摘要
研究了多灰度层次的特殊目标在复杂背景和噪声干扰下的多门限检测技术,该项技术考虑了视觉的灰度特性,空间特性以及噪声干扰引入的错误检测的统计特性。对多门限步级检测器作了理论上的分析和计算机实现,并将所研究的结果应用到 X 线片骨龄识别之中。实际测试结果表明:多门限步级检测器具有很好的检测性能。本文讨论的计算机自动目标检测技术在固定场景监测,工业无损探测以及生物医学工程中癌症、肿瘤、结核等病变的早期计算机诊断中有着十分广泛的应用。
Multi-Threshold Progressive Detection(MTPD)of special multi-graylevel image objects in complex background and interference is studied.In the technique,visu-al properties of gray contract,spatial resolution and minmum error detection probabilityare considered.The theoretical model and the computer software implementation ofMTPD and its application in X—ray bone age classification are presented.The Practicaltesting results show that the performance of MTPD is good.The developed computerautomatic object detection technique has wide applications in scene monitoring,non-destroying detection and computer automatic early-stage diagnosis of cancer,tu-mour etc.in biomedical engineering.
出处
《电子科技大学学报》
EI
CAS
CSCD
北大核心
1990年第2期134-138,共5页
Journal of University of Electronic Science and Technology of China
关键词
步级检测
骨龄识别
X线片
软件
progressive detection
software
visual characteristics
stochastic property
bone age classification
