Invention Grant
- Patent Title: Recognizing objects by utilizing dictionary data and probability data determined by machine learning
- Patent Title (中): 通过利用机器学习确定的字典数据和概率数据识别对象
-
Application No.: US12903584Application Date: 2010-10-13
-
Publication No.: US08634656B2Publication Date: 2014-01-21
- Inventor: Kotaro Yano
- Applicant: Kotaro Yano
- Applicant Address: JP Tokyo
- Assignee: Canon Kabushiki Kaisha
- Current Assignee: Canon Kabushiki Kaisha
- Current Assignee Address: JP Tokyo
- Agency: Fitzpatrick, Cella, Harper & Scinto
- Priority: JP2009-240820 20091019
- Main IPC: G06K9/62
- IPC: G06K9/62

Abstract:
An information processing apparatus acquires registration image data related to an object and input image data related to an object, and matches the acquired registration image data related to the object and each of a plurality of object dictionary data items stored in a storage device. Based on the matching result, the information processing apparatus selects an object dictionary data item relevant to the object related to the registration image data from the plurality of object dictionary data items, and matches the acquired input image data related to the object and the selected object dictionary data item. Based on the matching result, the information processing apparatus determines the object related to the input image data. Based on the determination result, the information processing apparatus updates the selected object dictionary data item. Thus, object recognition is easily and highly accurately performed.
Public/Granted literature
- US20110091116A1 OBJECT RECOGNITION APPARATUS AND OBJECT RECOGNITION METHOD Public/Granted day:2011-04-21
Information query