Abstract: The morphological characteristic of male and female flowers is one of the main parameters in the prediction methods of Korean pine cone production. This method is established on the basis of the image segmentation of Korean pine’s male and female flowers from field server-based remote. Image segmentation provides the precise data for the prediction methods of cone production by extracting the feature of Korean pine male and female flowers. The method of extracting width to length is improved which assumes that the disconnected area of the contour diagram is connected by calculating their shortest distance. Width to length from the contour diagram is extracted by calculating minimum enclosing rectangle. The method is proved by the experiments.

Key words: Korean pine’s male and female flowers, feature, forecast, minimum enclosing rectangle

摘要： 在红松结实量的各种预测方法中，红松雌雄花的形态特征是主要参数之一。依托原野服务器采集的红松图像，对红松雌雄花进行图像分割，提取红松雌雄花的特征值，为红松结实量的预测方法提供精确的数据。改进了宽长比的提取方法，将轮廓图中不连通的区域通过最短距离的计算，使得不连通的区域假定连通，计算得到最小外接矩，在轮廓图中成功地提取出了宽长比，并通过实验验证了该方法。

关键词: 红松雌雄花, 特征值, 预测, 最小外接矩