People involved: Damian J. Matuszewski, Rubens M. Lopes, Roberto M. Cesar Jr.

Description: The goal of the SAMBA project is to design, develop, construct, and test an instrument that allows scanning large volumes of water for detecting a range of planktonic organisms, with a primary application in ballast water monitoring but extending its potential use in other environmental applications. The system should be able to discriminate organisms from other suspended particles, to count them and measure their sizes, and, ideally, provide information on whether or not the organisms are alive (or viable). We are exploring several options in both hardware and software development for visualization and data extraction. SAMBA partners are Prof. Rudi Strickler from the University of Wisconsin Milwaukee, and Prof. Roberto César from the Institute of Mathematics and Statistics (USP). The project is carried out under a technical and financial partnership with Petrobras and Transpetro. Read More

Posters:

People involved: Maysa Macedo and Marcel Jackowski and Choukri Mekkaoui

Description: Vascular disease is characterized by any condition that affects the circulatory system. Currently there is a need for sophisticated software tools that can characterize the integrity and functional state of vascular networks from different vascular imaging modalities. These tools face significant challenges such as: large datasets, similarity in intensity distributions of other organs and structures, and the presence of complex vessel branches. In my PhD course I present a new approach to automatically track, and quantify the properties of vascular networks based on CTA and MRA images. Our methodology is based on the Hough Transform to dynamically estimate the center of masse and vessel size along the abdominal aorta. Furthermore, the vessel architecture and orientation is determined by the analysis of the Hessian matrix of the CTA and MRA intensity distribution. Read More

Posters: 

People involved: Miguel Galarreta-Valverde and Marcel Jackowski

Description: Magnetic resonance angiography (MRA) or computed tomography angiography (CTA) images allow for a thorough analysis of the blood vessels. Vessel segmentation from MRA or CTA is thus the primary task in the diagnosis of vascular diseases such as stenosis and aneurysms. The wide architectural variability of the blood vessels, however, hinders the validation of vascular segmentation methods. The construction of synthetic realistic vascular architecture trees will aid in the validation of new vessel segmentation methodologies. This thesis describes a three-dimensional synthetic blood vessel generation methodology that employs stochastic Lindenmayer systems (L-systems). For this purpose, we implemented a parser and a generator of L-systems to create grammars that represent blood vessel architectures. The parameterization of the grammar allows one to simulate natural features of real vessels such as bifurcation angle, average length and diameter, and also accounts for vascular anomalies. The resulting expressions are used to create synthetic angiographic images that mimic real vessel intensity distributions in MRA and CTA. Blood vessel growth can also be delimited by arbitrary 3D surfaces that may represent organ geometries. The flexibility in the parameterization and stochastic nature of this methodology makes it an ideal tool for the validation of blood vessel segmentation algorithms from angiographic images. Read More

Posters:

People involved: Talita Perciano, Roberto Hirata Jr. and Roberto M. Cesar Jr.

Description: Line- curve-like, elongated and ramified structures are commonly found inside many known ecosystems. In biomedicine and biosciences, for instance, different applications can be observed. Therefore, the process to extract this kind of structure is a constant challenge in image analysus problems. However, various difficulties are involved in this process. Their spectral and spatial characteristics are usually very complex and variable. Considering specifically the thinner ones, they are very "fragile" to any kind of process applied to the image, and then, it becomes easy the loss of crucial data. Another very common problem is the absence of part of the structures, either because of low image resolution and image acquisition problems or because of occlusion problems. This work aims to explore, describe and develop techniques for detection/segmentation of thin and ramified structures. Different methods are used in a combined way, aiming to reach a better topological and perceptual representation of the structures and, therefore, better results. Graphs are used to represent the structures. This data structure has been successfully used in the literature for the development of solutions for many image processing and analysis problems. Because of the fragility of the kind of structures we are dealing with, some computer vision principles are used besides usual image processing techniques. In doing so, we search for a better "perceptual understanding" of these structures in the image. This perceptual information along with contextual information about the structures are used in a Markov random field, searching for a final detection through an optimization process. Lastly, we propose the combined use of different image modalities simultaneously. A software is produced from the implementation of the developed framework and it is used in two application in order to evaluate the proposed approach: extraction of road networks from satellite images and extraction of plant roots from soil profile images. Results using the proposed approach for the extraction of road networks show a better performance if compared with an existent method from the literature. Besides that, the proposed fusion technique presents a meaningful improvement according to the presented results. Original and promising results are presented for the extraction of plant roots from soil profile images.

Posters: 

People involved: Martha Dornelles and Nina Hirata

Description: Combining several binary image operators, each one based on different windows, has proven to be an effective way to produce operators with better performance than designing single operators based on one window only. To facilitate the combination task that so far is done manually, we propose a genetic algorithm (GA) based approach. It consists of the definition of a collection of candidate windows and the use of a GA to select a subset of them that will determine the operators to be combined. Experimental results show that the proposed GA based approach produces combinations that are consistently better than those obtained manually, and indicate that the proposed window collections do contain relevant windows. Read More

Posters: