Title:
ANNOTATION DEVICE, LEARNING MODEL, IMAGE SENSOR, ANNOTATION METHOD, AND COMPUTER PROGRAM
Document Type and Number:
Japanese Patent JP2020107170
Kind Code:
A
Abstract:
To provide an annotation device that can efficiently generate data for learning for a discriminator.SOLUTION: An annotation device for generating data for learning used for learning a discriminator for identifying an object in image data, includes: a display control unit that displays object image data, which is image data to be learned, on a screen; a coordinate information acquisition unit that acquires coordinate information of a position specified by a user on the object image data; and a generation unit for data for learning that extracts, based on the acquired coordinate information, image data for learning, which is image data of an object region that is a region including the object, from the object image data for each coordinate information, and generates data for learning including image data for learning and a label of the object. Size of the object region is smaller as a distance from an imaging device that images the object to the object is longer.SELECTED DRAWING: Figure 2
More Like This:
Inventors:
TAKUBO SHINYA
Application Number:
JP2018246787A
Publication Date:
July 09, 2020
Filing Date:
December 28, 2018
Export Citation:
Assignee:
SUMITOMO ELECTRIC INDUSTRIES
International Classes:
G06T7/00
Attorney, Agent or Firm:
Atsumi Motoyuki
Mine straight road
Mine straight road