John Reidar Mathiassen

About

John Reidar Mathiassen is a leading researcher in the application of robotics, machine vision, and artificial intelligence to the food processing industry, with a particular focus on automating the handling and processing of fish. His work bridges the gap between advanced computer vision and practical industrial robotics. Mathiassen’s major contributions include pioneering the use of deep convolutional neural networks for tasks such as segmenting blood defects in cod fillets and guiding robots to grasp and process chicken and fish products. He has been instrumental in developing automated slaughter lines for salmonids and robotic systems for post-trimming salmon fillets, directly addressing labor costs and efficiency in Norwegian seafood processing. A standout achievement is his innovative approach to teaching robots to grasp real fish by training a virtual robot in virtual reality, a method that combines imitation learning with domain randomization. With his most cited paper, “GRIBBOT – Robotic 3D vision-guided harvesting of chicken fillets,” accumulating 60 citations, Mathiassen’s work has had a tangible impact on both academic research and industrial automation, making him a key figure in the field of robotic food processing.

Research Focus

Key Achievements

Top Papers

1. 1
   GRIBBOT – Robotic 3D vision-guided harvesting of chicken fillets

   60 citations · 2015
2. 2
   Teaching a Robot to Grasp Real Fish by Imitation Learning from a Human Supervisor in Virtual Reality

   24 citations · 2018
3. 3
   Robust classification approach for segmentation of blood defects in cod fillets based on deep convolutional neural networks and support vector machines and calculation of gripper vectors for robotic processing

   23 citations · 2017
4. 4
   An automated salmonid slaughter line using machine vision

   21 citations · 2011
5. 5
   Towards robotic post-trimming of salmon fillets

   20 citations · 2016
6. 6
   Grasping virtual fish: A step towards robotic deep learning from demonstration in virtual reality

   16 citations · 2017
7. 7
   Adaptive Snake Robot Locomotion: A Benchmarking Facility for Experiments

   15 citations · 2008
8. 8
   AUTOMATIC WEIGHT AND QUALITY GRADING OF WHOLE PELAGIC FISH

   12 citations · 2006
9. 9
   Bin Picking of Reflective Steel Parts Using a Dual-Resolution Convolutional Neural Network Trained in a Simulated Environment

   9 citations · 2018

Key Collaborators

Contact & Links