水稻是亞洲國家的重要農作物和栽培品種之一。在台灣，幾乎一半的耕地都用於種植水稻。水稻的生命週期可以分為幾個階段：營養期、生殖期和成熟期。這三個主要階段可以分為更詳細的階段。但是，階段之間的轉換很難被確定，且需要經驗的累積。因此，即使提供了SOP，對於沒有經驗的人來說，水稻種植也是一項挑戰。此外，高齡化和缺乏勞動力的問題對農業產生了極大的影響。此外，近年來智慧農業發展迅速，並在許多方面改善了農業行業。為了降低種植難度，以及幫助新農戶更好地理解作物，我們實作了基於隨機森林（RF）的水稻生長階
段分期模型。我們在實驗場域安裝了一些物聯網(IOT) 設備來收集數據和圖像。本論文采用影像處理技術得到稻田高度和冠蓋率（CC）或綠覆蓋率（GC）。這些數據是我們模型的輸入之一。我們使用與生長相關的因素，如水稻株高、CC、溫度和生育天數（DAT）來開發我們的模型。生長階段數據標註經由專家的協助進行標記。我們的最優模型達到了0.98772 的accuracy 和0.98653 的macro F1-score。因此，我們的模型擁有高性能的準確性，可以在現實世界中應用。

Rice is one of the significant crops cultivated in Asian countries. In Taiwan, almost half of the arable land is used for cropping rice. The life cycle of paddy rice can be divided into several stages: vegetative stage, reproductive stage, and ripening stage. These three main stages can be divided into more detailed stages. However, it requires experiences to determine the transitions between stages. Thus, rice cultivation is challenging for inexperienced, even with SOP provided. Additionally, aging and labor issues have had an impact on agriculture. Furthermore, smart farming has been growing rapidly recent years and has improved the agriculture in many ways. To lower the entry requirements and help novices better understand, we proposed a random forest (RF)-based machine learning (ML) classification model for rice growth stages. We have also installed some IoT devices in the experiment fields to collect data and images. Different image processing technique are used to get paddy’s height and canopy cover
(CC) or green coverage (GC). These values are one of the inputs for our purposed model. Furthermore, we used growth-related factors such as height, CC, accumulative temperature, and DAT to develop our model. Stages data are consulted with agronomist for labeling. Our optimal model has achieved an accuracy of 0.98772, and macro F1-score of 0.98653. Thus, the developed model produces high-performance accuracy and can be employed in real-world scenarios.

[1] Y. Hsing, “Rice in taiwan,” Encyclopaedia of the history of science, technology,
and medicine in non-western cultures, pp. 3769–3772, 2016.
[2] 鄭清煥, 稻米生長. [Online]. Available: https : / / www . baphiq . gov . tw /
Publish/plant_protect_pic_8/ricePDF/01-0.pdf.
[3] F. Ramadhani, R. Pullanagari, G. Kereszturi, and J. Procter, “Mapping a
cloud-free rice growth stages using the integration of proba-v and sentinel-1
and its temporal correlation with sub-district statistics,” Remote Sensing,
vol. 13, p. 1498, Apr. 2021. doi: 10.3390/rs13081498.
[4] 產銷履歷農產品生產過程臺灣良好農業規範（tgap）-米類-水稻, Jan. 2021.
[Online]. Available: https : / / www . afa . gov . tw / cht / index . php ? act =
download&ids=117668.
[5] R. U. Onyeneke, M. U. Amadi, C. L. Njoku, and E. E. Osuji, “Climate
change perception and uptake of climate-smart agriculture in rice production
in ebonyi state, nigeria,” Atmosphere, vol. 12, no. 11, 2021, issn: 2073-4433.
doi: 10.3390/atmos12111503. [Online]. Available: https://www.mdpi.
com/2073-4433/12/11/1503.
[6] R. Alfred, J. H. Obit, C. P.-Y. Chin, H. Haviluddin, and Y. Lim, “Towards
paddy rice smart farming: A review on big data, machine learning, and
rice production tasks,” IEEE Access, vol. 9, pp. 50 358–50 380, 2021. doi:
10.1109/ACCESS.2021.3069449.
[7] A. K. Thakur and N. T. Uphoff, “How the system of rice intensification
can contribute to climate-smart agriculture,” Agronomy Journal, vol. 109,
no. 4, pp. 1163–1182, 2017. doi: https://doi.org/10.2134/agronj2016.
03.0162. eprint: https://acsess.onlinelibrary.wiley.com/doi/pdf/
10 . 2134 / agronj2016 . 03 . 0162. [Online]. Available: https : / / acsess .
onlinelibrary.wiley.com/doi/abs/10.2134/agronj2016.03.0162.
[8] M. Muthusinghe, P. S.T., W. Weerakkody, A. H. Saranga, and W. Rankothge,
“Towards smart farming: Accurate prediction of paddy harvest and rice demand,”
in 2018 IEEE Region 10 Humanitarian Technology Conference (R10-
HTC), 2018, pp. 1–6. doi: 10.1109/R10-HTC.2018.8629843.
[9] R. Wang, F. Han, and W. Wu, “Estimation of paddy rice maturity using
digital imaging,” International Journal of Food Properties, vol. 24, no. 1,
pp. 1403–1415, 2021. doi: 10 . 1080 / 10942912 . 2021 . 1970581. eprint:
https : / / doi . org / 10 . 1080 / 10942912 . 2021 . 1970581. [Online]. Available:
https://doi.org/10.1080/10942912.2021.1970581.
[10] F. Vesali, M. Omid, A. Kaleita, and H. Mobli, “Development of an android
app to estimate chlorophyll content of corn leaves based on contact imaging,”
Computers and Electronics in Agriculture, vol. 116, pp. 211–220, 2015.
[11] C. L. Haw, W. I. W. Ismail, S. Kairunniza-Bejo, A. Putih, and R. Shamshiri,
“Colour vision to determine paddy maturity,” International Journal of Agricultural
and Biological Engineering, vol. 7, no. 5, pp. 55–63, 2014.
[12] K.-J. Lee and B.-W. Lee, “Estimating canopy cover from color digital camera
image of rice field,” Journal of Crop Science and Biotechnology, vol. 14,
pp. 151–155, Jun. 2011. doi: 10.1007/s12892-011-0029-z.
[13] ——, “Estimation of rice growth and nitrogen nutrition status using color
digital camera image analysis,” European Journal of Agronomy, vol. 48,
pp. 57–65, 2013, issn: 1161-0301. doi: https : / / doi . org / 10 . 1016 / j .
eja.2013.02.011. [Online]. Available: https://www.sciencedirect.com/
science/article/pii/S1161030113000294.
[14] Y. Zhang, D. Xiao, and Y. Liu, “Automatic identification algorithm of the
rice tiller period based on pca and svm,” IEEE Access, vol. 9, pp. 86 843–
86 854, 2021. doi: 10.1109/ACCESS.2021.3089670.
[15] Marsujitullah, Z. Zainuddin, S. Manjang, and A. S. Wijaya, “Rice farming
age detection use drone based on svm histogram image classification,” Journal
of Physics: Conference Series, vol. 1198, p. 092 001, Apr. 2019. doi:
10.1088/1742-6596/1198/9/092001.
[16] M. Hearst, S. Dumais, E. Osuna, J. Platt, and B. Scholkopf, “Support vector
machines,” IEEE Intelligent Systems and their Applications, vol. 13, no. 4,
pp. 18–28, 1998. doi: 10.1109/5254.708428.
[17] I. H. Ikasari, V. Ayumi, M. I. Fanany, and S. Mulyono, “Multiple regularizations
deep learning for paddy growth stages classification from landsat-8,”
in 2016 International Conference on Advanced Computer Science and Information
Systems (ICACSIS), 2016, pp. 512–517. doi: 10.1109/ICACSIS.
2016.7872790.[18] K. Murata, A. Ito, Y. Takahashi, and H. Hatano, “A study on growth stage
classification of paddy rice by cnn using ndvi images,” in 2019 Cybersecurity
and Cyberforensics Conference (CCC), 2019, pp. 85–90. doi: 10.1109/CCC.
2019.000-4.
[19] キヤノン、カメラで￿の生育￿況をデ￿タ化　￿証￿￿, Aug. 2022. [Online].
Available: shorturl.at/gqwY7.
[20] 湯皓茹, Canon 轉型跨足智慧農業！可幫水田自動拍照、上雲、數據化管理,
Aug. 2022. [Online]. Available: https://www.inside.com.tw/article/
28461-canon-smart-agriculture-gm-1.
[21] W. Abdulla, Splash of color: Instance segmentation with mask r-cnn and tensorflow,
Dec. 2018. [Online]. Available: https://engineering.matterport.
com/splash- of- color- instance- segmentation- with- mask- r- cnnand-
tensorflow-7c761e238b46.
[22] K. He, G. Gkioxari, P. Dollár, and R. Girshick, Mask r-cnn, 2017. doi:
10.48550/ARXIV.1703.06870. [Online]. Available: https://arxiv.org/
abs/1703.06870.
[23] M. Sewell, “Ensemble learning,” RN, vol. 11, no. 02, pp. 1–34, 2008.
[24] T. K. Ho, “Random decision forests,” in Proceedings of 3rd International
Conference on Document Analysis and Recognition, vol. 1, 1995, 278–282
vol.1. doi: 10.1109/ICDAR.1995.598994.
[25] J. R. Quinlan, “Induction of decision trees,” Machine learning, vol. 1, no. 1,
pp. 81–106, 1986.
[26] L. Rokach and O. Maimon, “Decision trees,” in Jan. 2005, vol. 6, pp. 165–
192. doi: 10.1007/0-387-25465-X_9.
[27] Y. Wu, A. Kirillov, F. Massa, W.-Y. Lo, and R. Girshick, Detectron2, https:
//github.com/facebookresearch/detectron2, 2019.
[28] S. Xie, R. Girshick, P. Dollár, Z. Tu, and K. He, “Aggregated residual
transformations for deep neural networks,” arXiv preprint arXiv:1611.05431,
2016.
[29] Nthu smart farming platform. [Online]. Available: https://nthu- smartfarming.
kits.tw/.
[30] J. Brooks, COCO Annotator, https : / / github . com / jsbroks / coco -
annotator/, 2019.
[31] E. Bisong, “Google colaboratory,” in Building Machine Learning and Deep
Learning Models on Google Cloud Platform: A Comprehensive Guide for
Beginners. Berkeley, CA: Apress, 2019, pp. 59–64, isbn: 978-1-4842-4470-
8. doi: 10 . 1007 / 978 - 1 - 4842 - 4470 - 8 _ 7. [Online]. Available: https :
//doi.org/10.1007/978-1-4842-4470-8_7.
[32] G. Bradski, “The OpenCV Library,” Dr. Dobb’s Journal of Software Tools,
2000.
[33] Z. Zhang, “A flexible new technique for camera calibration,” IEEE Transactions
on Pattern Analysis and Machine Intelligence, vol. 22, no. 11, pp. 1330–
1334, 2000. doi: 10.1109/34.888718.
[34] Camera calibration. [Online]. Available: https://docs.opencv.org/4.x/
dc/dbb/tutorial_py_calibration.html.
[35] K. S. S. Mallick, K. Sadekar, and S. Mallick, Camera calibration using
opencv: Learnopencv, May 2021. [Online]. Available: https://learnopencv.
com/camera-calibration-using-opencv/.
[36] S. Mallick, Geometry of image formation: Learnopencv, Apr. 2021. [Online].
Available: https://learnopencv.com/geometry-of-image-formation/.
[37] Camera calibration with python - opencv, Jun. 2021. [Online]. Available:
https://www.geeksforgeeks.org/camera-calibration-with-pythonopencv/.
[38] F. Pedregosa, G. Varoquaux, A. Gramfort, et al., “Scikit-learn: Machine
learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–
2830, 2011.
[39] L. Jeni, J. Cohn, and F. De la Torre, “Facing imbalanced data - recommendations
for the use of performance metrics,” vol. 2013, Sep. 2013. doi:
10.1109/ACII.2013.47.
[40] T. Fawcett, “An introduction to roc analysis,” Pattern Recognition Letters,
vol. 27, no. 8, pp. 861–874, 2006, ROC Analysis in Pattern Recognition,
issn: 0167-8655. doi: https://doi.org/10.1016/j.patrec.2005.10.010.
[Online]. Available: https://www.sciencedirect.com/science/article/
pii/S016786550500303X.
[41] M. Majnik and Z. Bosnic, “Roc analysis of classifiers in machine learning:
A survey,” Intelligent Data Analysis, vol. 17, pp. 531–558, May 2013. doi:
10.3233/IDA-130592.
[42] J. Davis and M. Goadrich, “The relationship between precision-recall and
roc curves,” vol. 06, Jun. 2006. doi: 10.1145/1143844.1143874.
[43] G. Lemaître, F. Nogueira, and C. K. Aridas, “Imbalanced-learn: A python
toolbox to tackle the curse of imbalanced datasets in machine learning,”
Journal of Machine Learning Research, vol. 18, no. 17, pp. 1–5, 2017. [Online].
Available: http://jmlr.org/papers/v18/16-365.
[44] K. W. Bowyer, N. V. Chawla, L. O. Hall, and W. P. Kegelmeyer, “SMOTE:
synthetic minority over-sampling technique,” CoRR, vol. abs/1106.1813,
2011. arXiv: 1106 . 1813. [Online]. Available: http : / / arxiv . org / abs /
1106.1813.
[45] D. L. Wilson, “Asymptotic properties of nearest neighbor rules using edited
data,” IEEE Transactions on Systems, Man, and Cybernetics, vol. SMC-2,
no. 3, pp. 408–421, 1972. doi: 10.1109/TSMC.1972.4309137.
[46] T. Lu, Y. Huang, W. Zhao, and J. Zhang, “The metering automation system
based intrusion detection using random forest classifier with smote+enn,” in
2019 IEEE 7th International Conference on Computer Science and Network
Technology (ICCSNT), 2019, pp. 370–374. doi: 10 . 1109 / ICCSNT47585 .
2019.8962430.
[47] M. Lamari, N. Azizi, N. E. Hammami, et al., “Smote–enn-based data sampling
and improved dynamic ensemble selection for imbalanced medical data
classification,” in Advances on Smart and Soft Computing, F. Saeed, T. Al-
Hadhrami, F. Mohammed, and E. Mohammed, Eds., Singapore: Springer
Singapore, 2021, pp. 37–49, isbn: 978-981-15-6048-4.
[48] X. Zhang, J. Ran, and J. Mi, “An intrusion detection system based on convolutional
neural network for imbalanced network traffic,” in 2019 IEEE
7th International Conference on Computer Science and Network Technology
(ICCSNT), 2019, pp. 456–460. doi: 10.1109/ICCSNT47585.2019.8962490.
[49] V. N. G. Raju, K. P. Lakshmi, V. M. Jain, A. Kalidindi, and V. Padma,
“Study the influence of normalization/transformation process on the accuracy
of supervised classification,” in 2020 Third International Conference on
Smart Systems and Inventive Technology (ICSSIT), 2020, pp. 729–735. doi:
10.1109/ICSSIT48917.2020.9214160.
[50] M. M. Ahsan, M. A. P. Mahmud, P. K. Saha, K. D. Gupta, and Z. Siddique,
“Effect of data scaling methods on machine learning algorithms and model
performance,” Technologies, vol. 9, no. 3, 2021, issn: 2227-7080. doi: 10.
3390/technologies9030052. [Online]. Available: https://www.mdpi.com/
2227-7080/9/3/52.
[51] A. Mucherino, P. J. Papajorgji, and P. M. Pardalos, “K-nearest neighbor
classification,” in Data Mining in Agriculture. New York, NY: Springer New
York, 2009, pp. 83–106, isbn: 978-0-387-88615-2. doi: 10.1007/978-0-387-
88615-2_4. [Online]. Available: https://doi.org/10.1007/978-0-387-
88615-2_4.
[52] C. Cortes and V. Vapnik, “Support-vector networks,” Machine learning,
vol. 20, no. 3, pp. 273–297, 1995.
[53] B. E. Boser, I. M. Guyon, and V. N. Vapnik, “A training algorithm for
optimal margin classifiers,” in Proceedings of the Fifth Annual Workshop
on Computational Learning Theory, ser. COLT ’92, Pittsburgh, Pennsylvania,
USA: Association for Computing Machinery, 1992, pp. 144–152, isbn:
089791497X. doi: 10 . 1145 / 130385 . 130401. [Online]. Available: https :
//doi.org/10.1145/130385.130401.
[54] C.-C. Chang and C.-J. Lin, “LIBSVM: A library for support vector machines,”
ACM Transactions on Intelligent Systems and Technology, vol. 2,
27:1–27:27, 3 2011, Software available at http://www.csie.ntu.edu.tw/
~cjlin/libsvm.
[55] Y. Freund and R. E. Schapire, “Experiments with a new boosting algorithm,”
in ICML, 1996.
[56] D. Berrar, “Bayes’theorem and naive bayes classifier,” in Encyclopedia of
Bioinformatics and Computational Biology, S. Ranganathan, M. Gribskov,
K. Nakai, and C. Schönbach, Eds., Oxford: Academic Press, 2019, pp. 403–
412, isbn: 978-0-12-811432-2. doi: https://doi.org/10.1016/B978-0-
12-809633-8.20473-1. [Online]. Available: https://www.sciencedirect.
com/science/article/pii/B9780128096338204731.
[57] J. Peng, K. Lee, and G. Ingersoll, “An introduction to logistic regression
analysis and reporting,” Journal of Educational Research - J EDUC RES,
vol. 96, pp. 3–14, Sep. 2002. doi: 10.1080/00220670209598786.