摘要
细胞学检查自动化是生物医学工程研究的重要课题。文中介绍了雅可比-傅里叶矩,它具有多畸不变性。多种临床细胞涂片模式识别表明,该矩不但是一种高度浓缩的图像特征,还有较好的抗畸变、抗噪声性能,可用于快速准确的细胞学自动化识别。
Automatic cytology test is a major research issue in the field of biomedical engineering. The Radial-Harmonic-Fourier Moment presented in this paper is multi-distortion invariant. Pattern recognition experiment on clinical cell slice shows that Jacobi-Fourier Moment is not only a highly concentrated image feature but also has good property of anti-distortion and anti-noise. It can be used in fast and accurate automatic cell recognition.
出处
《中国生物医学工程学报》
CAS
CSCD
北大核心
2005年第5期531-534,545,共5页
Chinese Journal of Biomedical Engineering
基金
国家自然科学基金资助项目(No.60262001)
内蒙古自然科学基金资助项目(No.201301)
关键词
模式识别
细胞学
矩
cell recognition
cytology
moment
