Development of a new principle of multi-spectrum-image-sensor that reduces the cost and improves sensitivity

2016 Fiscal Year Final Research Report

• Project/Area Number: 26540080
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Perceptual information processing
• Research Institution: Tokyo Institute of Technology
• Principal Investigator: Kumazawa Itsuo 東京工業大学, 科学技術創成研究院, 教授 (70186469)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: マルチスペクトル画像センサ / 画像処理 / マイクロプリズムアレイ / 回折格子 / 分光 / 高感度 / 逆問題 / 不良設定問題

Outline of Final Research Achievements

We investigated a new principle of multi-spectrum-image-sensor that uses a very simple optical system but a sophisticated post-numerical-processing to compensate for the incomplete capability of the optical system. The conventional multi-spectrum-image-sensor has been using a slit to avoid cross talks from other part of the object. This slit blocks the most of the optical energy and causes the low photographic sensitivity of the multi-spectrum-image-sensor. The principle proposed in this study does not use the slit and improves the sensitivity remarkably. In addition, the optical system can be small and light enough for a drone to carry. However, the principle needs an innovative post-numerical-processing to remove the cross talks. Various form of such simplified optical systems are prototyped. In addition, the latest techniques for the inverse-problem or deconvolution are applied to remove cross talks for each prototypes.

Free Research Field

画像処理、画像認識、ユーザインターフェイス