Impact of the Radar Image Resolution of Military Objects on the Accuracy of their Classification by a Deep Convolutional Neural Network
Introduction. Deep convolutional neural networks are considered as one of the most promising tools for classifying small-sized objects on radar images. However, no systemic study has been reported so far on the dependence between the classification accuracy achieved by convolutional neural networks...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | Russian |
| Published: |
Saint Petersburg Electrotechnical University "LETI"
2022-02-01
|
| Series: | Известия высших учебных заведений России: Радиоэлектроника |
| Subjects: | |
| Online Access: | https://re.eltech.ru/jour/article/view/604 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839584174665629696 |
|---|---|
| author | I. F. Kupryashkin |
| author_facet | I. F. Kupryashkin |
| author_sort | I. F. Kupryashkin |
| collection | DOAJ |
| description | Introduction. Deep convolutional neural networks are considered as one of the most promising tools for classifying small-sized objects on radar images. However, no systemic study has been reported so far on the dependence between the classification accuracy achieved by convolutional neural networks and such an important image characteristic as resolution.Aim. Determination of a dependence between of the accuracy of classifying military objects by a deep convolutional neural network and the resolution of their radar images.Materials and methods. An eight-layer convolutional neural network was designed, trained and tested using the Keras library and Tensorflow 2.0 framework. For training and testing, the open part of the standard MSTAR dataset comprising ten classes of military objects radar images was used. The initial weight values of the MobileNetV1 and Xception networks used for a comparative assessment of the achieved classification accuracy were obtained from the training results on the Imagenet.Results. The accuracy of classifying military objects decreases rapidly along with a deterioration in resolution, amounting to 97.91, 90.22, 79.13, 52.2 and 23.68 % at a resolution of 0.3, 0.6, 0.9, 1.5 and 3 m, respectively. It is shown that the use of pretrained MobileNetV1 and Xception networks does not lead to an improvement in the classification accuracy compared to a simple VGG-type network.Conclusion. Effective recognition of military objects at a resolution worse than one meter is practically impossible. The classification accuracy of deep neural networks depends significantly on the difference in the image resolution of the training and test sets. A significant increase in the resistance of the classification accuracy to changes in the resolution can be achieved by training on a set of images with different resolutions. |
| format | Article |
| id | doaj-art-a558a7dd3d0f40c29346edb1cfbeee04 |
| institution | Matheson Library |
| issn | 1993-8985 2658-4794 |
| language | Russian |
| publishDate | 2022-02-01 |
| publisher | Saint Petersburg Electrotechnical University "LETI" |
| record_format | Article |
| series | Известия высших учебных заведений России: Радиоэлектроника |
| spelling | doaj-art-a558a7dd3d0f40c29346edb1cfbeee042025-08-03T19:50:27ZrusSaint Petersburg Electrotechnical University "LETI"Известия высших учебных заведений России: Радиоэлектроника1993-89852658-47942022-02-01251364610.32603/1993-8985-2022-25-1-36-46429Impact of the Radar Image Resolution of Military Objects on the Accuracy of their Classification by a Deep Convolutional Neural NetworkI. F. Kupryashkin0Military Educational and Scientific Center of the Air Force "N. E. Zhukovsky and Y. A. Gagarin Air Force Academy"Introduction. Deep convolutional neural networks are considered as one of the most promising tools for classifying small-sized objects on radar images. However, no systemic study has been reported so far on the dependence between the classification accuracy achieved by convolutional neural networks and such an important image characteristic as resolution.Aim. Determination of a dependence between of the accuracy of classifying military objects by a deep convolutional neural network and the resolution of their radar images.Materials and methods. An eight-layer convolutional neural network was designed, trained and tested using the Keras library and Tensorflow 2.0 framework. For training and testing, the open part of the standard MSTAR dataset comprising ten classes of military objects radar images was used. The initial weight values of the MobileNetV1 and Xception networks used for a comparative assessment of the achieved classification accuracy were obtained from the training results on the Imagenet.Results. The accuracy of classifying military objects decreases rapidly along with a deterioration in resolution, amounting to 97.91, 90.22, 79.13, 52.2 and 23.68 % at a resolution of 0.3, 0.6, 0.9, 1.5 and 3 m, respectively. It is shown that the use of pretrained MobileNetV1 and Xception networks does not lead to an improvement in the classification accuracy compared to a simple VGG-type network.Conclusion. Effective recognition of military objects at a resolution worse than one meter is practically impossible. The classification accuracy of deep neural networks depends significantly on the difference in the image resolution of the training and test sets. A significant increase in the resistance of the classification accuracy to changes in the resolution can be achieved by training on a set of images with different resolutions.https://re.eltech.ru/jour/article/view/604deep convolutional neural networkradar imageclassification accuracy |
| spellingShingle | I. F. Kupryashkin Impact of the Radar Image Resolution of Military Objects on the Accuracy of their Classification by a Deep Convolutional Neural Network Известия высших учебных заведений России: Радиоэлектроника deep convolutional neural network radar image classification accuracy |
| title | Impact of the Radar Image Resolution of Military Objects on the Accuracy of their Classification by a Deep Convolutional Neural Network |
| title_full | Impact of the Radar Image Resolution of Military Objects on the Accuracy of their Classification by a Deep Convolutional Neural Network |
| title_fullStr | Impact of the Radar Image Resolution of Military Objects on the Accuracy of their Classification by a Deep Convolutional Neural Network |
| title_full_unstemmed | Impact of the Radar Image Resolution of Military Objects on the Accuracy of their Classification by a Deep Convolutional Neural Network |
| title_short | Impact of the Radar Image Resolution of Military Objects on the Accuracy of their Classification by a Deep Convolutional Neural Network |
| title_sort | impact of the radar image resolution of military objects on the accuracy of their classification by a deep convolutional neural network |
| topic | deep convolutional neural network radar image classification accuracy |
| url | https://re.eltech.ru/jour/article/view/604 |
| work_keys_str_mv | AT ifkupryashkin impactoftheradarimageresolutionofmilitaryobjectsontheaccuracyoftheirclassificationbyadeepconvolutionalneuralnetwork |