SRGAN-PyTorch 概述 该存储库包含对进行的逐点PyTorch重新实现。 目录 关于使用生成对抗网络的逼真的单图像超分辨率 如果您不熟悉SRGAN，请直接从本文中摘录以下内容： 尽管使用更快，更深的卷积神经网络在单图像超分辨率的准确性和速度方面取得了突破，但仍然存在一个主要问题仍未解决：当在较大的放大比例下进行超分辨率处理时，如何恢复更精细的纹理细节？ 基于优化的超分辨率方法的行为主要由目标函数的选择决定。 最近的工作主要集中在最小化均方重构误差上。 得出的估计值具有很高的峰值信噪比，但是它们通常缺少高频细节，并且在某种意义上说它们无法满足更高分辨率下的保真度，因此在感觉上并不令人满意。 在本文中，我们介绍了SRGAN，这是一种用于图像超分辨率（SR）的生成对抗网络（GAN）。 据我们所知，它是第一个能够为4倍放大因子推断出逼真的自然图像的框架。 为此，我们提出了一种感知损失函

Photonics-based radar with a photonic de-chirp receiver has the advantages of broadband operation and real-time signal processing, but it suffers from interference from image frequencies and other undesired frequency-mixing components, due to single-channel real-valued photonic frequency mixing. In this paper, we propose a photonics-based radar with a photonic frequency-doubling transmitter and a balanced in-phase and quadrature (I/Q) de-chirp receiver. This radar transmits broadband linearly fr

为了从高分辨率遥感影像中高精度的提取各种工业固体废物，提出了基于支持向量机（ＳＶＭ）的工业固体废物特征快速提取算法，首先对遥感影像进行各向异性扩散滤波，然后针对高分辨率遥感影像信息量丰富和工业固体废物堆复杂度高的特点，提出基于1-v-1的SVM的多种工业固体废物的多分类策略，在提高精度的同时，兼顾了工业固体废物特征识别的效率。研究结果表明：ＳＶＭ对工业固体废物判别能力比常规方法有更强的优势，从遥感影像中不仅能准确地辨别出工业固体废物堆的线形特征，还能识别出其材质和类型；该算法能同时识别出多种工业固体废物，执行效率更高。

The sensitivity of diffuse optical tomography (DOT) imaging exponentially decreases with the increase of photon penetration depth, which leads to a poor depth resolution for DOT. In this letter, an exponential adjustment method (EAM) based on maximum singular value of layered sensitivity is proposed

It is well known that a light spot of sub-wavelength will diverge in all directions. In this letter, A method is presented for generating sub-wavelength (0.44\lambda) longitudinally polarized beam, which propagates without divergence over lengths of about 2\lambda in free space. This is achieved by a high numerical aperture (NA) lens axicon that utilizes spherical aberration to duplicate the performance of an axicon and to create an extended focal line.

A study for texture feature extraction of high-resolution satellite images based on a direction measure and gray level co-occurrence matrix fusion algorithm

kscript：Kotlin的脚本增强

Image Super-Resolution Via Sparse Representation的中文版本

ISO国际标准，英文版，Resolution and SFR 影像分辨率空间频率响应的测量标准，介绍了视觉分辨率、e-SFR、s-SFR的测量方法。

本书是Donald经典之作High Resolution Radar的中文版，影印可能不太清晰，但是可以看。