Environment adaptive image recognition using environment specific information by real world sensing

2015 Fiscal Year Final Research Report

• Project/Area Number: 25730110
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Perceptual information processing
• Research Institution: Nagoya University
• Principal Investigator: Deguchi Daisuke 名古屋大学, 情報連携統括本部, 准教授 (20437081)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 画像認識 / 環境適応 / 車載カメラ映像処理

Outline of Final Research Achievements

The aim of this research is to develop environment adaptive image recognition and object detection methods by finding environment specific information obtained from real world sensing. This research considered this task as the combination of two sub-tasks: (i) identification of environment specific information from real world sensing, and (ii) adaptation / reconstruction of object detection and recognition system according to an environment. To solve these tasks, we developed methods of location adaptive image recognition using GPS information, Parts Selective DPM for detecting pedestrians possessing umbrella, false positive mining for eliminating environment specific false positives.

Free Research Field

知覚情報処理・ロボティクス