Adversarial Robustness Enhancement of UAV-Oriented Automatic Image Recognition Based on Deep Ensemble Models
Open Access
- 8 June 2023
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 15 (12) , 3007
Abstract
Deep neural networks (DNNs) have been widely utilized in automatic visual navigation and recognition on modern unmanned aerial vehicles (UAVs), achieving state-of-the-art performances. However, DNN-based visual recognition systems on UAVs show serious vulnerability to adversarial camouflage patterns on targets and well-designed imperceptible perturbations in real-time images, which poses a threat to safety-related applications. Considering a scenario in which a UAV is suffering from adversarial attack, in this paper, we investigate and construct two ensemble approaches with CNN and transformer for both proactive (i.e., generate robust models) and reactive (i.e., adversarial detection) adversarial defense. They are expected to be secure under attack and adapt to the resource-limited environment on UAVs. Specifically, the probability distributions of output layers from base DNN models in the ensemble are combined in the proactive defense, which mainly exploits the weak adversarial transferability between the CNN and transformer. For the reactive defense, we integrate the scoring functions of several adversarial detectors with the hidden features and average the output confidence scores from ResNets and ViTs as a second integration. To verify their effectiveness in the recognition task of remote sensing images, we conduct experiments on both optical and synthetic aperture radar (SAR) datasets. We find that the ensemble model in proactive defense performs as well as three popular counterparts, and both of the ensemble approaches can achieve much more satisfactory results than a single base model/detector, which effectively alleviates adversarial vulnerability without extra re-training. In addition, we establish a one-stop platform for conveniently evaluating adversarial robustness and performing defense on recognition models called AREP-RSIs, which is beneficial for the future research of the remote sensing field.Keywords
Funding Information
- National Natural Science Foundation of China (61971426)
This publication has 51 references indexed in Scilit:
- Detecting adversarial example attacks to deep neural networks2017
- AID: A Benchmark Data Set for Performance Evaluation of Aerial Scene ClassificationIEEE Transactions on Geoscience and Remote Sensing, 2017
- UAV PHOTOGRAMMETRY IMPLEMENTATION TO ENHANCE LAND SURVEYING, COMPARISONS AND POSSIBILITIESThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2016
- Deep Residual Learning for Image Recognition2016
- Deep learningNature, 2015
- Use of Unmanned Aerial Vehicles for Medical Product TransportAir Medical Journal, 2015
- A Study on Relationship Between Generalization Abilities and Fuzziness of Base Classifiers in Ensemble LearningIEEE Transactions on Fuzzy Systems, 2014
- Diversity Regularized Ensemble Pruning2012
- Bag-of-visual-words and spatial extensions for land-use classification2010
- Standard SAR ATR evaluation experiments using the MSTAR public release data set1998