Systematising clustering techniques through cross-disciplinary research

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K11964
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61010:Perceptual information processing-related
• Research Institution: Kyushu University
• Principal Investigator: Inoue Kohei 九州大学, 芸術工学研究院, 准教授 (70325570)
• Co-Investigator(Kenkyū-buntansha): 原 健二 九州大学, 芸術工学研究院, 教授 (50380712)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: クラスタリング / 画像処理 / 非写実的レンダリング

Outline of Final Research Achievements

Clustering is used in a wide range of fields because it enables the automatic classification of data without the use of supervised data, and although various clustering techniques have been developed, the systematisation of these techniques has not progressed sufficiently. Therefore, in this study, we have worked to elucidate the relationships among clustering techniques and their applications in the hope that the systematisation of clustering techniques as a whole will improve their overall usefulness and lead to the development of new methods. Specifically, we have developed new methods of non-photorealistic halftoning and image processing. These methods can also be interpreted in a unified manner when viewed from the perspective of clustering.

Translated with DeepL.com (free version)

Free Research Field

パターン認識、画像処理

Academic Significance and Societal Importance of the Research Achievements

クラスタリングは、データ要約のための基本的技術の一つであり、多くの分野で利用されている。個々の分野で開発された方法を統一的な視点で眺め、それぞれの方法の相違点を明らかにすることは、各方法をより深く理解するのに役立つだけでなく、新たな方法を生み出すきっかけにもなる。