Presenters-0452: Difference between revisions
From CSDMS
(Created page with "{{Presenters temp |CSDMS meeting event title=CSDMS 2020: Linking Ecosphere and Geosphere |CSDMS meeting event year=2020 |CSDMS meeting presentation type=Clinic |CSDMS meeting...") |
No edit summary |
||
| (8 intermediate revisions by 2 users not shown) | |||
| Line 5: | Line 5: | ||
|CSDMS meeting first name=Daniel | |CSDMS meeting first name=Daniel | ||
|CSDMS meeting last name=Buscombe | |CSDMS meeting last name=Buscombe | ||
|CSDMS meeting institute= | |CSDMS meeting institute=Marda Science | ||
|Country member=United States | |Country member=United States | ||
|CSDMS meeting state=Arizona | |CSDMS meeting state=Arizona | ||
|CSDMS meeting email address=daniel | |CSDMS meeting email address=daniel@mardascience.com | ||
|CSDMS meeting title presentation=Landcover and landform classification using deep neural networks | |CSDMS meeting title presentation=Part I: Landcover and landform classification using deep neural networks | ||
}} | }} | ||
{{Presenters presentation | {{Presenters presentation | ||
|CSDMS meeting abstract presentation=This two-part clinic will introduce deep learning methods for semantic segmentation of high-resolution aerial imagery for the purposes of landuse/cover/form classification. The datasets we will use consist of images of shoreline environments, with a focus on general-purpose classification in terrestrial, fluvial and coastal ecology and geomorphology. | |CSDMS meeting abstract presentation=This two-part clinic will introduce deep learning methods for semantic segmentation of high-resolution aerial imagery for the purposes of landuse/cover/form classification. The datasets we will use consist of images of shoreline environments, with a focus on general-purpose classification in terrestrial, fluvial and coastal ecology and geomorphology.<br><br>Deep neural networks are the current state-of-the-art for discrete classification of remotely sensed imagery from Earth observation platforms. The clinic will guide users through the process of preparing training datasets, training models, and evaluation. A number of different deep convolutional neural network architectures for image feature extraction and pixel-scale classifications will be explored and compared. The clinic will use the keras and tensorflow libraries within the python programming language. This hands-on class will be taught using Google colab through a browser, with the materials hosted on github. Participants will require a working knowledge of python. Some working knowledge of machine learning would be helpful, but we will assume no prior experience with machine/deep learning, neural networks, tensorflow, or keras.<br><br>Both the concepts and specific software would apply to many similar classification tasks at landscape scales. This clinic is composed of two, 2-hr sessions. You should sign up to both; the first clinic introduces the topic, data, and technology we use to solve the problem, and the second clinic implements these ideas and evaluates the results. | ||
'''Clinic materials can be found at:''' | |||
|CSDMS meeting youtube code= | * https://mardascience.gitlab.io/deep_learning_landscape_classification | ||
* https://colab.research.google.com/drive/1krjeCmKoAng0BWy-4mzHVX-eAqQ9qy22?usp=sharing | |||
* https://colab.research.google.com/drive/1_ddXkrZCRne7qJ2RXHV5l3qOnk98KIyp?usp=sharing | |||
|CSDMS meeting youtube code=eHikjEJ6D2M | |||
|CSDMS meeting participants=0 | |CSDMS meeting participants=0 | ||
}} | }} | ||
Latest revision as of 13:18, 26 May 2020
CSDMS 2020: Linking Ecosphere and Geosphere
Part I: Landcover and landform classification using deep neural networks
Abstract
This two-part clinic will introduce deep learning methods for semantic segmentation of high-resolution aerial imagery for the purposes of landuse/cover/form classification. The datasets we will use consist of images of shoreline environments, with a focus on general-purpose classification in terrestrial, fluvial and coastal ecology and geomorphology.
Deep neural networks are the current state-of-the-art for discrete classification of remotely sensed imagery from Earth observation platforms. The clinic will guide users through the process of preparing training datasets, training models, and evaluation. A number of different deep convolutional neural network architectures for image feature extraction and pixel-scale classifications will be explored and compared. The clinic will use the keras and tensorflow libraries within the python programming language. This hands-on class will be taught using Google colab through a browser, with the materials hosted on github. Participants will require a working knowledge of python. Some working knowledge of machine learning would be helpful, but we will assume no prior experience with machine/deep learning, neural networks, tensorflow, or keras.
Both the concepts and specific software would apply to many similar classification tasks at landscape scales. This clinic is composed of two, 2-hr sessions. You should sign up to both; the first clinic introduces the topic, data, and technology we use to solve the problem, and the second clinic implements these ideas and evaluates the results.
Clinic materials can be found at:
Please acknowledge the original contributors when you are using this material. If there are any copyright issues, please let us know (CSDMSweb@colorado.edu) and we will respond as soon as possible.
Of interest for:
