Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Oct 28;12(1):18199.
doi: 10.1038/s41598-022-20799-z.

An integrated learning algorithm for early prediction of melon harvest

Chunyang Qian  1   2 , Taihang Du  3 , Shuguang Sun  3 , Wei Liu  4 , Haiguang Zheng  5 , Jianchun Wang  4
Affiliations

An integrated learning algorithm for early prediction of melon harvest

Chunyang Qian et al. Sci Rep. .

Abstract

Different modeling techniques must be applied to manage production and statistical estimation to predict the expected harvest. By calculating advanced production methods and the rational valuation of different factors, we can accurately capture the variety of growth characteristics and the expected yield. This paper obtained 32 feature variables related to melons, including phenological features, shape features, and color features. The Gradient Boosted Decision Tree (GBDT) network and the Grid Search (GS) hyperparameter seeking method was applied to calculate the degree of importance of all melon fruits' characteristics and construct prediction models for three expected harvest indexes of melon yield, sugar content, and endocarp hardness. To facilitate growers to carry out prediction and estimation in the field without destroying the melon fruits. The reduced feature variables were selected as inputs. The GBDT model was used to provide a significant advantage in prediction compared to both Random Forest (RF) and Support Vector Regression (SVR) methods. In addition, to verify the feasibility of using only reduced feature variables as input for the evaluation work, this study also compares the predictive effects of the model when all feature variables and only reduced feature variables are used. The GBDT prediction model proposed in this paper predicted melon yield, sugar content, and hardness using reduced features as input, and the model R2 could reach more than 90%. Therefore, this method can effectively help growers carry out early non-destructive inspection and growth prediction of melons in the field.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1
Melon fruit shape features.
Figure 2
Figure 2
Importance order of feature variables related to yield, sugar content, and hardness.
Figure 3
Figure 3
Importance of yield, sugar content, and hardness feature variables.
Figure 4
Figure 4
GBDT prediction effectiveness.
Figure 5
Figure 5
Comparison of yield (a), sugar content (b), and hardness (c) prediction results of different models with the reduced features as inputs.
Figure 6
Figure 6
Comparison of R2 indicators for different models before and after feature reduction.
See this image and copyright information in PMC

References

    1. Sun, L. X., Wang, X. J., Wu, J. X. & Mao, S. P. Research on regional distribution changes and driving factors of watermelon and melon production in China. Chin. J. Agric. Resour. Region. Plan. https://kns.cnki.net/kcms/detail/11.3513.S.20220705.1954.008.html. (2022).
    1. Wang K. High efficient cultivation techniques of muskmelon in winter and spring in solar. J. Agric. Catastrophol. 2021;11(3):1.
    1. Schwarzacher HH. Domestication, genomics, and the future for banana. Ann. Bot. 2007;100(5):1073–1084. doi: 10.1093/aob/mcm191. - DOI - PMC - PubMed
    1. Sun JT. Study on Nondestructive Dection of Hami Melon Maturity Based on Information Fusion of Spectrum and Image. Shihezi University; 2017.
    1. Lv C, et al. Research progress of melon ripeness detection technology. Jiangsu Agric. Sci. 2014;42(1):244–246.

Publication types