Image-based high-throughput phenotyping of agri-photonics
In the past decade, the development of phenotypic detection was greatly promoted by advanced photonics-based technologies. A variety of imaging techniques, including visible light imaging, hyperspectral imaging, structured light, X-ray computed tomography, have been applied in the case of rice, maize, rape, cotton and grapevine. Phenotypic data were extracted from crop images using specialized algorithm, which generally adopted classical image processing and machine learning methods. Traditional phenotyping that depends largely on manual measuring were tend to be replaced by automatic, non-destructive image-based phenotyping, bringing the functional analysis of crop genome into a high-throughput stage.
File with abstract
Huazhong University of Science and Technology, HHainan Universityuazhong Agricultural University,