DEEP LEARNING TECHNIQUE COMPRISING OBJECT DETECTION AND IDENTIFICATION USING MOVIDIUS NEURAL COMPUTE STICK
Keywords:
Neural Compute Stick (NCS), AI and Deep learning, CNN and DNN, YOLO (You Only Look Once), Single shot Detector, Vision processing Unit (VPU)Abstract
Due to the vagueness in the research and development of image processing techniques and applications, detecting and identifying real-time objects with high accuracy at faster rate is an efficient way to acquire a revolution in technology. In this study, the device used for the process of detecting and recognizing objects is Movidius neural compute stick (NCS) which is a tiny offline, low power VPU, fanless deep learning device very effective for deploying Artificial Intelligence and deep learning applications using Neural Networks (CNN and DNN). Main objective of the process is to classify, detect and recognize the objects in real time which is trained prior as datasets, later used for identification using SSD (Single Shot Detector) algorithm trained with Caffe or TensorFlow framework and MobileNet architecture. And it incorporates some of the procedures of YOLO (Tiny YOLO). It is very interesting to understand the VPU (vision as importing processing unit) of Neural compute stick that is myriad 2 VPU. The NCS is plugged into a host machine using the USB interface on the VPU. OpenCV libraries was an essential part of it and this process can be achieved using UBUNTU (16.04) platform which is designed for strong focus in computational efficiency.
References
Banumathi, K., Bharathi, B., Daisy Deve Priya, ,. S., & Gogulalakshmi, P. (2018). Real Time Vehicle Detection using Movidius Neural Compute Stick. International Journal of Engineering Research & Technology (IJERT), (p. 5).
Othman, N. A., & AYDIN, I. (2018, October). A New Deep Learning Application Based on Movidius NCS for Embedded Object Detection and Recognition. In 2018 2nd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT) (pp. 1-5). IEEE.
Combining MobileNets and Single Shot Detectors for fast, efficient deep-learning based object detection,(n.d.). Retrieved from pyimagesearch website: https://www.pyimagesearch.com/2017/09/11/object-detection-with-deep-learning-and-opencv/
Installing the Intel Movidius SDK, ,(n.d.). Retrieved from pyimagesearch website: https://www.pyimagesearch.com/2018/02/19/real-time-object-detection-on-the-raspberry-pi-with-the-movidius-ncs
Introduction , (n.d.). Retrieved from movidius.github.io website: https://movidius.github.io/ncsdk/tools/tools_overview.html
Object detection using the Intel Neural Compute Stick 2 and OpenVINO, (n.d.). Retrieved from Good Audience website: https://blog.goodaudience.com/a-test-drive-of-the-intel-neural-compute-stick-2-b694983e5f9b Intel® Movidius™ Myriad™ X VPUs Intel® Movidius™ Myriad™ X VPUs Intel® Movidius™ Myriad™ X VPUs Intel® Movidius™ Myriad™ X VPUs
Intel Movidius Myriad X VPUS,(n.d.). Retrieved from intel.ai website: https://www.intel.ai/intel-movidius-myriad-vpus/
Intel NCS 2, (n.d.). Retrieved from THENEWSTACK Website: https://thenewstack.io/a-closer-look-at-intel-movidius-neural-compute-stick/
(n.d.). Retrieved from academia website: https://www.academia.edu/40065515/Raspberry_PI3_and_a_Movidius_Neural_Compute_Stick_for_Real-Time
(n.d.). Retrieved from www.adlinktech.com: https://www.adlinktech.com/Products/Deep_Learning_Accelerator_Platform_and_Server/Deep_Learning_Accelerator/EDL-mPCIeMA2485?Lang=en
Darren Crews. (n.d.). Retrieved from https://qconsf.com/: https://qconsf.com/sf2017/system/files/presentation-slides/qconf_presentation.pdf
Rungta, M. (n.d.). Retrieved from http://www.cse.iitd.ac.in/: http://www.cse.iitd.ac.in/mavi/docs/thesis_sarvesh_mukund.pdf
Hochstetler, J., Padidela, R., Chen, Q., Yang, Q., & Fu, S. (2018, October). Embedded deep learning for vehicular edge computing. In 2018 IEEE/ACM Symposium on Edge Computing (SEC) (pp. 341-343). IEEE.
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., ... & Adam, H. (2017). Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861.
Mane, S., & Mangale, S. (2018, June). Moving Object Detection and Tracking Using Convolutional Neural Networks. In 2018 Second International Conference on Intelligent Computing and Control Systems (ICICCS) (pp. 1809-1813). IEEE.
Hayat, S., Kun, S., Tengtao, Z., Yu, Y., Tu, T., & Du, Y. (2018, June). A Deep Learning Framework Using Convolutional Neural Network for Multi-Class Object Recognition. In 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC) (pp. 194-198). IEEE.
Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C. (2016, October). Ssd: Single shot multibox detector. In European conference on computer vision (pp. 21-37). Springer, Cham.
Downloads
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.