论文信息 - Background Invariance by Adversarial Learning

Background Invariance by Adversarial Learning

Convolutional neural networks are shown to be vulnerable to changes in the background. The proposed method is an end-to-end method that augments the training set by introducing new backgrounds during the training process. These backgrounds are created by a generative network that is trained as an adversary to the model. A case study is explored based on overhead power line insulators detection using a drone – a training set is prepared from photographs taken inside a laboratory and then evaluated using photographs that are harder to collect from outside the laboratory. The proposed method improves performance by over 20% for this case study.

Joaquim F. Pinto da Costa | Eduardo F. Simas Filho | Ricardo Cruz | Jaime S. Cardoso | Ricardo M. Prates

[1] Roland Vollgraf,et al. Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms , 2017, ArXiv.

[2] Yoshua Bengio,et al. Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[3] Matthias Bethge,et al. ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness , 2018, ICLR.

[4] Jonathon Shlens,et al. Explaining and Harnessing Adversarial Examples , 2014, ICLR.

[5] Yoshua Bengio,et al. Generative Adversarial Nets , 2014, NIPS.

[6] Thomas Brox,et al. U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[7] Jaime S. Cardoso,et al. Automatic Augmentation by Hill Climbing , 2019, ICANN.

[8] Sebastian Nowozin,et al. Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift , 2019, NeurIPS.

[9] Andrew Zisserman,et al. Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[10] Yi Yang,et al. Diagnose like a Radiologist: Attention Guided Convolutional Neural Network for Thorax Disease Classification , 2018, ArXiv.

[11] Yincheng Qi,et al. Multi-patch deep features for power line insulator status classification from aerial images , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[12] Jaime S. Cardoso,et al. Insulator visual non-conformity detection in overhead power distribution lines using deep learning , 2019, Comput. Electr. Eng..