Auzuir Ripardo de Alexandria

A New Solution for Automatic Microstructures Analysis from Images Based on a Backpropagation Artificial Neural Network

This article presents a new solution to segment and quantify the microstructures from images of nodular, grey, and malleable cast irons, based on an artificial neural network. The neural network topology used is the multilayer perception, and the algorithm chosen for its training was the backpropagation. This solution was applied to 60 samples of cast iron images and results were very similar to the ones obtained by visual human tests. This was better than the information obtained from a commercial system that is very popular in this area. In fact, this solution segmented the images of microstructures materials more efficiently. Thus, we can conclude that it is a valid and adequate option for researchers, engineers, specialists, and professionals from materials science field to realise a microstructure analysis from images faster and automatically.

Sistema de segmentação de imagens para quantificação de microestruturas em metais utilizando redes neurais artificiais

Digital Image Processing is an increasing expansion area in some field of application that uses the interpretation of images as tool. Quantitative Metallography area applied to Materials Sciences uses this technique for characterization of phase volumetric fractions, grain size, inclusion distribution determination and other parameters that influence the properties of the materials. The present paper has, as main objective, to present and validate the software Segmentation by Artificial Neural Network (SVRNA), developed by the authors. This software, based on artificial neural network, makes the percentile constituent counting in time reduced in relation to the conventional model. The study is carried out over ABNT 1020 and 1045 steel and nodular cast iron samples. Statistical analysis showed that this software is efficient for admitted degree of significance. It has concluded, therefore, that the program can be used in applications in the field of Material Sciences for determination of microstructures.

pSnakes: A new radial active contour model and its application in the segmentation of the left ventricle from echocardiographic images

Active contours are image segmentation methods that minimize the total energy of the contour to be segmented. Among the active contour methods, the radial methods have lower computational complexity and can be applied in real time. This work aims to present a new radial active contour technique, called pSnakes, using the 1D Hilbert transform as external energy. The pSnakes method is based on the fact that the beams in ultrasound equipment diverge from a single point of the probe, thus enabling the use of polar coordinates in the segmentation. The control points or nodes of the active contour are obtained in pairs and are called twin nodes. The internal energies as well as the external one, Hilbertian energy, are redefined. The results showed that pSnakes can be used in image segmentation of short-axis echocardiogram images and that they were effective in image segmentation of the left ventricle. The echo-cardiologists golden standard showed that the pSnakes was the best method when compared with other methods. The main contributions of this work are the use of pSnakes and Hilbertian energy, as the external energy, in image segmentation. The Hilbertian energy is calculated by the 1D Hilbert transform. Compared with traditional methods, the pSnakes method is more suitable for ultrasound images because it is not affected by variations in image contrast, such as noise. The experimental results obtained by the left ventricle segmentation of echocardiographic images demonstrated the advantages of the proposed model. The results presented in this paper are justified due to an improved performance of the Hilbert energy in the presence of speckle noise.

Identification and Quantification of Pulmonary Emphysema through Pseudocolors

Chronic Obstructive Pulmonary Disease (COPD) is a world health problem with high morbidity and mortality. High-Resolution Computed Tomography (HRCT), is an excellent tool for early detection of emphysema component of COPD. Despite this fact, HRCT presents limitations inherent to the subjective analysis of the gray scale image that directly compromises the accuracy for both diagnosis and precise determination of the disease extension. The objective of this paper is present a colored mask algorithm (CMA) to identify and quantify the emphysema, enhancing its visualization through pseudocolors. We studied 21 images of 7 patients with COPD and 1 healthy volunteer. The CMA applies colors to the segmented lungs according to pre-defined ranges of Hounsfield units. CMA automatically calculates the relative area occupied by tomographic densities within the pre-defined ranges, allowing precise quantification of diseased and normal parenchyma. Future works are needed in order to validate the incorporation of the CMA in the image assessment of emphysema in COPD patients.

Techniques of Binarization, Thinning and Feature Extraction Applied to a Fingerprint System

A large volume of images of fingerprints are collected and stored to be used in various systems such as in access control and identification records (ID). Systems for automatic fingerprint recognition perform searches and comparisons with a database. Biometric recognition is based on two fundamental premises: the first is that digital printing must have permanent details, and the second is the information unit. From these premises, a system analyzes the fingerprint image to extract the information and then compares the data in the verification mode or identification mode. , Extraction techniques must be used to obtain the fingerprint data. These techniques use binarization, thinning and features extraction algorithms which are computational methods that can be applied to digital image processing used in scientific research and security issues. This paper presents a comparative analysis of four thresholding techniques (Niblack, Bernsen, Fisher, Fuzzy), two thinning techniques (Stentiford and Holt) and a feature extraction (Cross Number) technique to evaluate the best performance of the algorithms in fingerprint images. To develop this project a set of 160 fingerprint images was used in experiments and analysis. The results point out the positive and negative points of the different algorithms. The system was developed in the C/C++ language.

An OCR System for Numerals Applied to Energy Meters

This work describes a prototype of an OCR (Optical Character Recognition) system designed for reading digits of power measurement devices, using Artificial Neural Networks. The motivation for this work is the implementation of an alternative automatic measurement system to be used in a prepaid power system - SEPPRA. Prototype software using Computer Vision techniques and pattern recognition through Neural Networks is implemented in the C++/Windows platform. Considering this application, adaptive threshold methods are compared in order to choose the more appropriated algorithm of binarization. Several algorithms are implemented and evaluated under different conditions of zoom and camera focus. The system works satisfactorily and can be carried to other platforms, making possible its production in commercial scale.

Global location of mobile robots using Artificial Neural Networks in omnidirectional images

This paper presents a comparison of Mobile Robots localization methods through Artifical Neural Networks in omnidirectional images. After an overview about Mobile Robotics, this work focuses on omnidirectional vision. The motivation for this work is the implementation and comparison of feature extraction techniques that can be used in omnidirectional images seeking invariance to rotation and building descriptors that can be used in Neural Networks. Five feature extraction techniques with their adaptations for omnidirectional image were presented and compared. The results were shown in order to choose the most suitable feature extractor for this application. The feature extractorsare evaluated with respect to time processing and quality of scene description (accuracy of Artificial Neural Network) The results are satisfactory and elect GIST descriptor as the most suitable for the application.

Radial snakes

To perform a comparative study of methods based on radial active contour methods.To evaluate the new method pSnakes.To apply at the segmentation of noisy image with different levels of noise.To apply Hilbert transform to calculate the external energy used on radial snakes. There has been a growing use of digital image processing since the 90s. To process an image it must be transformed successively in order to extract information more easily. The first steps in an image analysis are the acquisition followed by the preprocessing to prepare the image for the next step. This step is called image segmentation which is the process of separating different regions of the image according to their properties. The segmentation process is fundamental for all image analyses, as the final result is essentially dependent on the quality of the segmentation. Highlighted among these techniques are the active contour systems, known as snakes. The active contour methods can be subdivided in two main groups: two-dimensional search (traditional) and one-dimensional search (radial). The radial active contours were developed in order to obtain a smaller computational cost. The aim of this work was to study, evaluate and compare algorithms of radial active contours in synthetic noisy images and thus identify the advantages and disadvantages of each method in order to point out the most appropriate method for a given application. This work makes a quantitative and qualitative comparison of three methods: Traditional Radial Snakes, Hilbert Radial Snakes and pSnakes. The results of this research are suitable for academic research as they show that the recently developed pSnakes method is effective in image segmentation with noise. This paper also considered the processing time of the different methods.

Advances in Photopletysmography Signal Analysis for Biomedical Applications

Heart Rate Variability (HRV) is an important tool for the analysis of a patient’s physiological conditions, as well a method aiding the diagnosis of cardiopathies. Photoplethysmography (PPG) is an optical technique applied in the monitoring of the HRV and its adoption has been growing significantly, compared to the most commonly used method in medicine, Electrocardiography (ECG). In this survey, definitions of these technique are presented, the different types of sensors used are explained, and the methods for the study and analysis of the PPG signal (linear and nonlinear methods) are described. Moreover, the progress, and the clinical and practical applicability of the PPG technique in the diagnosis of cardiovascular diseases are evaluated. In addition, the latest technologies utilized in the development of new tools for medical diagnosis are presented, such as Internet of Things, Internet of Health Things, genetic algorithms, artificial intelligence and biosensors which result in personalized advances in e-health and health care. After the study of these technologies, it can be noted that PPG associated with them is an important tool for the diagnosis of some diseases, due to its simplicity, its cost–benefit ratio, the easiness of signals acquisition, and especially because it is a non-invasive technique.

Autonomous Underwater Vehicle to Inspect Hydroelectric Dams

Driven by the rising demand for underwater operations in the fields of dam structure monitoring, ecosystems of reservoir lakes from Hydropower Plants (HPP) and mining and oil, underwater robotics is increasing rapidly. The increase in exploration, prospecting, monitoring and security in lakes, rivers and sea, both in commercial applications such as scientific applications, has led large companies and research centers to invest in the development of underwater vehicles. The purpose of this work is to develop and evaluate the performance of a dedicated expert system for an Autonomous Underwater Vehicle (AUV) to inspect hydroelectric dams, focusing efforts on mechatronic project based on dimensioning structural elements and machinery and elaborating the sensory part, which includes navigation sensors and sensors of environment conditions, as well as its vision system to detect and measure cracks on hydroelectric dams. The integration of sensors in an intelligent platform provides a satisfactory control of the vehicle, allowing the movement of the submarine on the three spatial axes. Because of the satisfactory fast response of the sensors, it is possible to determine the acceleration and inclination besides his attitude in relation to the trajectory instantaneously taken, and geometry and depth of the cracks. This vehicle will be able to monitor the physical integrity of dams, making acquisition and storage of environment parameter such as temperature, dissolved oxygen, pH and conductivity as well as document images of the biota from reservoir lakes HPP, with minimized cost, high availability and low dependence on a skilled workforce to operate it.