Impact of Light and Shadow on Robustness of Deep Neural Networks
Not on the map. Why?
Abstract (arXiv)
Deep neural networks (DNNs) have made remarkable strides in various computer vision tasks, including image classification, segmentation, and object detection. However, recent research has revealed a vulnerability in advanced DNNs when faced with deliberate manipulations of input data, known as adversarial attacks. Moreover, the accuracy of DNNs is heavily influenced by the distribution of the training dataset. Distortions or perturbations in the color space of input images can introduce out-of-distribution data, resulting in misclassification. In this work, we propose a brightness-variation dataset, which incorporates 24 distinct brightness levels for each image within a subset of ImageNet. This dataset enables us to simulate the effects of light and shadow on the images, so as is to investigate the impact of light and shadow on the performance of DNNs. In our study, we conduct experiments using several state-of-the-art DNN architectures on the aforementioned dataset. Through our analysis, we discover a noteworthy positive correlation between the brightness levels and the loss of accuracy in DNNs. Furthermore, we assess the effectiveness of recently proposed robust training techniques and strategies, including AugMix, Revisit, and Free Normalizer, using the ResNet50 architecture on our brightness-variation dataset. Our experimental results demonstrate that these techniques can enhance the robustness of DNNs against brightness variation, leading to improved performance when dealing with images exhibiting varying brightness levels.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 27 of its 40 references that are in the map. First posted on arXiv 23 May 2023.
Builds on
27 of 40 references.
References · 27
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
- 2016Densely Connected Convolutional Networkscited by 6,720
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
- 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
- 2013Intriguing properties of neural networkscited by 5,155
- 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
- 2014Going Deeper with Convolutionscited by 4,241
- 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,877
- 2019Natural Adversarial Examplescited by 1,408
- 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 901
- 2015The Limitations of Deep Learning in Adversarial Settingscited by 790
15 more
- 2016Universal adversarial perturbationscited by 784
- 2017One pixel attack for fooling deep neural networkscited by 582
- 2021High-Performance Large-Scale Image Recognition Without Normalizationcited by 360
- 2014Caffe: Convolutional Architecture for Fast Feature Embeddingcited by 288
- 2015Inverting Visual Representations with Convolutional Networkscited by 170
- 2021Revisiting ResNets: Improved Training and Scaling Strategiescited by 158
- 2017EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examplescited by 155
- 2016Understanding How Image Quality Affects Deep Neural Networkscited by 119
- 2018Maximal Jacobian-based Saliency Map Attackcited by 46
- 2017On Classification of Distorted Images with Deep Convolutional Neural Networkscited by 26
- 2017DeepCorrect: Correcting DNN models against Image Distortionscited by 15
- 2018Distortion Robust Image Classification using Deep Convolutional Neural Network with Discrete Cosine Transformcited by 10
- 2020Color Channel Perturbation Attacks for Fooling Convolutional Neural Networks and A Defense Against Such Attackscited by 10
- 2020Revisiting Edge Detection in Convolutional Neural Networkscited by 8
- 2021Natural Adversarial Objectscited by 7
Built on by
Not cited yet.