Classification robustness to common optical aberrations

• URL: http://arxiv.org/abs/2308.15499v1
• Date: Tue, 29 Aug 2023 08:36:00 GMT
• Title: Classification robustness to common optical aberrations
• Authors: Patrick M\"uller, Alexander Braun, Margret Keuper
• Abstract summary: This paper proposes OpticsBench, a benchmark for investigating robustness to realistic, practically relevant optical blur effects. Experiments on ImageNet show that for a variety of different pre-trained DNNs, the performance varies strongly compared to disk-shaped kernels. We show on ImageNet-100 with OpticsAugment that can be increased by using optical kernels as data augmentation.
• Score: 64.08840063305313
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Computer vision using deep neural networks (DNNs) has brought about seminal changes in people's lives. Applications range from automotive, face recognition in the security industry, to industrial process monitoring. In some cases, DNNs infer even in safety-critical situations. Therefore, for practical applications, DNNs have to behave in a robust way to disturbances such as noise, pixelation, or blur. Blur directly impacts the performance of DNNs, which are often approximated as a disk-shaped kernel to model defocus. However, optics suggests that there are different kernel shapes depending on wavelength and location caused by optical aberrations. In practice, as the optical quality of a lens decreases, such aberrations increase. This paper proposes OpticsBench, a benchmark for investigating robustness to realistic, practically relevant optical blur effects. Each corruption represents an optical aberration (coma, astigmatism, spherical, trefoil) derived from Zernike Polynomials. Experiments on ImageNet show that for a variety of different pre-trained DNNs, the performance varies strongly compared to disk-shaped kernels, indicating the necessity of considering realistic image degradations. In addition, we show on ImageNet-100 with OpticsAugment that robustness can be increased by using optical kernels as data augmentation. Compared to a conventionally trained ResNeXt50, training with OpticsAugment achieves an average performance gain of 21.7% points on OpticsBench and 6.8% points on 2D common corruptions.

Related papers

• Utilizing Generative Adversarial Networks for Image Data Augmentation and Classification of Semiconductor Wafer Dicing Induced Defects [0.21990652930491852]
  In semiconductor manufacturing, the wafer dicing process is central yet vulnerable to defects that significantly impair yield. Deep neural networks are the current state of the art in (semi-)automated visual inspection. We explore the application of generative adversarial networks (GAN) for image data augmentation and classification of semiconductor wafer dicing induced defects.
  arXiv  Detail & Related papers  (2024-07-24T20:44:16Z)
• Impact of Light and Shadow on Robustness of Deep Neural Networks [5.015796849425367]
  Deep neural networks (DNNs) have made remarkable strides in various computer vision tasks, including image classification, segmentation, and object detection. Recent research has revealed a vulnerability in advanced DNNs when faced with deliberate manipulations of input data, known as adversarial attacks. We propose a brightness-variation dataset, which incorporates 24 distinct brightness levels for each image within a subset of ImageNet.
  arXiv  Detail & Related papers  (2023-05-23T15:30:56Z)
• Physics-aware Roughness Optimization for Diffractive Optical Neural Networks [15.397285424104469]
  diffractive optical neural networks (DONNs) have shown promising advantages over conventional deep neural networks. We propose a physics-aware diffractive optical neural network training framework to reduce the performance difference between numerical modeling and practical deployment.
  arXiv  Detail & Related papers  (2023-04-04T03:19:36Z)
• Deep Dynamic Scene Deblurring from Optical Flow [53.625999196063574]
  Deblurring can provide visually more pleasant pictures and make photography more convenient. It is difficult to model the non-uniform blur mathematically. We develop a convolutional neural network (CNN) to restore the sharp images from the deblurred features.
  arXiv  Detail & Related papers  (2023-01-18T06:37:21Z)
• LoCI: An Analysis of the Impact of Optical Loss and Crosstalk Noise in Integrated Silicon-Photonic Neural Networks [8.930237478906266]
  Integrated silicon-photonic neural networks (SP-NNs) promise higher speed and energy efficiency for emerging artificial-intelligence applications. This paper presents the first comprehensive and systematic optical loss and crosstalk modeling framework for SP-NNs.
  arXiv  Detail & Related papers  (2022-04-08T04:22:39Z)
• A Deep Variational Bayesian Framework for Blind Image Deblurring [46.585763459441154]
  Blind image deblurring is an important yet very challenging problem in low-level vision. We present a deep variational Bayesian framework for blind image deblurring.
  arXiv  Detail & Related papers  (2021-06-05T12:47:36Z)
• Universal and Flexible Optical Aberration Correction Using Deep-Prior Based Deconvolution [51.274657266928315]
  We propose a PSF aware plug-and-play deep network, which takes the aberrant image and PSF map as input and produces the latent high quality version via incorporating lens-specific deep priors. Specifically, we pre-train a base model from a set of diverse lenses and then adapt it to a given lens by quickly refining the parameters.
  arXiv  Detail & Related papers  (2021-04-07T12:00:38Z)
• Improving robustness against common corruptions with frequency biased models [112.65717928060195]
  unseen image corruptions can cause a surprisingly large drop in performance. Image corruption types have different characteristics in the frequency spectrum and would benefit from a targeted type of data augmentation. We propose a new regularization scheme that minimizes the total variation (TV) of convolution feature-maps to increase high-frequency robustness.
  arXiv  Detail & Related papers  (2021-03-30T10:44:50Z)
• Calibrating Deep Neural Networks using Focal Loss [77.92765139898906]
  Miscalibration is a mismatch between a model's confidence and its correctness. We show that focal loss allows us to learn models that are already very well calibrated. We show that our approach achieves state-of-the-art calibration without compromising on accuracy in almost all cases.
  arXiv  Detail & Related papers  (2020-02-21T17:35:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.