| The problem of identifying and counting rolling leukocytes within intravital microscopy images is of both theoretical and practical interest. In order to identify the leukocytes with both bright and dark appearances, we propose detection method based on Bayesian classification. The classification depends on a feature score, the gradient inverse coefficient of variation (GICOV), which serves to discriminate rolling leukocytes from a possibly cluttered environment. The leukocyte detection process consists of three sequential steps. The first step utilizes an ellipse matching algorithm to coarsely identify the leukocytes by finding the ellipses with a locally maximal GICOV value. In the second step, starting from each of the ellipses found in the first step, a B-spline snake is evolved to refine the leukocytes boundaries by maximizing the associated GICOV. The third and final step retains only the extracted contours that have a GICOV value above the analytically determined threshold.; As an extension, we utilize the marked point process (MPP) framework while aiming at improving both the accuracy and efficiency of the detection process by means of exploiting the spatial information and leukocyte inter-relationships. The MPP provides a useful and theoretically well-established tool for integrating spatial information into the image analysis process. We construct a Markov chain Monte Carlo algorithm to obtain the maximum a posteriori (MAP) estimation of a set of candidate points corresponding to the centroids of leukocytes observed in the image. The optimal solution, in terms of MAP principle, is computed with respect to all leukocytes, rather than a single leukocyte. A quantitative study of our detection approach demonstrates results that exceed the solution quality given by two other competing detection methods, using a dataset consisting of 60 intravital microscopic video sequences each 31 frames long. Our approach can serve as a fully automated substitute to the tedious and time-consuming manual rolling leukocyte detection process.; The real-time rolling leukocyte detection is implemented using the Mercury AdapDev multiprocessor architecture. Additionally, the problems of associated feature extraction with micro-particle tracer detection in microvessels are discussed in this thesis. |
