Feature space optimization prior to fuzzy image classification
msra(2004)
摘要
This paper presents a method for features space optimization in a context of fuzzy image classification. Based on membership functions intersections, the method allows to select the most appropriate features for objects discrimination. Comparison of the eCognition nearest neighbor algorithms and fuzzy classification is provided with the use of un-optimized and optimized features sets.
更多查看译文
关键词
classification.,fuzzy sets,feature space optimization,dimensionality reduction
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要