A PADDY GRAIN SIZE ANALYSIS SYSTEM FOR VARIETY BREEDING

Hà Quang Hưng; Phùng Trường Trinh; Lê Hoàng Vũ; Nguyễn Minh Quân; Nguyễn Thị Kim Cúc; Nguyễn Văn Quyết; Chu Đức Hà; Nguyễn Thị Hồng; Phạm Minh Triển

doi:10.34238/tnu-jst.12049

A PADDY GRAIN SIZE ANALYSIS SYSTEM FOR VARIETY BREEDING

About this article

Received: 14/02/25 Revised: 06/03/25 Published: 07/03/25

Authors

1. Ha Quang Hung, 1VNU University of Engineering and Technology
2. Phung Truong Trinh, 1VNU University of Engineering and Technology
3. Le Hoang Vu, 1VNU University of Engineering and Technology
4. Nguyen Minh Quan, 1VNU University of Engineering and Technology
5. Nguyen Thi Kim Cuc, 1VNU University of Engineering and Technology
6. Nguyen Van Quyet, 1VNU University of Engineering and Technology
7. Chu Duc Ha, 1VNU University of Engineering and Technology
8. Nguyen Thi Hong, 2Agricultural Genetics Institute - Vietnam Academy of Agricultural Sciences
9. Pham Minh Trien , 1VNU University of Engineering and Technology

Abstract

In rice breeding, the size and shape of rice grains are key indicators of yield potential; however, manually measuring the large number of grains from dozens to hundreds of hybrid pairs is very time-consuming, making the automation of this process essential. This study develops a grain size analysis system comprising three components: (1) an industrial computer; (2) a fixed image capture system; and (3) software running on the Linux operating system. Among the evaluated grain recognition models, including U-Net, U-Net++, ResNet, and YOLOv8, the YOLOv8 model achieved mAP50, mAP50:95, and an average Dice coefficient of 0.99, 0.91, and 0.98, respectively. Two extraction methods were applied to estimate grain size, with the approach that calculates width based on the perpendicular distance from the center to the nearest point yielding an MAE of 0.38. The average time the system takes to process an image is 3.4 seconds, and it remains stable for up to 500 images (16,489 grains). With its high performance and scalability, the system can be widely applied in research institutes, breeding companies, and training organizations, contributing to the automation of grain analysis and the enhancement of rice production quality.

Keywords

Paddy grain; Grain size; Breeding; Machine learning; Paddy grain Grain size Breeding Machine learning Automation

Full Text:

PDF (Tiếng Việt) PDF (Tiếng Việt)

References

[1] N. Childs and P. Jarrell, “Rice Outlook: January 2025,” USDA’s World Agricultural Outlook Board, Jan. 2025.

[2] M. S. Alam, M. A. Wazed, S. M. K. Hasan, M. Ahmed, M. S. Alam, and M. S. H. Sarker, “Quality aspects of paddy grain and seed dried in HSTU mobile grain and seed dryer integrated with a dual heating system,” Heliyon, vol. 10, no. 23, Dec. 2024, doi: 10.1016/j.heliyon.2024.e40835.

[3] R. Wimalasekera, “Role of seed quality in improving crop yields,” in Crop Production and Global Environmental Issues, Springer International Publishing, 2015, pp. 153–168, doi: 10.1007/978-3-319-23162-4_6.

[4] S. Chowdhury, M. Chowdhury, S. Bhattacherjee, and K. Ghosh, “Quality assessment of rice seed using different storage techniques,” Journal of the Bangladesh Agricultural University, vol. 12, no. 2, pp. 297-305, Jul. 2016, doi: 10.3329/jbau.v12i2.28688.

[5] D. Ren, C. Ding, and Q. Qian, “Molecular bases of rice grain size and quality for optimized productivity,” Sci Bull (Beijing), vol. 68, no. 3, pp. 314–350, Feb. 2023, doi: 10.1016/j.scib.2023.01.026.

[6] L. Yan et al., “Control of grain size and weight by the RNA-binding protein EOG1 in rice and wheat,” Cell Rep., vol. 43, no. 11, Nov. 2024, doi: 10.1016/j.celrep.2024.114856.

[7] K. Kiratiratanapruk et al., “Development of Paddy Rice Seed Classification Process using Machine Learning Techniques for Automatic Grading Machine,” J. Sens., vol. 2020, 2020, doi: 10.1155/2020/7041310.

[8] A. Feng et al., “Research on a rice counting algorithm based on an improved mcnn and a density map,” Entropy, vol. 23, no. 6, Jun. 2021, doi: 10.3390/e23060721.

[9] J. Peng, Z. Yang, D. Lv, and Z. Yuan, “A dynamic rice seed counting algorithm based on stack elimination,” Measurement, vol. 227, Mar. 2024, Art. no. 114275, doi: 10.1016/ j.measurement.2024.114275.

[10] O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” Springer, Cham, pp. 234-241, May 2015, doi: 10.1007/978-3-319-24574-4_28.

[11] K. Zhang, L. Zhang, and H. Pan, “UoloNet: based on multi-tasking enhanced small target medical segmentation model,” Artif. Intell. Rev., vol. 57, no. 2, Feb. 2024, doi: 10.1007/s10462-023-10671-5.

[12] S. Liu et al., “High-throughput measurement method for rice seedling based on improved UNet model,” Comput. Electron. Agric., vol. 219, Apr. 2024, doi: 10.1016/j.compag.2024.108770.

[13] Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, “UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation,” IEEE Trans Med Imaging, vol. 39, no. 6, pp. 1856–1867, Jun. 2020, doi: 10.1109/TMI.2019.2959609.

[14] H. Peng, J. Zhong, H. Liu, J. Li, M. Yao, and X. Zhang, “ResDense-focal-DeepLabV3+ enabled litchi branch semantic segmentation for robotic harvesting,” Comput. Electron. Agric., vol. 206, Mar. 2023, doi: 10.1016/j.compag.2023.107691.

[15] B. Yang et al., “FRPNet: An improved Faster-ResNet with PASPP for real-time semantic segmentation in the unstructured field scene,” Comput. Electron. Agric., vol. 217, Feb. 2024, doi: 10.1016/j.compag.2024.108623.

[16] R. Varghese and S. M., “YOLOv8: A Novel Object Detection Algorithm with Enhanced Performance and Robustness,” in 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS), IEEE, Apr. 2024, pp. 1–6, doi: 10.1109/ADICS58448.2024.10533619.

[17] E. Cervantes, J. J. Martín, and E. Saadaoui, “Updated Methods for Seed Shape Analysis,” Scientifica, vol. 2016, 2016, doi: 10.1155/2016/5691825.

DOI: https://doi.org/10.34238/tnu-jst.12049

Refbacks

There are currently no refbacks.



Remember me