【作者】 曹晓光； 崔林艳；

【Author】 Cao Xiaoguang Cui Linyan (Image Processing Center School of Astronautics BeiHang University Beijing 100083 China)

【机构】 北京航空航天大学宇航学院图像中心；

【摘要】 针对尿沉渣图像识别过程中传统阈值方法的问题,提出了一种新的组合算法。组合算法的主要思想是:利用阈值分割算法对图像进行预分割,再采用水平集的方法进行精确的分割。初步实验结果表明,组合算法可以解决阈值分割产生的问题,尿沉渣图像中有形成分被完整精确的分割出来,高精度的分割可以提高后续的尿沉渣图像有形成分识别的正确率。更多还原

【Abstract】 To solve the segmentation problems of typical threshholding algorithm in urinary sediment image processing and recognition,a new combined algorithm of threshholding and level set method is proposed.The key point of the combined algorithm is that threshholding is chosen to do pre-segmentation,then the level set method is chosen to do precise segmentation.Experimental results show that the segmentation problems of threshholding has been solved,With the result of precise segmentation of visible objects in urinary sediment images,the correct rate of final recognition of the objects might be improved..更多还原