International Journal of Advanced Networking and Applications

Open Access Open Access  Restricted Access Subscription Access

Development of Object Recognition Model Using Machine Learning Algorithms on MobileNet V2


Affiliations
1 Department of Electronics and Communication Engineering, India
2 Department of Electronics and Communication Engineering Government Engineering College, Haveri-581110, Karnataka, India
3 Department of Electronics and Communication Engineering Nitte Meenakshi Institute of Technology, Yelahanka, Bangalore-560064, Karnataka, India
 

The proposed model is focused on achieving high accuracy and real-time performance in object detection using a deep learning-based approach. The two types of state-of-the-art methods for object detection were discussed: onestage methods prioritizing inference speed, such as YOLO, SSD, and RetinaNet, and two-stage methods prioritizing detection accuracy, such as Faster R-CNN, Mask R-CNN, and Cascade R-CNN. The Faster R-CNN and SSD have better accuracy, while YOLO performs better when speed is given preference over accuracy. The proposed model uses a deep learning-based approach that combines SSD and MobileNet to efficiently implement detection and tracking. The SSD eliminates the feature resampling stage and combines all calculated results as a single component, while MobileNetV2 is a lightweight network model that uses depth-wise separable convolution to perform efficient object detection without compromising on performance. The model aims to elaborate on the accuracy of the SSD object detection method and the importance of the pre-trained deep learning model MobileNetV2. The experiments were conducted on the COCO dataset to recognize objects, and the model was also tested on real-time images for object recognition. The resulting system is fast and accurate, making it suitable for applications that require object detection.

Keywords

RetinaNet, Object recognition, object detection, convolution network, MobileNetV2.
User
Notifications
Font Size

  • Murthy CB, Hashmi MF, Bokde ND, Geem ZW,“Investigations of Object Detection in Images/Videos Using Various Deep Learning Techniques and Embedded Platforms—A Comprehensive Review”, Applied Sciences,vol. 10, no. 9, pp. 1-8, 2020.

  • Girshick R, Donahue J, Darrell T, Malik J. Rich, “Feature Hierarchies for Accurate Object Detection and Semantic Segmentation”, IEEE Conference on Computer Vision and Pattern Recognition, pp. 580– 587, 2014.

  • He K, Zhang X, Ren S, Sun J, “Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition”,IEEEInternational Conference on Computer Vision, pp. 346–361, 2014.

  • Girshick R, “Fast R-CNN”,IEEE International Conference on Computer Vision, pp.1440–1448, 2015.

  • Ren S, He K, Girshick R, Sun J. Faster R-CNN, “Towards Real-Time Object Detection with Region Proposal Networks”,IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 2017.

  • Yoo D, Park S, Lee JY, Paek AS, Kweon IS, “Attention net: Aggregating Weak Directions for Accurate Object Detection”, IEEE Conference on Computer Vision and Pattern Recognition, pp. 2659- 2667, 2016.

  • Zhao W, Du S, Emery WI, “Object based convolutional neural network for high resolution imagery classification”,IEEE Journal of Selected Topics in Applied Earth Observation and Remote Sensing, vol. 10, no.7, pp. 3386-3396, 2017.

  • Rothe R, Timofte R, Goo LV, “Deep expectation of real and apparent age from a single image without facial landmarks”, International Journal of Computer Vision. vol. 126, pp. 144–157, 2018.

  • Mau L, Ghamisi P, Zhu XX, “Unsupervised spectral spatial feature learning via deep residual convdeconv network for hyperspectral image classification”, IEEE Transactions on Geoscience and Remote Sensing, vol. 56, no. 1, pp. 391-406, 2018.

  • Redmon J, Divvala S, Girshick R, Farhadi A, You Only Look Once: Unified, Real-Time Object Detection. IEEE Conference on Computer Vision and Pattern Recognition pp. 779-788, 2016.

  • Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu CY, Berg AC, SSD: Single Shot Multibox Detector”, IEEE European Conference on Computer Vision,vol. 9905, pp. 21-37, 2016.

  • Tsung-Yi L, & Priyal G, Ross G, Kaiming He, Piotr D, “Focal Loss for Dense Object Detection”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 42, no. 2, pp.318-327, 2018.

  • Chaudhary PK, Pachori RB,“FBSED based automatic diagnosis of COVID-19 using X-ray and CT images”, Computers Biology and Medicine, vol. 134, pp. 104454, 2021.

  • Shangsheng Z, Jiangzhou Z, Xiaobo C, Yanqiang L, ‘Road Information Detection Method Based on Deep Learning”, International Conference on Electronic Technology and Information Science, vol. 1847, 2021.

  • Chaudhary PK, Pachori RB, “Automatic diagnosis of glaucoma using two dimensional Fourier-Bessel series expansion based empirical wavelet transform. Biomedical Signal Processing and Control”,vol. 64, pp. 1-17, 2021.

  • https://cocodataset.org/#home


Abstract Views: 147

PDF Views: 1




  • Development of Object Recognition Model Using Machine Learning Algorithms on MobileNet V2

Abstract Views: 147  |  PDF Views: 1

Authors

Shridhar H.
Department of Electronics and Communication Engineering, India
Sumalata M. V.
Department of Electronics and Communication Engineering Government Engineering College, Haveri-581110, Karnataka, India
Thushara M.
Department of Electronics and Communication Engineering Government Engineering College, Haveri-581110, Karnataka, India
Ashwini D. N.
Department of Electronics and Communication Engineering Government Engineering College, Haveri-581110, Karnataka, India
M. Suma
Department of Electronics and Communication Engineering Government Engineering College, Haveri-581110, Karnataka, India
R. Premananda
Department of Electronics and Communication Engineering Government Engineering College, Haveri-581110, Karnataka, India
Sunil S. Harakannanavar
Department of Electronics and Communication Engineering Nitte Meenakshi Institute of Technology, Yelahanka, Bangalore-560064, Karnataka, India

Abstract


The proposed model is focused on achieving high accuracy and real-time performance in object detection using a deep learning-based approach. The two types of state-of-the-art methods for object detection were discussed: onestage methods prioritizing inference speed, such as YOLO, SSD, and RetinaNet, and two-stage methods prioritizing detection accuracy, such as Faster R-CNN, Mask R-CNN, and Cascade R-CNN. The Faster R-CNN and SSD have better accuracy, while YOLO performs better when speed is given preference over accuracy. The proposed model uses a deep learning-based approach that combines SSD and MobileNet to efficiently implement detection and tracking. The SSD eliminates the feature resampling stage and combines all calculated results as a single component, while MobileNetV2 is a lightweight network model that uses depth-wise separable convolution to perform efficient object detection without compromising on performance. The model aims to elaborate on the accuracy of the SSD object detection method and the importance of the pre-trained deep learning model MobileNetV2. The experiments were conducted on the COCO dataset to recognize objects, and the model was also tested on real-time images for object recognition. The resulting system is fast and accurate, making it suitable for applications that require object detection.

Keywords


RetinaNet, Object recognition, object detection, convolution network, MobileNetV2.

References