Eduardo Todt

Efficient Implementation of Canny Edge Detection Filter for ITK Using CUDA

This work presents an efficient CUDA implementation of the Canny edge detection Filter for the Insight Segmentation and Registration Toolkit (ITK). The algorithm is tested on three generations of NVidia GPGPUs, showing performance gains of 3.6 to 50 times when compared to the standard ITK Canny running on two CPU models. The CUDA-enabled Canny is also compared to a more efficient Canny implementation from the OpenCV library. Examples of coding strategies to avoid warp serialization in CUDA are shown on a smart implementation of the Sobel filter, as well as on other algorithms.

A Harris Corner Detector Implementation in SoC-FPGA for Visual SLAM

The present paper discusses the implementation of the Harris and Stephen corner detector algorithm optimized for an embedded system-on-a-chip (SOC) platform that integrates a multicore ARM processor and FPGA fabric in a single chip, the Xilinx Zynq-7000. The algorithm is implemented as a hardware co-processor on the FPGA portion of the SoC. As a whole, the SoC is used as a stereo vision pre-processing module to retrieve depth information from the features in order to compose 3D landmark points for Visual SLAM, speeding up feature extraction and relieving this highly parallelizable process from the main embedded processor. The optimizations of the algorithm’s hardware implementation take into account the particularities of the SoC, such as compliance with its I/O requirements and FPGA’s constraints on the amount of logical elements available for hardware synthesis. Also, optimizations done in order to reduce the time of execution of the algorithm in hardware, such as parallelization and introduction of a pipeline, are also presented in the article. A speedup of 1.77 was achieved when comparing the time of execution of the algorithm in the hardware coprocessor with the algorithm running in software in the dual-core ARM processor.

Applying global time Petri net analysis on the embedded software context

This paper presents an application of Global Time technique (GTT) which is an approach to construct class graphs of Time Petri nets based on relative and global time. Besides the constructing of GTT class graph we propose a schedulability analysis of quantitative time type for fixed priority policies and Earliest Deadline First (EDF). We propose that analysis of scenarios, or behavior itineraries duration of a system, can be done using this approach.

A SoC with FPGA Landmark Acquisition System for Binocular Visual SLAM

The present paper analyzes the use of an embedded system-on-a-chip (SoC) platform, integrating a multicore ARM processor with FPGA fabric in a single chip, as a stereo vision pre-processing module, used to retrieve depth information from the features to compose 3D landmark points. The Harris and Stephens corner detector is applied to an image pair acquired from a stereo camera setup using a hardware co-processor synthesized in the FPGA to speed up feature extraction and relieve this highly parallelizable process from the main embedded processor. The other tasks that compose the stereo vision processing module, such as image correction from camera calibration, finding a unique descriptor for the detected features, determining the correspondence between POIs in the stereo pair and retrieving depth are solved in software running on the main processor. The proposed hardware implementation enabled the corner extraction task to be performed in about half the time taken solely by the embedded ARM processor, speeding up the vision pre-processing task as a whole.

Educational Open Government Data: From Requirements to End Users

The large availability of open government data raises enormous opportunities for open big data analytics. However, providing an end-to-end framework able to handle tasks from data extraction and processing to a web interface involves many challenges. One critical factor is the existence of many players with different knowledge, who need to interact, such as application domain experts, database designers, and web developers. This represents a knowledge gap that is difficult to overcome. In this paper, we present a case study for big data analytics over Brazilian educational data, with more than 1 billion records. We show how we organized the data analytics phase, starting from the analytics requirements, data evolution, development and deployment in a public interface.

Development of a low cost weather station using free hardware and software

Weather stations are a valuable tool to many different areas, like agriculture, aviation, construction, sports and recreation, specially because they collect weather data, which can be stored and processed to gather specific information to correlate events to the weather action. This work shows a development model of a low cost weather station, using only free hardware and software, connected to the internet, giving real time data taken from the station, as well as history of the stored data. This station, even though using low cost or handcrafted sensors, like those that measure wind speed, wind direction and rainfall index, showed consistent results costing only 10% of the price, in comparison with some professional weather stations, allowing its popularization in agriculture, homes and institutions.

Landmark-based Facial Expression Parametrization for Sign Languages Avatar Animation

Facial expressions and associates emotions important factors defining the message tone and context information in both spoken and sign language communication. Sign language virtual systems use structural models that define control values specifying body configurations in the animation process, where facial parameters are generally relegated to simple templates or completely neglected. In this work, a facial expression parametrization for avatars is proposed through an procedure that aims to identify the most relevant facial landmarks and emotions in the context of sign languages, in order to enhance automatic sign synthesis systems. An analysis of the influence of landmarks on a geometric mesh based on MPEG-4 model of human face is performed in this research, aiming to identify the principal components and their relationships, in order to allow further optimization of the animation process, supporting faster and lighter avatar animation.

Hierarchical Sensor Fusion Method Based on Fingerprint kNN and Fuzzy Features Weighting for Indoor Localization of a Mobile Robot Platform

In this paper, a method of sensor fusion is presented, a hierarchical sensor fusion (HSF) method with fingerprint kNN and fuzzy features weighting to solve the problem of localization of a mobile robotic platform with sonars octagon, digital compass, and wireless network signal strength measure to determine the location of an autonomous mobile robot. The fingerprint technique has the goal of determining the robot localization in a map. A kNN matching algorithm is used in two main steps: training and positioning. For the improvement of features acquired from low cost sensors, used in the mobile robot platform, techniques were applied regarding the value of each feature weighting based on a fuzzy inference system and sensor availability in the matching algorithm. In brief, the HSF method proposed in this paper can determine the robot localization in the indoor environment, in different grid sizes, using low cost sensors.

PROINFODATA: Monitoring a Large Park of Computational Laboratories

This paper briefly presents a model for monitoring a large, heterogeneous and geographically scattered computer park. The data collection is performed by a software agent. The collected data are sent to the central server over the Internet, and stored by the storage system. An on-line portal makes up the visualization system, featuring charts, reports, and other tools for assessing the state of the park. This system is currently monitoring circa 150,000 machines.