Juan Pablo D’Amato

Color Based Fruits classification using GPU

In fruit packaging companies, the color is a metric used to determine the fruit quality, maturation, healthiness, etc.. For apples, the color also indicates their class or variety (red delicious, granny smith, etc.). Nowadays the distinction between the different qualities of apples is based on empirical measures proposed by experts. In this paper a mechanical-digital system for capturing and classifying fruit by color in real time is presented. Its supposed to be incorporated to an existent fruit transport line. The tracking algorithms were implemented using graphics cards, which allow processing a very large number of lines in a single commercial PC. The solution is able to emulate the human criterion in the fruits classification. Based on the images captured from the transport line, the system generates an output indicating the quality, which is used to distribute the fruit in different packages

Multi-object segmentation framework using deformable models for medical imaging analysis

AbstractSegmenting structures of interest in medical images is an important step in different tasks such as visualization, quantitative analysis, simulation, and image-guided surgery, among several other clinical applications. Numerous segmentation methods have been developed in the past three decades for extraction of anatomical or functional structures on medical imaging. Deformable models, which include the active contour models or snakes, are among the most popular methods for image segmentation combining several desirable features such as inherent connectivity and smoothness. Even though different approaches have been proposed and significant work has been dedicated to the improvement of such algorithms, there are still challenging research directions as the simultaneous extraction of multiple objects and the integration of individual techniques. This paper presents a novel open-source framework called deformable model array (DMA) for the segmentation of multiple and complex structures of interest in different imaging modalities. While most active contour algorithms can extract one region at a time, DMA allows integrating several deformable models to deal with multiple segmentation scenarios. Moreover, it is possible to consider any existing explicit deformable model formulation and even to incorporate new active contour methods, allowing to select a suitable combination in different conditions. The framework also introduces a control module that coordinates the cooperative evolution of the snakes and is able to solve interaction issues toward the segmentation goal. Thus, DMA can implement complex object and multi-object segmentations in both 2D and 3D using the contextual information derived from the model interaction. These are important features for several medical image analysis tasks in which different but related objects need to be simultaneously extracted. Experimental results on both computed tomography and magnetic resonance imaging show that the proposed framework has a wide range of applications especially in the presence of adjacent structures of interest or under intra-structure inhomogeneities giving excellent quantitative results.

Automatic rectum limit detection by anatomical markers correlation

Several diseases take place at the end of the digestive system. Many of them can be diagnosed by means of different medical imaging modalities together with computer aided detection (CAD) systems. These CAD systems mainly focus on the complete segmentation of the digestive tube. However, the detection of limits between different sections could provide important information to these systems. In this paper we present an automatic method for detecting the rectum and sigmoid colon limit using a novel global curvature analysis over the centerline of the segmented digestive tube in different imaging modalities. The results are compared with the gold standard rectum upper limit through a validation scheme comprising two different anatomical markers: the third sacral vertebra and the average rectum length. Experimental results in both magnetic resonance imaging (MRI) and computed tomography colonography (CTC) acquisitions show the efficacy of the proposed strategy in automatic detection of rectum limits. The method is intended for application to the rectum segmentation in MRI for geometrical modeling and as contextual information source in virtual colonoscopies and CAD systems.

Video Analytics on a Mixed Network of Robust Cameras with Processing Capabilities

Public safety is, to a greater or lesser extent, a significant concern in most modern cities. In many of these cities, video surveillance is employed to prevent and deter crime, often building systems with hundreds of cameras and sensors. Such systems proved to be effective in crime fighting and prevention, but they have high bandwidth requirements in order to bring a real-time monitoring. In countryside areas, where only low speed connections are available, the installation of such systems is not suitable and a new approach is required.

A Shadow Removal Approach for a Background Subtraction Algorithm

This paper presents preliminary results of an algorithm for shadow detection and removal in video sequences. The proposal is that from the base of the background subtraction with the Visual Background Extraction (ViBE), which identifies areas of movement, to apply a post processing to separate pixels from the real object and those of the shadow. As the areas of shadows have similar characteristics to those of the objects in movement, the separation becomes a difficult task. Consequently, the algorithms used for this classification may produce several false positives. To solve this problem, we set to use information of the object involved such as the size and movement direction, to estimate the most likely position of the shadow. Furthermore, the analysis of similarity between the present frame and the background model are realized, by means of the traditional indicator of normalized cross correlation to detect shadows. The algorithm may be used to detect both people and vehicles in applications for safety of cities, traffic monitoring, sports analysis, among others. The results obtained in the detection of objects show that it is highly likely to separate the shadow in a high percentage of effectiveness and low computational cost; allowing improving steps of further processing, such as object recognition and tracking.

Tridimensional Scenes Management and Optimization for Virtual Reality Simulators

Virtual Reality (VR) simulation offers a new paradigm for realistic control and operation training, employing 3D digital objects and environments to create immersive experiences. The virtual scenarios should be as similar to the real ones as possible, to improve training. Furthermore, obtaining a digital version of these scenarios is a complex process that involves several tasks including mathematical modeling and 3D detailed geometry generation, among others. This paper presents some novel software tools that easier the development of a 3D virtual subterranean simulator. By providing tight integration between traditional CAD software and visualization engines, these tools are successfully applied to the construction of several virtual scenarios. Moreover, some strategies to optimize digital resources to allow loading and visualizing large maneuver exercises along realistic train trajectories in desktop computers are presented.

From a Serious Training Simulator for Ship Maneuvering to an Entertainment Simulator

This paper presents a ship-handling entertainment simulator that was developed to be used as a virtual reality experience in science exhibitions. It is a low-cost implementation that allows navigating a ship through a simple interface. Realistic 3D graphics area projected on a three panel screen implemented with computer monitors or HD LED TV. This simulator is an adaptation of a previous set of serious ship handling training simulators -called MELIPAL- that were developed for the Argentina Army. We describe how we adapted the original simulator to the new entertainment version, particularly the system architecture, the hardware, the 3D visualization and the user interface aspects.