论文标题
使用卷积自动编码器的图像denoing
Image Denoising Using Convolutional Autoencoder
论文作者
论文摘要
随着现代世界的不可阻碍的数字化，技术领域的每个子集都会不断发展。这样的子集就是如此受欢迎的数字图像。图像并不总是像您希望的那样在视觉上令人愉悦或清晰，并且经常被噪音扭曲或掩盖。随着岁月的流逝，已经出现了许多增强图像的技术，所有这些技术都具有各自的利弊。在本文中，我们研究了一种特殊的技术，该技术借助通常称为自动编码器的神经网络模型来完成此任务。我们为模型构建不同的体系结构，并比较结果，以决定最适合该任务的架构。简短地讨论了模型的特征和工作，知道哪个可以帮助为将来的研究树立途径。
With the inexorable digitalisation of the modern world, every subset in the field of technology goes through major advancements constantly. One such subset is digital images which are ever so popular. Images can not always be as visually pleasing or clear as you would want them to be and are often distorted or obscured with noise. A number of techniques to enhance images have come up as the years passed, all with their own respective pros and cons. In this paper, we look at one such particular technique which accomplishes this task with the help of a neural network model commonly known as an autoencoder. We construct different architectures for the model and compare results in order to decide the one best suited for the task. The characteristics and working of the model are discussed briefly knowing which can help set a path for future research.
