基于自适应阈值的舌像分割方法

中医舌诊是目前医学领域的重要前沿课题之一,而舌体轮廓的正确分割是实现中医舌诊信息化的重要前提,目前主流的阈值分割方法对对比度较小的舌像仍不能实现很好的分割。为此提出了一种舌像的自适应阈值分割算法,该算法是以VC＋＋为开发平台,首先把图像分成多个子块,然后运用迭代的方法计算每个子块的最佳阈值,根据每个局部最佳阈值构成的阈值矩阵进行分割。实验结果表明,该算法对背景和目标分界不明显的舌像有很好的分割效果,对中医舌诊的继续发展有很强的现实意义。

Tongue Image Segmentation Method Based on Adaptive Thresholds

Traditional Chinese medicine tongue diagnosis is one of the important medicine topics.The correct segmentation of the body of tongue is premise to the information on diagnosis by the feature of tongue in Traditional Chinese Medicine.The main method at present is the threshold algorithm.But the method can not deal with some tongue images with low contrast.Proposed an adaptive segementation algorithm to segment tongue image efficiently.The algorithm is based on VC ＋ ＋ as a development platform.First,divided the image into several parts,and then use an iterative approach to calculate each sub-block threshold.According to each local threshold to segment.The experimental results show that the algorithm can well segment the tongue image whose background and boundaries of the objectives is not obvious.