在图像识别技术的实现过程中，图像分割是一个重要的预处理环节，图像分割效果，直接影响着后续的分类、目标识别、图像分析、图像理解等过程的结果。针对着不同的图像特点，目前已经提出了错综复杂的图像分割算法。其中基于图论的图像分割算法是近几年研究的热点，这类算法着眼于全局，更注重局部数据的处理，比一般方法可以获得更佳的效果，并且图论理论有着比较完备的数学理论基础，将其用于图像处理有着较好的应用前景。-In the implementation process of image recognition technology, image segmentation is an important pre-processing aspects of image segmentation directly affect the subsequent classification, target recognition, image analysis, image understanding the results of the process. For different image characteristics, the intricate image segmentation algorithm has been proposed. The hot image segmentation algorithm based on graph theory research in recent years, this type of algorithm looks at the overall situation and focus more on the local data processing can get better results than the average, and graph theory, the theory has a more complete mathematical theory the basis for image processing has good application prospects.