An Embedded Cascade SVM Approach for face detection in the IoT Edge Layer

Abstract

In last years, the Internet of Things (IoT) has gained attention in relevant fields of application as Industry 4.0. IoT is usually structured around three layers according to computing capacity and energy cost: cloud, fog, and edge. This paper focuses on the edge layer which is close to the end-user. Specifically, the authors fully address a binary image classification problem in the edge without going toward upper layers, i.e., the intelligence and the computation is brought to the edge layer, instead of being simple sensing devices. To this end, the authors propose a cascade Support Vector Machine (SVM) embedded implementation specially designed to be executed within a low-cost FPGA, which could be embedded in an IoT edge device. The experimental results performed show the feasibility and correct functionality of the proposal when comparing to a regular computer.

Fuzzy ARTMAP-Based Fast Object Recognition for Robots Using FPGA

“…Early design of VLSI circuits based on adaptive resonance theory were reported in [17]. Other methods such as convolutional neural networks (CNNs) have also been reported [18]; support vector machines (SVM) [19] and IoT approaches are also reported in [20,21]. Late developments in hardware have culminated with Dynamic Vision Sensors (DVSs) which provide a new strategy for representing real-time visual information as event-flows such as spike Neural Networks [17].…”

Field Programmable Gate Array Applications—A Scientometric Review

“…IoT in this list is the application with the highest PDLY, with 85% of the documents published in the last three years. For IoT applications, FPGAs have been used in improving devices security [1263][1264][1265], data encryption [460,1144,[1266][1267][1268], edge computing [1269][1270][1271], FPGA-based gateways [1272,1273] and fog computing platforms [1274,1275].…”