arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1604.04004cs.CV
收藏

了解图像质量如何影响深度神经网络

Understanding How Image Quality Affects Deep Neural Networks

Arizona State University(亚利桑那州立大学)

Samuel Dodge, Lina Karam

摘要:

图像质量是一个重要的实际挑战,在机器视觉系统的设计中经常被忽视。通常,机器视觉系统是在高质量的图像数据集上训练和测试的,然而在实际应用中,不能假设输入图像是高质量的。最近,深度神经网络在许多机器视觉任务上获得了最先进的性能。在本文中,我们对4种最先进的深度神经网络模型进行了评估,用于质量失真下的图像分类。我们考虑五种类型的质量失真:模糊,噪声,对比度,JPEG和JPEG 2000压缩。我们表明,现有的网络容易受到这些质量失真,特别是模糊和噪声。这些结果使未来的工作能够开发对质量失真更不变性的深度神经网络。

英文摘要:

Image quality is an important practical challenge that is often overlooked in the design of machine vision systems. Commonly, machine vision systems are trained and tested on high quality image datasets, yet in practical applications the input images can not be assumed to be of high quality. Recently, deep neural networks have obtained state-of-the-art performance on many machine vision tasks. In this paper we provide an evaluation of 4 state-of-the-art deep neural network models for image classification under quality distortions. We consider five types of quality distortions: blur, noise, contrast, JPEG, and JPEG2000 compression. We show that the existing networks are susceptible to these quality distortions, particularly to blur and noise. These results enable future work in developing deep neural networks that are more invariant to quality distortions.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑