Watermelon seeds germination study by shortwave infrared-based hyperspectral imaging techniques
Session
Mechatronics, System Engineering and Robotics
Description
The germination capability of watermelon (Citrullus Lanatus) represents the proportion of seeds that produce healthy seedlings within the parameters and time frames specified by the approved techniques of analysis. Rapid and non-invasive analytical processes for seed quality control must be established in order to guarantee the quality and germination of seed production up until the point of use. In this scenario, Near Infrared (NIR)-based HyperSpectral Imaging (HSI) represents an optimal solution for quality control applications in different sectors. The purpose of this study was to explore the utilization of methods based on the use of HyperSpectral Imaging (HSI) in the Short-Wave Infrared (SWIR) spectral region (1000–2500 nm) to test the germinability of watermelon seeds. A Partial Least Squares–Discriminant Analysis (PLS-DA) model was built in order to recognize non-viable seeds from viable ones. The obtained classification results are promising, reaching a sensitivity of 0.72 and a specificity of 0.98 for not germinated seed.
Keywords:
Seeds; Germinability; Watermelon; HyperSpectral Imaging; Short-Wave Infrared; HIS; SWIR; Partial Least Squares–Discriminant Analysis.
Proceedings Editor
Edmond Hajrizi
ISBN
978-9951-550-50-5
Location
Lipjan, Kosovo
Start Date
29-10-2022 12:00 AM
End Date
30-10-2022 12:00 AM
DOI
10.33107/ubt-ic.2022.82
Recommended Citation
Bonifazi, Giuseppe; Serranti, Silvia; and Gasbarrone, Riccardo, "Watermelon seeds germination study by shortwave infrared-based hyperspectral imaging techniques" (2022). UBT International Conference. 82.
https://knowledgecenter.ubt-uni.net/conference/2022/all-events/82
Watermelon seeds germination study by shortwave infrared-based hyperspectral imaging techniques
Lipjan, Kosovo
The germination capability of watermelon (Citrullus Lanatus) represents the proportion of seeds that produce healthy seedlings within the parameters and time frames specified by the approved techniques of analysis. Rapid and non-invasive analytical processes for seed quality control must be established in order to guarantee the quality and germination of seed production up until the point of use. In this scenario, Near Infrared (NIR)-based HyperSpectral Imaging (HSI) represents an optimal solution for quality control applications in different sectors. The purpose of this study was to explore the utilization of methods based on the use of HyperSpectral Imaging (HSI) in the Short-Wave Infrared (SWIR) spectral region (1000–2500 nm) to test the germinability of watermelon seeds. A Partial Least Squares–Discriminant Analysis (PLS-DA) model was built in order to recognize non-viable seeds from viable ones. The obtained classification results are promising, reaching a sensitivity of 0.72 and a specificity of 0.98 for not germinated seed.