acm-header
Sign In

Communications of the ACM

ACM TechNews

Computer Vision and Robotics Used to Survey Cardigan Bay


View as: Print Mobile App Share:
Scientists Matt Pugh and Phil Hughes on board a Friends of Cardigan Bay dinghy.

Computer scientists from Aberystwyth University have helped marine experts to map and record the ocean floor.

Credit: Aberystwyth University

Aberystwyth University computer scientists assisted marine experts in mapping and recording the sea bed. The team enabled the marine conservation group Friends of Cardigan Bay to work more efficiently by utilizing software programs to review video images of the sea floor.

"The key question we wanted to address was that of 'substrate classification,'" says Aberystwyth's Hannah Dee. "Put simply, can we wave a video camera around underwater, and map the sea floor from the video we get back?"

The use of computer vision and machine-learning techniques enabled the marine experts to automatically build a classification system that categorizes the sea floor into different habitats and classes, such as fine sands and rocky areas. "The project obtained good classifications and was a major achievement, considering it was a one-year project," Dee says.

Traditionally, researchers use a clipboard and a survey vehicle, and sit for hours evaluating videos. However, this approach is costly and time-consuming, and also creates a significant carbon footprint. "It is hoped that this [method] combined with...robotics...could be developed further which would change significantly how current data is obtained," notes marine expert Phil Hughes.

From Aberystwyth University
View Full Article

 

Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account