UM students receive best paper awards at iFuzzy2016
強化計算機理解圖像能力 澳大獲國際最佳論文獎

25 Nov 2016

(From left) Dr Chen Long, UM students Guo Li and Wu Yingwen, and Prof Philip Chen
（左起）陳龍、郭莉、吳穎文、陳俊龍

A paper written by PhD student Guo Li and master's student Wu Yingwen from the Faculty of Science and Technology (FST), University of Macau (UM), received the Best Student Paper Award - First Place and the only Best Paper Award in Theory, respectively, at the 2016 International Conference on Fuzzy Theory and Its Applications (iFuzzy2016).

The students' paper, titled ‘Image Guided Fuzzy C-Means for Image Segmentation’, focuses on image segmentation, a fundamental technique in computer vision that is widely used in vision-based artificial intelligence tasks, as in the case of self-driving cars and medical image understanding. However, images of poor quality like the medical ones obtained via X-Ray or Ultrasound are difficult for computers to analyse. This paper proposes an image segmentation method based on fuzzy theory to solve this problem and increase the computers' capability to understand low-quality images.

The paper is co-authored with FST Assistant Professor Chen Long and Dean Philip Chen. Approximately 100 papers have been presented at this year's conference.

澳門大學科技學院博士生郭莉、碩士生吳穎文的論文“針對圖像分割的圖像引導模糊C均值聚類算法＂賦予計算機思考能力，理解圖像內容。相關論文於“2016國際模糊理論及應用大會＂上獲專家學者一致良評，奪“最佳學生論文第一名＂及全場唯一的“最佳理論論文獎＂，顯示澳大計算機領域的研究水平獲得國際認可。

圖像分割是計算機視覺中的基礎任務，被廣泛應用於各種基於視覺的人工智能技術，如自動駕駛和醫學圖像理解中。但在一些實際應用，如對由X射線或超音波獲得的醫學影像進行分析時，不佳的圖像質量往往阻礙了對影像的理解。針對這個問題，論文利用模糊數學理論，提出了一種對低質量圖像也有較好效果的圖像分割方法，有效增強了計算機理解圖像的能力。論文由郭莉、吳穎文、澳大科技學院助理教授陳龍和院長陳俊龍共同撰寫。

今屆國際模糊理論及應用大會共有近百篇論文發表，旨在於提升與擴展模糊理論之廣泛應用。