Normalized spectral indices of seed coat as a predictor of germination in shelterbelt crops for adaptive farming technology

In adaptive farming, focused on stress tolerance and resource efficiency, pre-sowing seed quality assessment of shelterbelt crops is critical. The study evaluates the predictive potential of normalized spectral indices of the seed coat (epidermis) in the RGB space for forecasting laboratory-container germination. The research uses as an example a breeding form of Scots pine (Pinus sylvestris L. cv. Negorelskaya), which is used in protective afforestation. Seven normalized indices minimizing the influence of overall brightness were analyzed: [(R-G-B)/(R+G)]², [(R-B)/(R+B)]², [(G-B)/(G+B)]², [(G-R)/(G+R)]², [(R+G-B)/(R+G+B)]², [(R+B-G)/(R+G+B)]², [(G+B-R)/(R+G+B)]². Scanner images were obtained for more than 1 000 individual seeds, and container germination was assessed on the 50th day. Statistically significant differences (p<0.0001-0.0165) in the distributions of all indices were established between zero- and non-zero-germination groups. Stable spectral patterns associated with low germination potential were identified: increased relative reflectance in the red and green channels compared to the blue channel and decreased reflectance in the blue channel relative to the red and green ones. The findings demonstrate the possibility of integrating simple, low-cost spectral markers based on RGB scanning into seed technology passports. This approach enables non-destructive selection of viable seed material with predictable properties. It is a key element of the “Target Plant” concept for establishing adaptive, stress-tolerant forest shelterbelts.

1. Zhuchenko A.A. Adaptive plant growing. Moscow, Russia: Kolos, 2009:520. (In Russ.)

2. Nikolaev M.V. The impact of climate change on crop farming in the drained lands of the European Nonchernozem region of Russia: vulnerability and adaptation assessment. Agricultural Biology. 2023;58(1):60-74. (In Russ.) https://doi.org/10.15389/agrobiology.2023.1.60rus

3. Neufeld V.V., Kadomtseva M.E. Mechanisms of adaptation of crop production in the regions of the Volga Federal District to the consequences of global climate change. The Agrarian Science Journal. 2022;(4):37-43. (In Russ.) https://doi.org/10.28983/asj.y2022i4pp37-43

4. Ivonin M.V., Voskoboynikova I.V. Landscape agroforestry reclamation. Land Reclamation and Hydraulic Engineering. 2021;11(3):54-77. (In Russ.) https://doi.org/10.31774/2712-9357-2021-11-3-54-77

5. Primakov N.V., Tanyukevich V.V. Forestry measures in shelterbelts of the Krasnodar Territory. Russian Forestry Journal. 2024;(4(400)):185-192. (In Russ.) https://doi.org/10.37482/0536-1036-2024-4-185-192

6. Dubenok N.N., Kuzmichev V.V., Lebedev A.V. Analysis of ecological functions of birch and oak stands in conditions of an urbanized environment based on the materials of long-term observations. Rossiiskaia selskokhoziaistvennaia nauka. 2018;(5):29-31. (In Russ.) https://doi.org/10.31857/S250026270000632-0

7. Novikova T.P., Tylek P., Mastrangelo C.B., Drapalyuk M.V. et al. The root collar diameter growth reveals a strong relationship with the height growth of juvenile Scots pine trees from seeds differentiated by spectrometric feature. Forests. 2023;14(6):1164. https://doi.org/10.3390/f14061164

8. Novikova T.P., Novikov A.I., Petrishchev E.P. FLR-library reference information system for adaptive forest restoration: cluster analysis of descriptors. Forestry Engineering Journal. 2023;(3(51)):164-179. (In Russ.) https://doi.org/10.34220/issn.2222-7962/2023.3/12

9. ElMasry G., ElGamal R., Mandour N., Gou P. et al. Emerging thermal imaging techniques for seed quality evaluation: Principles and applications. Food Research International. 2020;131:109025. https://doi.org/10.1016/j.foodres.2020.109025

10. Xia Y., Xu Y., Li J., Zhang C. et al. Recent advances in emerging techniques for non-destructive detection of seed viability: A review. Artificial Intelligence in Agriculture. 2019;1: 35-47. https://doi.org/10.1016/j.aiia.2019.05.001

11. ElMasry G., Mandour N., Al-Rejaie S., Belin E. et al. Recent Applications of Multispectral Imaging in Seed Phenotyping and Quality Monitoring – An Overview. Sensors. 2019;19:1090. https://doi.org/10.3390/s19051090

12. ElMasry G., Mandour N., Wagner M.-H., Demilly D. et al. Utilization of computer vision and multispectral imaging techniques for classification of cowpea (Vigna unguiculata) seeds. Plant Methods. 2019;15:24. https://doi.org/10.1186/s13007-019-0411-2

13. Loddo A., Di Ruberto C. On the efficacy of handcrafted and deep features for seed image classification. Journal of Imaging. 2021;7:171. https://doi.org/10.3390/jimaging7090171

14. Bernardes R.K., De Medeiros A., da Silva L., Cantoni L. et al. Deep-Learning Approach for Fusarium Head Blight Detection in Wheat Seeds Using Low-Cost Imaging Technology. Agriculture. 2022;12:1801. https://doi.org/10.3390/agriculture12111801

15. Nehoshtan Y., Carmon E., Yaniv O. Robust seed germination prediction using deep learning and RGB image data. Scientific Reports. 2021;11:17126. https://doi.org/10.1038/s41598-021-01712-6

16. Reddy P., Panozzo J., Guthridge K.M. Single Seed Near-Infrared Hyperspectral Imaging for Classification of Perennial Ryegrass Seed. Sensors. 2023;23:1820. https://doi.org/10.3390/s23041820

17. Chen C., Bai M., Wang T. An RGB image dataset for seed germination prediction and vigor detection – maize. Frontiers in Plant Science. 2024;15:1341335. https://doi.org/10.3389/fpls.2024.1341335

18. Li J., Xu F., Song S. Hyperspectral RGB Imaging Combined with Deep Learning for Maize Seed Variety Identification. IEEE Access. 2024;12:83788-83800. https://doi.org/10.1109/access.2024.3419006

19. Tigabu M., Daneshvar A., Jingjing R., Wu P. et al. Multivariate discriminant analysis of single seed near infrared spectra for sorting dead-filled and viable seeds of three pine species: does one model fit all species? Forests. 2019;10:469. https://doi.org/10.3390/f10060469

20. Tigabu M., Daneshvar A., Wu P., Ma X. et al. Rapid and non-destructive evaluation of seed quality of Chinese fir by near infrared spectroscopy and multivariate discriminant analysis. New Forests. 2020;51:395-408. https://doi.org/10.1007/s11056-019-09735-8

21. Sokolov S.V., Kamensky V.V., Novikov A.I., Ivetić V. How to increase the analog-to-digital converter speed in optoelectronic systems of the seed quality rapid analyzer. Inventions. 2019;4(4):61. https://doi.org/10.3390/inventions4040061

22. Davis A.S., Pinto J.R. The Scientific Basis of the Target Plant Concept: An Overview. Forests. 2021;12:1293. https://doi.org/10.3390/f12091293