Hyperspectral satellite image classification Using Deep CNNs

Name: Hyperspectral satellite image classification Using Deep CNNs
Rating: 3.5 (77 reviews)

Hyperspectral satellite imagery classification Using 1-D, 2-D, 3-D, and Hybrid Convolutional Neural Networks (CNNs)

Created byAli Jamali

Last updated 2/2023

English

What you'll learn

Concepts/basics of Convolutional Neural Networks
Visualizing Hyperspectral data in Google Colab
1-Dimensional Convolutional Neural Network development
2-Dimensional Convolutional Neural Network development
3-Dimensional Convolutional Neural Network development
Deep machine learning models training and validation in Google Colab
Developing different advanced machine learning algorithms in Google Colab
Land Use Land Cover (LULC) mapping with Hyperspectral satellite imagery

Course content

4 sections • 19 lectures • 2h 0m total length

Introduction1:06
Welcome message.
Required Concepts2:31
Required knowledge for this course.
Basics of Convolutional Neural Networks (CNNs)12:13
Concepts of Convolutional Neural Networks (CNNs).

Load Hyperspectral data15:41
Includes how you can load a Hyperspectral satellite imagery into Google Colab.
1-Dimensional CNN development15:08
Includes how you can develop a 1-D CNN model.
1-D CNN model Training15:54
Includes how you can Train a 1-D CNN model.
Hyperspectral satellite imagery classification using a 1-D CNN model7:22
Includes the Hyperspectral satellite imagery classification using a 1-D CNN model.
Points0:56
Number of classes in Pavia dataset.

Load Hyperspectral satellite imagery3:18
Includes how you can load Hyperspectral satellite imagery data.
3-D CNN development2:16
Includes how you can develop a 3-D CNN model.
3-D CNN model training1:48
Includes how you can train a 3-D CNN model.
3-D CNN model validation1:25
Includes how you can validate a 3-D CNN model.
Hyperspectral satellite imagery classifiaction-12:07
Includes how you can classify a Hyperspectral satellite imagery using a 3-D CNN model.
Hyperspectral satellite imagery classifiaction-21:12
Includes how you can classify a Hyperspectral satellite imagery using a 3-D CNN model.
Hybrid CNN development8:05
Includes how you can develop a Hybrid 3D-2D CNN model.

Requirements

Basics of Remote Sensing
Basics of Programming

Description

Land cover mapping is a critical aspect of Earth’s surface monitoring and mapping. In this course, Land Use Land Cover Mapping utilizing Hyperspectral satellite imagery is covered. You will learn how to develop 1-Dimensional, 2-Dimensional, 3-Dimensional, and Hybrid Convolutional Neural Networks (CNNs) using Google Colab. The discussed and developed methods can be utilized for different object/feature extraction and mapping (i.e., urban region extraction from high-resolution satellite imagery). Remote sensing is a powerful tool that can be used to identify and classify different land types, assess vegetation conditions, and estimate environmental changes. The use of Google Colab will significantly help you to decrease the issues encountered by software and platforms, such as Anaconda. There is a much lower need for library installation in the Google Colab, resulting in faster and more reliable classification map generation. The validation of the developed models is also covered. In summary, remote sensing and GIS technologies are widely used for land cover mapping. They provide accurate and timely information that is critical for monitoring and managing natural resources.

Highlights:

1. Learn the concepts of Convolutional Neural Networks (CNNs)

2. Learn how to develop CNN models

3. Learn how to classify Hyperspectral satellite imagery using python programming language

4. Learn how to validate a CNN model

5. Learn to read and import your data from your Google Drive into Google Colab

6. Map Land use land covers utilizing Hyperspectral satellite data with different variations of CNN models

7. Learn how to validate a machine-learning model

Who this course is for:

Students and professionals interested in writing and publishing high-quality papers
Remote sensing engineers
GIS engineers
Govt sector agriculture scientists
Master students of GIS and Remote Sensing
Data scientists interested in Remote Sensing image processing
Ph.D. students of Data science, GIS, and Remote Sensing

Hyperspectral satellite image classification Using Deep CNNs

What you'll learn

Explore related topics

Course content

Introduction3 lectures • 16min

1-Dimensional CNNs5 lectures • 55min

2-Dimensional CNNs4 lectures • 29min

3-Dimensional CNNs7 lectures • 20min

Requirements

Description

Who this course is for: