Automatic particle detection through efficient Hough transforms

摘要: Manual selection of single particles in images acquired using cryo-electron microscopy (cryoEM) will become a significant bottleneck when very large number are required to achieve three-dimensional reconstructions at near atomic resolution. Investigation fast, accurate approaches for automatic particle detection has one the current challenges cryoEM community. At same time, investigation is hampered by fact that few benchmark or image datasets exist The unavailability such data makes it difficult evaluate newly developed algorithms and leverage expertise from other disciplines. paper presents our recent contribution this effort. It also describes computational framework detection, through application edge sequence ordered Hough transforms. Experimental results keyhole limpet hemocyanin (KLH) as model promising. In addition, introduces established web site, designed support providing an annotated dataset KLH available general scientific