Large Scale Object Group Tracking And Neuron Circuit 3D Reconstruction

Complex motion patterns of natural systems, such as fish schools, bird flocks and cell groups, have attracted great attention of scientists for years. Trajectory measure-ment of individuals is vital for quantitative and high-throughput study of their collec-tive behaviors. However, such data is rare mainly due to the challenges of detection and tracking of large numbers of objects with similar visual features and frequent oc-clusions. We here present an automatic and effective framework to measure trajectories of large numbers of crowded oval-shaped objects, such as fish and cell. We first use a novel dual ellipse locator to detect coarse position of each individual and then propose a variance minimization active contour method to obtain the optimal segmentation result-s. For tracking, cost matrix of assignment between consecutive frames are trainable via a random forest classifier with many spatial, texture and shape features. The optimal trajectories are found for the whole image sequence by solving two linear assignment problems (LAP). To evaluate the proposed method, we perform our system on many challenging data sets.To reveal the real decision mechanism of each object, deeper study of three-dimensional (3D) structure of the brain is an essential work. Furthermore, mitochondria in neuron circuit play an important role in cellular physiology and synaptic function. Recent electron microscopy (EM) advances make it possible to observe mitochondri-al structure on nanoscale, but the attendant massive EM data unfortunately requires months of tedious manual labor. In this paper, we present an automatic approach for the 3D reconstruction of mitochondria from anisotropic EM stack. We first extract a novel local intensity distribution signature (LIDS) feature and learn a random forest classifier (RF) in x-y directions to obtain coarse superpixels. A random disjoint-set for-est algorithm can then cluster these superpixels into supervoxels. And then we use a 3D erosion and dilation method to discard unsatisfying structures, e.g., neural membrane or synapse. At last, the second random forest classifier is learned combining with mito-chondrial shape and texture features, which can select the real mitochondria effectively. The confidence values are given to help human experts decide which mitochondrion to be reviewed first. We evaluate the proposed approach on two different anisotropic EM stacks of drosophila brain and compare against the current state-of-the-art methods. In addition, we present an improved mitochondria 3D reconstruction algorithm and also give our latest research results on membrane detection.