Research on image recognition and retrieval using space-filling curves

2014 Fiscal Year Final Research Report

• Project/Area Number: 24500216
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: Waseda University
• Principal Investigator: KAMATA Seiichiro 早稲田大学, 理工学術院, 教授 (00204602)
• Research Collaborator: SUGIMOTO Kenjiro ; ZHANG Qieshi ; RYU Jegoon ; YANG Zhuo ; HAO Pengyi ; ZHOU Wei ; YU Jiya ; WANG Chengjie ; WU Renjie
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 画像認識 / 画像検索 / 空間充填曲線 / 薬剤パッケージ / ヒルベルト走査

Outline of Final Research Achievements

In order to realize a large-scale image inspection system for preventing human error of medicine dispensing to patients at pharmacy, research on image recognition and retrieval using space-filling curves is introduced. First a linear manifold color descriptor is proposed for medicine package recognition. And then a new shapr descriptor called hyper-complex polar Fourier analysis is developed for medicine feature extraction. Finally using several space filling curves, an adaptive Hilbert scan based Bag-of-Features is proposed for image retrieval. This method is based on adaptive selection of several scanning methods like Hilbert scan, etc., and several experiments show that the proposed method performs better than other retrieval methods.

Free Research Field

画像処理、パターン認識