中文摘要：

利用图像处理技术,检测荧光显微序列中的活性线粒体是生物医学领域重要的研究手段之一.受到荧光显微镜成像技术的限制,序列中每帧图像均包含细胞质阴影和荧光标记的线粒体,具有很低的信噪比,难以满足一般粒子检测算法的要求.为了精确检测活细胞中的线粒体,提出一种基于矩阵分解的荧光显微序列线粒体检测算法,并利用增广拉格朗日乘子法,快速准确地实现该算法,将线粒体从细胞质阴影中有效分离出来,实现线粒体的精确检测.实验结果表明,此方法为活细胞中线粒体的精确检测提供了快速、高效的分析工具。

英文摘要：

Detection of mitochondria in fluorescent microscopic images is one of the most important methods in studies concerning apoptosis and the nature of life phenomena in the area of biomedical image processing. Limited by fluorescence microscopy, fluorescent microscopic images contain two parts which are the shadow of cytoplasm and live mitochondria, and the signal-to-noise ratio （SNR） of live mitochondria time sequence images is low, which cannot meet the requirements of general particle algorithm. A new detection algorithm was proposed for live mitochondria in fluorescent microscopic images. To realize this method rapidly, augmented Lagrange multiplier algorithm was used. Mitochondria was be separated from the cytoplasm and accurately detected in fluorescent microscopic images. Therefore, the proposed algorithm provides an efficient and accurate tool to detect mitochondria in live cell.