CNN-Based Target Recognition and Identification for Infrared Imaging in Defense Systems
Convolutional neural networks (CNNs) have rapidly become the state-of-the-art models for image classification applications. They usually require large groundtruthed datasets for training. Here, we address object identification and recognition in the wild for infrared (IR) imaging in defense applicat...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-04-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/19/9/2040 |
| _version_ | 1811300449339310080 |
|---|---|
| author | Antoine d’Acremont Ronan Fablet Alexandre Baussard Guillaume Quin |
| author_facet | Antoine d’Acremont Ronan Fablet Alexandre Baussard Guillaume Quin |
| author_sort | Antoine d’Acremont |
| collection | DOAJ |
| description | Convolutional neural networks (CNNs) have rapidly become the state-of-the-art models for image classification applications. They usually require large groundtruthed datasets for training. Here, we address object identification and recognition in the wild for infrared (IR) imaging in defense applications, where no such large-scale dataset is available. With a focus on robustness issues, especially viewpoint invariance, we introduce a compact and fully convolutional CNN architecture with global average pooling. We show that this model trained from realistic simulation datasets reaches a state-of-the-art performance compared with other CNNs with no data augmentation and fine-tuning steps. We also demonstrate a significant improvement in the robustness to viewpoint changes with respect to an operational support vector machine (SVM)-based scheme. |
| first_indexed | 2024-04-13T06:52:20Z |
| format | Article |
| id | doaj.art-7a8d2fae9e8f46ffbf3176f34958aba8 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-13T06:52:20Z |
| publishDate | 2019-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-7a8d2fae9e8f46ffbf3176f34958aba82022-12-22T02:57:22ZengMDPI AGSensors1424-82202019-04-01199204010.3390/s19092040s19092040CNN-Based Target Recognition and Identification for Infrared Imaging in Defense SystemsAntoine d’Acremont0Ronan Fablet1Alexandre Baussard2Guillaume Quin3ENSTA-Bretagne, UMR 6285 labSTICC, 29806 Brest, FranceInstitut Mines-Télécom, UMR 6285 labSTICC, 29238 Brest, FranceENSTA-Bretagne, UMR 6285 labSTICC, 29806 Brest, FranceMBDA France, 92350 Le Plessis-Robinson, FranceConvolutional neural networks (CNNs) have rapidly become the state-of-the-art models for image classification applications. They usually require large groundtruthed datasets for training. Here, we address object identification and recognition in the wild for infrared (IR) imaging in defense applications, where no such large-scale dataset is available. With a focus on robustness issues, especially viewpoint invariance, we introduce a compact and fully convolutional CNN architecture with global average pooling. We show that this model trained from realistic simulation datasets reaches a state-of-the-art performance compared with other CNNs with no data augmentation and fine-tuning steps. We also demonstrate a significant improvement in the robustness to viewpoint changes with respect to an operational support vector machine (SVM)-based scheme.https://www.mdpi.com/1424-8220/19/9/2040deep learningCNNtarget identification and recognitioninfrared imaging |
| spellingShingle | Antoine d’Acremont Ronan Fablet Alexandre Baussard Guillaume Quin CNN-Based Target Recognition and Identification for Infrared Imaging in Defense Systems Sensors deep learning CNN target identification and recognition infrared imaging |
| title | CNN-Based Target Recognition and Identification for Infrared Imaging in Defense Systems |
| title_full | CNN-Based Target Recognition and Identification for Infrared Imaging in Defense Systems |
| title_fullStr | CNN-Based Target Recognition and Identification for Infrared Imaging in Defense Systems |
| title_full_unstemmed | CNN-Based Target Recognition and Identification for Infrared Imaging in Defense Systems |
| title_short | CNN-Based Target Recognition and Identification for Infrared Imaging in Defense Systems |
| title_sort | cnn based target recognition and identification for infrared imaging in defense systems |
| topic | deep learning CNN target identification and recognition infrared imaging |
| url | https://www.mdpi.com/1424-8220/19/9/2040 |
| work_keys_str_mv | AT antoinedacremont cnnbasedtargetrecognitionandidentificationforinfraredimagingindefensesystems AT ronanfablet cnnbasedtargetrecognitionandidentificationforinfraredimagingindefensesystems AT alexandrebaussard cnnbasedtargetrecognitionandidentificationforinfraredimagingindefensesystems AT guillaumequin cnnbasedtargetrecognitionandidentificationforinfraredimagingindefensesystems |