No abstract

Fluorescence microscopy imaging has become one of the essential tools used by biologists to visualize and study intracellular particles within a cell. Studying these particles is a long-term research effort in the field of microscopy image analysis, consisting of discovering the relationship between the dynamics of particles and their functions. However, biologists are faced with challenges such as the counting and tracking of these intracellular particles. To overcome the issues faced by biologists, tools which can extract the location and motion of these particles are essential. One of the most important steps in these analyses is to accurately detect particle positions in an image, termed spot detection. The detection of spots in microscopy imaging is seen as a critical step for further quantitative analysis. However, the evaluation of these microscopic images is mainly conducted manually, with automated methods becoming popular. This work presents some advances in fluorescence microscopy image analysis, focusing on the detection methods needed for quantifying the location of these spots. We review several existing detection methods in microscopy imaging, along with existing synthetic benchmark datasets and evaluation metrics.

No abstract

No abstract

-A novel segmentation technique which may be useful for two dimensional (2D) magnetic resonance (MR) image segmentation is presented. The technique utilizes a dynamic target tracking algorithm and a Kalman filter and permits edges to be followed in the presence of intensity variation similar to that found in MR images. Segmentation of two synthetic test images, one with intensity nonuniformity and one without, is performed. Fuzzy c-means clustering with pixel intensity features is used to segment the same test images for qualitative comparison.Keywords -Image segmentation, edge tracing, Kalman filter, intensity nonuniformity. I. INTRODUCTIONMagnetic Resonance (MR) images are excellent sources of patient-specific anatomical information. Automatic segmentation of these images into component tissue classes provides a method for reproducible extraction of this information. One problem that complicates this process, however, is intensity nonuniformity, an artifact in MR images which is evident as a gradual variation in intensity over otherwise identical tissue classes. Intensity nonuniformity has several causes, notably, inhomogeneity in radio frequency (RF) transmitter and receiver coils during image acquisition [1].MR images provide excellent soft tissue contrast so that intensity-related features are natural choices for use with automatic segmentation methods. However, compensation for intensity nonuniformity must be included in order for such methods to be effective.Although it is possible to perform some compensation during image acquisition, equipment or protocol modifications are typically required. Furthermore, retrospective application of these corrective measures is not possible. Therefore, compensation applied as a post-processing step is considered to be desirable [2].Adaptive Determining the object boundaries in two dimensional images can be done by application of active contours [10] or by edge tracing [11]. We describe a technique for edge tracing which includes a Kalman filter and a dynamic target tracking algorithm to associate edge pixels into object boundaries. Fig. 1 panels (a), and (c) show the two synthetic test images. The shapes of the objects in the images have been chosen to resemble cortical gray matter and white matter in MR images of the brain. Each image has size of 200x200 pixels with 256 gray levels. The unbiased image was formed by interpolating a small set of points with cubic splines to form boundaries of closed regions. These boundaries were then converted to discrete pixels and the enclosed regions were filled with a selected gray level value. The unbiased test image has three gray levels with a difference of fifty gray levels between the brightest region and the intermediate intensity level. II. METHODOLOGY A. Synthetic ImagesIn MR images, intensity nonuniformity has been approximated by exponential functions [1]. For the test images, the gain field ( g ), which simulates the intensity nonuniformity, was formed using a two dimensional exponential function: Abstract Subj...