CAJ | 학술논문

细胞显微图像计数在现代生物医学领域发挥着非常重要的作用。本文基于数字图像处理技术，研究一种密集型细胞显微图像高精度快速计数方法。针对密集型细胞重叠，粘连较多，细胞间距不明显等特点，提出了一种快速连通域标记算法及基于数学统计学的细胞计数方法。大量实验验证结果表明：该方法计数精度能达到95%以上，弱化了由图像分割不到位造成的统计误差，可应用于多个不同种类细胞的计数，通用性较强。并且由于结合了快速图像预处理及分割优化方法，极大缩短了算法运行时间，可满足一般工程应用中细胞实时计数的要求。
세포현미도상계수재현대생물의학영역발휘착비상중요적작용。본문기우수자도상처리기술，연구일충밀집형세포현미도상고정도쾌속계수방법。침대밀집형세포중첩，점련교다，세포간거불명현등특점，제출료일충쾌속련통역표기산법급기우수학통계학적세포계수방법。대량실험험증결과표명：해방법계수정도능체도95%이상，약화료유도상분할불도위조성적통계오차，가응용우다개불동충류세포적계수，통용성교강。병차유우결합료쾌속도상예처리급분할우화방법，겁대축단료산법운행시간，가만족일반공정응용중세포실시계수적요구。
Cell microscopic image count plays an important role in modern biomedicine. In this paper, an fast counting method with high accuracy of intensive cell microscopic image based on digital image processing techniques is re-searched. Aiming at the features that the intensive cell features are overlapped,adhesion and spacing are not obvious,a fast connected component labeling algorithm and a method of cell counting based on mathematical statistics are put for-ward in consider of the real-time performance. A large number of experimental results show that the method of count-ing accuracy can reach more than 95%. The statistical error caused by image segmentation which is not in place is weakened. This method can be applied in many different types of cell counting and have strong universal.As a combina-tion of fast image and segmentation method with optimization,the running time of the algorithm is shortened greatly by using this method,which can meet the engineering application requirements in real-time cell counting.