Ongoing Projects

Back to Falcão's home page

The design of image operators for image filtering, image segmentation, pixel classification, object representation, and object description have been efficiently reduced to the choice of an adjacency relation (image graph) and connectivity function between pixels. The minimization/maximization of this connectivity function for every image pixel results into an optimum-path forest, whose attributes are used to complete the image operation. The method is called Image Foresting Transform (IFT). This project aims at new image operators based on the IFT in the context of several major projects which involve biomedical image analysis, digital video processing and content-based image retrieval in biodiversity systems. See selected publications in Image Foresting Transform, Image Filtering, Image Segmentation, Shape Representation and Description. There are also free softwares for downloading , which use the IFT for interactive segmentation, connected filtering, and skeletonization.

The IFT has been extended from the image domain to the feature space, where the same idea applies to the design of optimum-path forest classifiers, under the unsupervised, supervised, and semi-supervised learning paradigms. In this case, the nodes of the graph can be images, pixels, objects, or any other abstract entity. This project aims at the design of new OPF classifiers in the context of biomedical image analysis, digital video processing, content-based image retrieval in biodiversity systems, and other applications, such as image compression, restoration, and super-resolution. See selected publications in Pattern Recognition, Texture Description, and Image Analysis. There is also the code of the OPF supervised classifier for free downloading.

Magnetic Resonance images (MRA, MRI, DTI, MRS and fMRI) of the human brain provide visual information about tissues and structures, whose anatomy and function may be affected by neuronal disturbs and cerebral diseases. The visualization and quantification of such information are important to understand the natural course of the disease, plan a treatment, and study the effects of the treatment. This project investigates methods for filtering, segmentation, registration, visualization and morphometric analysis of cerebral structures in patients with epilepsy and other degenerative diseases. See selected publications in Image Segmentation, Alignment, Registration, Visualization, and Analysis. There are also free softwares for downloading , which allow medical image filtering, segmentation, and visualization.

Intestinal parasites in humans have been responsible for several diseases and, in the worst case, death. However, the diagnosis of intestinal parasites is still precarious and error prone. This project aims at automation of the diagnosis of intestinal parasites in humans through the computer analysis of microscopy images. The idea is to further extend the work to the diagnosis of intestinal parasites in animals (e.g., dogs and cats) and in other medium (e.g., water and blood). See selected publications in Image Analysis.