Product inspection system, product inspection method, learning device, and recognition device

Abstract

To provide a learning apparatus, an estimation apparatus, a dimension measurement apparatus, and a dimension measurement method capable of measuring a dimension of an object with high accuracy.SOLUTION: A dimension measurement method includes: a step of constructing a model which outputs lighting condition information on an optimal lighting condition for measuring a dimension of an object when object information is input, the model being obtained by machine learning using the lighting condition information on the lighting condition for a lighting device to emit light for measuring the dimension of the object and the object information for identifying the object, as training data; a step of inputting the object information for identifying the object; a step of inputting the input object information to the model, estimating the optimal lighting condition for measuring the dimension of the object on the basis of the lighting condition information output from the model, and controlling the lighting device on the basis of a result of the estimation; and a step of measuring the dimension of the object.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a learning device, an estimation device, a dimension measurement device, and a dimension measurement method. [Background technology]

[0002] 2. Description of the Related Art Conventionally, a packaging material called an embossed carrier tape for storing minute electronic components has been known (see Patent Document 1 below). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-136545 A Summary of the Invention [Problem to be solved by the invention]

[0004] It is desirable to manufacture the above-mentioned embossed carrier tape with high dimensional accuracy, for example, in the shape of the recesses that house electronic components, but in order to measure the dimensions, workers must check the measurement position and manually align the position. On the other hand, dimensional measurement using an image of the embossed carrier tape can reduce the amount of work required by workers. However, in order to perform highly accurate dimensional measurement using an image, it is necessary to use a clear image, and since there are various shapes and types of embossed carrier tapes, it is necessary to carefully set the lighting conditions of the lighting device that are optimal for measuring the dimensions of each embossed carrier tape.

[0005] An object of the present invention is to provide a learning device, an estimation device, a dimension measuring device, and a dimension measuring method that are capable of measuring the dimensions of an object with high accuracy. [Means for solving the problem]

[0006] A learning device according to one aspect of the present invention includes: An embossed carrier tape made of plastic with continuous pockets for storing electronic components.lighting condition information regarding a lighting condition under which a lighting device emits light to measure the dimensions of the Embossed Carrier Tape and a model that is machine-learned using the illumination condition information and the object information input by the input unit as training data, Embossed Carrier Tape a model constructing unit that constructs the model and outputs the illumination condition information relating to an optimum condition for dimension measurement of the object; the object information indicates a portion of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape, and the model outputs optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape when information indicating the portion of the embossed carrier tape is input. It is a learning device. A learning device according to one aspect of the present invention includes: An embossed carrier tape made of plastic with continuous pockets for storing electronic components. lighting condition information regarding a lighting condition under which a lighting device emits light to measure the dimensions of the Embossed Carrier Tape and a model that is machine-learned using the illumination condition information and the object information input by the input unit as training data, Embossed Carrier Tape a model constructing unit that constructs the model and outputs the illumination condition information relating to an optimum condition for dimension measurement of the object; The object information indicates physical property information including a molding method and a mold shape of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for dimensional measurement of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape, and the model outputs optimal lighting conditions for dimensional measurement of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape when information indicating the molding method and mold shape of the embossed carrier tape is input. It is a learning device.

[0007] An estimation device according to one aspect of the present invention comprises: An embossed carrier tape made of plastic with continuous pockets for storing electronic components. An input unit for inputting object information for identifying the object; Embossed Carrier Tape The object information input by the input unit is input to a model that has been machine-learned using lighting condition information regarding lighting conditions under which a lighting device irradiates light in order to measure the dimensions of the object and the object information as training data, and based on the lighting condition information output from the model, Embossed Carrier Tape and an estimation unit that estimates optimal lighting conditions for dimension measurement of the object. the object information indicates a portion of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape, and the model outputs optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape when information indicating the portion of the embossed carrier tape is input. An estimation device. An estimation device according to one aspect of the present invention comprises: An embossed carrier tape made of plastic with continuous pockets for storing electronic components. An input unit for inputting object information for identifying the object; Embossed Carrier Tape The object information input by the input unit is input to a model that has been machine-learned using lighting condition information regarding lighting conditions under which a lighting device irradiates light in order to measure the dimensions of the object and the object information as training data, and based on the lighting condition information output from the model, Embossed Carrier Tape and an estimation unit that estimates optimal lighting conditions for dimension measurement of the object. Embossed Carrier Tape of Molding method and mold shape The illumination condition information indicates physical property information including Embossed Carrier Tape of The embossed carrier tape according to the molding method and the mold shape The optimum lighting conditions for measuring the dimensions of death , the model outputs, when information indicating a molding method and a mold shape of the embossed carrier tape is input, an optimum lighting condition for measuring the dimensions of the embossed carrier tape corresponding to the molding method and the mold shape of the embossed carrier tape. An estimation device.

[0008] A dimension measuring device according to one aspect of the present invention comprises: An embossed carrier tape made of plastic with continuous pockets for storing electronic components. A lighting device for irradiating light onto the Embossed Carrier Tape A measuring unit for measuring the dimensions of the Embossed Carrier Tape and a control unit that controls the lighting device based on the object information input by the input unit. Embossed Carrier Tape The object information input by the input unit is input to a model that has been machine-learned using lighting condition information regarding lighting conditions under which a lighting device irradiates light in order to measure the dimensions of the object and the object information as training data, and based on the lighting condition information output from the model, Embossed Carrier Tape and estimating optimal lighting conditions for measuring the dimensions of the object and controlling the lighting device based on the estimation result. the object information indicates a portion of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape, and the model outputs optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape when information indicating the portion of the embossed carrier tape is input. A dimension measuring device. A dimension measuring device according to one aspect of the present invention comprises: An embossed carrier tape made of plastic with continuous pockets for storing electronic components. A lighting device for irradiating light onto the Embossed Carrier Tape A measuring unit for measuring the dimensions of the Embossed Carrier Tape and a control unit that controls the lighting device based on the object information input by the input unit. Embossed Carrier Tape The object information input by the input unit is input to a model that has been machine-learned using lighting condition information regarding lighting conditions under which a lighting device irradiates light in order to measure the dimensions of the object and the object information as training data, and based on the lighting condition information output from the model, Embossed Carrier Tape and estimating optimal lighting conditions for measuring the dimensions of the object and controlling the lighting device based on the estimation result. The object information indicates physical property information including a molding method and a mold shape of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for dimensional measurement of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape, and the model outputs optimal lighting conditions for dimensional measurement of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape when information indicating the molding method and mold shape of the embossed carrier tape is input. A dimension measuring device.

[0009] A dimension measuring method according to one aspect of the present invention includes the steps of: An embossed carrier tape made of plastic with continuous pockets for storing electronic components. lighting condition information regarding a lighting condition under which a lighting device emits light to measure the dimensions of the Embossed Carrier Tape A model that has been machine-learned using training data as target information for identifying the Embossed Carrier Tape constructing the model to output the lighting condition information regarding optimal lighting conditions for measuring the dimensions of the Embossed Carrier Tape inputting object information for identifying the object; inputting the input object information into the model, and based on the lighting condition information output from the model, Embossed Carrier Tape and controlling the illumination device based on the result of the estimation. Embossed Carrier Tape and measuring a dimension of the the object information indicates a portion of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape, and the model outputs optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape when information indicating the portion of the embossed carrier tape is input. Dimension measurement method. A dimension measuring method according to one aspect of the present invention includes the steps of: An embossed carrier tape made of plastic with continuous pockets for storing electronic components. lighting condition information regarding a lighting condition under which a lighting device emits light to measure the dimensions of the Embossed Carrier Tape A model that has been machine-learned using training data as target information for identifying the Embossed Carrier Tape constructing the model to output the lighting condition information regarding optimal lighting conditions for measuring the dimensions of the Embossed Carrier Tape inputting object information for identifying the object; inputting the input object information into the model, and based on the lighting condition information output from the model, Embossed Carrier Tape and controlling the illumination device based on the result of the estimation. Embossed Carrier Tape and measuring a dimension of the The object information indicates physical property information including a molding method and a mold shape of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for dimensional measurement of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape, and the model outputs optimal lighting conditions for dimensional measurement of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape when information indicating the molding method and mold shape of the embossed carrier tape is input. Dimension measurement method. Effect of the Invention

[0010] According to the present invention, it is possible to provide a learning device, an estimation device, a dimension measuring device, and a dimension measuring method that are capable of measuring the dimensions of an object with high accuracy. [Brief description of the drawings]

[0011] [Figure 1] 1 is a diagram illustrating an example of a schematic configuration of a dimension measuring system according to an embodiment. [Diagram 2] 1 is a block diagram showing an example of a functional configuration of a dimension measuring system according to an embodiment; [Diagram 3] FIG. 2 is a block diagram showing an example of a learning unit in the embodiment. [Figure 4] FIG. 1 is a diagram illustrating the relationship between input variables and output variables in a convolutional neural network. [Diagram 5] FIG. 2 is a block diagram illustrating an example of an estimation unit in the embodiment. [Figure 6] FIG. 2 is a top view showing an example of an arrangement of lighting devices in the embodiment. [Figure 7] FIG. 11 is a top view showing another example of the arrangement of the lighting device in the embodiment. [Figure 8] FIG. 2 is a side view of the lighting device according to the embodiment. [Figure 9] FIG. 11 is another side view of the lighting device according to the embodiment. [Figure 10] FIG. 2 is a top view showing an example of an embossed carrier tape according to an embodiment. [Figure 11] FIG. 2 is a diagram for explaining an example of a mold for molding an embossed carrier tape in an embodiment. [Figure 12] 13 is a diagram showing an example of a change in a measurement value in an embodiment, and is a diagram showing a change in a value measured in a state where an embossed carrier tape is irradiated with light under illumination condition A. FIG. [Figure 13] 13 is a diagram showing an example of a change in a measurement value in the embodiment, and is a diagram showing a change in a value measured in a state where an embossed carrier tape is irradiated with light under illumination condition B. FIG. [Figure 14] 1 is a diagram showing the relationship between sites A, B, and C of an embossed carrier tape in an embodiment, ideal values, and measured values ​​for each of illumination conditions A, B, and C. FIG. [Figure 15]FIG. 11 is a diagram for explaining a process of performing an estimation process using a convolutional neural network in the embodiment. [Figure 16] FIG. 1 is a diagram for explaining a learning process using a convolutional neural network in an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of a learning device, an estimation device, a dimension measuring device, and a dimension measuring method will be described with reference to the drawings.

[0013] [Dimensional measurement overview] The dimension measuring system 1 of the embodiment measures the dimensions of an object. In the embodiment, the object is, for example, an embossed carrier tape. The embossed carrier tape is made of plastic and has storage pockets (recesses) for storing chip-type electronic components formed continuously. The dimension measuring system 1 measures the dimensions of the storage pockets by acquiring information based on the dimensions of the storage pockets of the embossed carrier tape while irradiating the embossed carrier tape with light. Note that, although the present embodiment measures the dimensions of shapes such as recesses, the present invention is not limited thereto and can also be applied to applications for measuring shapes other than recesses.

[0014] [Configuration of Dimension Measurement System 1] FIG. 1 is a diagram showing an example of a schematic configuration of a dimension measuring system 1 in an embodiment. The dimension measuring system 1 includes, for example, an illumination device 100 and an imaging device 200 arranged above and below a production line L. The production line L is a moving means for delivering an embossed carrier tape 10 on which recesses 10a are formed. When measuring the dimensions of the embossed carrier tape 10, the dimension measuring system 1 images the embossed carrier tape 10 using the imaging device 200 while irradiating the embossed carrier tape 10 with light from the illumination device 100. The dimension measuring system 1 measures the dimensions of the recesses 10a based on the captured image or the like.

[0015] 2 is a block diagram showing an example of a functional configuration of the dimension measurement system 1 according to the embodiment. The dimension measurement system 1 includes, for example, an illumination device 100, an imaging device 200, a dimension measurement device 300, an information input unit 400, a teacher data storage unit 500, and a notification unit 600.

[0016] The dimension measuring device 300 includes, for example, a measuring unit 310, a learning unit 320, an estimating unit 330, and an illumination control unit 340. The functional units, such as the measuring unit 310, the learning unit 320, the estimating unit 330, and the illumination control unit 340, are realized by a processor, such as a CPU (Central Processing Unit), executing a program stored in a program memory. In addition, some or all of these functional units may be realized by hardware, such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array), or may be realized by software and hardware working together. The program may be stored in advance in a storage device (a storage device having a non-transient storage medium), such as an HDD (Hard Disc Drive) or a flash memory, of the dimension measuring device 300, or may be stored in a removable storage medium, such as a DVD or a CD-ROM, and installed in the HDD or flash memory of the dimension measuring device 300 by mounting the storage medium (non-transient storage medium) in a drive device.

[0017] The measurement unit 310 measures the dimensions of the embossed carrier tape 10 using the image captured by the imaging device 200. The learning unit 320 constructs an estimation model for estimating optimal lighting conditions for measuring the dimensions of the embossed carrier tape 10. The estimation unit 330 performs calculations using the estimation model to estimate optimal lighting conditions for measuring the dimensions of the embossed carrier tape 10. The lighting control unit 340 controls the lighting device 100 according to the lighting conditions estimated by the estimation unit 330.

[0018] The information input unit 400 is, for example, a user interface device operated by a user. The information input unit 400 receives object information based on a user's operation and outputs the object information to the dimension measuring device 300.

[0019] The teacher data storage unit 500 is realized by a storage device such as a HDD, a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), a ROM (Read Only Memory), or a RAM (Random Access Memory). The teacher data storage unit 500 may be realized by a SAN (Storage Area Network) or a NAS (Network Attached Storage). The teacher data storage unit 500 stores lighting condition information and object information as teacher data. The lighting condition information includes at least one lighting condition of the arrangement of a plurality of lights constituting the lighting device 100, the light intensity, the light angle, and the light wavelength. The object information is information indicating a portion of the embossed carrier tape 10 as the object, but is not limited thereto. The object information may be physical property information including at least one of the shape, color, transmittance, surface properties, molding method, and mold shape of the embossed carrier tape 10. The object information may be an image of the embossed carrier tape 10, or measurement information including a peak value of reflected light from the embossed carrier tape 10.

[0020] The notification unit 600 is, for example, a display or a speaker, and notifies, for example, the results of dimensional measurement, the results of estimating the lighting conditions, the results of controlling the lighting device 100, and the like.

[0021] 3 is a block diagram showing an example of the learning unit 320 in the embodiment. The learning unit 320 includes, for example, a teacher data acquisition unit 321, a model construction unit 322, a Convolutional Neural Network (CNN) 323, and a learning result storage unit 324.

[0022] FIG. 4 is a diagram showing the relationship between input variables and output variables in the convolutional neural network 323. The input variables are object information, and the output variables are lighting condition information. In FIG. 4, the input variables and output variables included in each row correspond to one embossed carrier tape 10. For example, the shape information of the embossed carrier tape 10 having product information "1" is a1, the transparency information is b1, and the surface property information is c1, and the arrangement information of the lighting device 100 optimal for measuring the dimensions of the embossed carrier tape 10 having product information "1" is A1, the wavelength information is B1, and the intensity information is C1.

[0023] The teacher data acquisition unit 321 acquires object information and illumination condition information as teacher data from the teacher data storage unit 500. The model construction unit 322 inputs object information to the convolutional neural network 323 and learns processing parameters of the convolutional neural network 323 so as to output output variables from the convolutional neural network 323. Specifically, the model construction unit 322 uses the teacher data to recursively calculate (update) processing parameters so as to reduce the difference between the output variables corresponding to the input variables shown in FIG. 4 and the target output. The model construction unit 322 performs, for example, deep learning to acquire the processing parameters. Deep learning is machine learning using a multi-layered neural network, particularly a neural network with three or more layers. In the embodiment, the convolutional neural network 323 is used as the neural network with the multi-layered structure. The model construction unit 322 stores the updated processing parameters in the learning result storage unit 324.

[0024] FIG. 5 is a block diagram showing an example of the estimation unit 330 in the embodiment. The estimation unit 330 includes an object information acquisition unit 331, an estimation processing unit 332, a convolutional neural network 333, and an estimation result storage unit 334. The object information acquisition unit 331 acquires object information from the information input unit 400 and / or a captured image from the imaging device 200. The convolutional neural network 333 is an estimation model in which processing parameters stored in the learning result storage unit 324 are set. When object information and / or an image are input, the convolutional neural network 333 outputs illumination condition information. The estimation processing unit 332 stores the estimation result output from the convolutional neural network 333 in the estimation result storage unit 334. The estimation result includes information such as the arrangement, wavelength, and intensity of the illumination device 100 as information corresponding to the illumination condition information shown in FIG. 4.

[0025] FIG. 6 is a top view showing an example of the arrangement of the lighting device 100 in the embodiment. In FIG. 6, for example, one circle indicates one lighting element 100a. The lighting device 100 is configured, for example, by arranging a plurality of lighting elements 100a radially from a virtual central circle. The lighting device 100 is arranged so that the center of the central circle intersects with the production line L. FIG. 7 is a top view showing another example of the arrangement of the lighting device 100 in the embodiment. The lighting device 100 is configured, for example, by arranging a plurality of lighting elements 100a along a plurality of straight lines extending vertically from each side of a virtual quadrangle. The lighting device 100 is arranged so that the center of gravity of the quadrangle intersects with the production line L. The dimension measurement system 1 can, for example, switch the arrangement of FIG. 6 or FIG. 7 based on the estimation result according to the object information of the embossed carrier tape 10, adjust the wavelength of the light irradiated from each lighting element 100a, or adjust the intensity of the light irradiated by each lighting element 100a.

[0026] 8 and 9 are side views of the lighting device 100 in the embodiment. As shown in FIG. 8, the lighting device 100 has a plurality of lighting elements 100a arranged along upper and lower arcs with respect to the embossed carrier tape 10 conveyed on the manufacturing line L. Each lighting element 100a may be positioned so that the irradiated light is directed toward a predetermined range of the manufacturing line L, and may be dimmed so that the irradiated light is directed toward the predetermined range of the manufacturing line L. A certain lighting element 100a irradiates the irradiated light to the predetermined range via a mirror 110. The imaging device 200 is arranged at a position to image the predetermined range of the manufacturing line L. The dimension measurement system 1 may also irradiate the embossed carrier tape 10 with the irradiated light from a part of the lighting elements 100a among the plurality of lighting elements based on the estimation result, as shown in FIG. 9.

[0027] FIG. 10 is a top view showing an example of an embossed carrier tape 10 of an embodiment. A plurality of recesses 10a-1 and 10a-2 are formed on the embossed carrier tape 10. The dimension measuring system 1 measures the dimensions of the recesses 10a-1 and 10a-2, for example, by acquiring the coordinate values ​​(X, Y) of points P1, P2, P3, P4, P5, and P6 and calculating the distance between each point. The dimension measuring system 1 measures the width direction dimension of the embossed carrier tape 10, for example, the Y direction distance from point P1 to point P6, measures the X direction dimension and Y direction dimension of the recess 10a-1 by calculating the X direction distance and Y direction distance from point P2 to P3, and measures the X direction dimension and Y direction dimension of the recess 10a-2 using points P4 and P5.

[0028] FIG. 11 is a diagram for explaining an example of a mold for molding the embossed carrier tape 10 in the embodiment. The inner wall surface of the embossed carrier tape 10 changes depending on various factors such as the molding method and the mold shape. When the embossed carrier tape 10 is molded by pressing the mold X against the inner wall of the recess as shown in FIG. 11(a), the inner wall surface is formed into a regulated shape by the mold X, and as a result, light and dark lines used for measurement appear at positions such as X and Y. On the other hand, when the embossed carrier tape 10 is molded by pressing the mold Y against the outer wall of the recess as shown in FIG. 11(b), the outer wall surface is formed into a regulated shape by the mold Y, and as a result, light and dark lines used for measurement appear at positions such as X and Y. Positions such as X and Y vary greatly depending on the shape, color, transparency, surface properties, molding method, and mold shape. In this way, the target information includes the molding method and mold shape of the embossed carrier tape 10, and it is desirable to estimate the optimal lighting conditions for dimension measurement based on the molding method and mold shape of the embossed carrier tape 10.

[0029] 12 and 13 are diagrams showing an example of changes in the measurement value in the embodiment, where FIG. 12 shows changes in values ​​measured in a state where the embossed carrier tape 10 is irradiated with light under illumination condition A, and FIG. 13 shows changes in values ​​measured in a state where the embossed carrier tape 10 is irradiated with light under illumination condition B. When the embossed carrier tape 10 is measured under illumination condition A, as shown in FIG. 12, a peak value in the appropriate range cannot be obtained at position P on the embossed carrier tape 10. However, when the embossed carrier tape 10 is measured under illumination condition B, as shown in FIG. 13, a peak value in the appropriate range can be obtained at position P on the embossed carrier tape 10. In this way, there are cases where the peak of the measurement value cannot be detected due to differences in illumination conditions. The measurement value is, for example, a pixel value corresponding to the position of the embossed carrier tape 10. It is desirable that the dimension measurement system 1 learns an estimation model using an image of the embossed carrier tape 10 or measurement information including a peak value of reflected light from the embossed carrier tape 10 as object information.

[0030] Fig. 14 is a diagram showing the relationship between the sites A, B, and C on the embossed carrier tape 10 in the embodiment, the ideal values, and the measured values ​​for each of the illumination conditions A, B, and C. The measured values ​​show the average value and the variation of values ​​measured multiple times; for example, the average value of the measured values ​​of the site A under the illumination condition A is 12.5, and the variation is ±0.02. Fig. 14 shows that the illumination condition B is the optimum illumination condition for measuring the dimension of the site A, the illumination condition A is the optimum illumination condition for measuring the dimension of the site B, and the illumination condition C is the optimum illumination condition for measuring the dimension of the site C.

[0031] [Convolutional Neural Network] Hereinafter, a description will be given of an example of the convolutional neural network 323 and the convolutional neural network 333 in the embodiment. In this description, the convolutional neural network 323 and the convolutional neural network 333 will be collectively referred to as a "convolutional neural network."

[0032] As described above, the model construction unit 322 sets the teacher data as an input variable to be input to the input layer of the learning convolutional neural network 323, and sets the lighting condition information as an output variable to be output from the output layer. The model construction unit 322 performs machine learning using a learning data set of object information and lighting condition information. The estimation processing unit 332 inputs the object information to the input layer of the trained convolutional neural network 333, and acquires the lighting condition information from the output layer.

[0033] FIG. 15 is a diagram for explaining a process of performing an estimation process using a convolutional neural network in an embodiment. The convolutional neural network includes, for example, a layer L0, a layer L1, a layer L2, a layer Li, and a layer LI. The layer L0 is also called an input layer, the layers L1 to Li are also called intermediate layers or hidden layers, and the layer LI is also called an output layer. In the convolutional neural network, object information is input to the input layer L0. The object information is represented by a matrix D11 in which parameters of various information are the positions of the matrix. Each element of the matrix D11 includes an R (red) sub-value, a G (green) sub-value, and a B (blue) sub-value as sub-values ​​of parameters corresponding to the positions of the matrix. The first intermediate layer L1 is a layer in which convolution processing (also called filtering processing) and pooling processing are performed.

[0034] (Convolution) An example of the convolution process of the intermediate layer L1 will be described. The convolution process is a process of filtering the original parameters and outputting a feature map. Specifically, the input parameters are divided into a sub-matrix D121 of R, a sub-matrix D122 of B, and a sub-matrix D123 of G. For each of the sub-matrices D121, D122, and D123 (each of which is also referred to as a "sub-matrix D12"), the elements of the sub-matrix are multiplied by the elements of a convolution matrix CM1 (also called a kernel) of s rows and t columns for each sub-matrix of s rows and t columns, and then added together to calculate a first parameter value. The first parameter values ​​calculated in each of the sub-matrixes D12 are multiplied by a weighting coefficient and then added together to calculate a second parameter value. The second parameter values ​​are set as the elements of the convolution matrix D131 as matrix elements corresponding to the positions of the sub-matrix. In each sub-matrix D12, the position of the submatrix is ​​shifted for each element (sub-parameter), so that the second parameter value at each position is calculated, and all matrix elements of the convolution matrix D131 are calculated.

[0035] FIG. 15 is an example of a convolutional neural network using a 3-row, 3-column convolution matrix CM (convolution matrix)1, and the first parameter value D1311 is calculated for the 3-row, 3-column submatrix from the 2nd row to the 4th row and the 2nd column to the 4th column of each submatrix D12. A weighting factor is calculated and added to each of the first parameter values ​​of the submatrices D121, D122, and D123, thereby calculating a second parameter value as the matrix element in the 2nd row, 2nd column of the convolution matrix D131. Similarly, the second parameter value of the matrix element in the 3rd row, 2nd column of the convolution matrix D131 is calculated from the submatrix from the 3rd row to the 5th row and the 2nd column to the 4th column. Similarly, the convolution matrices D132, ... are calculated using other weighting factors or other convolution matrices.

[0036] (Pooling) An example of the pooling process of the intermediate layer L1 will be described. The pooling process is a process for reducing a parameter group (matrix) while retaining the features of the object information. Specifically, for each region PM (pooling matrix) of u rows and v columns in the convolution matrix D131, a representative value of the matrix elements in the region is calculated. The representative value is, for example, the maximum value of the matrix elements in the region. The representative value is set for each element of the CNN matrix D141 as a matrix element corresponding to the position of the region PM. The region in the convolution matrix D131 is shifted for each region PM, so that a representative value at each position is calculated, and all elements of the CNN matrix D141 are calculated.

[0037] FIG. 15 is an example of a convolutional neural network using a region PM of 2 rows and 2 columns, for example, and for the region PM of 2 rows and 2 columns from the 3rd row to the 4th row and the 3rd column to the 4th column of the convolution matrix D131, the maximum value in the region PM is calculated as a representative value. This representative value is set to the matrix element of the 2nd row and the 2nd column of the CNN matrix D141. Similarly, a representative value of the matrix element of the 3rd row and the 2nd column of the CNN matrix D141 is calculated from the submatrix of the 5th row to the 6th row and the 2nd column to the 4th column. Similarly, CNN matrices D142, ... are calculated from the convolution matrices D132, ....

[0038] Each matrix element (N elements) of the CNN matrices D141, D142, ... is generated as a vector X by arranging them in a predetermined order. The vector X is generated by arranging N elements x n (n=1, 2, 3, . . . N). Vector X is a vector u (0) is equivalent to.

[0039] The hidden layer Li represents the i-th (i = 2, ) hidden layer. The hidden layer Li is a vector u (i) Each node in the hidden layer Li receives a vector z (i) The vector z (i) is the vector u (i) is the activation function f(u (i) ) has the values ​​entered in the vector u (i) is the vector z output from the i-1th hidden layer node. (i-1) The weight matrix W (i) and the vector b (i) The vector b has a value obtained by adding (i) is the bias.

[0040] The output layer LI is a vector z (I-1) The output of the output layer LI is M y m (m=1, 2, . . . M). In other words, the output layer LI is a set of M y m A vector Y(y1, y2, y3, y M ) is output. As described above, when parameters of object information are input as input variables, the convolutional neural network outputs vector Y as an output variable. Vector Y in the embodiment represents lighting condition information.

[0041] Fig. 16 is a diagram for explaining the learning process using a convolutional neural network in the embodiment. The vector output from the first hidden layer L1 for the parameter values ​​of the object information in the learning data set is defined as vector [X]. The vector representing the determined class in the learning data set is defined as vector [Y]. The determined class is the output coefficient (illumination condition information such as the arrangement of the lighting elements, wavelength, intensity, etc.) shown in Fig. 4 in the embodiment.

[0042] Weight matrix W (i) An initial value is set for . When a vector [X] is input to the second hidden layer L2 based on the object information input to the input layer L0, a vector Y(X) based on the vector [X] is output from the output layer LI. The error E between vector Y(X) and vector [Y] is calculated using a loss function. The gradient ΔE of the i-th layer i is the output from each layer z i and the error signal δ i The error signal δ is calculated using i is the error signal δ i-1 The process of calculating the error signal of the input layer side from the error signal of the output layer side from the output layer side to the input layer side from the output layer side is also called backpropagation. (i) is the gradient ΔE i Similarly, in the first hidden layer L1, the convolution matrix CM or the weighting coefficients are updated.

[0043] [Estimation model settings] The model construction unit 322 sets processing parameters for the convolutional neural network, such as the number of layers, the number of nodes in each layer, the node connection method between each layer, the activation function, the error function, the gradient descent algorithm, the pooling region, the kernel, the weighting coefficient, and the weighting matrix. For example, the model construction unit 322 sets the number of layers to three (I=3). As the number of nodes in each layer (also referred to as the "node number"), the model construction unit 322 sets the number of elements of the vector X (node ​​number N) to 800, the number of nodes in the second intermediate layer (i=2) to 500, and the number of nodes in the output layer (i=3) to 10. However, the embodiment is not limited to this, and the total number may be four or more layers, and a different value may be set for the number of nodes.

[0044] The model construction unit 322 sets 20 convolution matrices CM of 5 rows and 5 columns, and sets a region PM of 2 rows and 2 columns. However, the embodiment is not limited to this, and a different number of matrices or a different number of convolution matrices CM may be set. Also, a region PM of a different number of matrices may be set. The model construction unit 322 may perform more convolution processes or pooling processes.

[0045] The model construction unit 322 sets a full connection as the connection of each layer of the convolutional neural network. However, the embodiment is not limited to this, and the connections of some or all layers may be set to non-full connection. The model construction unit 322 sets a sigmoid function as the activation function of all layers. However, the embodiment is not limited to this, and the activation function of each layer may be other activation functions such as a step function, linear combination, soft sine, soft plus, ramp function, truncated power function, polynomial, absolute value, radial basis function, wavelet, maxout, etc. Also, the activation function of a certain layer may be a different type from that of other layers.

[0046] The model construction unit 322 sets squared loss (mean squared error) as the error function. However, the present invention is not limited to this, and the error function may be cross entropy, τ-quantile loss, Huber loss, or ε sensitivity loss (ε tolerance error function). In addition, the model construction unit 322 sets SGD (stochastic gradient descent) as an algorithm (gradient descent algorithm) for calculating the gradient. However, the present invention is not limited to this, and the gradient descent algorithm may be Momentum (inertia term) SDG, AdaGrad, RMSprop, AdaDelta, Adam (adaptive moment estimation), or the like.

[0047] The model construction unit 322 may set other neural networks such as a perceptron neural network, a recurrent neural network (RNN), a residual network (ResNet), etc. The model construction unit 322 may set a part or all of a supervised learning estimation model such as a decision tree, a regression tree, a random forest, a gradient boosting tree, a linear regression, a logistic regression, or an SVM (support vector machine).

[0048] [Effects of the embodiment] As described above, according to the embodiment, it is possible to realize a learning device (learning unit 320) including: a teacher data acquisition unit 321 that inputs lighting condition information related to the lighting conditions under which the lighting device 100 irradiates light to measure the dimensions of the embossed carrier tape 10, and object information for identifying the embossed carrier tape 10; and a model construction unit 322 that constructs a convolutional neural network 323, which is a model trained by machine learning using the lighting condition information and object information input by the teacher data acquisition unit 321 as teacher data and outputs lighting condition information related to the lighting condition optimal for measuring the dimensions of the object when the object information is input. According to this embodiment, it is possible to obtain lighting condition information optimal for measuring the dimensions of the embossed carrier tape 10 to be measured simply by inputting information about the embossed carrier tape 10 to be measured into the model, so that the dimensions of the embossed carrier tape 10 can be measured with high accuracy.

[0049] However, since the accuracy and reproducibility of the dimension measurement results change significantly depending on the lighting conditions, it may be difficult to find the optimal lighting conditions for dimension measurement. Conventionally, the time required for the work of adjusting the optimal lighting conditions for dimension measurement depending on the characteristics of the embossed carrier tape 10 varies depending on the person performing the dimension measurement. However, since the optimal lighting conditions vary depending on the characteristics of the embossed carrier tape 10 such as the material such as the color and transparency of the embossed carrier tape 10, the surface property (reflective, glossy or non-glossy), the shape (flat or uneven), and the part of the embossed carrier tape 10, it may take a long time to estimate the optimal lighting conditions with high accuracy. In contrast, according to the embodiment, the estimation model described above is constructed, so that the estimation time for the lighting conditions can be shortened.

[0050] Moreover, according to the embodiment, the arrangement of the lighting device 100, the light intensity, the light angle, and the light wavelength are used as the lighting condition information, so that it is possible to estimate the arrangement of the lighting device 100, the light intensity, the light angle, and the light wavelength that are optimal for the dimensional measurement. Furthermore, according to the embodiment, the part of the embossed carrier tape 10 is used as the object information, so that it is possible to estimate the optimal lighting condition for the dimensional measurement of the part of the embossed carrier tape 10. Furthermore, according to the embodiment, the physical property information of the embossed carrier tape 10, such as the shape, color, transmittance, surface property, molding method, and mold shape, is used as the object information, so that it is possible to estimate the optimal lighting condition for the dimensional measurement of the embossed carrier tape 10 indicated by the physical property information. Furthermore, according to the embodiment, the image of the embossed carrier tape 10 or the measurement information including the peak value of the reflected light from the embossed carrier tape 10 is used as the object information, so that it is possible to estimate the optimal lighting condition for the dimensional measurement using the measurement information.

[0051] Also, according to the embodiment, it is possible to realize an estimation device (estimation unit 330) including an object information acquisition unit 331 that inputs object information for identifying the embossed carrier tape 10, and an estimation processing unit 332 that inputs the object information input by the object information acquisition unit 331 into a model that has been machine-learned using training data of object information and illumination condition information related to the illumination conditions under which the illumination device 100 irradiates light to measure the dimensions of the embossed carrier tape 10, and estimates optimal illumination conditions for measuring the dimensions of the embossed carrier tape 10 based on illumination condition information output from the model. According to this embodiment, it is possible to acquire optimal illumination condition information for measuring the dimensions of the embossed carrier tape 10 simply by inputting information about the embossed carrier tape 10 to be measured into the model, and therefore it is possible to measure the dimensions of the embossed carrier tape 10 with high accuracy.

[0052] Furthermore, according to the embodiment, it is possible to realize a dimension measurement system 1 including an illumination device 100 that irradiates light onto the embossed carrier tape 10, a measurement unit 310 that measures the dimensions of the embossed carrier tape 10, an information input unit 400 that inputs object information for identifying the embossed carrier tape 10, and an illumination control unit 340 that controls the illumination device 100 based on the object information input by the information input unit 400. According to this embodiment, it is possible to obtain optimal illumination condition information for measuring the dimensions of the embossed carrier tape 10 that is the measurement target simply by inputting information about the embossed carrier tape 10 that is the measurement target into a model, and to control the illumination device 100 to achieve the illumination conditions. As a result, according to the dimension measurement system 1, it is possible to measure the dimensions of the embossed carrier tape 10 with high accuracy.

[0053] The program of the dimension measuring device 300 in one aspect of the present invention may be a program that controls one or more processors such as a CPU (a program that makes a computer function) so as to realize the functions shown in each of the above-mentioned embodiments and modifications related to one aspect of the present invention. Information handled by each of these devices may be temporarily stored in a RAM during processing, and then stored in various storage devices such as a flash memory, SSD, or HDD, and may be read, modified, or written by the CPU as necessary.

[0054] In addition, a part or all of the learning unit 320 or the estimation unit 330 included in the dimension measuring device 300 in each of the above-mentioned embodiments and modifications may be realized by a computer having one or more processors. In that case, the control function may be realized by recording a program for realizing the control function in a computer-readable recording medium, reading the program recorded in the recording medium into a computer system, and executing the program. In addition, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, an optical magnetic disk, a ROM, a CD-ROM, and a storage device such as a hard disk built into a computer system. Furthermore, the "computer-readable recording medium" may also include a medium that dynamically holds a program for a short time, such as a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, and a medium that holds a program for a certain period of time, such as a volatile memory inside a computer system that is a server or client in that case. In addition, the above-mentioned program may be a program for realizing a part of the above-mentioned functions, and may further be a program that can realize the above-mentioned functions in combination with a program already recorded in the computer system.

[0055] In addition, a part or the whole of the learning unit 320 or the estimation unit 330 included in the dimension measuring device 300 in each of the above-mentioned embodiments and modifications may be realized as an LSI, which is typically an integrated circuit, or may be realized as a chip set. The integrated circuit method is not limited to LSI, and may be realized by a dedicated circuit and / or a general-purpose processor. In addition, when an integrated circuit technology that replaces LSI appears due to the progress of semiconductor technology, it is also possible to use an integrated circuit based on that technology.

[0056] Although the embodiments and modifications have been described above in detail with reference to the drawings as one aspect of the present invention, the specific configuration is not limited to the embodiments and modifications, and includes design changes within the scope of the gist of the present invention. In addition, one aspect of the present invention can be modified in various ways within the scope of the claims, and embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. In addition, configurations in which elements described in the above embodiments and modifications are replaced with elements that have the same effect are also included.

[0057] For example, one aspect of the present invention may be realized by combining some or all of the above-described embodiments.

[0058] The above describes the form for carrying out the present invention using an embodiment, but the present invention is not limited to such an embodiment, and various modifications and substitutions can be made within the scope that does not deviate from the gist of the present invention. [Explanation of symbols]

[0059] 1 Dimensional measurement system 100 Lighting Equipment 200 Imaging device 300 Dimensional Measuring Device 310 Measuring section 320 Learning Department 330 Estimation Department 340 Lighting control section 400 Information input section 500 Teacher data storage unit 600 Notification Department

Claims

1. An input section for inputting lighting condition information relating to lighting conditions under which a lighting device irradiates light in order to measure the dimensions of an embossed carrier tape made of plastic and having successively formed storage pockets for storing electronic components, and object information for identifying the embossed carrier tape; a model construction unit that constructs a model that has been machine-learned using the illumination condition information and the object information input by the input unit as training data, and that outputs the illumination condition information relating to optimal conditions for measuring the dimensions of the embossed carrier tape when the object information is input; A learning device in which the object information indicates a portion of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape, and the model outputs optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape when information indicating the portion of the embossed carrier tape is input.

2. An input section for inputting lighting condition information relating to lighting conditions under which a lighting device irradiates light to measure the dimensions of an embossed carrier tape made of plastic and having continuously formed storage pockets for storing electronic components, and object information for identifying the embossed carrier tape; a model construction unit that constructs a model that has been machine-learned using the illumination condition information and the object information input by the input unit as training data, and that outputs the illumination condition information relating to optimal conditions for measuring the dimensions of the embossed carrier tape when the object information is input; The object information indicates physical property information including the molding method and mold shape of the embossed carrier tape, the lighting condition information indicates the optimal lighting conditions for measuring the dimensions of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape, and the model outputs the optimal lighting conditions for measuring the dimensions of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape when information indicating the molding method and mold shape of the embossed carrier tape is input.

3. An input unit for inputting object information for identifying an embossed carrier tape made of plastic and having successively formed storage pockets for storing electronic components; an estimation unit that inputs the object information input by the input unit into a machine-learned model that uses lighting condition information related to lighting conditions under which a lighting device irradiates light to measure the dimensions of the embossed carrier tape and the object information as training data, and estimates optimal lighting conditions for measuring the dimensions of the embossed carrier tape based on the lighting condition information output from the model, The object information indicates a portion of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape, and the model outputs optimal lighting conditions for measuring the dimensions of the portion of the embossed carrier tape when information indicating the portion of the embossed carrier tape is input.

4. An input unit for inputting object information for identifying an embossed carrier tape made of plastic and having successively formed storage pockets for storing electronic components; an estimation unit that inputs the object information input by the input unit into a machine-learned model that uses lighting condition information related to lighting conditions under which a lighting device irradiates light to measure the dimensions of the embossed carrier tape and the object information as training data, and estimates optimal lighting conditions for measuring the dimensions of the embossed carrier tape based on the lighting condition information output from the model, The object information indicates physical property information including the molding method and mold shape of the embossed carrier tape, the lighting condition information indicates the optimal lighting conditions for measuring the dimensions of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape, and the model outputs the optimal lighting conditions for measuring the dimensions of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape when information indicating the molding method and mold shape of the embossed carrier tape is input.

5. An illumination device that irradiates light onto an embossed carrier tape made of plastic and having continuously formed storage pockets for storing electronic components; a measuring unit for measuring the dimensions of the embossed carrier tape; an input unit for inputting object information for identifying the embossed carrier tape; a control unit that controls the lighting device based on the object information input by the input unit, the control unit inputs the object information input by the input unit into a model that has been machine-learned using lighting condition information related to lighting conditions under which an illumination device irradiates light to measure the dimensions of the embossed carrier tape and the object information as training data, estimates optimal lighting conditions for measuring the dimensions of the embossed carrier tape based on the lighting condition information output from the model, and controls the illumination device based on the estimation result; the object information indicates a portion of the embossed carrier tape, the lighting condition information indicates an optimum lighting condition for measuring the dimensions of the portion of the embossed carrier tape, and the model outputs the optimum lighting condition for measuring the dimensions of the portion of the embossed carrier tape when information indicating the portion of the embossed carrier tape is input. Dimensional measuring device.

6. An illumination device that irradiates light onto an embossed carrier tape made of plastic and having continuously formed storage pockets for storing electronic components; a measuring unit for measuring the dimensions of the embossed carrier tape; an input unit for inputting object information for identifying the embossed carrier tape; a control unit that controls the lighting device based on the object information input by the input unit, the control unit inputs the object information input by the input unit into a model that has been machine-learned using lighting condition information related to lighting conditions under which an illumination device irradiates light to measure the dimensions of the embossed carrier tape and the object information as training data, estimates optimal lighting conditions for measuring the dimensions of the embossed carrier tape based on the lighting condition information output from the model, and controls the illumination device based on the estimation result; the object information indicates physical property information including a molding method and a mold shape of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for measuring the dimensions of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape, and the model outputs optimal lighting conditions for measuring the dimensions of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape when information indicating the molding method and mold shape of the embossed carrier tape is input. Dimensional measuring device.

7. A model that has been machine-learned using training data as training data lighting condition information regarding the lighting conditions under which a lighting device emits light to measure the dimensions of an embossed carrier tape made of plastic and having continuously formed storage pockets for storing electronic components, and object information for identifying the embossed carrier tape, the model outputting lighting condition information regarding the lighting conditions that are optimal for measuring the dimensions of the embossed carrier tape when the object information is input; inputting object information for identifying the embossed carrier tape; inputting the input object information into the model, estimating optimal lighting conditions for measuring the dimensions of the embossed carrier tape based on the lighting condition information output from the model, and controlling the lighting device based on the estimation result; measuring the dimensions of the embossed carrier tape; the object information indicates a portion of the embossed carrier tape, the lighting condition information indicates an optimum lighting condition for measuring the dimensions of the portion of the embossed carrier tape, and the model outputs the optimum lighting condition for measuring the dimensions of the portion of the embossed carrier tape when information indicating the portion of the embossed carrier tape is input. Dimensional measurement method.

8. A model that has been machine-learned using training data as training data lighting condition information regarding the lighting conditions under which a lighting device emits light to measure the dimensions of an embossed carrier tape made of plastic and having continuously formed storage pockets for storing electronic components, and object information for identifying the embossed carrier tape, the model outputting lighting condition information regarding the lighting conditions that are optimal for measuring the dimensions of the embossed carrier tape when the object information is input; inputting object information for identifying the embossed carrier tape; inputting the input object information into the model, estimating optimal lighting conditions for measuring the dimensions of the embossed carrier tape based on the lighting condition information output from the model, and controlling the lighting device based on the estimation result; measuring the dimensions of the embossed carrier tape; the object information indicates physical property information including a molding method and a mold shape of the embossed carrier tape, the lighting condition information indicates optimal lighting conditions for measuring the dimensions of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape, and the model outputs optimal lighting conditions for measuring the dimensions of the embossed carrier tape corresponding to the molding method and mold shape of the embossed carrier tape when information indicating the molding method and mold shape of the embossed carrier tape is input. Dimensional measurement method.

Images