|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/deep-learning-application-for-earth-observation/
课程评论：没有评论
课程名称：地球观测的深度学习应用 课程概述：深度学习是机器学习的一个子集，它利用数学函数将输入映射到输出。这些函数能够从数据中提取非冗余的信息或模式，从而形成输入和输出之间的关系。这一过程被称为学习，学习的过程称为训练。随着计算技术的快速发展，自动计算机辅助处理技术在科学和工程中的兴趣、能力及优势愈加明显，尤其是在计算机视觉（CV）技术与深度学习（DL，亦称计算智能）系统结合的情况下，能够实现高自动化和高准确度。本课程将探讨AI算法在地球观测（EO）应用中的使用。参与者将熟悉AI概念、深度学习及卷积神经网络（CNN）。课程将展示CNN在目标检测、语义分割和分类等方面的应用。本课程分为六个不同部分，参与者将在每个部分学习深度学习在地球观测应用中的最新趋势。课程中将使用以下技术：TensorFlow（使用Keras进行模型训练）、Google Colab（Jupyter Notebook的替代品）、GeoTile包（用于创建深度学习训练数据集）、ArcGIS Pro（另一种创建训练数据集的方法）以及QGIS（用于可视化输出）。
Deep Learning is a subset of Machine Learning that uses mathematical functions to map the input to the output. These functions can extract non-redundant information or patterns from the data, which enables them to form a relationship between the input and the output. This is known as learning, and the process of learning is called training.With the rapid development of computing, the interest, power, and advantages of automatic computer-aided processing techniques in science and engineering have become clear-in particular, automatic computer vision (CV) techniques together with deep learning (DL, a.k.a. computational intelligence) systems, in order to reach both a very high degree of automation and high accuracy.This course is addressing the use of AI algorithms in EO applications. Participants will become familiar with AI concepts, deep learning, and convolution neural network (CNN). Furthermore, CNN applications in object detection, semantic segmentation, and classification will be shown. The course has six different sections, in each section, the participants will learn about the recent trend of deep learning in the earth observation application. The following technology will be used in this course,Tensorflow (Keras will be used to train the model)Google Colab (Alternative to Jupiter notebook)GeoTile package (to create the training dataset for DL)ArcGIS Pro (Alternative way to create the training dataset)QGIS (Simply to visualize the outputs)
