A recently published study has made significant advancements in the automated detection of internal seed defects using artificial intelligence combined with 2D X-ray imaging. This innovative approach employs advanced deep learning models to analyze 2D X-ray images with remarkable accuracy and time efficiency.
A key technological challenge in utilizing artificial intelligence is obtaining a large, balanced database of data. Certain classes of seed defects were underrepresented, which was further complicated by the lack of standardized radiography protocols, leading to variations in image quality, according to a press release.
To address these challenges, a strategy was developed based on artificial data augmentation (X-Robustifier) to account for the natural variability in seed morphology, fluctuations in X-ray imaging conditions, and the typically low occurrence of defective seeds. These data augmentation techniques enhance the robustness of the deep learning models, enabling them to adapt to variations in physical imaging parameters while effectively compensating for the rarity of defects. Some tests even demonstrated the model’s robustness and its ability to perform well on coated seeds or in 2D projections derived from 3D tomography images.
The findings reveal that this model matches human performance in both computation time and error rates, while also demonstrating strong robustness against common challenges such as physical noise and variations in morphology. The method, successfully applied to sugar beet (Beta vulgaris L.) and faba bean (Vicia faba L.) seeds, proves to be an effective and scalable solution for automated seed testing.
This project, conducted in collaboration with partners from Angers University and the ImHorPhen research group, aims to encourage further partnerships with organizations interested in applying artificial intelligence to seed and plant research, promoting innovation and collaboration in both research and technology.
The post Advancements in Seed Quality Evaluation Using Artificial intelligence and 2D X-Ray Imaging appeared first on Seed World.