Application of Improved Snake Model in Segmentation of Korean Pine Cone Image

In the Korean Pine solid quantity´s forecast technique, the characteristic of Korean Pine cone´s shape is one of main parameters. This paper can provide the precise data for the Korean Pine solid quantity´s forecast technique by segmenting the image of Korean Pine which is taken by the Filed Server. Considering of the complex background of Korean Pine image and the target of hollow contours, we proposed a Snake model image segmentation algorithm that is improved by Ant Colony Algorithm. First, the overall robustness advantage of the Ant Colony Algorithm is used to gain target contours. Then the contours are set to the improved Snake model´s starting value and overcome the original Snake model´s drawbacks. Finally, we can obtain the complete target. The experiment has proven the algorithm´s validity and precision.