Improving Deep CNN-Based Radar Target Classification Performance by Applying a Denoise Filter
This paper presents a novel method for removing noise from range-Doppler images by using a filter prior to conducting target classification using a deep neural network. Specifically, Kuan, Frost, and Lee filters are employed to eliminate speckle noise components from radar data images. Furthermore,...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The Korean Institute of Electromagnetic Engineering and Science
2024-03-01
|
| Series: | Journal of Electromagnetic Engineering and Science |
| Subjects: | |
| Online Access: | https://www.jees.kr/upload/pdf/jees-2024-2-r-220.pdf |
| _version_ | 1797213505147895808 |
|---|---|
| author | Van-Tra Nguyen Chi-Thanh Vu Van-Sang Doan |
| author_facet | Van-Tra Nguyen Chi-Thanh Vu Van-Sang Doan |
| author_sort | Van-Tra Nguyen |
| collection | DOAJ |
| description | This paper presents a novel method for removing noise from range-Doppler images by using a filter prior to conducting target classification using a deep neural network. Specifically, Kuan, Frost, and Lee filters are employed to eliminate speckle noise components from radar data images. Furthermore, a neural network that combines residual and inception blocks (RINet) is proposed. The RINet model is trained and tested on the RAD-DAR dataset—a collection of range-Doppler feature maps. The analysis results show that the application of a Lee filter with a window size of 7 in the RAD-DAR dataset demonstrates the most improvement in the model’s classification performance. On applying this noise filter to the dataset, the RINet model successfully classified radar targets, exhibiting a 4.51% increase in accuracy and a 14.07% decrease in loss compared to the classification results achieved for the original data. Furthermore, a comparison of the RINet model with the noise filtering solution with five other networks was conducted, the results of which show that the proposed model significantly outperforms the others. |
| first_indexed | 2024-04-24T10:59:20Z |
| format | Article |
| id | doaj.art-1a1bafccd8a848d8b3abb384cc2c2e23 |
| institution | Directory Open Access Journal |
| issn | 2671-7255 2671-7263 |
| language | English |
| last_indexed | 2024-04-24T10:59:20Z |
| publishDate | 2024-03-01 |
| publisher | The Korean Institute of Electromagnetic Engineering and Science |
| record_format | Article |
| series | Journal of Electromagnetic Engineering and Science |
| spelling | doaj.art-1a1bafccd8a848d8b3abb384cc2c2e232024-04-12T04:10:31ZengThe Korean Institute of Electromagnetic Engineering and ScienceJournal of Electromagnetic Engineering and Science2671-72552671-72632024-03-0124219820510.26866/jees.2024.2.r.2203658Improving Deep CNN-Based Radar Target Classification Performance by Applying a Denoise FilterVan-Tra Nguyen0Chi-Thanh Vu1Van-Sang Doan2 Department of Signal Processing, Radar Institute, Academy of Military Science and Technology, Hanoi, Vietnam Department of Signal Processing, Radar Institute, Academy of Military Science and Technology, Hanoi, Vietnam Department of Radar Systems, Faculty of Communication and Radar, Vietnam Naval Academy, Nha Trang, VietnamThis paper presents a novel method for removing noise from range-Doppler images by using a filter prior to conducting target classification using a deep neural network. Specifically, Kuan, Frost, and Lee filters are employed to eliminate speckle noise components from radar data images. Furthermore, a neural network that combines residual and inception blocks (RINet) is proposed. The RINet model is trained and tested on the RAD-DAR dataset—a collection of range-Doppler feature maps. The analysis results show that the application of a Lee filter with a window size of 7 in the RAD-DAR dataset demonstrates the most improvement in the model’s classification performance. On applying this noise filter to the dataset, the RINet model successfully classified radar targets, exhibiting a 4.51% increase in accuracy and a 14.07% decrease in loss compared to the classification results achieved for the original data. Furthermore, a comparison of the RINet model with the noise filtering solution with five other networks was conducted, the results of which show that the proposed model significantly outperforms the others.https://www.jees.kr/upload/pdf/jees-2024-2-r-220.pdfdeep neural networkdenoise filterinception-residual moduleradar target classificationrange-doppler image |
| spellingShingle | Van-Tra Nguyen Chi-Thanh Vu Van-Sang Doan Improving Deep CNN-Based Radar Target Classification Performance by Applying a Denoise Filter Journal of Electromagnetic Engineering and Science deep neural network denoise filter inception-residual module radar target classification range-doppler image |
| title | Improving Deep CNN-Based Radar Target Classification Performance by Applying a Denoise Filter |
| title_full | Improving Deep CNN-Based Radar Target Classification Performance by Applying a Denoise Filter |
| title_fullStr | Improving Deep CNN-Based Radar Target Classification Performance by Applying a Denoise Filter |
| title_full_unstemmed | Improving Deep CNN-Based Radar Target Classification Performance by Applying a Denoise Filter |
| title_short | Improving Deep CNN-Based Radar Target Classification Performance by Applying a Denoise Filter |
| title_sort | improving deep cnn based radar target classification performance by applying a denoise filter |
| topic | deep neural network denoise filter inception-residual module radar target classification range-doppler image |
| url | https://www.jees.kr/upload/pdf/jees-2024-2-r-220.pdf |
| work_keys_str_mv | AT vantranguyen improvingdeepcnnbasedradartargetclassificationperformancebyapplyingadenoisefilter AT chithanhvu improvingdeepcnnbasedradartargetclassificationperformancebyapplyingadenoisefilter AT vansangdoan improvingdeepcnnbasedradartargetclassificationperformancebyapplyingadenoisefilter |