Emanuele Torti

Real-Time Implementation of the Vertex Component Analysis Algorithm on GPUs

In this letter, we present a new parallel implementation of the vertex component analysis (VCA) algorithm for spectral unmixing of remotely sensed hyperspectral data on commodity graphics processing units. We first developed a C serial version of the VCA algorithm and three parallel versions: one using NVIDIAs Compute Unified Device Architecture (CUDA), another using CUDA basic linear algebra subroutines library CUBLAS, and the last using the CUDA linear algebra library CULA. Experimental results, based on the analysis of hyperspectral images acquired by a variety of hyperspectral imaging sensors, show the effectiveness of our implementation, which satisfies the real-time constraints given by the data acquisition rate.

A Hybrid CPU–GPU Real-Time Hyperspectral Unmixing Chain

Hyperspectral images are used in different applications in Earth and space science, and many of these applications exhibit real- or near real-time constraints. A problem when analyzing hyperspectral images is that their spatial resolution is generally not enough to separate different spectrally pure constituents (endmembers); as a result, several of them can be found in the same pixel. Spectral unmixing is an important technique for hyperspectral data exploitation, aimed at finding the spectral signatures of the endmembers and their associated abundance fractions. The development of techniques able to provide unmixing results in real-time is a long desired goal in the hyperspectral imaging community. In this paper, we describe a real-time hyperspectral unmixing chain based on three main steps: 1) estimation of the number of endmembers using the hyperspectral subspace identification with minimum error (HySime); 2) estimation of the spectral signatures of the endmembers using the vertex component analysis (VCA); and 3) unconstrained abundance estimation. We have developed new parallel implementations of the aforementioned algorithms and assembled them in a fully operative real-time unmixing chain using graphics processing units (GPUs), exploiting NVIDIAs compute unified device architecture (CUDA) and its basic linear algebra subroutines (CuBLAS) library, as well as OpenMP and BLAS for multicore parallelization. As a result, our real-time chain exploits both CPU (multicore) and GPU paradigms in the optimization. Our experiments reveal that this hybrid GPU-CPU parallel implementation fully meets real-time constraints in hyperspectral imaging applications.

Parallel real-time virtual dimensionality estimation for hyperspectral images

One of the most important tasks in hyperspectral imaging is the estimation of the number of endmembers in a scene, where the endmembers are the most spectrally pure components. The high dimensionality of hyperspectral data makes this calculation computationally expensive. In this paper, we present several new real-time implementations of the well-known Harsanyi–Farrand–Chang method for virtual dimensionality estimation. The proposed solutions exploit multi-core processors and graphic processing units for achieving real-time performance of this algorithm, together with better performance than other works in the literature. Our experimental results are obtained using both synthetic and real images. The obtained processing times show that the proposed implementations outperform other hardware-based solutions.

The Human Brain Project

Studying and understanding human brain is one of the main challenges of 21st century scientists.The Human Brain Project was conceived for addressing this challenge in an innovative way, enabling collaborations between 112 partners spread in 24 European countries.The project is funded by the European Commission and will last until 2023.This paper describes the ongoing activity at one of the Italian units focused on innovative brain simulation through high performance computing technologies. Simulations concern realistic models of neurons belonging to the cerebellar cortex. Due to the level of biological realism, the computational complexity of this model is high, requiring suitable technologies. In this work, simulations have been conducted on high-end Graphical Processing Units (GPUs) and Field Programmable Gate Arrays (FPGAs). The first technology is used during model tuning and validation phases, while the latter allows to achieve real time elaboration, aiming at a possible development of embedded implantable systems. Simulations performance evaluations are discussed in the result section.

Custom FPGA processing for real-time fetal ECG extraction and identification

Monitoring the fetal cardiac activity during pregnancy is of crucial importance for evaluating fetus health. However, there is a lack of automatic and reliable methods for Fetal ECG (FECG) monitoring that can perform this elaboration in real-time. In this paper, we present a hardware architecture, implemented on the Altera Stratix V FPGA, capable of separating the FECG from the maternal ECG and to correctly identify it. We evaluated our system using both synthetic and real tracks acquired from patients beyond the 20th pregnancy week. This work is part of a project aiming at developing a portable system for FECG continuous real-time monitoring. Its characteristics of reduced power consumption, real-time processing capability and reduced size make it suitable to be embedded in the overall system, that is the first proposed exploiting Blind Source Separation with this technology, to the best of our knowledge.

The Human Brain Project: High Performance Computing for Brain Cells Hw/Sw Simulation and Understanding

This paper describes the challenge of understanding brain function through high performance computing dealt with by the Human Brain Project, the European Commission Future and Emerging Technologies Flagship involving a consortium of 112 partners spread in 24 European countries. In particular, we describe the activity, performed by one of the Italian units involved into the project, aiming at identifying very accurate models of cerebellum neurons. These models are processed through high end Graphic Processing Units (GPUs) during the tuning phase and later implemented on FPGA-based application specific processors for respecting real time requirements together with embedded implantability. Models and performance of granular neurons implementations are given in the results section.

Acceleration of brain cancer detection algorithms during surgery procedures using GPUs

Abstract The HypErspectraL Imaging Cancer Detection (HELICoiD) European project aims at developing a methodology for tumor tissue classification through hyperspectral imaging (HSI) techniques. This paper describes the development of a parallel implementation of the Support Vector Machines (SVMs) algorithm employed for the classification of hyperspectral (HS) images of in vivo human brain tissue. SVM has demonstrated high accuracy in the supervised classification of biological tissues, and especially in the classification of human brain tumor. In this work, both the training and the classification stages of the SVMs were accelerated using Graphics Processing Units (GPUs). The acceleration of the training stage allows incorporating new samples during the surgical procedures to create new mathematical models of the classifier. Results show that the developed system is capable to perform efficient training and real-time compliant classification.

A suite of parallel algorithms for efficient band selection from hyperspectral images

AbstractThe analysis of hyperspectral images is usually very heavy from the computational point-of-view, due to their high dimensionality. In order to avoid this problem, band selection (BS) has been widely used to reduce the dimensionality before the analysis. The aim is to extract a subset of the original bands of the hyperspectral image, preserving most of the information contained in the original data. The BS technique can be performed by prioritizing the bands on the basis of a score, assigned by specific criteria; in this case, BS turns out in the so-called band prioritization (BP). This paper focuses on BP algorithms based on the following parameters: signal-to-noise ratio, kurtosis, entropy, information divergence, variance and linearly constrained minimum variance. In particular, an optimized C serial version has been developed for each algorithm from which two parallel versions have been derived using OpenMP and NVIDIA’s compute unified device architecture. The former is designed for a multi-core CPU, while the latter is designed for a many-core graphics processing unit. For each version of these algorithms, several tests have been performed on a large database containing both synthetic and real hyperspectral images. In this way, scientists can integrate the proposed suite of efficient BP algorithms into existing frameworks, choosing the most suitable technique for their specific applications.

Accelerating the K-Nearest Neighbors Filtering Algorithm to Optimize the Real-Time Classification of Human Brain Tumor in Hyperspectral Images

The use of hyperspectral imaging (HSI) in the medical field is an emerging approach to assist physicians in diagnostic or surgical guidance tasks. However, HSI data processing involves very high computational requirements due to the huge amount of information captured by the sensors. One of the stages with higher computational load is the K-Nearest Neighbors (KNN) filtering algorithm. The main goal of this study is to optimize and parallelize the KNN algorithm by exploiting the GPU technology to obtain real-time processing during brain cancer surgical procedures. This parallel version of the KNN performs the neighbor filtering of a classification map (obtained from a supervised classifier), evaluating the different classes simultaneously. The undertaken optimizations and the computational capabilities of the GPU device throw a speedup up to 66.18× when compared to a sequential implementation.

Development of a real-time heart rate estimation algorithm on a low-power device

This paper introduces an algorithm for the estimation of heart rate during physical activities through photoplethysmographic signals, acquired by a wearable device. Firstly, the algorithm has been developed in Matlab and then ported in C language. In this way, it has been possible to test it on a low-power micro-controller (STM32F4DISCOVERY) and to check the similarity of the results, given by the two algorithm versions. Moreover, tests conducted on a public dataset show that the algorithm is real-time compliant. Finally, the power consumption estimation conducted proves the feasibility of a wearable device since the micro-controllers requires less than 20 mW.