Skip to content

Building a CNN to get accurate counts of Sea Lions, divided by gender and age, from Satellite images

Notifications You must be signed in to change notification settings

be-ns/counting_sea_lions

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Counting Sea Lions

Analyzing NOAA aerial images to get more accurate counts for Sea Lion populations in Alaska


Steller sea lions populations in the western Aleutian Islands have declined 94 percent in the last 30 years, making them an endangered species, at least for the population found in the North Pacific. The data used in this analysis was obtained from unoccupied aircraft systems, and downloaded from this Kaggle Competition. The desired result would be to subdivide the population and get accurate adult male and female populations as well as juveniles and pups.

"Currently, it takes biologists up to four months to count sea lions from the thousands of images NOAA Fisheries collects each year. Once individual counts are conducted, the tallies must be reconciled to confirm their reliability. The results of these counts are time-sensitive." - From Kaggle

This is an example of the images we are using.

  • The Sea Lion population is relatively difficult to pick out by hand - making this a prime job for Machine Learning. s image example

Method

  • build a convolutional neural network using Keras and Theano on preprocessed images to get accurate counts for each category.

About

Building a CNN to get accurate counts of Sea Lions, divided by gender and age, from Satellite images

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages