Low Quality Face Image Recognition Using Learning-Based Feature Transform

2014 Fiscal Year Final Research Report

• Project/Area Number: 23700219
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: Gifu Shotoku Gakuen University
• Principal Investigator: TAKAHASHI Tomokazu 岐阜聖徳学園大学, 経済情報学部, 准教授 (90397448)
• Project Period (FY): 2011-04-28 – 2015-03-31
• Keywords: 画像、文章、音声等認識 / 機械学習 / アルゴリズム

Outline of Final Research Achievements

To achieve accurate person identification from low quality face images and person images taken by a surveillance camera, we have developed five methods; a face pose transformation method a face image super-resolution method, a transformation method from far-infrared face image to visible light face image, a baggage detection method from person image, and a person image super-resolution method. We confirmed through experiments on actual face images and person images that each developed method works effectively for accurate person identification from low quality face images and person images

Free Research Field

パターン認識、画像処理