Title:
画像処理装置、画像処理方法及びプログラム
Document Type and Number:
Japanese Patent JP7350622
Kind Code:
B2
Abstract:
To provide an image processing apparatus which can present information relating to an estimation result obtained by a learned machine learning model, in such a form that it is easy for a user to recognize the information.SOLUTION: The image processing apparatus of an embodiment includes a partial state estimation unit, a consolidated state estimation unit, and a display processing unit. The partial state estimation unit individually inputs a plurality of partial images obtained by dividing an original image to a learned machine learning model to acquire a partial estimation result allowing specification of positions and the number of areas of interest and to generate partial state explanation information. The consolidated state estimation unit uses at least one of the partial state estimation result and the partial state explanation information to perform processing of a rule base for generating consolidated state explanation information. The display processing unit performs processing for displaying both non-linguistic visual information allowing the areas of interest to be visually recognized and a consolidated state explanation sentence generated in accordance with a character string included in the consolidated state explanation information.SELECTED DRAWING: Figure 2
Inventors:
Renshu Liang
Hiroyuki Mizutani
Hiroyuki Mizutani
Application Number:
JP2019196727A
Publication Date:
September 26, 2023
Filing Date:
October 29, 2019
Export Citation:
Assignee:
Toshiba Corporation
Toshiba Digital Solutions Corporation
Toshiba Digital Solutions Corporation
International Classes:
G06T7/00; G06F3/0484
Domestic Patent References:
JP2018072240A | ||||
JP2014109479A |
Attorney, Agent or Firm:
Patent Attorney Corporation Itosin International Patent Office