logo
Social Media
twitter
youtube

  • Home
  • Team
  • Vehicle
    • SubjuGator 9
    • SubjuGator 8
    • SubjuGator 7
    • SubjuGator 6
    • SubjuGator 5
    • SubjuGator 4
    • SubjuGator 3
    • SubjuGator 2
    • SubjuGator 1
  • Media
  • Press
  • Sponsors
    • Sponsors 2018
    • Sponsors 2017
    • Sponsors 2016
    • Sponsors 2015
    • Sponsors 2014
    • Sponsors 2013
    • Sponsors 2012
  • Blog
  • Outreach
  • Contact



Begining Stages of Machine Learning

Mar 23, 2016
by SubjuGator
Comments are off

This Week We

  • Made a tool for segmenting training data
  • Experimented with using an SVM for segmentation
  • Worked on stuff for easily segmenting/labelling planes in PCL
    • + distinguishing things “on top” of the plane from the plane itself
The training tool: The plot on the right is HSV->XYZ. You draw a box around the desired color groups, and use the resulting segmentation.

This is less annoying than manually drawing on images.

easy_labeller

A SVM was trained on the segmentation data and it ran in ~near~ real-time on a super downsampled input image.
This is one frame of the SVM running on the video. It was just a 2-second demo, the threshold segmentation looks a lot better.
Sci-kit learn’s svm implementation struggles to fit at more than 10k samples, which is another problem.
svm_trainer
About the Author
Social Share

Categories

  • Competition News (10)
  • Sponsorship News (3)
  • Team News (64)

Tag Cloud

anodized assembly auvsi battery pods buoy calibration cnc competition computer controller dvl ethernet fischer grabber gumstix hydrophone imu JDSquared kona kai manufacturing merge board motor drivers navigation navigation computer new paper pcb pipe pool power rain robosub shipping shooter testing thruster thrusters trajectory generator travel trays tuning Twitter videoray website wiring

New Members

Visit us in the Machine Intelligence Lab (MIL), which is located in MAE-C 126. Check out the 2016 RoboSub Final Missions and Tasks. If you are interested in software, see our GitHub.

About Us

SubjuGator is an autonomous underwater vehicle project designed and built by students of the Machine Intelligence Laboratory (MIL) at the University of Florida. The University of Florida has participated in the RoboSub underwater vehicle competitions started in 1998, placing in the top 3 seven times, including first place in 2005, 2006, and in 2007. The SubjuGator team is comprised of graduate and undergraduate students of the Department of Electrical and Computer Engineering, the Department of Mechanical and Aerospace Engineering, and the Department of Computer and Information Science and Engineering.

Recent Posts

2019 Media
Nov 04, 2020
2018 Media
Nov 04, 2020
Sponsors 2019
Nov 04, 2020




RoboSub 2017 begins in:

1276
Days
20
Hours
15
Mins
12
Secs

Machine Intelligence Laboratory | University of Florida
Login