Hyperspectral images clustering on reconfigurable hardware using the k-means algorithm
Abel Guilhermino da Silva Filho,Alejandro C. Frery,Cristiano Araujo,Haglay Alice,Jorge Cerqueira,Juliana A. Loureiro,Manoel Eusebio de Lima,Maria das Graças S. Oliveira,Michelle M. Horta +8 more
- pp 99-104
Reads0
Chats0
TLDR
An automatic and parameterized implementation for hyperspectral images has been developed in a hardware/software codesign approach and an unsupervised clustering technique k-means that uses the Euclidian distance to calculate the pixel to centers distance was used as a case study to validate the methodology.Abstract:
Unsupervised clustering is a powerful technique for understanding multispectral and hyperspectral images, k-means being one of the most used iterative approaches. It is a simple though computationally expensive algorithm, particularly for clustering large hyperspectral images into many categories. Software implementation presents advantages such as flexibility and low cost for implementation of complex functions. However, it presents limitations, such as difficulties in exploiting parallelism for high performance applications. In order to accelerate the k-means clustering, a hardware implementation could be used. The disadvantage in this approach is that any change in the project requires previous knowledge of the hardware design process and can take several weeks to be implemented. In order to improve the design methodology, an automatic and parameterized implementation for hyperspectral images has been developed in a hardware/software codesign approach. An unsupervised clustering technique k-means that uses the Euclidian distance to calculate the pixel to centers distance was used as a case study to validate the methodology. Two implementations, a software and a hardware/software codesign one, have been implemented. Although the hardware component operates at 40 MHz, being 12.5 times less than the software operating frequency (PC), the codesign implementation was approximately 2 times faster than software one.read more
Citations
More filters
Proceedings ArticleDOI
PuDianNao: A Polyvalent Machine Learning Accelerator
Liu Daofu,Tianshi Chen,Liu Shaoli,Jinhong Zhou,Shengyuan Zhou,Olivier Teman,Xiaobing Feng,Xuehai Zhou,Yunji Chen +8 more
TL;DR: An ML accelerator called PuDianNao is presented, which accommodates seven representative ML techniques, including k-means, k-nearest neighbors, naive bayes, support vector machine, linear regression, classification tree, and deep neural network, and can perform up to 1056 GOP/s, and consumes 596 mW only.
Journal ArticleDOI
DianNao family: energy-efficient hardware accelerators for machine learning
TL;DR: A series of hardware accelerators designed for ML (especially neural networks), with a special emphasis on the impact of memory on accelerator design, performance, and energy are introduced.
Proceedings ArticleDOI
TABLA: A unified template-based framework for accelerating statistical machine learning
Divya Mahajan,Jongse Park,Emmanuel Amaro,Hardik Sharma,Amir Yazdanbakhsh,Joon Kyung Kim,Hadi Esmaeilzadeh +6 more
TL;DR: TABLA provides a template-based framework that generates accelerators for a class of machine learning algorithms and rigorously compares the benefits of FPGA acceleration to multi-core CPUs and many-core GPUs using real hardware measurements.
Journal ArticleDOI
Clustering of Hyperspectral Images Based on Multiobjective Particle Swarm Optimization
TL;DR: The obtained results show that interesting classification performances can be achieved by the proposed methodology despite its completely unsupervised nature.
Proceedings ArticleDOI
FPGA implementation of K-means algorithm for bioinformatics application: An accelerated approach to clustering Microarray data
TL;DR: This work proposes a highly parallel hardware design to accelerate the K-means clustering of Microarray data by implementing the K -means algorithm in Field Programmable Gate Arrays (FPGA).
References
More filters
Proceedings Article
Refining Initial Points for K-Means Clustering
Paul S. Bradley,Usama M. Fayyad +1 more
TL;DR: A procedure for computing a refined starting condition from a given initial one that is based on an efficient technique for estimating the modes of a distribution that allows the iterative algorithm to converge to a “better” local minimum.
Journal ArticleDOI
K-Means-Type Algorithms: A Generalized Convergence Theorem and Characterization of Local Optimality
TL;DR: It is shown that under certain conditions the K-means algorithm may fail to converge to a local minimum, and that it converges under differentiability conditions to a Kuhn-Tucker point.
Journal ArticleDOI
The roles of FPGAs in reprogrammable systems
TL;DR: The promise and problems of reprogrammable systems are discussed, including an overview of the chip and system architectures of repprogrammable systems as well as the applications of these systems.
Proceedings ArticleDOI
Algorithmic transformations in the implementation of K- means clustering on reconfigurable hardware
TL;DR: In mapping the k-means algorithm to FPGA hardware, this work examined algorithm level transforms that dramatically increased the achievable parallelism and also examined the effects of using fixed precision and truncated bit widths in the algorithm.
Proceedings ArticleDOI
Applying reconfigurable hardware to the analysis of multispectral and hyperspectral imagery
TL;DR: In this article, a reconfigurable hardware implementation of k-means clustering for multispectral and hyperspectral images has been presented, which is parameterized by the number of pixels in an image, number of channels per pixel, and number of bits per channel as well as number of clusters.