Abstract: A fuzzy classification scheme for river channel image interpretation is proposed.Accurate extraction of river channel is very important for river channel image interpretation of pattern classification.Application-sensitive factor combination method，region growing and mathematical morphology are integrated to extract river channel.The feature vector is determined and the selected feature vector is calculated.Fuzzy recognition theory is used to identify the pattern of river channel.Experiments show satisfaction with the results of river channel image interpretation of pattern classification.

Key words: image interpretation, sensitive factor combination method, seeds growing, water system pattern

摘要： 研究了遥感图片的水体类型解译识别问题。水体图形的准确提取是遥感图片解译识别的关键问题。综合应用敏感因子组合法、种子点增长法和数学形态学的方法提取水体，在对提取的水体图像分析基础上，进一步确定了特征向量和采用模糊识别理论进行识别，并对识别结果进行了分析，实验结果显示提取的水体获得较满意的识别效果。

关键词: 图像解译, 敏感因子组合法, 种子点增长法, 水体类型