Electrodeposited board evaluation apparatus and electrodeposited board evaluation system

Abstract

To easily and appropriately perform quality determination of the surface state of an electrodeposition region, about an electrodeposition plate on which nonferrous metal is electrodeposited by electrolytic sampling.SOLUTION: An electrodeposition plate is obtained by a wet type electrolysis method of nonferrous metal. Electrodeposition plate evaluation devices 30a, 30b comprise: image acquisition units 31a, 31b which acquire surface images of an electrodeposition region of the electrodeposition plate taken out from an electrolytic solution; data holding units 33a, 33b which hold sample images related to a surface state of the electrodeposition region of the electrodeposition plate as teacher data; learning function units 34a, 34b which learn a determination reference when performing quality determination of the surface state of the electrodeposition region of the electrodeposition plate by using the teacher data held by the data holding units 33a, 33b; and quality determination units 35a, 35b which perform the quality determination of the surface state of the electrodeposition region of the electrodeposition plate related to the surface image, based on the determination reference which has been learned by the learning function units 34a, 34b, while comparing the surface images acquired by the image acquisition units 31a, 31b with the teacher data held by the data holding units 33a, 33b.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an electrodeposited plate evaluation apparatus, an electrodeposited plate evaluation system, an electrodeposited plate evaluation program, and an electrodeposited plate evaluation method.

Background Art

[0002] Non-ferrous metals such as zinc and copper are sometimes recovered by applying an electrowinning method as a so-called wet electrolysis method (see, for example, Patent Document 1). Electrowinning is a method in which electrode plates functioning as an anode and a cathode are immersed in an electrolyte containing the non-ferrous metal to be recovered, and the non-ferrous metal is electrodeposited on the cathode electrode plate by electrolytic dissociation for recovery.

Prior Art Literature

Patent Literature

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When recovering non-ferrous metals by electrowinning, it is preferable to sort out defective electrodeposited plates with poor electrodeposition conditions. However, when sorting defective electrodeposited plates by human visual inspection, skill (advanced technology) is required, and there are concerns such as oversights and misjudgments. On the other hand, it is also conceivable to detect the surface state of the electrodeposited plate by some means and sort out defective electrodeposited plates based on the detection results. However, considering the unevenness variation (size, size distribution, shape, etc.) on the detected surface, changes in plate thickness, the degree of light scattering, etc., it is difficult to create a simple defect model-based criterion. That is, it is not always possible to appropriately sort out defective electrodeposited plates simply based on the detection results of the surface state.

[0005] The present invention provides a technique that enables easy and appropriate quality determination of the surface state of an electrodeposited region of an electrodeposited plate on which a non-ferrous metal is electrodeposited by electrowinning. [Means for solving the problem]

[0006] A first aspect of the present invention is: Regarding electrodeposited plates obtained by a wet electrolytic method for non-ferrous metals, an image acquisition unit acquires a surface image of the electrodeposited region of the electrodeposited plate after it has been removed from the electrolyte, A data holding unit that holds sample images relating to the surface state of the electrodeposited region of the electrodeposited plate as training data, A learning function unit learns the criteria for determining the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate using the training data held by the data holding unit, A quality determination unit compares the surface image acquired by the image acquisition unit with the training data held by the data holding unit, and performs a quality determination of the surface state of the electrodeposited area of ​​the electrodeposited plate related to the surface image based on the judgment criteria learned by the learning function unit. This is an electrodeposited plate evaluation device equipped with [specific features / features].

[0007] A second aspect of the present invention is: The image acquisition unit is comprised of a narrow-angle camera that captures the surface state of a representative region of the electrodeposited area of ​​the electrodeposited plate. The aforementioned training data and the aforementioned judgment criteria are set to correspond to the imaging results obtained by the narrow-angle camera. This is an electrodeposited plate evaluation apparatus as described in the first embodiment.

[0008] A third aspect of the present invention is: The training data corresponding to the narrow-angle camera consists of multiple sample images with different surface smoothness levels of the electrodeposited plate. This is an electrodeposited plate evaluation apparatus according to the second embodiment.

[0009] A fourth aspect of the present invention is: The image acquisition unit is comprised of a wide-angle camera that captures the surface condition of the entire electrodeposited area of ​​the electrodeposited plate. The aforementioned training data and judgment criteria are set to correspond to the imaging results from the wide-angle camera. This is an electrodeposited plate evaluation apparatus as described in the first embodiment.

[0010] A fifth aspect of the present invention is: The training data corresponding to the wide-angle camera comprises at least one sample image of a specific form of surface defect that may occur on the electrodeposited plate. This is an electrodeposited plate evaluation apparatus according to the second embodiment.

[0011] A sixth aspect of the present invention is: The learning function unit is configured to update the judgment criteria in accordance with information regarding the accuracy of the judgment made by the pass / fail judgment unit. This is an electrodeposited plate evaluation apparatus as described in the first embodiment.

[0012] A seventh aspect of the present invention is: An electrodeposited plate evaluation apparatus according to the first embodiment, An electrolytic cell containing the electrolyte in a state in which the electrodeposited plate can be immersed, A transport mechanism for transporting the electrodeposited plate removed from the electrolytic cell to at least the image acquisition location by the image acquisition unit of the electrodeposited plate evaluation device, An information output unit that outputs information regarding the result of the quality determination by the quality determination unit of the electrodeposited plate evaluation device for the electrodeposited plate that has been transported to the image acquisition location and whose image has been acquired, This is an electrodeposited board evaluation system equipped with [features / equipment].

[0013] An eighth aspect of the present invention is: The image acquisition unit of the electrodeposited plate evaluation apparatus is configured to include a narrow-angle camera that captures the surface state of a representative region of the electrodeposited area of ​​the electrodeposited plate, and a wide-angle camera that captures the surface state of the entire electrodeposited area of ​​the electrodeposited plate. The training data and the judgment criteria are set to correspond to the imaging results from the narrow-angle camera and the imaging results from the wide-angle camera, respectively. This is the electrodeposited board evaluation system described in the seventh aspect.

[0014] A ninth aspect of the present invention is: The electrolytic cell is configured to be capable of immersing a plurality of said electrodeposition plates, The image acquisition unit is configured to sequentially acquire surface images of the plurality of electrodeposition plates conveyed from the electrolytic cell, The quality determination unit is configured to associate identification information of each electrodeposition plate for which a surface image has been acquired by the image acquisition unit, and perform quality determination on the surface condition of the electrodeposition region of each said electrodeposition plate The electrodeposition plate evaluation system according to the 7th or 8th aspect.

[0015] The 10th aspect of the present invention provides: The identification information is information relating to the immersion position of each said electrodeposition plate in the electrolytic cell The electrodeposition plate evaluation system according to the 9th aspect.

[0016] The 11th aspect of the present invention provides: The information output unit is configured to output information relating to the result of quality determination by the quality determination unit in a mode that allows visual recognition of the immersion position of each said electrodeposition plate in the electrolytic cell The electrodeposition plate evaluation system according to the 10th aspect.

[0017] The 12th aspect of the present invention provides: The quality determination unit is configured to determine the quality of the surface condition of the electrodeposition region of each said electrodeposition plate in a plurality of stages or a plurality of items, The information output unit is configured to output information relating to the result of quality determination by the quality determination unit in a mode that allows visual recognition of differences among the plurality of stages or the plurality of items The electrodeposition plate evaluation system according to the 11th aspect.

[0018] The 13th aspect of the present invention provides: A computer comprising: for an electrodeposition plate obtained by a hydrometallurgical electrolysis method for non-ferrous metals, an image acquisition unit that acquires a surface image of an electrodeposition region of said electrodeposition plate taken out from an electrolytic solution; a data holding unit that holds, as teacher data, sample images relating to the surface condition of the electrodeposition region of the electrodeposition plate; A learning function unit learns the criteria for determining the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate using the training data held by the data holding unit, A quality determination unit compares the surface image acquired by the image acquisition unit with the training data held by the data holding unit, and performs a quality determination of the surface state of the electrodeposited area of ​​the electrodeposited plate related to the surface image based on the judgment criteria learned by the learning function unit. This is an electrodeposited board evaluation program that functions as such.

[0019] A fourteenth aspect of the present invention is: Regarding electrodeposited plates obtained by a wet electrolytic method for non-ferrous metals, an image acquisition procedure is provided for acquiring a surface image of the electrodeposited region of the electrodeposited plate after it has been removed from the electrolyte. A data retention procedure for retaining sample images relating to the surface state of the electrodeposited region of the electrodeposited plate as training data, A learning procedure for learning the criteria for determining the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate using the training data held in the data holding procedure, A quality determination procedure that compares the surface image acquired in the image acquisition procedure with the training data held in the data holding procedure, and determines the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate related to the surface image based on the judgment criteria learned in the learning procedure, This is an electrodeposited board evaluation method that includes [a specific feature / feature]. [Effects of the Invention]

[0020] According to the present invention, it becomes possible to easily and appropriately determine the quality of the surface condition of the electrodeposited region of an electrodeposited plate on which a non-ferrous metal is electrodeposited by electrolytic extraction. [Brief explanation of the drawing]

[0021] [Figure 1] This is a schematic diagram illustrating an example configuration of an electrodeposited board evaluation system according to one embodiment of the present invention. [Figure 2] This is an explanatory diagram showing a specific example of training data for an electrodeposited plate evaluation device according to one embodiment of the present invention. [Figure 3]This is an explanatory diagram showing another specific example of training data in an electrodeposited plate evaluation device according to one embodiment of the present invention. [Figure 4] This is an explanatory diagram showing an example of information output in an electrodeposited plate evaluation system according to one embodiment of the present invention. [Modes for carrying out the invention]

[0022] The following describes, with reference to the drawings, an electrodeposited board evaluation apparatus, an electrodeposited board evaluation system, an electrodeposited board evaluation program, and an electrodeposited board evaluation method according to one embodiment of the present invention.

[0023] (1) Example of system configuration First, an example of the configuration of the electrodeposited board evaluation system according to this embodiment will be described.

[0024] The electrodeposited plate evaluation system according to this embodiment is configured to recover non-ferrous metals such as zinc and copper by electrolytic extraction using a wet electrolytic method, and further comprises a function to determine the quality of the surface condition of the electrodeposited region of the electrode plate on which the non-ferrous metal to be recovered is electrodeposited.

[0025] The following explanation uses zinc as an example of the non-ferrous metal (electrodeposited metal). Electrode plates to which zinc is electrodeposited are sometimes called "zinc electrodeposited plates" or simply "electrodeposited plates."

[0026] Figure 1 is a schematic diagram illustrating an example of the configuration of the electrodeposited plate evaluation system according to this embodiment.

[0027] The electrodeposited plate evaluation system 1 shown in the figure includes an electrolytic cell 10 that contains an electrolyte solution containing zinc and sulfuric acid. Each electrode plate is immersed in the electrolyte solution of the electrolytic cell 10, serving as the anode and cathode. Metallic zinc is electrodeposited onto the electrode plate serving as the cathode. In other words, the electrolytic cell 10 is configured to contain the electrolyte solution in a state where at least the zinc electrodeposited plates 2, onto which zinc is electrodeposited, can be immersed. Furthermore, the electrolytic cell 10 is configured so that multiple zinc electrodeposited plates 2 can be immersed in the electrolyte solution contained within the cell while aligned in a predetermined configuration. The cathode material is aluminum, and it is plate-shaped with a smooth surface.

[0028] Furthermore, the electrodeposited plate evaluation system 1 is equipped with a transport mechanism for transporting the electrodeposited plates 2 and a transport conveyor 20 which serves as a washing and drying mechanism. The transport conveyor 20 is fitted with a transport device 21 for transporting the electrodeposited plates 2 from the transport conveyor 20 to the electrolytic cell 10, an unloading device 22 for removing the electrodeposited plates 2 from the electrolyte in the electrolytic cell 10 and transporting them to the transport conveyor 20, and a pre-treatment device 23 for drying and washing the electrodeposited plates 2 removed by the unloading device 22. In addition, the transport conveyor 20 is fitted with a peeling device 24 for peeling off the zinc electrodeposited on the electrodeposited plates 2, a zinc recovery unit 25 for recovering the zinc after peeling, a defective plate extraction unit 26 for extracting defective plates from which the zinc peeling failed, and a good plate supply unit 27 for supplying good electrodeposited plates 2 to the transport conveyor 20.

[0029] In the electrodeposited plate evaluation system 1 with this configuration, current is applied to the anode and cathode immersed in the electrolyte of the electrolytic cell 10 to electrodeposit zinc onto the cathode. The zinc electrodeposited plate 2, with the zinc electrodeposited onto it, is then removed from the electrolyte by the removal device 22, dried and washed in the pretreatment device 23, and then transported by the conveyor belt 20 to the stripping device 24, where the electrodeposited zinc is stripped from the electrodeposited plate 2. The stripped zinc is transported and recovered by the zinc recovery unit 25. After the zinc has been stripped, the electrodeposited plate 2, excluding any defective plates extracted by the defective plate extraction unit 26, is transported by the conveyor belt 20 to the loading device 21 and immersed again in the electrolyte of the electrolytic cell 10.

[0030] Multiple electrodeposited plates 2 are immersed in the electrolytic cell 10, and zinc is electrodeposited onto each of them in parallel. Then, each electrodeposited plate 2 is sequentially lifted out of the electrolytic cell 10 by the unloading device 22 and transported one by one on the transport conveyor 20. The transport speed at this time is, for example, about 5 to 10 seconds per plate.

[0031] Each electrodeposited plate 2, as it is sequentially removed from the electrolytic cell 10, is assigned identification information to identify it at the time of removal. Examples of such identification information include information regarding the immersion position of each electrodeposited plate 2 within the electrolytic cell 10 (specifically, information that identifies the position from which the removal device 22 removed each electrodeposited plate 2 within the electrolytic cell 10). However, the identification information is not limited to this, as long as it allows for the identification of each electrodeposited plate 2.

[0032] Incidentally, some electrodeposited plates 2 that are removed from the electrolytic cell 10 may have poor zinc deposition. It is preferable to sort and remove such defective electrodeposited plates. However, sorting defective electrodeposited boards is extremely rare, and because the electrodeposited area on board 2 is large (for example, about 1000 x 1500 mm) and the transport speed after lifting is fast, visual inspection by humans requires skill (advanced technique) and raises concerns about oversights and misjudgments. In this regard, it is conceivable to detect the surface condition of electrodeposited board 2 by some means, rather than relying on human visual inspection, and to select defective electrodeposited boards based on the detection results. However, there are many different types of defects in the electrodeposition state of defective electrodeposited boards, and considering variations in surface irregularities, changes in board thickness, and the degree of light scattering, it is difficult to create a simple defect model criterion. In other words, simply basing the selection of defective electrodeposited boards on the detection results alone does not necessarily guarantee proper selection of defective electrodeposited boards. Furthermore, it is preferable that the judgment environment for sorting defective electrodeposited boards can be carried out in a normal work area without requiring a special environment such as a cleanroom.

[0033] Based on the above, the electrodeposited plate evaluation system 1 according to this embodiment includes electrodeposited plate evaluation devices 30a and 30b, and an information processing device 40 that is communicatively connected to them, in order to enable easy and appropriate determination of the quality of the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 on which zinc is electrodeposited by electrolytic extraction. Both electrodeposited plate evaluation devices 30a and 30b are arranged in accordance with the transport path of the electrodeposited plate 2 by the transport conveyor 20, specifically the transport path from the pre-processing device 23 to the peeling device 24.

[0034] (2) Example of configuration of an electrodeposited plate evaluation device Next, we will describe the configuration examples of the electrodeposited plate evaluation devices 30a and 30b according to this embodiment. Here, we will first describe an example configuration of one electrodeposited plate evaluation device 30a, and then describe an example configuration of the other electrodeposited plate evaluation device 30b.

[0035] The electrodeposited plate evaluation device 30a comprises an image acquisition unit 31a and a control unit 32a connected thereto. The image acquisition unit 31a and the control unit 32a may be configured as a single integrated device, or they may be configured as separate units. The control unit 32a is composed of a small computer device having various hardware resources, such as a CPU (Central Processing Unit), and functions as a data holding unit 33a, a learning function unit 34a, and a pass / fail judgment unit 35a by executing a predetermined program (software). Each of these units will be described in turn below.

[0036] The image acquisition unit 31a acquires a surface image of the electrodeposited area of ​​the electrodeposited plate 2, which is removed from the electrolyte in the electrolytic cell 10 and transported to the transport conveyor 20, between the pretreatment device 23 and the peeling device 24. Specifically, the image acquisition unit 31a is composed of a narrow-angle camera that captures the surface state of a representative area within the electrodeposited area of ​​the electrodeposited plate 2. The representative area captured by the narrow-angle camera is, for example, a predetermined-size area (for example, about 150 x 150 mm) near the center of the entire electrodeposited area of ​​the electrodeposited plate 2 (for example, about 1000 x 1500 mm) where a typical surface state is expected to appear. However, it is not limited to this, and any other area within the electrodeposited area may be used as a representative area. Furthermore, the data format of the image acquired by the image acquisition unit 31a is not limited, and it can be a still image or a video, as long as it can be used for determining the quality of the surface state, as described later.

[0037] The data holding unit 33a holds sample images relating to the surface state of the electrodeposited region of the electrodeposited plate 2 as training data. The training data is used to determine the quality of the surface state of the electrodeposited region of the electrodeposited plate 2, for which the image acquisition unit 31a has acquired images. Therefore, since the image acquisition unit 31a is configured with a narrow-angle camera, the data holding unit 33a holds sample images of a representative region of the electrodeposited region of the electrodeposited plate 2 as training data, corresponding to the imaging results from the narrow-angle camera. The training data held by the data holding unit 33a consists of multiple sample images, and specific examples of each sample image will be described in detail later.

[0038] The learning function unit 34a learns the criteria for determining the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate 2 using the training data held by the data holding unit 33a. Here, "learning" means setting criteria for determining quality, or updating the set criteria to new criteria. Learning of the criteria can be performed, for example, in response to information input to the electrodeposited plate evaluation device 30a from an external source, such as information input by a user of the electrodeposited plate evaluation device 30a. The criteria learned in this way are criteria corresponding to the imaging results by the narrow-angle camera, since the image acquisition unit 31a is configured as a narrow-angle camera. Specific examples of the criteria will be described in detail later.

[0039] The quality determination unit 35a determines the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate 2, whose image has been acquired by the image acquisition unit 31a. The quality determination is performed by comparing the surface image of the electrodeposited plate 2 acquired by the image acquisition unit 31a with the training data held by the data holding unit 33a, and based on the judgment criteria learned by the learning function unit 34a. Specific examples of quality determination will be described in detail later.

[0040] The electrodeposited plate evaluation device 30b, like the electrodeposited plate evaluation device 30a described above, is equipped with an image acquisition unit 31b and a control unit 32b, and furthermore, the control unit 32b functions as a data holding unit 33b, a learning function unit 34b, and a quality determination unit 35b. In other words, the electrodeposited plate evaluation device 30b is configured the same as the electrodeposited plate evaluation device 30a, except for the points described below.

[0041] In the electrodeposited plate evaluation apparatus 30b, the image acquisition unit 31b differs from the image acquisition unit 31a in the electrodeposited plate evaluation apparatus 30a in that it is composed of a wide-angle camera that captures the surface condition of the entire electrodeposited area of ​​the electrodeposited plate 2. Since the image acquisition unit 31b is configured with a wide-angle camera, the data holding unit 33b is configured to hold sample images of the entire electrodeposited area of ​​the electrodeposited plate 2 as training data, corresponding to the imaging results from the wide-angle camera. The training data held by the data holding unit 33b consists of at least one, preferably multiple, sample images, and specific examples of such sample images will be described in detail later. Furthermore, since the image acquisition unit 31b is composed of a wide-angle camera, the learning function unit 34b learns criteria corresponding to the imaging results from the wide-angle camera as judgment criteria for determining whether an image is good or bad. Specific examples of judgment criteria will be described in detail later.

[0042] In the electrodeposited board evaluation apparatus 30a, 30b configured as described above, the program (software) that realizes the functions of each of the above-described parts 31a to 35a, 31b to 35b corresponds to an example of an electrodeposited board evaluation program according to this embodiment. In this case, such an electrodeposited board evaluation program may be provided by being stored on a recording medium (e.g., semiconductor memory, magnetic disk, optical disk, magneto-optical disk, etc.) that can be read by the control units 32a, 32b of the electrodeposited board evaluation apparatus 30a, 30b, or it may be provided from an external source via a network such as the Internet or a dedicated line.

[0043] (3) Example of information processing device configuration Next, we will describe an example configuration of the information processing device 40 connected to the electrodeposited plate evaluation devices 30a and 30b described above.

[0044] The information processing device 40 is a computer device composed of various hardware resources, such as a CPU. By executing a predetermined program (software), software-based information processing is concretely realized using the hardware resources. As a result, the information processing device 40 functions as an information management unit 41, an information output unit 42, and an image processing unit 43. Each of these units will be described in turn below.

[0045] The information management unit 41 manages the information acquired from the electrodeposited plate evaluation devices 30a and 30b. The information acquired from the electrodeposited plate evaluation devices 30a and 30b includes at least information regarding the results of the quality judgment by the quality judgment units 35a and 35b, and identification information for identifying the electrodeposited plate 2 that has been judged as quality. The acquired information may also include image data (so-called raw data) related to the images acquired by the image acquisition units 31a and 31b that formed the basis for the quality judgment. The information management unit 41 manages such acquired information by storing and accumulating it (for example, by creating a database) using the memory function of the information processing device 40.

[0046] The information output unit 42 outputs various types of information managed by the information management unit 41 to an external device as needed. The information output by the information output unit 42 includes at least information regarding the results of the pass / fail judgment by the pass / fail judgment units 35a and 35b, and identification information for identifying the electrodeposited board 2 that has been judged as pass / fail. This information is output, for example, by display output using the display panel of the information processing device 40. However, it is not limited to this, and output may also be performed by an external device not shown.

[0047] The image processing unit 43 performs image processing necessary for information output by the information output unit 42. The image processing performed by the image processing unit 43 includes, for example, the generation of images to be displayed on the display panel. By performing such image processing, the information output unit 42 can display and output information related to the pass / fail judgment results of the pass / fail judgment units 35a and 35b, along with various information associated with the pass / fail judgment results, in a visually perceptible manner. The content of the display output by the information output unit 42 (including the content of the associated information), i.e., specific examples of images generated by the image processing unit 43, will be described in detail later.

[0048] In the information processing device 40 configured as described above, the functions of each of the parts 41 to 43 described above are realized by the execution of a predetermined program (software). In this case, such predetermined program may be provided stored on a recording medium readable by the information processing device 40 (for example, semiconductor memory, magnetic disk, optical disk, magneto-optical disk, etc.), as long as it can be installed on the information processing device 40, or it may be provided from an external source via a network such as the Internet or a dedicated line.

[0049] (4) Procedure for determining quality in the electrodeposited plate evaluation device Next, we will describe an example of the processing operation of the electrodeposited plate evaluation devices 30a and 30b, which constitute the electrodeposited plate evaluation system 1 according to this embodiment.

[0050] Both the electrodeposited plate evaluation devices 30a and 30b are used to determine the quality of the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 as it is transported on the conveyor belt 20. The quality determination is made based on pre-set training data. In other words, the training data necessary for determining the quality of the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 is pre-set in the training devices 30a and 30b and stored in the data holding units 33a and 33b. Here, a specific example of the training data for quality determination will be explained.

[0051] First, we will explain the training data used in the electrodeposited plate evaluation device 30a. Figure 2 is an explanatory diagram showing a specific example of training data.

[0052] When zinc is electrodeposited onto the electrodeposited plate 2, the particles that make up its surface become larger, and the degree of surface smoothness decreases. This increases the probability of contact between the anode and cathode, making electrical short circuits more likely to occur. When a short circuit occurs, electrical energy is converted into thermal energy, leading to power loss during electrolysis and increased electricity costs. To address this, it is conceivable to prevent short circuits by, for example, checking the electrolyte supply condition of the electrolytic cell 10 and increasing the amount of glue, an additive, added to the electrolyte. To achieve this, it is preferable to understand the surface smoothness of the electrodeposited plate 2 after it has been removed from the electrolyte and reflect the results of this understanding in the electrolyte supply condition.

[0053] Therefore, in the electrodeposited plate evaluation device 30a, as shown in Figure 2, the training data is composed of multiple sample images of the electrodeposited plate 2 with different surface smoothness. These sample images are then stored in the data holding unit 33a of the electrodeposited plate evaluation device 30a as training data for the electrodeposited plate evaluation device 30a, in which the image acquisition unit 31a is a narrow-angle camera.

[0054] Specifically, as training data corresponding to a narrow-angle camera, for example, the data holding unit 33a holds the following: a first sample image with an average grain size of 0 mm or less, as shown in Figure 2(a); a second sample image with an average grain size in the range of 1 to 2 mm, as shown in Figure 2(b); a third sample image with an average grain size in the range of 2 to 5 mm, as shown in Figure 2(c); a fourth sample image with an average grain size in the range of 5 to 7 mm, as shown in Figure 2(d); a fifth sample image with an average grain size in the range of 7 to 9 mm, as shown in Figure 2(e); and a sixth sample image with an average grain size of 9 mm or more, as shown in Figure 2(f). In other words, the smoothness of the surface of the electrodeposited plate 2 is classified into, for example, six stages, and the training data corresponding to the narrow-angle camera is composed of each sample image for each stage. Note that although six sample images are given as an example here, the number of stages to be classified is not limited to a specific value.

[0055] These sample images can be obtained by preparing multiple electrodeposited plates 2 with different surface smoothness and capturing images of the surface state of each electrodeposited plate 2 with the image acquisition unit 31a. However, the system is not limited to this; for example, the control unit 32a of the electrodeposited plate evaluation device 30a may be connected to an external device (not shown) in a communication manner, and the images may be acquired from that external device.

[0056] Furthermore, in response to the retention of training data in the data holding unit 33a, the electrodeposited plate evaluation device 30a sets judgment criteria for determining whether a plate is good or bad in the learning function unit 34a. Examples of judgment criteria corresponding to imaging results from a narrow-angle camera include comparing image features extracted from the image acquisition unit 31a to be evaluated, according to predetermined parameters, with the image features of the first to sixth sample images described above, determining which sample image's image features match or which sample image's image features are closest. The image feature extraction method and the predetermined parameters for it may be based on publicly known technology. Additionally, for example, if short circuits are unlikely to occur when the average particle size is around 2 mm, the acquired image matching or similar to the first or second sample image may be judged as good, and the others as defective. Such judgment criteria can be set, for example, in response to information input to the electrodeposited plate evaluation device 30a from an external source, such as information input by the operator of the electrodeposited plate evaluation device 30a.

[0057] After the training data is stored and the judgment criteria are set as described above, the electrodeposited plate evaluation device 30a can determine whether the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 is good or bad. The determination of whether the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 is good or bad is performed in the following procedure.

[0058] When the electrodeposited plate 2 to be judged is transported by the conveyor belt 20 to the image acquisition point by the image acquisition unit 31a, the image acquisition unit 31a acquires a surface image of the electrodeposited area of ​​the electrodeposited plate 2. At this time, since the image acquisition unit 31a is composed of a narrow-angle camera, it acquires a surface image of a predetermined size area near the center of the electrodeposited area of ​​the electrodeposited plate 2, for example. The reason for targeting imaging a partial area is that a typical surface state is expected to appear near the center of the electrodeposited area compared to the peripheral area. Furthermore, by limiting the imaging target to a partial area, the amount of data in the acquired image can be reduced compared to when the entire electrodeposited area is targeted for imaging. The same applies to the sample image compared with the acquired image. Therefore, by limiting the imaging target to a partial area, the processing load when comparing the acquired image with the sample image can be reduced. The partial area targeted for imaging is 1.0 to 10%, preferably 2.0%, of the entire electrodeposited area, which is necessary and sufficient for performing quality judgment using the above-mentioned training data.

[0059] When the image acquisition unit 31a acquires a surface image of the electrodeposited area of ​​the electrodeposited plate 2 to be judged, the quality determination unit 35a then performs a quality determination on the surface state of that electrodeposited area. At this time, the quality determination unit 35a compares the image acquired by the image acquisition unit 31a with each sample image that constitutes the training data in the data holding unit 33a. Then, the quality determination unit 35a performs a quality determination on the surface state of the electrodeposited area of ​​the electrodeposited plate 2 to be judged based on the judgment criteria set in the learning function unit 34a. Specifically, for example, it determines which of the 1st to 6th sample images the image features of the acquired image match. In other words, it determines the quality of the surface state of the electrodeposited area of ​​each electrodeposited plate 2 in multiple stages. Furthermore, for example, if classification criteria for good and defective products are set, it determines whether the surface state of the electrodeposited area of ​​the electrodeposited plate 2 related to the acquired image is good or defective depending on which sample image's image features match.

[0060] The judgment result from the pass / fail judgment unit 35a is output to, for example, the information processing device 40, and managed by the information management unit 41 of the information processing device 40.

[0061] Incidentally, in the electrodeposited board evaluation device 30a that performs the processing operations described above, the learning function unit 34a can not only set judgment criteria for determining whether a board is good or bad, but also update the set judgment criteria to new judgment criteria. Therefore, the electrodeposited board evaluation device 30a can perform the following processing operations.

[0062] When the quality determination unit 35a outputs a determination result, the user of the electrodeposited plate evaluation device 30a can compare the output result with the surface condition of the electrodeposited area of ​​the actual electrodeposited plate 2 and recognize whether there is a discrepancy between them (i.e., whether the determination result matches the actual surface condition). This means that information regarding the accuracy of the quality determination by the quality determination unit 35a is recognizable.

[0063] Upon recognizing information regarding the accuracy of the pass / fail judgment, the user of the electrodeposited plate evaluation device 30a determines, based on the recognition result, whether or not it is necessary to update the judgment criteria for pass / fail judgment. For example, if the degree of conformance (conformance rate) is lower than a predetermined tolerance range, the user will determine that it is necessary to update the existing judgment criteria to new criteria in order to improve the conformance rate. The update of the judgment criteria can be performed by inputting information to the electrodeposited plate evaluation device 30a from an external source.

[0064] Examples of updates to the judgment criteria include modifying the parameters used when extracting image features from acquired images, and correcting the correspondence between good / bad products and each sample image. However, the updates to the judgment criteria are not limited to these, and other content may be used as long as it contributes to improving the precision.

[0065] Thus, in the electrodeposited board evaluation device 30a, the learning function unit 34a is configured to update the judgment criteria according to information regarding the accuracy of the pass / fail judgment, in order to improve the accuracy of the pass / fail judgment results. In other words, the learning function unit 34a can learn the pass / fail judgment criteria as needed. Therefore, for example, by increasing the number of learning iterations, the electrodeposited board evaluation device 30a can improve the accuracy of the pass / fail judgment results to the extent that it can ensure a defect detection rate equivalent to that of a visual surface condition check.

[0066] Furthermore, the learning function unit 34a improves the precision of the pass / fail judgment result by learning the criteria for pass / fail judgment, so there is no need to modify each sample image that makes up the training data in order to improve the precision. Therefore, for example, there is no need to increase the number of sample images in order to improve the precision, and it is possible to improve the precision of the pass / fail judgment result without making the processing for pass / fail judgment more complex or difficult.

[0067] Next, we will explain an example of the processing operation of the electrodeposited plate evaluation device 30b. Here again, we will first explain the training data used in the electrodeposited plate evaluation device 30b. Figure 3 is an explanatory diagram showing another specific example of training data.

[0068] In the electrodeposited plate 2, the electrodeposition on its surface may be uneven (i.e., surface defects). The specific types of surface defect modes that can occur in the electrodeposited plate 2 can be broadly classified into electrodeposition defects, point shorts, and line shorts. Defective electrodeposition refers to a state where electrodeposition on the surface of the electrodeposited plate 2 is uneven, as shown in Figure 3(a). Ideally, the electrodeposited material should be uniformly deposited in the current-carrying region of the cathode, but for some reason, the current may not flow properly, resulting in areas with insufficient electrodeposition area or amount. In addition, irregularities may be observed at the edges of the electrodeposition. Such unevenness in electrodeposition can manifest as defects on the surface of the electrodeposited plate 2. A point short circuit, as shown in Figure 3(b), refers to a condition where zinc does not electrodeposit onto a portion of the surface of the electrodeposited plate 2, resulting in a hole. An electrical short circuit occurs when the anode and cathode come into contact for some reason, converting electrical energy into thermal energy. This thermal energy can melt the anode, causing zinc to fail to electrodeposit onto that portion, leaving a hole. In this way, a point short circuit can occur. A linear short circuit, unlike a point short circuit which is a hole-like exposed area, refers to a condition where a linear exposed area occurs, as shown in Figure 3(c). A linear short circuit may have a length of 1000 mm or more, a width of about 50 mm (though this varies), and discoloration may be observed, with the color being the same as the electrodeposited material on the anode side (manganese dioxide in the case of zinc electrolysis). Such linear short circuits can occur.

[0069] To address the surface defect modes described above, the electrodeposited plate evaluation device 30b uses sample images as shown in Figure 3 to construct the training data. These sample images are then stored in the data holding unit 33b of the electrodeposited plate evaluation device 30b, where the image acquisition unit 31b is a wide-angle camera, as training data.

[0070] Specifically, as training data corresponding to the wide-angle camera, the data holding unit 33a holds, for example, a sample image of the electrodeposited plate 2 with an electrodeposited defect on its surface as shown in Figure 3(a), a sample image of the electrodeposited plate 2 with a point short circuit on its surface as shown in Figure 3(b), and a sample image of the electrodeposited plate 2 with a line short circuit on its surface as shown in Figure 3(c). In other words, the training data corresponding to the wide-angle camera is composed of sample images for each of the electrodeposited defects, point short circuits, and line short circuits that can occur on the surface of the electrodeposited plate 2. Here, we have given an example where the training data is composed of all three types of sample images, but it is not limited to these, and it is sufficient for the training data to be composed of at least one of them, or it may also be composed of sample images other than these. In other words, the training data in the electrodeposited plate evaluation device 30b only needs to include at least one sample image of a specific form of surface defect that can occur on the electrodeposited plate 2.

[0071] Each of these sample images can be obtained by preparing an electrodeposited plate 2 in which at least electrodeposition defects, point shorts, and line shorts have occurred, and capturing an image of the surface state of each electrodeposited plate 2 with the image acquisition unit 31b. However, it is not limited to this, and for example, the control unit 32b of the electrodeposited plate evaluation device 30b may be connected to an external device (not shown) in a communication manner, and the images may be acquired from that external device.

[0072] Furthermore, in conjunction with the retention of training data in the data retention unit 33b, the electrodeposited plate evaluation device 30b sets judgment criteria for determining whether the plate is good or bad in the learning function unit 34b. As judgment criteria corresponding to the imaging results from the wide-angle camera, for example, the image features extracted from the image acquisition unit 31b to be evaluated according to predetermined parameters are compared with the image features of each sample image of electrodeposition defects, point shorts, and line shorts mentioned above, and it is determined which sample image's image features match or which sample image's image features are closest to those of the acquired image to be evaluated. Moreover, for example, the degree to which the image features match or are close is determined, and according to that degree, it is determined whether or not electrodeposition defects, point shorts, or line shorts have occurred in the acquired image to be evaluated, and any of these are judged as defective. The setting of such judgment criteria can be done, for example, in response to information input to the electrodeposited plate evaluation device 30b from an external source, such as information input by the operator of the electrodeposited plate evaluation device 30b.

[0073] After the training data is stored and the judgment criteria are set as described above, the electrodeposited plate evaluation device 30b can determine whether the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 is good or bad. The determination of whether the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 is good or bad is performed in the following procedure.

[0074] When the electrodeposited plate 2 to be evaluated is transported by the conveyor belt 20 to the image acquisition point of the image acquisition unit 31b, the image acquisition unit 31b acquires a surface image of the electrodeposited area of ​​the electrodeposited plate 2. At this time, since the image acquisition unit 31b is composed of a wide-angle camera, it acquires a surface image of the entire electrodeposited area of ​​the electrodeposited plate 2, for example. The reason for targeting the entire electrodeposited area for imaging is that electrodepositing defects, point shorts, or line shorts can occur in any part of the electrodeposited area, and it is considered necessary to ensure that they can be reliably detected regardless of where they occur. The same applies to the sample image compared with the acquired image. In other words, unlike the electrodeposited plate evaluation device 30a described above, by targeting the entire electrodeposited area for imaging, it becomes possible to reliably detect any electrodepositing defects, point shorts, or line shorts if they occur.

[0075] When the image acquisition unit 31b acquires a surface image of the electrodeposited area of ​​the electrodeposited plate 2 to be judged, the quality determination unit 35b then performs a quality determination on the surface state of that electrodeposited area. At this time, the quality determination unit 35b compares the image acquired by the image acquisition unit 31b with each sample image that constitutes the training data in the data holding unit 33b. Then, the quality determination unit 35b performs a quality determination on the surface state of the electrodeposited area of ​​the electrodeposited plate 2 to be judged based on the judgment criteria set in the learning function unit 34b. Specifically, for example, it determines which of the sample images for electrodepositing defects, point shorts, or line shorts the image features of the acquired image match. Furthermore, for example, depending on the degree to which the image features match or are similar, it determines whether or not electrodepositing defects, point shorts, or line shorts have occurred in the surface state of the electrodeposited area of ​​the electrodeposited plate 2 related to the acquired image, and determines that any of these have occurred as a defective product. In other words, the quality of the surface state of the electrodeposited area of ​​each electrodeposited plate 2 is judged based on multiple items.

[0076] The judgment result from the pass / fail judgment unit 35b is output to, for example, the information processing device 40, and managed by the information management unit 41 of the information processing device 40.

[0077] Incidentally, in the electrodeposited plate evaluation device 30b that performs the processing operations described above, the learning function unit 34b can not only set judgment criteria for determining whether the plate is good or bad, but can also update the set judgment criteria to new judgment criteria. This is the same as in the case of the electrodeposited plate evaluation device 30a described above. Since the defect characteristics of the electrodeposition state change depending on the electrolysis conditions and equipment specifications, it is possible to make a judgment of whether the plate is good or bad that is adapted to the electrolysis equipment and conditions.

[0078] Therefore, in the electrodeposited plate evaluation device 30b, the learning function unit 34b is capable of updating the judgment criteria according to information regarding the accuracy of the pass / fail judgment in order to improve the accuracy of the pass / fail judgment result. As a result, for example, by increasing the number of learning iterations, the electrodeposited plate evaluation device 30b can improve the accuracy of the pass / fail judgment result to the extent that it can ensure a defect detection rate equivalent to that of a visual surface condition check. Moreover, since it is not necessary to modify each sample image that makes up the training data in order to improve the accuracy, it is possible to improve the accuracy of the pass / fail judgment result without causing the processing for pass / fail judgment to become more complex or difficult.

[0079] (5) Procedures for information processing in information processing equipment Next, we will describe an example of processing operation in the information processing device 40, which outputs the respective judgment results from the electrodeposited plate evaluation devices 30a and 30b.

[0080] When the electrodeposited plate evaluation devices 30a and 30b determine the quality of the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 being transported by the conveyor belt 20, information regarding the determination result is output to the information processing device 40 from the electrodeposited plate evaluation devices 30a and 30b. The information processing device 40 then manages the output information from the electrodeposited plate evaluation devices 30a and 30b in the information management unit 41. At this time, the information management unit 41 manages the information regarding the quality determination result of each electrodeposited plate evaluation device 30a and 30b for the electrodeposited plate 2 in association with the identification information of the electrodeposited plate 2.

[0081] The information processing device 40 outputs various types of information managed by the information management unit 41 at predetermined timings as needed, via the information output unit 42. These predetermined timings include, for example, when a user of the information processing device 40 requests information output, or after all of the electrodeposited plates 2 (i.e., multiple electrodeposited plates 2 constituting a single lot) immersed in the electrolytic cell 10 have been judged for quality by the electrodeposited plate evaluation devices 30a and 30b, but are not necessarily limited to these timings.

[0082] When the information output unit 42 outputs information, a user of the information processing device 40 who refers to the output content can recognize the result of the quality judgment of the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 that has been removed from the electrolyte of the electrolytic cell 10. Moreover, if the information regarding the judgment result is associated with the identification information of the electrodeposited plate 2, it is possible to determine which electrodeposited plate 2 the information output judgment result pertains to, even when the conveyor belt 20 is transporting multiple electrodeposited plates 2 in succession.

[0083] When the information output unit 42 outputs information as described above, it is preferable that the user of the information processing device 40 who accesses the information output can easily and appropriately recognize the content of the information output. Therefore, the information processing device 40 performs the following processing operations when outputting information.

[0084] Figure 4 is an explanatory diagram showing examples of information output. In the figure, (a) shows an example of information output regarding the pass / fail judgment result by the electrodeposited plate evaluation device 30a, and (b) shows an example of information output regarding the pass / fail judgment result by the electrodeposited plate evaluation device 30b.

[0085] In the information processing device 40, when the information output unit 42 outputs information, the image processing unit 43 performs the process of generating a display image.

[0086] The image processing unit 43 generates an image that visually shows the immersion position of each electrodeposited plate 2 in the electrolytic cell 10 for each electrodeposited plate 2 that has been judged as good or bad by the electrodeposited plate evaluation devices 30a and 30b. Specifically, as shown in Figures 4(a) and (b), the image is represented by a two-dimensional map of the electrolytic cell 10 viewed from above, and the immersion position of each electrodeposited plate 2 is identified by its display position on the two-dimensional map.

[0087] Furthermore, the images generated by the image processing unit 43 show the differences between multiple stages or multiple items in the quality judgment results for multiple stages or multiple items regarding the surface condition of the electrodeposited area of ​​each electrodeposited plate 2 in a manner that makes these differences visually apparent. Specifically, as shown in the legend in Figures 4(a) and (b), the display color of each electrodeposited plate 2 displayed on the two-dimensional map is differentiated according to which of the multiple stages of electrodeposition rank it corresponds to, or according to which of the multiple items of electrodeposition defect it corresponds to.

[0088] Once the image processing unit 43 generates an image, the information output unit 42 then outputs the image generated by the image processing unit 43. As a result, the information output unit 42 outputs information regarding the quality judgment results for each electrodeposited plate 2 in a manner that allows the immersion position of each electrodeposited plate 2 in the electrolytic cell 10 to be visually confirmed, and in a manner that allows the differences between multiple stages or multiple items to be visually confirmed.

[0089] Therefore, users of the information processing device 40 who refer to the output content from the information output unit 42 will be able to easily and appropriately recognize the result of the quality judgment of the surface condition of the electrodeposited area of ​​each electrodeposited plate 2 removed from the electrolyte of the electrolytic cell 10.

[0090] (6) Effects of this embodiment The electrodeposited board evaluation apparatus, electrodeposited board evaluation system, electrodeposited board evaluation program, and electrodeposited board evaluation method described in this embodiment provide one or more of the following effects.

[0091] In this embodiment, in the electrodeposited plate evaluation devices 30a and 30b, the surface images acquired by the image acquisition units 31a and 31b are compared with the training data held by the data holding units 33a and 33b, and the quality determination units 35a and 35b determine the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate 2 related to the surface image, based on the judgment criteria learned by the learning function units 34a and 34b. Therefore, it becomes easy to select and eliminate defective electrodeposited plates based on the quality determination results by the quality determination units 35a and 35b. Moreover, unlike when the quality of the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 is determined by human visual inspection, this method does not require expertise (advanced skills) in the determination process, nor does it raise concerns about oversights or judgment errors. Furthermore, since the image acquisition units 31a and 31b perform quality judgment by comparing the acquired surface images with training data, the quality judgment can be performed appropriately regardless of the transport speed or area size of the electrodeposited plate 2. Moreover, since the system can learn the judgment criteria used for quality judgment by comparing the surface images with training data, it can flexibly and appropriately respond to various types of defects in the electrodeposited state, such as variations in surface irregularities, changes in plate thickness, and light scattering. In other words, even when based on the detection results of the surface state of the electrodeposited plate 2, defective electrodeposited plates can be appropriately sorted. Furthermore, the judgment environment for sorting defective electrodeposited boards can be operated in a normal work area without requiring a special environment such as a cleanroom. As described above, according to this embodiment, it becomes possible to easily and appropriately determine the quality of the surface condition of the electrodeposited region of the electrodeposited plate 2, on which zinc, a non-ferrous metal, is electrodeposited by electrolytic extraction.

[0092] As described in this embodiment, in the electrodeposited plate evaluation devices 30a and 30b, if the learning function units 34a and 34b are capable of learning judgment criteria (especially updating to new judgment criteria), then, for example, by increasing the number of learning iterations according to the information on the accuracy (e.g., conformance rate) of the quality judgment by the quality judgment units 35a and 35b, it becomes possible to improve the conformance rate of the quality judgment results to the extent that a defect detection rate equivalent to that of a visual surface condition check can be ensured. Therefore, this is highly preferable for easily and appropriately determining the quality of the surface condition of the electrodeposited area of ​​the electrodeposited plate 2. Furthermore, if the learning function units 34a and 34b are compatible with learning the judgment criteria, there is no need to modify each sample image that makes up the training data, even when aiming to improve the precision of the pass / fail judgment result. Therefore, for example, there is no need to increase the number of sample images to improve the precision, and it becomes possible to improve the precision of the pass / fail judgment result without increasing the complexity or difficulty of the processing for pass / fail judgment.

[0093] In this embodiment, the electrodeposited plate evaluation device 30a has an image acquisition unit 31a which is configured as a narrow-angle camera that captures the surface state of a representative region of the electrodeposited area of ​​the electrodeposited plate 2. The training data in the data holding unit 33a is composed of multiple sample images of the electrodeposited plate 2 with different levels of surface smoothness. Therefore, according to this embodiment, for example, even if the degree of surface smoothness of the electrodeposited plate 2 decreases and short circuits are more likely to occur, this can be easily and appropriately detected. In other words, for example, it becomes possible to grasp the surface smoothness of the electrodeposited plate 2 after it has been removed from the electrolyte and reflect the result of this grasp in the electrolyte supply state of the electrolytic cell 10.

[0094] In this embodiment, the electrodeposited plate evaluation device 30b is configured with a wide-angle camera as the image acquisition unit 31b, which captures the surface condition of the entire electrodeposited area of ​​the electrodeposited plate 2. The training data in the data holding unit 33b is configured to include at least one sample image of a specific form of surface defect that may occur on the electrodeposited plate 2, preferably a sample image of the electrodeposited plate 2 with an electrodeposited defect, a sample image of the electrodeposited plate 2 with a point short, and a sample image of the electrodeposited plate 2 with a line short. Therefore, according to this embodiment, even if an electrodeposited defect, a point short, or a line short occurs on the surface of the electrodeposited plate 2, it becomes possible to easily and appropriately detect it. In other words, it becomes possible to determine whether, for example, an electrodeposited defect, a point short, or a line short has occurred on the surface condition of the electrodeposited area of ​​the electrodeposited plate 2, and to determine that any of these is a defective product.

[0095] In this embodiment, in the electrodeposited plate evaluation system 1, for electrodeposited plates 2 that have been transported to the image acquisition location by the image acquisition units 31a and 31b of the electrodeposited plate evaluation devices 30a and 30b and whose images have been acquired, the information output unit 42 outputs information regarding the results of the quality judgment performed by the quality judgment units 35a and 35b of the electrodeposited plate evaluation devices 30a and 30b. Therefore, a user of the information processing device 40 who refers to the output content of the information output unit 42 can recognize the result of the quality judgment of the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 after it has been removed from the electrolyte of the electrolytic cell 10. As a result, it is possible to easily and appropriately change the electrolyte supply state of the electrolytic cell 10 according to the surface condition of the electrodeposited plate 2, or to sort out defective electrodeposited plates.

[0096] In this embodiment, when the information output unit 42 outputs information in the electrodeposited plate evaluation system 1, it displays information regarding the pass / fail judgment results by the electrodeposited plate evaluation devices 30a and 30b in a manner that allows the immersion position of each electrodeposited plate 2 in the electrolytic cell 10 to be visually confirmed. Furthermore, it displays the differences between multiple stages or multiple items of pass / fail judgment results regarding the surface state of the electrodeposited area of ​​each electrodeposited plate 2 in a manner that allows the differences between those multiple stages or multiple items to be visually confirmed. Therefore, by referring to the two-dimensional map output content from the information output unit 42, users of the information processing device 40 can easily and appropriately recognize from which position in the electrolytic cell 10 each electrodeposited plate 2 was removed and what the pass / fail judgment results are. As a result, it is extremely convenient for users.

[0097] In this embodiment, the electrodeposited plate evaluation system 1 includes both an electrodeposited plate evaluation device 30a having a narrow-angle camera and an electrodeposited plate evaluation device 30b having a wide-angle camera. In other words, it includes both an electrodeposited plate evaluation device 30a that determines whether the electrodeposited plate 2 is good or bad using the smoothness of the surface as an indicator, and an electrodeposited plate evaluation device 30b that determines whether the electrodeposited plate 2 is good or bad using the surface defect mode as an indicator. By having both electrodeposited plate evaluation devices 30a and 30b that determine whether the electrodeposited plate is good or bad using different indicators, it becomes possible to determine whether the surface condition of the electrodeposited area of ​​the electrodeposited plate 2 is good or bad more appropriately and with greater accuracy compared to having only one of them.

[0098] (7) Variant Although embodiments of the present invention have been described above, the disclosures described above represent exemplary embodiments of the present invention. In other words, the technical scope of the present invention is not limited to the exemplary embodiments described above.

[0099] In the embodiments described above, the case where the non-ferrous metal (electrodeposited metal) is zinc was given as an example, but the present invention is not limited to this. That is, if the electrolytic extraction method is applied, the present invention can be applied in exactly the same way even when recovering other non-ferrous metals such as copper.

[0100] In the above-described embodiment, an example is given in which two electrodeposited plate evaluation devices 30a and 30b are arranged in the electrodeposited plate evaluation system 1, but the present invention is not limited thereto. That is, it is sufficient to have at least one electrodeposited plate evaluation device in the system, and there may be three or more. [Explanation of Symbols]

[0101] 1…Electrodeposited plate evaluation system, 2…Electrodeposited plate, 10…Electrolytic cell, 20…Conveyor belt, 30a,30b…Electrodeposited plate evaluation device, 31a,31b…Image acquisition unit, 32a,32b…Control unit, 33a,33b…Data storage unit, 34a,34b…Learning function unit, 35a,35b…Pass / fail judgment unit, 40…Information processing unit, 41…Information management unit, 42…Information output unit, 43…Image processing unit

Claims

1. An image acquisition unit that acquires a surface image of the electrodeposited region of an electrodeposited plate obtained by a wet electrolytic method of non-ferrous metals, after the electrodeposited plate has been removed from the electrolyte, A data holding unit that holds sample images relating to the surface state of the electrodeposited region of the electrodeposited plate as training data, A learning function unit learns the criteria for determining the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate using the training data held by the data holding unit, A quality determination unit compares the surface image acquired by the image acquisition unit with the training data held by the data holding unit, and performs a quality determination of the surface state of the electrodeposited area of ​​the electrodeposited plate related to the surface image based on the judgment criteria learned by the learning function unit. Equipped with, The image acquisition unit is comprised of a narrow-angle camera that captures the surface state of a representative region of the electrodeposited area of ​​the electrodeposited plate. The aforementioned training data and the aforementioned judgment criteria are set to correspond to the imaging results obtained by the narrow-angle camera. Electrodeposited plate evaluation device.

2. The training data corresponding to the narrow-angle camera consists of multiple sample images with different surface smoothness levels of the electrodeposited plate. The electrodeposited plate evaluation apparatus according to claim 1.

3. An image acquisition unit that acquires a surface image of the electrodeposited region of an electrodeposited plate obtained by a wet electrolytic method of non-ferrous metals, after the electrodeposited plate has been removed from the electrolyte, A data holding unit that holds sample images relating to the surface state of the electrodeposited region of the electrodeposited plate as training data, A learning function unit learns the criteria for determining the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate using the training data held by the data holding unit, A quality determination unit compares the surface image acquired by the image acquisition unit with the training data held by the data holding unit, and performs a quality determination of the surface state of the electrodeposited area of ​​the electrodeposited plate related to the surface image based on the judgment criteria learned by the learning function unit. Equipped with, The image acquisition unit is comprised of a wide-angle camera that captures the surface condition of the entire electrodeposited area of ​​the electrodeposited plate. The aforementioned training data and judgment criteria are set to correspond to the imaging results from the wide-angle camera. Electrodeposited plate evaluation device.

4. The training data corresponding to the wide-angle camera comprises at least one sample image of a specific form of surface defect that may occur on the electrodeposited plate. The electrodeposited plate evaluation apparatus according to claim 3.

5. An image acquisition unit that acquires a surface image of the electrodeposited region of an electrodeposited plate obtained by a wet electrolytic method of nonferrous metals, after the electrodeposited plate has been removed from the electrolyte, A data holding unit that holds sample images relating to the surface state of the electrodeposited region of the electrodeposited plate as training data, A learning function unit learns the criteria for determining the quality of the surface state of the electrodeposited area of ​​the electrodeposited plate using the training data held by the data holding unit, A quality determination unit compares the surface image acquired by the image acquisition unit with the training data held by the data holding unit, and performs a quality determination of the surface state of the electrodeposited area of ​​the electrodeposited plate related to the surface image based on the judgment criteria learned by the learning function unit. Equipped with, The learning function unit is configured to update the judgment criteria in accordance with information regarding the accuracy of the judgment made by the pass / fail judgment unit. Electrodeposited plate evaluation device.

6. An electrodeposited plate evaluation apparatus comprising: an image acquisition unit for acquiring a surface image of the electrodeposited region of an electrodeposited plate obtained by a wet electrolytic method of non-ferrous metals after it has been removed from the electrolyte; a data holding unit for holding sample images relating to the surface state of the electrodeposited region of the electrodeposited plate as training data; a learning function unit for learning criteria for determining the quality of the surface state of the electrodeposited region of the electrodeposited plate using the training data held by the data holding unit; and a quality determination unit for determining the quality of the surface state of the electrodeposited region of the electrodeposited plate relating to the surface image, based on the criteria learned by the learning function unit, while comparing the surface image acquired by the image acquisition unit with the training data held by the data holding unit. An electrolytic cell containing the electrolyte in a state in which the electrodeposited plate can be immersed, A transport mechanism for transporting the electrodeposited plate removed from the electrolytic cell to at least the image acquisition location by the image acquisition unit of the electrodeposited plate evaluation device, An information output unit that outputs information regarding the result of the quality determination by the quality determination unit of the electrodeposited plate evaluation device for the electrodeposited plate that has been transported to the image acquisition location and whose image has been acquired, Equipped with, The image acquisition unit of the electrodeposited plate evaluation apparatus is configured to include a narrow-angle camera that captures the surface state of a representative region of the electrodeposited area of ​​the electrodeposited plate, and a wide-angle camera that captures the surface state of the entire electrodeposited area of ​​the electrodeposited plate. The training data and the judgment criteria are set to correspond to the imaging results from the narrow-angle camera and the imaging results from the wide-angle camera, respectively. Electrodeposited board evaluation system.

Images