Abstract: Object-oriented method of image analysis is effective to process high-resolution remote sensing data.Based on object-oriented analysis of imagery，a method of bridge target extraction is proposed in this paper.First，the original image is segmented by region-growing technique to produce image objects，which are treated as the basic processing units in image classification.After image classification，the water class is obtained.Second，the features of bridge are given and the recognition rule of bridge is analyzed.Using shape features of image objects and their contextual information，the bridge targets are extracted from a high spatial resolution remote sensing image.Finally the effeteness of the proposed method is tested in the experiment.

摘要： 面向对象的分析方法是一种有效的高分辨率遥感影像处理技术，提出一种基于图像对象的水上桥梁识别方法。首先采用区域生长方法对影像进行分割，以分割后产生的图像对象为基本处理单元进行分类，提取出水体类别。然后在分析桥梁目标特征的基础上，利用图像对象的形状特征，以及桥梁和水体的上下文关系特征，提取影像中的桥梁目标。最后以实验验证了所提出方法的有效性。