Ship Detection from Satellite Imagery || Machine Learning Project for Space and Sustainability

Shipping traffic is growing fast. More ships increase the chances of infractions at sea like environmentally devastating ship accidents, piracy, illegal fishing, drug trafficking, and illegal cargo movement. This has compelled many organizations, from environmental protection agencies to insurance companies and national government authorities, to have a closer watch over the open seas.
In this Project, we are building a Machine Learning and Deep learning system to accurately detect Ships in the Ocean from Satellite Imagery. This Project uses Semantic Segmentation Methodology with custom build UNet Architecture.
Udemy Certified Course On this Project: www.udemy.com/...
#machinelearning #mlprojects #space #deeplearning
You can submit the tasks at the end of this project and submit your code notebook in the below form. You will get a certificate for this project by submitting the solution for the tasks.
Project Notebook: github.com/Spa...
Submission Form Link: forms.gle/viPS...
RUclips Academic Projects Series is a video series by Spartificial that provides free certified Machine Learning related student projects focusing on Space technology and Sustainable Development.
** TIME STAMPS - SDSI0922 **
-------------------------------
00:00:55 Training Programs at Spartificial
00:03:33 Intro to Dataset
00:05:02 Applications of Machine Learning in Computer Vision
00:11:55 Exploring the data
00:22:37 Run Length Encoding and Decoding
00:36:15 Convert RLE Masks to Images
01:04:16 Preparing Train and Validation Data
01:14:44 Random Undersampling for ship counts
01:22:55 Parameters that you may want to change later
01:38:52 Augmenting Images and Masks
01:52:07 Intro to Convolutions, Upsampling, Pooling
01:57:08 Understanding UNET and understanding our model
02:20:11 Understanding Metric and Loss for compiling the model
02:25:32 Preparing Callbacks for training
02:30:08 Tasks for you to get chance to earn the certificate for this project
-------------------------------------------------------------------------------------------------
We are Spartificial, an AI and Space Research company where we majorly focus on project-based learning. We are currently working on deploying different cutting-edge technologies like AI and Robotics in Space Exploration, Earth Observation and more. We have a community of AI and Space enthusiasts looking for opportunities in the domain of AI and Space technologies.
Follow us to know more:
Website: www.spartifici...
Instagram: / spartificial_
Linkedin: / spar. .
Discord: / discord
-------------------------------------------------------------------------------------------------------
machine learning for space-tech
ai projects
artificial intelligence
ML projects
real world machine learning projects
video tutorial machine learning
certification project
project certificate

[@ninjalife11]

I understand it, thank you very much! The Run length encoding explanation was superb!

[@rohanshah8129]

Thank you 🤗

[@ramganesh3689]

you guys are doing a really good job. educating the students with intersection of two major industry technologies. thank you soo much for these kinds of videos. and i feel that these kinds of videos should roll out every now and then for educating the mass

[@rohanshah8129]

💗💗

[@cemrealdogan7163]

Hi, thanks for this informative video. Do you also classify ships like for instance, general Cargo, Containership, Bulk carrier, etc.? If so, should we implement some sort of preprocessing to satellite images before annotating train images?

[@rohanshah8129]

Hey @cemrealdogan7163,
Surely you can perform the classification task too. However, for that you will be required to label the images as in what images contain what class of ships.
Then you can use that model itself to classify remaining images if it performs well during evaluation phase.
If you give it a try do let us know how the experience was for you.
All the best!

[@sajansudhir1859]

Can u do project on rocket landing pin point detection from satellite images????

[@rohanshah8129]

All we need is its data!!
Once we have it then we can surely play around 🥂