In Field Technologies

3D Machine Vision

The Technology

Known as ‘robot eyes’, 3D machine vision technologies construct high resolution digital 3D models (known as point clouds) of textured and contoured objects in real time, independently of ambient light conditions. These 3D images are captured and processed by combining 3D laser scanners with image processing and deep learning software.

How it Works

3D vision systems can either be comprised of multiple cameras which provide part-orientation information by ‘triangulation on an objective position in space.  These are mostly used in robot guidance systems. Highly accurate 3D Laser displacement sensor technology is more suited to surface inspection and volume measurement, with typically just one camera generating a height map from the displacement of the reflected laser on the target object.

Farmer / Agronomist Benefits

3D machine vision has been already been deployed in weed recognition and spot spraying systems as well as direct weeding robots, driverless tractors automated fruit and vegetable harvesters, tuber grading systems and plant phenotyping. Whilst still in its infancy in UK arable crop protection, 3D vision systems supported by machine learning and combined with other new technologies and machine learning will improve the efficiency and quality of agricultural production.

Key Researchers / Stakeholders

Ecorobotix

Basler

Canon

 

Video / Social Media Links

https://youtu.be/XlXSGqvP-A8  (tomato picker)

Subscribe for alerts

By submitting this form you are agreeing to our Privacy Policy.

Get in Touch

We're here to help and answer any questions you might have. We look forward to hearing from you...