Check out this new paper that Jordan contributed to, led by Song-Quan Ong (Aarhus University and Universiti Malaysia Sabah): ScannerVision: Scanner-based image acquisition of medically important arthropods for the development of computer vision and deep learning models
The paper demonstrates how flatbed CCD scanners can be used to digitise insect traps and how these digital images can be used with computer vision and deep learning to identify whole communities. The results are comparable to the relatively laborious and poorly reproducible process of imaging insects using cameras and microscopes. This should provide a significant boost to the scalability and reproducibility of insect biomonitoring. We are currently applying these ideas to everything from marine biofouling to soil communities, so we should have more exciting things to come in this space!


One thought on “NEW PAPER: ScannerVision: Scanner-based image acquisition of medically important arthropods for the development of computer vision and deep learning models”