Jorcy de Oliveira Neto- BEng
- Federal University of Espírito Santo
Jorcy de Oliveira Neto
- BEng
- Federal University of Espírito Santo
About
19
Publications
58,964
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
97
Citations
Introduction
Current institution
Additional affiliations
March 2008 - August 2012
Position
- Researcher
Description
- The High Performance Computing Laboratory Group at the Federal University of Espirito Santo in Vitória - Brazil.
Publications
Publications (19)
![Description of the CAM, SAM, and CBAM Blocks. Adapted from [26].](publication/370743055/figure/fig3/AS:11431281414854791@1746021019711/Description-of-the-CAM-SAM-and-CBAM-Blocks-Adapted-from-26_Q320.jpg)
Agricultural losses due to post-harvest diseases can reach up to 30% of total production. Detecting diseases in fruits at an early stage is crucial to mitigate losses and ensure the quality and health of fruits. However, this task is challenging due to the different formats, sizes, shapes, and colors that the same disease can present. Convolutional...
Hydrocarbon industry works with seismic reflection as a chief indirect method for investigating subsurface data. Seismic volume provides inaccurate geometries of geological structures, inferring rock properties based on seismic amplitude. Ones consider it as an ill posed inverse problem. In order to obtain reliable information, we need to perform s...
In recent years, approaches based on machine learning, more specifically Deep Neural Networks (DNN), have gained prominence as a solution to computer vision problems in the most diverse areas. However, this type of approach requires a large number of samples of the problem to be treated, which often makes this type of approach difficult. In compute...
In the Oil and Gas industry, estimating a subsurface velocity field is an essential step in seismic processing, reservoir characterization, and hydrocarbon volume calculation. Full-waveform inversion (FWI) velocity modeling is an iterative advanced technique that provides an accurate and detailed velocity field model, although at a very high comput...
We present a biologically inspired long-term object tracking system based on Virtual Generalizing Random Access Memory (VG-RAM) Weightless Neural Networks (WNN). VG-RAM WNN is an effective machine learning technique that offers simple implementation and fast training. Our system models the biological saccadic eye movement, the transformation suffer...
A rigorous initiative was undertaken in Brazil 3 years back to identify invisible lost time with the objective to improve safety and performance of the pre-salt drilling program. This was only possible by combining the following:
-Rigorous drilling data quality assurance and recording.
-Use of automation to correctly identify the rig state using a...
The Virtual Generalizing Random Access Memory Weightless Neural Network (VG-RAM WNN) is a type of WNN that only requires storage capacity proportional to the training set. As such, it is an effective machine learning technique that offers simple implementation and fast training – it can be made in one shot. However, the VG-RAM WNN test time for app...
We present a biologically inspired approach to traffic sign detection based on Virtual Generalizing Random Access Memory Weightless Neural Networks (VG-RAM WNN). VG-RAM WNN are effective machine learning tools that offer simple implementation and fast training and test. Our VG-RAM WNN architecture models the saccadic eye movement system and the tra...
![Figure 2: Retinotopic mapping of an image from the retina to V1 [34].](profile/Jorcy-De-Oliveira-Neto/publication/270273697/figure/fig2/AS:294998613479440@1447344559026/Retinotopic-mapping-of-an-image-from-the-retina-to-V1-34_Q320.jpg)
We present two biologically inspired approaches to traffic sign recognition based on Weightless Neural Networks (WNN): one based on Virtual Generalizing Random Access Memory (VG-RAM) neurons and the other on the Wilkes, Stonham and Aleksander Recognition Device (WiSARD) neurons. Both approaches employ the same neural architecture that models the tr...
Virtual Generalizing Random Access Memory Weightless Neural Networks (VG-RAM WNN) is an effective machine learning technique that offers simple implementation and fast training and test. We examined the performance of VG-RAM WNN on binocular dense stereo matching using the Middlebury Stereo Datasets. Our experimental results showed that, even witho...
Virtual Generalizing Random Access Memory Weightless Neural Networks (VG-RAM WNN) is an effective machine learning technique that offers simple implementation and fast training and test. In this paper, we present a new approach for traffic sign recognition based on VG-RAM WNN. We evaluate its performance using the German Traffic Sign Recognition Be...
As imagens projetadas dentro de nossos olhos mudam todo o tempo por conta do movimento dos olhos ou do nosso corpo como um todo. Contudo, em um aparente paradoxo, percebemos o mundo retratado nas imagens capturadas pelos olhos como estável.
Neste projeto, aprofundamos nossos estudos sobre a biologia da cognição visual e sobre novos modelos matemáti...
We developed two parallel versions of the Constant Space Belief Propagation algorithm (CSBP - one of the best stereo algorithms currently known): one in OpenMP and one in C+CUDA. For images with 640x480 pixels, the sequential version has a performance of 1.16 frames per second (FPS), the OpenMP parallel version has a performance of 3.7 FPS, while t...
We developed two parallel versions of the Constant Space Belief Propagation algorithm (CSBP - one of the best stereo algorithms currently known) [1]: one in OpenMP and one in C+CUDA. For images with 640x480 pixels, the sequential version has a performance of 1.16 frames per second (FPS), the OpenMP parallel version has a performance of 3.7 FPS, whi...
A arquitetura Dynamically Trace Scheduling Dataflow (DTSD), desenvolvida na UFES, possui um modelo de execução híbrido que incorpora, na mesma arquitetura, os mecanismos de fluxo de controle e de fluxo de dados (control-flow/data-flow). Ela inclui: um processador primário que executa instruções escalares de modo tradicional (fluxo de controle); uma...
Questions
Questions (30)
Since most of the theoretical and conceptual designs of Single Stage to Orbit (SSTO) aircraft are depicted through Scramjet-based vehicles (such as the X-30 concept from the National Aerospace Plane Program - https://en.wikipedia.org/wiki/Rockwell_X-30), would it actually be possible (and also economically feasible) to reach an orbit (at least a LEO-grade orbit) using an air-breathing engine as a Scramjet or even an hybrid approach like the Turboramjet in the SR-71 ?
Best Regards.
According to Wikipedia, the human population on Earth is estimated around 7,4 Billion as 2016 ( https://en.wikipedia.org/wiki/World_population ) and we're expected to reach a population of over 10 Billion until the end of this century.
However, what would be the actual definion of overpopulation ?
Is is there any estimation of a "theoretical peak capacity" for the human population due to the finite availability of resources (water, food, power sources ...) on Earth ?
Best Regards
In a very specific way, how could one, for example, confront Engle's ARCH-LM test against the Ljung-Box (over the squared residuals) test?
Which would be the most commonly used (i.e. Often Referenced) techniques for detecting/assessing Structural Breaks within ARFIMA-GARCH modelled Time Series?
Does the Temperature in the Thermal Noise Power equation (N = kTB), refers itself to the signal source/reciever/medium or what in specific ?
-Best Regards
Surely, it is widely knows that TCP/IP satellite networks have a very high latency, specially for GEO links. However, which factors would be the most influent on the jitter for TCP/IP over satellite links ?
I'm looking for a better understanding on how good/ill optimized is the native Oracle JVM regarding the usage of processor/architecture specific features (eg: Cache layout, SIMD instructions, OOO Execution Pipelines, etc ... ).
Also. Which would be a good example of a processor-optimized JVM implementation ?
Of course one could implement current limiters within the source induction coils inverters and/or current drivers and also matching the impedances between the power source and reciever coils, however would be these measures enough on guaranteeing that the power source would't accidentally induce parasitic currents within metallic materials nearby ?
Both IETF RFC2488 (Enhancing TCP Over Satellite Channels) and RFC3135 (Performance Enhancing Proxies Intended to Mitigate Link-Related
Degradations) provides some reccomendation/advice on usage of TCP over long latency links (e.g. Satelltes bridges), however, as of 2015, which are the most commonly adopted techniques in commercially available equipment ?
I'm looking for some introductory material on Finite Fields Arithmetics, specially on GF(2), as a way to have a better understanding on the basic principles of some Finite Fields applications, such as CRC, FEC/ECC and Cryptography.
Which would be the most commonly used Channel Equalization techniques for Satellite Links (regardless of the Medium Acess Schemes used, SCPC, DVB-S, DVB-RCS, etc .... ) which are publicly known/available ? Here i'm also considering both linear and non-linear techniques and algorithms.
