Development and Test of a Millimetre-Wave ISAR Images ATR System
Abstract
The Defense R &D Canada - Valcartier is developing an efficient ATR algorithm capable of classifying millimetre-wave images using a back-propagation neural network. This neural network was developed and tested using the MATRIX 2005 ISAR data bank featuring the signatures of T-72, BMP-2, T-62 and the ZSU-23-4. The neural network was developed with MATLAB Toolbox Neural Network 3.02 and the results are compared with the HNet3 ATR system. Specialized M files were developed to evaluate the performance of the neural network providing the confusion matrix and the ROC curves.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 2005
- Accession Number
- ADA471173
Entities
People
- Yves De Villers
Organizations
- DRDC Valcartier