Header menu link for other important links
X
Edge Intelligence-Based Object Detection System Using Neural Compute Stick for Visually Impaired People
Published in Springer Science and Business Media Deutschland GmbH
2020
Volume: 195
   
Pages: 433 - 439
Abstract
There are around 16 million people who are blind, and this system is to support them for object detection and as well as person recognition in future. This is future of artificial intelligence. This system used Intel neural compute stick which has OpenVINO API and supports various frameworks like TensorFlow, Caffe, MXNet, ONNX, Kaldi, etc. This system can detect object from the frame correctly. For testing, implementation is done on raspberry pi 3B+ which uses Raspbian operating system, Intel’s neural compute stick, and Web camera. This can be artificial eye for blind people, and beauty of project is edge intelligence, without Internet and cloud. When data is analyzed and aggregated in a spot, where it is captured in a network is known as edge intelligence. There are many advantages of intelligence at the edge like it will minimize latency and help to reduce bandwidth, and cost reduction is also one of the important features addition to reducing threats which also minimizes duplication as well as improves reliability. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About the journal
JournalData powered by TypesetInternational Conference on Information and Communication Technology for Intelligent Systems
PublisherData powered by TypesetSpringer Science and Business Media Deutschland GmbH
ISSN21903018
Open AccessNo