Recognition of very low quality images using active preprocessing and environment adaptive learning

2016 Fiscal Year Final Research Report

• Project/Area Number: 24240028
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: Nagoya University
• Principal Investigator: Murase Hiroshi 名古屋大学, 情報科学研究科, 教授 (90362293)
• Co-Investigator(Kenkyū-buntansha): 井手 一郎 名古屋大学, 情報科学研究科, 准教授 (10332157) ; 出口 大輔 名古屋大学, 情報連携統括本部, 准教授 (20437081) ; 川西 康友 名古屋大学, 情報科学研究科, 助教 (50755147) ; 平山 高嗣 名古屋大学, 情報科学研究科, 准教授 (10423021) ; 高橋 友和 岐阜聖徳学園大学, 経済情報学部, 准教授 (90397448) ; 目加田 慶人 中京大学, 工学部, 教授 (00282377)
• Project Period (FY): 2012-04-01 – 2017-03-31
• Keywords: パタン認識 / 画像処理 / 監視カメラ / 車載カメラ

Outline of Final Research Achievements

The purpose of this research is to develop an accurate recognition algorithm for low quality images such as image sequences from in-vehicle cameras or surveillance cameras. We proposed several new algorithms including pre-processing methods such as super-resolution or region selective feature extraction, and location adaptive recognition methods using on-line learning. We also proposed a visibility estimation method for appropriate feedback of recognition results. Experimental results on real-world data demonstrated the superiority of the proposed methods in terms of recognition rate.

Free Research Field

総合領域