Identification code reading apparatus and machine learning device
An identification code reading apparatus includes a machine learning device for learning estimation of an ideal image of an identification code with respect to an image of the identification code read by the visual sensor. The machine learning device observes read image data related to an image of the identification code read by the visual sensor as a state variable representing a current state of an environment, and acquires ideal image data related to the ideal image of the identification code as label data. Then, the machine learning device learns the read image data and the ideal image data in association with each other using the observed state variable and the acquired label data.
1. An identification code reading apparatus for reading an identification code assigned to a surface of a product using a visual sensor, the identification code reading apparatus comprising:
a machine learning device for learning estimation of an ideal image of the identification code with respect to an image of the identification code read by the visual sensor,
wherein the machine learning device includes a processor configured to:
control positioning of the visual sensor relative to the identification code, or position the identification code relative to the visual sensor, and control the visual sensor to read the identification code,
observe read image data related to the image of the identification code read by the visual sensor as a state variable representing a current state of an environment,
acquire ideal image data related to the ideal image of the identification code as label data,
learn the read image data and the ideal image data in association with each other using the state variable and the label data, and estimate and output the ideal image of the identification code from the read image data.
2. The identification code reading apparatus according to claim 1 , wherein the processor is further configured to estimate and output the ideal image of the identification code based on a plurality of ideal images of the identification code estimated from each of a plurality of read image data.
3. The identification code reading apparatus according to claim 1 , wherein the processor is further configured to observe assignment state data related to a material of the product, a method of surface treatment, or a method of assigning the identification code as a state variable representing a current state of an environment.
4. The identification code reading apparatus according to claim 1 , wherein the visual sensor is fixed to an arm distal end portion or the product is gripped by the arm distal end portion to read the identification code.
5. A machine learning device for learning estimation of an ideal image of an identification code with respect to an image of the identification code read by a visual sensor, the machine learning device comprising:
a processor configured to:
control positioning of the visual sensor relative to the identification code, or position the identification code relative to the visual sensor, and control the visual sensor to read the identification code;
observe read image data related to the image of an identification code read by the visual sensor as a state variable representing a current state of an environment;
acquire ideal image data related to the ideal image of the identification code as label data;
learn the read image data and the ideal image data in association with each other using the state variable and the label data; and
estimate and output the ideal image of the identification code from the read image data.
6. A method for learning estimation of an ideal image of an identification code with respect to an image of the identification code read by a visual sensor, the method comprising:
controlling, by a processor of a machine learning device, a robot to position the visual sensor relative to the identification code, or position the identification code relative to the visual sensor, and controlling the visual sensor to read the identification code;
observing, by the processor of the machine learning device, read image data related to the image of the identification code read by the visual sensor as a state variable representing a current state of an environment;
acquiring, by the processor of the machine learning device, ideal image data related to the ideal image of the identification code as label data;
learning, by the processor of the machine learning device, the read image data and the ideal image data in association with each other using the state variable and the label data; and
estimating and outputting, by the processor of the machine learning device, the ideal image of the identification code from the read image data.