What you would learn in Deep Learning Application for Earth Observation course?
Deep Learning is a subset of Machine Learning that utilizes mathematical methods to connect the input and output. The functions can find patterns or non-redundant information from the data. This can help them form connections between the input to the output. This is also known as studying, and it is known as the process of training.
As technology advances, computing, the importance as well as the advantages of computer-aided, automatic processing techniques in engineering and science are now evident, in particular, automated computer vision (CV) techniques, in conjunction and deep learning (DL, a.k.a. computational intelligence) systems, to enable to attain a high level of automation and high-accuracy.
This course will focus on the application in the use of AI algorithms in EO applications. Participants will be familiar with AI concepts, deep learning, deep learning, and convolution neural networks (CNN). Additionally, CNN applications in object detection, semantic segmentation, and classification will be demonstrated. The course is divided into six sections. In each, the students will be taught about the current trends in deep learning for the application of earth observation. The following technologies are used during this course.
Tensorflow (Keras is used to create the model)
Google Colab (Alternative to Jupiter notebook)
GeoTile package (to make the training datasets for DL)
ArcGIS Pro (Alternative method to build your training data)
QGIS (Simply to show the outputs)
Course Content:
- The practical use case for deep learning to improve satellite imagery
- Analysis of satellite imagery
- Object detection
- Image classification
- Image segmentation
- Keras, Tensorflow
- ArcGIS Pro (Optional)
- QGIS (Optional)
Download Deep Learning Application for Earth Observation from below links NOW!
