紫色土彩色图像的H阈值分割

The Color Image Segmentation of Purple Soil with Its H Threshold

【目的】将研究对象紫色土自然断面图像从复杂背景中分割出来,以消除复杂背景的影响,为对紫色土视觉图像进一步分析处理、识别打下基础。【方法】通过对紫色土自然断面图像统计分析,证明它的H域有良好的聚集特征,且在95%和97%置信点近似于正态分布;因此,根据正态分布确定紫色土壤自然断面的H阈值,分割紫色土壤自然断面图像。【结果】用获得的二值分割图像与原图像计算哈达玛积,分割提取出了完整的紫色土自然断面区域图像。【结论】算法分析及仿真实验显示:相对于对比算法,新的算法分割速度更快,分割提取出的紫色土自然断面图像更完整;该算法有效。

[Purposes]The first job is to separate out the natural cross-section image of purple soil from the chaotic background in lhe purple soil images because it is a base of the processing in future.[Methods]It displays the clustering property which is palpable in the H values of the HSV color space to study the natural cross-section image of purple soil.The distribution of the pixel number in the H values is closer to a normal distribution at the points that confidence level is 95%or 97%.So it is feasible to compute the H threshold with the normal distribution and to separate out the natural cross-section image of purple soil with it.[Findings]The segmcn tation image of the purple soil is obtai ned by calculating the Hadamard prod net with the bin ary image that elim in ates the discrete small and small holes and the original image.[Conclusions]Theoretical analysis and simulation experiments show that the algorithm is effective,and its computing speed is faster and it improves the image segmentation effect compared with the contrast algorithms.