SU Jincheng, HU Yong, GONG Cailan. A pore detection algorithm based on joint feature constraint[J]. 2019, (5): 146-148+153. DOI: 10.13873/J.1000-9787(2019)05-0146-03.
一种基于联合特征约束的毛孔检测算法
摘要
在皮肤数码图像的毛孔检测中
受不均匀光照、斑点、油光、毛发等因素影响
常常出现错检。针对这种情况
提出了一种基于联合特征约束的毛孔检测算法
算法充分考虑各影响因素
通过背景抑制、形态学处理削弱了光照不均和背景复杂的影响。基于毛孔色调特征
改进传统大津法计算最佳毛孔分割阈值
利用阈值进行分块分割。基于毛孔的尺寸、形状特征
对分割后的候选目标进行精细筛选
从而得到精确的毛孔检测结果。实验结果表明
提出的算法检测精度高
误检率低
具备良好的鲁棒性。
Abstract
In pore detection of digital image of the skin
influenced by the factors such as uneven lighting
spots
oil and hair
etc
false detection often occur. Aiming at this problem
a pore detection algorithm based on joint feature constraint is proposed. The algorithm takes full account of influence factors. Based on feature of color
pore is segmented by the threshold value which is computed by the improved Ostu algorithm. Based on the pore size and shape characteristics
the selected candidate targets were carefully selected to obtain accurate pore test results.Experimental result shows that this algorithm has high detection accuracy