Cited by 30Neural Networks and ApplicationsEmbedded Systems Design TechniquesAdvanced Neural Network Applications

SOMprocessor: A high throughput FPGA-based architecture for implementing Self-Organizing Maps and its application to video processing

Sousa, Miguel Angelo de Abreu de, Pires, Ricardo, Del-Moral-Hernandez, Emilio·2020·Neural Networks

Abstract

The design of neuromorphic chips aims to develop electronic circuits dedicated to executing artificial neural networks, mainly by exploring parallel processing. Unsupervised learning models, such as Self-organizing Maps (SOM), may benefit from massively concurrent hardware-based implementations to meet the requirements of real-time and embedded applications. This work first presents a theoretical analysis of the algorithms implemented in hardware to compute SOM learning and recall phases. This is important because, albeit similar, the processing steps executed in hardware are not necessarily identical to those executed in software. Then, the proposed FPGA architecture entitled SOMprocessor is shown in detail. The circuit of the processor explores two different computational strategies for increasing the performance of current state-of-the-art works. These computational strategies aim to improve the data flow through the processor and its flexibility to implement different network topologies. Finally, this work presents the application of the SOMprocessor to a video categorization task. The results show that topographic and quantization errors are similar between hardware and software implementations, as well as the overall accuracy. Moreover, the proposed FPGA architecture achieves acceleration of 3 to 4 orders of magnitude as compared to CPU executions.

Integrity signals

Registry

Crossref · 10.1016/j.neunet.2020.02.019

Cites

110 of 110 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Using multiplexers allows resources to be reused for larger networks, such as reusing fixed-size blocks for computing large amounts of data and reusing the circuit for decision making and the weight update component.

    “The benefit of using multiplexers is due to the ability to reuse the resources for larger networks, such as reusing the fixed size blocks for computing large amounts of data, and reusing the circuit for decision making and the weight update component [52].”
  • The proposed FPGA architecture improves overall acceleration by three to four orders of magnitude compared to CPU execution.

    “The proposed architecture will use the FPGA platform and show that it is able to improve the overall acceleration by three to four orders of magnitude as compared to CPU execution.”
  • The proposed architecture of SOMs achieved a higher magnitude order relation than software implementation of the same order.

    “As a conclusion, the proposed architecture of SOMs was able to achieve a higher magnitude order relation than software implementation of the same order.”

Cites

Flagged first, then by citations

3 of the 113 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Authors
Sousa, M. A. D. A. D., Pires, R., & Del-Moral-Hernandez, E.
Affiliations
  • Federal Institute of São Paulo (BR)
  • Universidade de São Paulo (BR)
Corresponding author
Sousa, Miguel Angelo de Abreu de
Venue
Neural Networks
Publisher
Elsevier BV
Published
2020-05
Volume
125
Pages
349-362
ISSN
0893-6080 (ISSN-L)
OpenAlex
W3010523544
Cited by
30
References
113
Citations by year
2026: 1 · 2025: 3 · 2024: 3 · 2023: 5 · 2022: 5 · 2021: 8 · 2020: 5
FWCI
2.26
Type
journal article
Language
en
Open access
closed
Locations
  • Neural Networks (publishedVersion)
  • PubMed (publishedVersion)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
  • Physical Sciences › Computer Science › Hardware and Architecture › Embedded Systems Design Techniques
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
Keywords
self-organizing maps, FPGA architecture, hardware implementation of neural networks, neuromorphic hardware, video classification, topographic error, quantization error, hardware acceleration, parallel computing
MeSH terms
Machine Learning, Computers, Electronics, Image Processing, Computer-Assisted, Software

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review