking method are specified based on the image. And the quantity of goods is calculated | required with reference to the table prepared previously using the packing method specified at the end as a search key. If this inspection method is used, the worker can save labor for both the check of goods and the work of counting quantities.

JP 2013-67499 A

  However, the types of products handled in the logistics industry often exceed 10,000 types, and there is a high possibility that similar products exist on the image. Since an image of a product changes depending on various conditions at the time of imaging, if the number of similar images increases, the target product cannot often be specified by image matching. In such a case, it is conceivable to display images of these similar products as candidates, and the worker selects the target product visually. However, in this case, it is conceivable that the frequency of work relying on visual inspection increases and the work efficiency decreases. In the first place, it is conceivable that the verification time for extracting similar products from a large amount of products becomes long, waiting time is generated, and inspection efficiency does not increase.

  Therefore, an object of the present invention is to provide an inspection processing apparatus, an inspection system, an inspection processing method, and a program that solve the above-described problems.

  The present invention acquires a feature amount of an image obtained by capturing an image of a target product and a unit weight of the target product, and classifies each group of products whose feature amount and unit weight of a predetermined product image are similar to each other. Based on the feature amount of the captured image and the unit weight of the target product based on the group determination unit that selects the group to which the target product falls, and based on the feature amount of the captured image and the unit weight, A collation unit that identifies the target product from among the products belonging to the selected group, and the group determination unit corresponds to the target product when the group to which the target product corresponds cannot be selected. This is an inspection processing apparatus for newly creating a group to be processed.

  Moreover, this invention is an inspection system provided with the imaging device which images the image of the target goods to ship, the weight measuring device which measures the weight of the target goods to ship, and said inspection processing apparatus.

  Further, according to the present invention, the inspection processing apparatus acquires a feature amount of an image obtained by imaging the target product and a unit weight of the target product, and a product whose feature amount and unit weight of a predetermined product image are similar to each other A group corresponding to the target product is selected based on the feature amount of the captured image and the unit weight of the target product from the group classified for each, and based on the feature amount of the captured image and the unit weight An inspection processing method for newly creating a group to which the target product corresponds when the target product is specified from among the products belonging to the selected group and the group to which the target product corresponds cannot be selected. It is.

  In the present invention, the computer of the inspection processing apparatus acquires the feature amount of the image obtained by capturing the target product and the unit weight of the target product, and the feature amount and unit weight of the predetermined product image are similar to each other. Means for selecting a group corresponding to the target product based on the feature amount of the captured image and the unit weight of the target product from the groups classified for each product to be processed, the feature amount of the captured image and the unit A means for identifying the target product from among the products belonging to the selected group based on the weight, and if the group to which the target product falls cannot be selected, a new group to which the target product falls is created It is a program for making it function as a means to do.

  According to the present invention, the effect of increasing the efficiency of inspection work can be obtained.

It is a block diagram which shows the minimum structure of the inspection processing apparatus by 1st embodiment of this invention. It is a block diagram which shows an example of a structure of the inspection processing apparatus by 1st embodiment of this invention. It is a figure which shows an example of the inspection processing system by 1st embodiment of this invention. It is a 1st figure which shows the processing flow of the inspection processing apparatus by 1st embodiment of this invention. It is a figure which shows an example of the information which the database used with the inspection processing apparatus by 1st embodiment of this invention memorize | stores. It is a figure which shows an example of the inspection confirmation screen by 1st embodiment of this invention. It is a figure which shows an example of the goods selection screen by 1st embodiment of this invention. It is a figure which shows an example of the goods registration screen by 1st embodiment of this invention. It is a block diagram which shows an example of the hardware constitutions of the inspection processing apparatus by 1st embodiment of this invention.

<First embodiment>
Hereinafter, an inspection processing apparatus according to an embodiment of the present invention will be described with reference to FIGS.
FIG. 1 is a diagram showing a minimum configuration of an inspection processing apparatus according to the first embodiment.
In this figure, reference numeral 100 represents an inspection processing apparatus. As shown in FIG. 1, the inspection processing apparatus 100 includes at least a verification unit 110. The inspection processing apparatus 100 is, for example, a computer including a CPU (Central Processing Unit).
The collation unit 110 acquires a feature amount of an image obtained by capturing the target product to be inspected and a unit weight of the target product, and specifies the target product from at least the feature amount and the unit weight.

FIG. 2 is a block diagram showing an example of the configuration of the inspection processing apparatus according to the first embodiment.
As shown in FIG. 2, the inspection processing apparatus 100 includes a control determination unit 101, a weight acquisition unit 102, a product image acquisition unit 103, an image buffer unit 104, an input reception unit 105, a display information output unit 106, Feature amount calculation unit 107, unit weight quantity calculation unit 108, group information storage unit 109, verification unit 110, verification data storage unit 111, group determination unit 112, product information registration unit 113, and product information deletion Part 114.

The control determination unit 101 instructs information transmission and processing to other functional units.
The weight acquisition unit 102 acquires the weight of the inspection target product measured by the weight measuring instrument from the weight measuring instrument.
The product image acquisition unit 103 instructs the imaging device to capture an inspection target product, and acquires an image captured from the imaging device.
The image buffer unit 104 is an area for storing an image captured by the product image acquisition unit 103.
The input reception unit 105 acquires information input by the user and outputs the information to the control determination unit 101. The information input by the user is, for example, the quantity of the inspection target product.
The display information output unit 106 generates various information necessary for the inspection work and outputs the information to a display device connected to the inspection processing apparatus 100.
The feature amount calculation unit 107 calculates the feature amount of the image captured by the product image acquisition unit 103 by a general feature amount calculation method such as calculating a graphic feature amount.
The unit weight quantity calculation unit 108 calculates the unit weight of the target product by dividing the weight acquired from the weight acquisition unit 102 by the quantity of the inspection target product received by the input reception unit 105. The unit weight quantity calculation unit 108 divides the acquired weight by the unit weight to calculate the quantity of the product.

The collation data storage unit 111 stores, for each product, a product name, a product image, a feature amount of the product image, a unit weight, information indicating a part that is a feature that distinguishes the product in the captured image, and the like. It is a database.
The group information storage unit 109 divides each feature amount and unit weight of each product image registered in the verification data storage unit 111 into a certain range, and stores the group number in association with the group obtained by combining them. It is a database.
Based on the feature amount of the image calculated by the feature amount calculation unit 107 and the unit weight calculated by the unit weight quantity calculation unit 108, the group determination unit 112 determines the feature amount of the image in the predetermined range and the unit of the predetermined range. A group corresponding to the target product is selected from the groups classified by weight.
The product information registration unit 113 registers product information such as a product name and unit weight in the collation data storage unit 111.
The merchandise information deleting unit 114 deletes information on products that have not been inspected for a certain period or longer from the collation data storage unit 111.
Note that at least a part of the control determination unit 101 to the product information deletion unit 114 is a function provided by the CPU included in the inspection processing apparatus 100 reading and executing a program from a storage unit such as a hard disk.

FIG. 3 is a diagram showing an example of an inspection processing system according to the first embodiment of the present invention.
Hereinafter, the inspection processing system of this embodiment will be described with reference to FIG.
In this figure, reference numeral 1 represents an inspection processing system. As shown in FIG. 3, the inspection processing system 1 includes an inspection processing device 100, an imaging device 200, a weight measuring device 300, a display device 400, and an input device 500. Reference numeral 600 indicates a product to be inspected.

  The imaging device 200 is, for example, a camera that images an inspection target product. The weight measuring device 300 is a device that measures the weight of a product to be inspected. In FIG. 3, the imaging device 200 is attached to the support column 302 with the lens facing the ground so that an inspection target product placed on the weight measuring device 300 can be imaged from directly above. The support column 302 is fixed by a base 301, and the weight measuring device 300 is installed on the base 301. The attachment position of the imaging device 200 is not limited to the example of this figure. The inspection processing apparatus 100 is connected to the imaging apparatus 200 and acquires an image of the inspection target product captured by the imaging apparatus 200. The inspection processing apparatus 100 is connected to the weight measuring device 300 and acquires the weight of the inspection target product measured by the weight measuring device 300.

  The display device 400 is a display, for example. The display device 400 is connected to the inspection processing device 100 and displays information necessary for the inspection work generated by the inspection processing device 100. The input device 500 is a keyboard, a mouse, a touch panel, or the like. The worker operates the inspection processing apparatus 100 using the input device 500 while referring to information displayed on the display device 400 to perform inspection work.

Next, a rough flow of the inspection work will be described with reference to FIG. First, the worker places the inspection target product 600 on the weight measuring device 300. Then, the weight measuring device 300 transmits the measured weight data to the inspection processing apparatus 100. The inspection processing apparatus 100 transmits an imaging instruction signal to the imaging apparatus 200 at the timing when the weight data is received. When receiving the imaging instruction signal, the imaging apparatus 200 captures the inspection target product 600 and transmits the captured image to the inspection processing apparatus 100.
In parallel with these processes, the worker inputs the quantity of the inspection target product 600 to be inspected using the input device 500. The quantity to be inspected is, for example, the quantity of the product described in the order slip and the shipment quantity of the product.

Next, the inspection processing apparatus 100 calculates the unit weight of the inspection target product by dividing the weight data by the input quantity. Then, the inspection processing apparatus 100 uses the acquired image and the calculated unit weight to read a database included in the apparatus and search for a product. If the inspection target product 600 uniquely matches the registered product, the inspection processing device 100 displays the product image and product name registered in the display device 400.
Further, when the inspection target product 600 does not uniquely match the registered product, the inspection processing apparatus 100 displays a list of similar products. At this time, the inspection processing apparatus 100 outputs an image of each product together with an emphasis display for conspicuous the difference between them. For example, when the inspection target product 600 is a catalog booklet and the product number on the catalog is different from a similar product, the highlighting means displaying such as enclosing the part of the product number with a rectangular frame. The part to be highlighted is designated in advance in a product registration operation described later.
The worker visually checks the product images and product names of the plurality of product lists displayed on the display device 400, and selects a product that matches the product to be inspected. If the inspection object does not exist in the displayed image list, it is considered that the inspection object is not registered, so a product registration operation described later is performed.

The inspection processing apparatus 100 calculates the excess or deficiency of the inspection target product 600 placed on the weight measuring device 300 for the product specified through such a process, and outputs the calculation result to the display device 400 together with the product image. Calculation of excess and deficiency in quantity is as follows. First, the inspection processing apparatus 100 reads the unit weight of the specified product from the verification data storage unit 111. Next, the inspection processing apparatus 100 divides the weight acquired from the weight measuring device 300 by the read unit weight, and obtains the quantity of the inspection target product 600 placed on the weight measuring device 300. Then, the inspection processing apparatus 100 calculates the difference between the quantity obtained in this way and the quantity (inspection quantity) of the inspection target product 600 to be inspected input by the user. This difference is an excess or deficiency of the quantity described above. The unit weight of the product is registered in the verification data storage unit 111 in advance.
Next, if the excess / deficiency calculated in this way is not zero, the worker adjusts the quantity by adding or reducing the inspection target product 600 of the excess / deficiency quantity on the weight measuring device 300. Do. As a result of the above work and processing, if the worker determines that the displayed image matches the object to be inspected, and the excess / deficiency displayed on the display device 400 becomes zero, the inspection work for the product is completed. is there.

  Processing in which the inspection processing apparatus 100 searches for products will be described in more detail. As a specific example, it is assumed that, for example, the inspection target product 600 is a catalog booklet, and the order quantity (inspection quantity) of this catalog booklet is 100 copies. The operator places a catalog booklet of 100 copies on the weight measuring device 300 and inputs 100 as the inspection quantity from the input device 500. Then, the inspection processing apparatus 100 captures an image of the catalog booklet. Further, the inspection processing apparatus 100 calculates the weight per book and narrows down the group of products to which the catalog booklet belongs from the captured image and the calculated unit weight. The quantity of the product placed on the weight measuring device 300 may not necessarily be the correct quantity. However, if the quantity error is small, for example, about several percent, the error of the target unit weight is about several percent, If the unit weight grouping has an allowable value, product grouping can be performed correctly. Thereafter, the inspection processing apparatus 100 performs collation within the group, and refers to the correct unit weight from the registered data subjected to coincidence collation. Then, the inspection processing apparatus 100 displays the excess or deficiency of the catalog quantity based on the correct unit weight, inspection quantity, and weight data. The subsequent processing is as described above.

FIG. 4 is a diagram showing a processing flow of the inspection processing apparatus according to the first embodiment of the present invention.
The processing of the inspection processing apparatus in the inspection work described above will be described using the processing flow of FIG.
First, when the inspection target product 600 is placed on the weight measuring device 300, the weight acquisition unit 102 acquires the weight data of the inspection target product 600 measured by the weight measuring device 300 (step S21). The weight acquisition unit 102 outputs the acquired weight data to the control determination unit 101. The control determination unit 101 records weight data in a memory.
Next, the control determination unit 101 outputs an imaging instruction signal to the product image acquisition unit 103 when the weight data is acquired (step S22). Then, the product image acquisition unit 103 issues an imaging instruction to the imaging device 200, acquires an image of the inspection target product 600 captured by the imaging device 200 (step S23), and stores the image in the image buffer unit 104.

When the image is stored in the image buffer unit 104, the control determination unit 101 instructs the feature amount calculation unit 107 to calculate the feature amount of the stored image. The feature amount calculation unit 107 reads an image from the image buffer unit 104, calculates a feature amount by a predetermined known feature calculation method (step S24), and outputs the calculation result to the control determination unit 101. The control determination unit 101 records the feature amount in the memory.
Next, the control determination unit 101 obtains the quantity data to be inspected input by the worker from the input reception unit 105 and records the quantity data in the memory (step S25). Note that the process of step S25 may be performed in parallel with the processes of steps S21 to S24.
When the weight data and the quantity data are acquired, the control determination unit 101 instructs the unit weight quantity calculation unit 108 to calculate. The unit weight quantity calculation unit 108 reads the weight data and the quantity data from the memory, and divides the weight data by the quantity data to calculate a unit weight (step S26). This unit weight data is the unit weight of the inspection target product 600 when it is assumed that the quantity of the inspection target product 600 placed on the weight measuring device 300 is equal to the quantity data acquired in step S25. The unit weight quantity calculation unit 108 outputs the calculated unit weight data to the control determination unit 101, and the control determination unit 101 records the unit weight data in the memory.

Next, the control determination unit 101 outputs the feature amount and unit weight data to the group determination unit 112, and requests the group number to which the inspection target product 600 belongs. The group determination unit 112 searches the group information storage unit 109 using the acquired feature amount and unit weight data, and acquires a group number (step S27).
Here, information stored in the group information storage unit 109 will be described with reference to FIG. FIG. 5A is an example of information stored in the group information storage unit 109. In this example, it is assumed that the feature amount of the image calculated by the feature amount calculation unit 107 is represented by a certain one-dimensional value. Then, as shown in FIG. 5A, the group information storage unit 109 has at least items of a group number, a feature amount, and a unit weight. In the item “feature amount”, for example, a value representing a predetermined range of the feature amount is stored, such as “X100 to X199”. Further, in the item “unit weight”, a value representing a predetermined range of weight per product such as “100 to 119 (g)” is stored. One group number is assigned to a combination of values that define the range of the feature amount and the unit weight. For example, the feature amount acquired by the group determination unit 112 is “X110”, and the unit weight data is “110”. Further, it is assumed that the value “X110” is included in the range of “X100 to X199”. At this time, the group determination unit 112 acquires the group number “001” from the table illustrated in FIG.
The group determination unit 112 searches the group information storage unit 109 and outputs the acquired group number to the control determination unit 101. When the corresponding group number is not registered in the group information storage unit 109, the group determination unit 112 outputs an error code indicating failure of acquisition to the control determination unit 101. The control determination unit 101 records the acquired group number in the memory.

Next, the control determination unit 101 determines whether a group number has been acquired (step S28). If the group number could not be acquired (step S28 = No), the process proceeds to step S36. When the group number can be acquired (step S28 = Yes), the control determination unit 101 outputs the feature amount of the image, the unit weight data, and the acquired group number to the collation unit 110, and instructs to specify the product. The collation unit 110 searches for information on the inspection target product 600 placed on the weight measuring device 300 registered in the collation data storage unit 111 using the acquired three pieces of information (step S29).
Here, the information stored in the collation data storage unit 111 will be described with reference to FIGS. 5B and 5C are examples of information stored in the collation data storage unit 111. FIG. As shown in FIGS. 5B and 5C, the collation data storage unit 111 shows a product name, unit weight data, a product image, a feature amount calculated from the product image, and a part to be highlighted among the product images. This is a database in which information is divided into group numbers and stored. The matching unit 110 first selects an appropriate group based on the acquired group number. If the above example is used, since the group number is “001”, the collation unit 110 first selects the table of FIG. By this selection, the total amount of target data to be collated is narrowed down. Next, the collation unit 110 searches for FIG. 5B using the feature amount “X110” and the unit weight data “110”, and assigns the top data in FIG. 5B. Since the products to be inspected are collated using the database classified by group as described above, the processing speed can be increased. The collation unit 110 outputs the search result to the control determination unit 101. In addition, when there is no feature amount and unit weight data that completely match, the collating unit 110 may allocate data in a range where the feature amount and unit weight data are allowable.

  Next, the control determination unit 101 determines whether the product has been identified based on the search result acquired from the collation unit 110 (step S30). When it cannot be identified (step S30 = No), the process proceeds to step S34. If it can be identified (step S30 = Yes), the control determination unit 101 instructs the display information output unit 106 to output information of the searched product. The display information output unit 106 acquires the “product name”, “unit weight”, “product image”, and “information indicating the highlighted part of the product image” of the product searched by the matching unit 110 from the matching unit 110. To do. Further, the display information output unit 106 outputs the unit weight acquired from the verification unit 110, the weight data recorded in the memory, and the quantity data input by the worker to the unit weight quantity calculation unit 108, and outputs it to the weight measuring device 300. A difference value between the quantity of the inspection target product 600 placed and the inspection quantity is requested. The unit weight quantity calculation unit 108 divides the weight data by the unit weight to calculate the quantity (installed quantity) of the inspection target product 600 placed on the weight measuring device 300. Further, the unit weight quantity calculation unit 108 calculates the difference (quantity error) between the calculated installation quantity and the inspection quantity input by the worker. The unit weight quantity calculation unit 108 outputs the quantity error to the display information output unit 106. The display information output unit 106 uses the information on various products acquired from the collation unit 110 and the quantity error acquired from the unit weight quantity calculation unit 108 to use the confirmation screen (inspection confirmation screen) used by the worker for the inspection work. Are generated and displayed on the display device 400 (step S31).

Here, the inspection confirmation screen will be described with reference to FIG. FIG. 6 is a diagram illustrating an example of an inspection confirmation screen. Reference numeral 61 denotes a product name. Reference numeral 62 represents an inspection quantity. Reference numeral 64 denotes a product image. The worker can check whether the product or quantity to be inspected is correct by comparing it with an order form or the like. Further, the inspection target product 600 imaged by the imaging device 200 may be displayed so that the worker can compare both images and check whether the product has an error.
Reference numeral 63 denotes a quantity error calculated by the unit weight quantity calculation unit 108. This value is the difference between the quantity of the inspection target product 600 currently placed on the weight measuring device 300 and the inspection quantity described in the order form. The worker can grasp the excess and deficiency of the currently placed product by looking at the quantity error. Reference numeral 65 denotes an OK button. When the inspection is completed, the worker presses the button 65 to complete the inspection of this product.

  The worker sees the quantity error 63 and adjusts the quantity of the inspection target product 600 placed on the weight measuring device 300. Then, the unit weight quantity calculation unit 108 reacquires the weight measured by the weight measuring device 300, divides by the previously acquired unit weight, and recalculates the current installation quantity (step S32). The display information output unit 106 displays the recalculated installation quantity at the position indicated by reference numeral 63. The worker adjusts the installation quantity until the quantity error 63 becomes zero. Next, the end of the processing flow is determined (step S33). For example, when the input reception unit 105 detects that the worker presses the button 65 (step S33 = Yes), the process proceeds to step S37. When pressing of the button 65 is not detected (step S33 = No), the processing from step S32 is repeated.

  Steps S37 and S38 will be described. The control determination unit 101 instructs the product information deletion unit 114 to record the date and time when the inspection was performed. The merchandise information deletion unit 114 records the date and time when the inspection was performed in the verification data storage unit 111 (step S37). Next, the control determination unit 101 instructs the product information deletion unit 114 to delete product information that has not been inspected for a certain period of time. The product information deletion unit 114 refers to the date and time of inspection recorded in the collation data storage unit 111, deletes information on products that have not been inspected for a certain period of time (step S38), and the processing flow is as follows. finish.

  A case where Step S30 = No will be described. This is a case where the matching unit 110 cannot uniquely identify the product. In this case, the control determination unit 101 instructs the display information output unit 106 to output information on candidate similar products. The display information output unit 106 requests product information similar to the collation unit 110. The matching unit 110 determines, as a similar product, a product in which the difference between the feature amount or unit weight of the image and those values used in the search condition is within a predetermined range among the searched products. The product belonging to the group of the group number used in the above may be a similar product. The display information output unit 106 acquires a product name, a product image, information on a part to be highlighted, and the like of the similar product determined by the collation unit 110, generates an image of a product selection screen, and displays it on the display device 400 ( Step S34).

Here, the product selection screen will be described with reference to FIG. FIG. 7 is a diagram showing an example of a product selection screen according to the first embodiment of the present invention.
Reference numeral 71 </ b> A is an image of “product 1”. Reference numeral 71 </ b> B is the name of the product “product 1”. Reference numeral 71 </ b> C is highlighting indicating that there is a characteristic part that is distinguished from other products in the image of “product 1”. The highlighted display 71C is displayed in a rectangular frame surrounding a characteristic part as shown in FIG. The same applies to reference numerals 72A to 72C, 73A to 73C, and 74A to 74C. The products denoted by reference numerals 71 to 74 are products that the collation unit 110 has determined to be similar to the inspection target product 600.
Reference numeral 77 </ b> A is an image of a product under inspection stored in the image buffer unit 104. Reference numeral 77C is a highlighted display of this product. The worker compares the images 71A to 74A with the image 77A, and particularly refers to the highlights 71C to 74C and the highlight 77C, and selects the inspection target from among the candidates for “product 1” to “product 4”. Product 600 is selected. Then, for example, the worker selects the product 2 and presses the selection button 76. Then, the display information output unit 106 notifies the control determination unit 101 of information on the uniquely selected product. If an appropriate product is not displayed on the product selection screen, the worker presses the cancel button 75. Then, the display information output unit 106 notifies the control determination unit 101 that the product has not been specified.

  Next, the control determination unit 101 determines whether a product has been identified based on the product selection result acquired from the display information output unit 106 (step S35). If it can be identified (step S35 = Yes), the display information output unit 106 outputs the image of the inspection confirmation screen illustrated in FIG. 6 to the display device 400 for the product selected by the worker (step S31). The subsequent processing is the same as the processing from step S32 described above. If the product cannot be specified (step S36 = Yes), the process proceeds to step S36.

  Step S36 will be described. Step S36 is processing when the group number cannot be acquired at step S28 or when the product cannot be specified at step S35. In these cases, it is considered that the product is not registered in the collation data storage unit 111, so the product registration work is performed. The display information output unit 106 displays a product registration screen used by the worker for product information registration. Then, the product information registration unit 113 registers the information input from the product registration screen in the verification data storage unit 111 (step S36).

Here, the product registration process will be described with reference to FIG. FIG. 8 is a diagram showing an example of a product registration screen according to the first embodiment of the present invention.
Reference numeral 81 denotes a field for inputting a product name. The worker uses the input device 500 to input the product name of the newly registered product in the field 81 or uses the product name obtained from the manufacturer. Reference numeral 82 is a field for inputting a unit weight. The worker inputs the unit weight of the newly registered product in the field 82 using the input device 500. It is assumed that the unit weight information is input from, for example, the weight acquisition unit 102 or obtained from the manufacturer of the product. Reference numeral 83 denotes an image of a product stored in the image buffer unit 104. Note that the image 83 can be input for highlighting. Reference numeral 85 indicates a similar product. If the product cannot be specified in step S35, the similar product displayed on the product selection screen illustrated in FIG. The worker visually compares the displayed image of the similar product with the image 83. Then, the worker uses the input device 500 to perform an input of highlighting such as enclosing a portion that is particularly different from the similar products with a rectangle on the image 83. Reference numeral 86 denotes a registration button. The worker presses the button 86 after finishing the above input. Then, the product information registration unit 113 stores the product name, unit weight, and highlight information received by the input receiving unit 105, the image feature amount and group number recorded in the memory, and the image buffer unit 104. The registered product image is registered in the collation data storage unit 111. The highlight information is, for example, coordinate information of each vertex of the rectangle in the image 83 when the highlight is a rectangle. Further, as described above, the highlighting is used as an auxiliary for comparison in visual comparison between similar products and products to be inspected.

  If the group number cannot be acquired in step S28, the image of the similar product is not displayed in the area 85. In such a case, the group determination unit 112 creates a new group before the registration process by the product information registration unit 113. At this time, the group determination unit 112 may create a new group having a range including the feature amount of the image and a range including the unit weight of the newly registered product, or the group information storage unit 109 has already been created. A new group corresponding to this product may be created by reorganizing a part of the groups registered in. However, if the group is subdivided too much, there is a high possibility of assigning an incorrect group in step S27. Therefore, for example, an upper limit value of the number of groups is set in advance, and the group determination unit 112 does not exceed the upper limit. Create. When the product information registration process is completed, the display information output unit 106 outputs the image of the inspection confirmation screen illustrated in FIG. 6 to the display device 400 for the registered product (step S31). And the process after step S32 is performed.

FIG. 9 is a block diagram showing an example of a hardware configuration of the inspection processing apparatus according to the first embodiment of the present invention.
As shown in FIG. 9, the inspection processing apparatus 100 is realized by a computer apparatus including a CPU 901, a memory 902, a storage device 903 such as a hard disk, and a communication IF 904 (communication interface) for network connection. However, the configuration of the inspection processing apparatus 100 is not limited to the computer apparatus shown in FIG. For example, the inspection processing apparatus 100 may transmit the image or weight data to the outside via the communication IF 904, collate the product to be inspected outside, and receive the result.

According to this embodiment, for example, it is possible to automate the collation of a plurality of products such as catalog booklets and boxes and to omit the work of counting the number of products, thereby speeding up the inspection work and reducing the work time Can be obtained.
More specifically, according to the present embodiment, the product is placed on the weight measuring instrument and the weight is measured, and at the same time, the uppermost product is imaged by the camera provided directly above the weight measuring instrument, and the product is registered in advance. The product is identified by comparing the image and the unit weight data. Thereby, the worker can confirm the coincidence between the inspection product and the order product by comparing with the product on the order form. At the same time, the quantity of the product placed on the weight measuring instrument is calculated and displayed from the weight measured by the weight measuring instrument and the unit weight registered in advance, in comparison with the unit weight of the commodity registered in advance. . Thereby, the worker can confirm the coincidence between the quantity of the inspection product and the quantity of the order product by comparing with the quantity of the purchase order.

In addition, according to the present embodiment, not only images but also images and unit weights are used at the time of verification of products to be inspected, and verification efficiency can be improved by further narrowing down products in stages.
For example, in order to automatically check a bundle of products, it takes a very long time to compare the images of products to be inspected with all the images of products registered in advance. Therefore, the feature amount of the image is calculated at the time of image registration, and the images are classified into, for example, 10 groups according to the difference of the feature amount. At the time of inspection, the feature amount of the inspection product is calculated, and the feature amount is a group of 10 Determine where it applies. If a two-stage search is performed in which detailed comparison between images is performed in the determined group, the speed can be increased by about 10 times in this example. However, on the other hand, this speed-up method has a problem of how far the grouping is performed. For example, if it is classified into 100 groups, the time required for image collation can be expected to be reduced to nearly 1/100. On the other hand, if a group is subdivided, for example, depending on how the product image is captured, it is assigned to the adjacent group. With the problem of misclassification such as classification, the possibility of not having a correct inspection result increases.
As one of means for solving this problem, means for increasing the types of image feature values can be considered. However, considering the possibility that the image at the time of registration of the product image and the image at the time of inspection are affected by the difference in lighting and reflected light from the surface, and that the feature quantity selected from the image may not be independent of each other, Selection of the quantity is generally difficult. On the other hand, in this embodiment, in addition to the feature amount of the image, the unit weight of the product is used as the feature amount. Since the feature quantity of unit weight is independent of the feature quantity of the image, it is an effective index for selecting a group. For example, even products with similar images can be classified into different groups if their unit weights are different.

  In addition, according to the present embodiment, for example, unnecessary product information that has not been inspected for a certain period or more registered in the verification data storage unit 111 is deleted, and the number of products to be verified is reduced. The effect of shortening the time can be obtained.

  Note that the block diagram used in the present embodiment shows functional unit blocks, not hardware unit configurations. These functional blocks are realized by any combination of hardware and software. Further, the inspection processing apparatus 100 may be realized by one physically coupled apparatus, or may be realized by connecting two or more physically separated apparatuses by wire or wirelessly and by these plural apparatuses. Also good.

  The inspection processing apparatus 100 described above has a computer inside. Each process of the inspection processing apparatus 100 described above is stored in a computer-readable recording medium in the form of a program, and the above process is performed by the computer reading and executing the program. Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program.

  The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, what is called a difference file (difference program) may be sufficient. The unit weight quantity calculation unit 108 is an example of a quantity calculation unit.

DESCRIPTION OF SYMBOLS 1 ... Inspection processing system 100 ... Inspection processing apparatus 101 ... Control determination part 102 ... Weight acquisition part 103 ... Product image acquisition part 104 ... Image buffer part 105 ... Input reception part 106 ... Display information output unit 107 ... Feature amount calculation unit 108 ... Quantity calculation unit 109 ... Group information storage unit 110 ... Verification unit 111 ... Verification data storage unit 112 ... Group Determination unit 113 ... product information registration unit 114 ... product information deletion unit 200 ... imaging device 300 ... weight measuring device 400 ... display device 500 ... input device 600 ... inspection target product 901 ... CPU
902 ... Memory 903 ... Storage device 904 ... Communication IF

Claims (10)

The feature amount of the image obtained by capturing the target product and the unit weight of the target product are acquired, and the feature amount and the unit weight of the predetermined product image are classified for each product similar to each other. A group determination unit that selects a group corresponding to the target product based on a feature amount of the captured image and a unit weight of the target product;
A matching unit that identifies the target product from among the products belonging to the selected group based on the feature amount of the captured image and the unit weight;
With
The group determination unit creates a new group to which the target product corresponds when the group to which the target product corresponds cannot be selected.
Inspection processing equipment. The group determination unit includes a feature amount of the image of the target product within a predetermined range defined for a feature amount of an image obtained by capturing the product of the own group when the group to which the target product is applicable cannot be selected. , Create a new group that includes the unit weight of the target product within a predetermined range defined for the unit weight of the product of the own group,
The inspection processing apparatus according to claim 1. The group determination unit corresponds to the target product so that the number of all groups including the newly created group is equal to or less than a predetermined upper limit when the group to which the target product corresponds cannot be selected. Create groups,
The inspection processing apparatus according to claim 1 or 2. Obtaining the weight of the target product, the unit weight of the target product, and the inspection quantity, which is the quantity that must be inspected for the target product, and the quantity of the target product specified by the matching unit indicated by the weight and the A quantity calculation unit for calculating a difference value from the inspection quantity;
A display information output unit that obtains the difference value calculated by the quantity calculation unit and outputs the difference value;
The inspection processing apparatus according to any one of claims 1 to 3, further comprising: The display information output unit acquires the inspection quantity of the target product, and further outputs the inspection quantity.
The inspection processing apparatus according to claim 4. The quantity calculation unit recalculates the quantity of the target product according to a change in the weight of the target product, and calculates a difference value between the recalculated quantity and the inspection quantity,
The display information output unit outputs a difference value based on the quantity of the recalculated target product,
The inspection processing apparatus according to claim 4 or 5. The feature amount of the image obtained by capturing the target product and the unit weight of the target product are acquired, and the feature amount and the unit weight of the predetermined product image are classified for each product similar to each other. A group determination unit that selects a group corresponding to the target product based on a feature amount of the captured image and a unit weight of the target product;
A matching unit that identifies the target product from among the products belonging to the selected group based on the feature amount of the captured image and the unit weight;
Obtaining the weight of the target product, the unit weight of the target product, and the inspection quantity, which is the quantity that must be inspected for the target product, and the quantity of the target product specified by the matching unit indicated by the weight and the A quantity calculation unit for calculating a difference value from the inspection quantity;
A display information output unit that obtains the difference value calculated by the quantity calculation unit and outputs the difference value;
Inspection processing apparatus comprising: An imaging device for capturing an image of a target product to be shipped;
A weight measuring instrument for measuring the weight of the target product to be shipped;
The inspection processing apparatus according to any one of claims 1 to 7,
Inspection system with Inspection processing equipment
The feature amount of the image obtained by capturing the target product and the unit weight of the target product are acquired, and the feature amount and the unit weight of the predetermined product image are classified for each product similar to each other. Based on the feature amount of the captured image and the unit weight of the target product, select a group to which the target product corresponds,
Based on the feature amount of the captured image and the unit weight, the target product is specified from among the products belonging to the selected group,
When the group to which the target product is applicable cannot be selected, a group to which the target product is applicable is newly created.
Inspection processing method. Computer of inspection processing equipment,
The feature amount of the image obtained by capturing the target product and the unit weight of the target product are acquired, and the feature amount and the unit weight of the predetermined product image are classified for each product similar to each other. Means for selecting a group corresponding to the target product based on a feature amount of the captured image and a unit weight of the target product;
Means for identifying the target product from among the products belonging to the selected group based on the feature amount of the captured image and the unit weight;
Means for newly creating a group to which the target product corresponds when the group to which the target product corresponds cannot be selected;
Program to function as.

Images