No abstract

Algorithms are presented for fully automatic three-dimensional (3-D) tracing of neurons that are imaged by fluorescence confocal microscopy. Unlike previous voxel-based skeletonization methods, the present approach works by recursively following the neuronal topology, using a set of 4 x N2 directional kernels (e.g., N = 32), guided by a generalized 3-D cylinder model. This method extends our prior work on exploratory tracing of retinal vasculature to 3-D space. Since the centerlines are of primary interest, the 3-D extension can be accomplished by four rather than six sets of kernels. Additional modifications, such as dynamic adaptation of the correlation kernels, and adaptive step size estimation, were introduced for achieving robustness to photon noise, varying contrast, and apparent discontinuity and/or hollowness of structures. The end product is a labeling of all somas present, graph-theoretic representations of all dendritic/axonal structures, and image statistics such as soma volume and centroid, soma interconnectivity, the longest branch, and lengths of all graph branches originating from a soma. This method is able to work directly with unprocessed confocal images, without expensive deconvolution or other preprocessing. It is much faster that skeletonization, typically consuming less than a minute to trace a 70-MB image on a 500-MHz computer. These properties make it attractive for large-scale automated tissue studies that require rapid on-line image analysis, such as high-throughput neurobiology/angiogenesis assays, and initiatives such as the Human Brain Project.

Abstract-In contrast to holistic methods, local matching methods extract facial features from different levels of locality and quantify them precisely. To determine how they can be best used for face recognition, we conducted a comprehensive comparative study at each step of the local matching process. The conclusions from our experiments include: (1) additional evidence that Gabor features are effective local feature representations and are robust to illumination changes; (2) discrimination based only on a small portion of the face area is surprisingly good; (3) the configuration of facial components does contain rich discriminating information and comparing corresponding local regions utilizes shape features more effectively than comparing corresponding facial components; (4) spatial multiresolution analysis leads to better classification performance; (5) combining local regions with Borda Count classifier combination method alleviates the curse of dimensionality. We implemented a complete face recognition system by integrating the best option of each step. Without training, illumination compensation and without any parameter tuning, it achieves superior performance on every category of the FERET test: near perfect classification accuracy (99.5%) on pictures taken on the same day regardless of indoor illumination variations; and significantly better than any other reported performance on pictures taken several days to more than a year apart. The most significant experiments were repeated on the AR database, with similar results. I. INTRODUCTIONver the last decade, face recognition has become one of the most active applications of visual pattern recognition due to its potential value for law enforcement, surveillance and human-computer interaction.Although face recognition systems show striking improvement in successive competitions [35,36], the face recognition problem is still considered unsolved. Modern face recognition methods can be generally divided into two categories: holistic matching methods and local matching methods.After the introduction of Eigenfaces [21,44], holistic matching approaches, that use the whole face region as the input to a recognition system, have been extensively studied. The principle of holistic methods is to construct a subspace using Principal Component Analysis (PCA) [21,44,46] Careful comparative studies of different options in a holistic recognition system have been reported in the literature [38]. We believe that a similar comparative study on the options at each step in the local matching process will benefit face recognition through localized matching. Although several stand-alone local matching methods have been proposed, we have not found any studies on comparing different options. The aim of this paper is to fill this blank by presenting a general framework for the local matching approach, and then reviewing, comparing and extending the current methods for face recognition through localized matching. modeling illumination variations, has received increasing attention [...

Glossary AND-OR graph (or tree). Representation of a solution strategy in which a path from the start node to the solution node requires traversing any branch at an OR node and every branch at an AND node. In a related Min-Max search used in two-person games, a path from the start node to the solution node takes the lowest cost branch at a Min node and the highest cost branch at a Max node.Bitmap. Digital representation of an image in which points are mapped to an array of binary pixels. Branch-and-bound.A search technique that avoids paths certain to lead to higher cost solutions than the best solution obtained so far.

No abstract