AI radar (object classification using deep learning)
In this project, we aim to use the self-collected datasets which is fully labelled to train a Convolutional Neural Network (CNN) to reduce the computation cost and improve performance accuracy, to classify the targets detected by the radar as human beings or nonhuman objects, using the range-Doppler...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/157516 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | In this project, we aim to use the self-collected datasets which is fully labelled to train a Convolutional Neural Network (CNN) to reduce the computation cost and improve performance accuracy, to classify the targets detected by the radar as human beings or nonhuman objects, using the range-Doppler maps. For the radar used in this project, it is a frequency modulated continuous wave (FMCW) radar. The whole project can be divided into three parts. Firstly, Design and implementation of collecting training data process. Secondly, training data preprocessing using MATLAB. And lastly, the construction of Convolutional Neural Network based on a VGG-11 backbone using PyTorch. Results show the encouraging improvement on the classification accuracy. |
|---|