Tomato Growth Height Prediction Method by Phenotypic Feature Extraction Using Multi-modal Data

doi:10.12133/j.smartag.SA202410032

Abstract

Abstract:

[Objective] Accurate prediction of tomato growth height is crucial for optimizing production environments in smart farming. However, current prediction methods predominantly rely on empirical, mechanistic, or learning-based models that utilize either images data or environmental data. These methods fail to fully leverage multi-modal data to capture the diverse aspects of plant growth comprehensively. [Methods] To address this limitation, a two-stage phenotypic feature extraction (PFE) model based on deep learning algorithm of recurrent neural network (RNN) and long short-term memory (LSTM) was developed. The model integrated environment and plant information to provide a holistic understanding of the growth process, emploied phenotypic and temporal feature extractors to comprehensively capture both types of features, enabled a deeper understanding of the interaction between tomato plants and their environment, ultimately leading to highly accurate predictions of growth height. [Results and Discussions] The experimental results showed the model's effectiveness: When predicting the next two days based on the past five days, the PFE-based RNN and LSTM models achieved mean absolute percentage error (MAPE) of 0.81% and 0.40%, respectively, which were significantly lower than the 8.00% MAPE of the large language model (LLM) and 6.72% MAPE of the Transformer-based model. In longer-term predictions, the 10-day prediction for 4 days ahead and the 30-day prediction for 12 days ahead, the PFE-RNN model continued to outperform the other two baseline models, with MAPE of 2.66% and 14.05%, respectively. [Conclusions] The proposed method, which leverages phenotypic-temporal collaboration, shows great potential for intelligent, data-driven management of tomato cultivation, making it a promising approach for enhancing the efficiency and precision of smart tomato planting management.

Key words: tomato growth prediction, deep learning, phenotypic feature extraction, multi-modal data, recurrent neural network, long short-term memory, large language model

CLC Number:

TP181

GONG Yu, WANG Ling, ZHAO Rongqiang, YOU Haibo, ZHOU Mo, LIU Jie. Tomato Growth Height Prediction Method by Phenotypic Feature Extraction Using Multi-modal Data[J]. Smart Agriculture, 2025, 7(1): 97-110.

Figures/Tables 21

Table 1

Table 2

Fig. 1

Fig. 2

Fig. 3

Table 3

Table 4

Table 5

Table 6

Fig. 4

Table 7

Fig. 5

Table 8

Fig. 6

Table 9

Table 10

Table 11

Fig. 7

Fig. 8

Fig. 9

Table 12

References 31

1	QUINET M, ANGOSTO T, YUSTE-LISBONA F J, et al. Tomato fruit development and metabolism[J]. Frontiers in plant science, 2019, 10: ID 1554.
2	PASLEY H, BROWN H, HOLZWORTH D, et al. How to build a crop model. A review[J]. Agronomy for sustainable development, 2022, 43(1): ID 2.
3	JONES J W, DAYAN E, ALLEN L H, et al. A dynamic tomato growth and yield model (tomgro)[J]. Transactions of the ASAE, 1991, 34(2): 663-672.
4	AMTHOR J S. After photosynthesis, what then: Importance of respiration to crop growth and yield[J]. Field crops research, 2025, 321: ID 109638.
5	HAN KUNLIN, WANG ZHAOYING, YANG HUIMIN, et al. Prediction model of tomato fruit diameter in greenhouse based on PCA-BPNN[J]. Xinjiang agricultural sciences, 2022, 59(2): 485-492.
6	ZHANG M, LI T, JI Y H, et al. Optimization of CO₂ enrichment strategy based on BPNN for tomato plants in greenhouse[J]. Transactions of the Chinese society for agricultural machinery, 2015, 46(8): 239-245.
7	CHEN B, OUYANG Z. Prediction of winter wheat evapotranspiration based on BP neural networks[J]. Transactions of the Chinese society of agricultural engineering, 2010, 26(4): 81-86.
8	HAN L, LI R, ZHU H L. Comprehensive evaluation model of soil nutrient based on BP neural network[J]. Transactions of the Chinese society for agricultural machinery, 2011, 42(7): 109-115.
9	TIAN X Q, ZHU X J. Loop control stategies of CO₂ concentration based on BPNN for Pleurotus eryngii in a factory farm[J]. Edible fungi of China, 2016, 35(2): 46-49, 53.
10	CHEN D J, SHI R L, PAPE J M, et al. Predicting plant biomass accumulation from image-derived parameters[J]. GigaScience, 2018, 7(2): ID giy001.
11	YANG H, HUANG C, LIU X R, et al. Convolutional neural networksed feature extraction for non-destructive lettuce growth monitoring[J]. Computer technology and development, 2023, 33(8): 137-143.
12	CHANDEL N S, CHAKRABORTY S K, RAJWADE Y A, et al. Identifying crop water stress using deep learning models[J]. Neural computing and applications, 2021, 33(10): 5353-5367.
13	YALCIN H. An approximation for A relative crop yield estimate from field images using deep learning[C]// 2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics). Piscataway, New Jersey, USA: IEEE, 2019.
14	BALI N S, SINGLA A. Deep learning based wheat crop yield prediction model in punjab region of north India[J]. Applied artificial intelligence, 2021, 35(15): 1304-1328.
15	NIGAM A, GARG S, AGRAWAL A, et al. Crop yield prediction using machine learning algorithms[C]// 2019 Fifth International Conference on Image Information Processing (ICIIP). Piscataway, New Jersey, USA: IEEE, 2019.
16	ELAVARASAN D, DURAIRAJ VINCENT P M. Crop yield prediction using deep reinforcement learning model for sustainable agrarian applications[J]. IEEE access, 2020, 8: 86886-86901.
17	DE ALWIS S, ZHANG Y S, NA M, et al. Duo attention with deep learning on tomato yield prediction and factor interpretation[M]// PRICAI 2019: Trends in Artificial Intelligence. Cham: Springer International Publishing, 2019: 704-715.
18	ZHOU X H, LIU Q Z, KATZIN D, et al. Boosting the prediction accuracy of a process-based greenhouse climate-tomato production model by particle filtering and deep learning[J]. Computers and electronics in agriculture, 2023, 211: ID 107980.
19	OIKONOMIDIS A, CATAL C, KASSAHUN A. Deep learning for crop yield prediction: A systematic literature review[J]. New Zealand journal of crop and horticultural science, 2023, 51(1): 1-26.
20	WANG P X, TIAN H R, ZHANG Y, et al. Crop growth monitoring and yield estimation based on deep learning: State of the art and beyond[J]. Transactions of the Chinese society for agricultural machinery, 2022, 53(2): 1-14.
21	GONG L Y, YU M, JIANG S Y, et al. Deep learning based prediction on greenhouse crop yield combined TCN and RNN[J]. Sensors, 2021, 21(13): ID 4537.
22	XING H M, XU X G, LI Z H, et al. Global sensitivity analysis of the AquaCrop model for winter wheat under different water treatments based on the extended Fourier amplitude sensitivity test[J]. Journal of integrative agriculture, 2017, 16(11): 2444-2458.
23	MAIMAITIJIANG M, SAGAN V, SIDIKE P, et al. Soybean yield prediction from UAV using multimodal data fusion and deep learning[J]. Remote sensing of environment, 2020, 237: ID 111599.
24	YANG Y, WEI X B, WANG J, et al. Prediction of seedling oilseed rape crop phenotype by drone-derived multimodal data[J]. Remote sensing, 2023, 15(16): ID 3951.
25	LIN Z X, FU R M, REN G Q, et al. Automatic monitoring of lettuce fresh weight by multi-modal fusion based deep learning[J]. Frontiers in plant science, 2022, 13: ID 980581.
26	TAN K. Large language models for crop yield prediction[EB/OL]. [2025-10-11].
27	ANAMI B S, MALVADE N N, PALAIAH S. Deep learning approach for recognition and classification of yield affecting paddy crop stresses using field images[J]. Artificial intelligence in agriculture, 2020, 4: 12-20.
28	JIANG H H, ZHANG C Y, QIAO Y L, et al. CNN feature based graph convolutional network for weed and crop recognition in smart farming[J]. Computers and electronics in agriculture, 2020, 174: ID 105450.
29	SHARMA R, DAS S, GOURISARIA M K, et al. A model for prediction of paddy crop disease using CNN[M]// Progress in Computing, Analytics and Networking. Singapore: Springer Singapore, 2020: 533-543.
30	TANG Q, LIANG J, ZHU F Q. A comparative review on multi-modal sensors fusion based on deep learning[J]. Signal processing, 2023, 213: ID 109165.
31	YU Y, SI X S, HU C H, et al. A review of recurrent neural networks: LSTM cells and network architectures[J]. Neural computation, 2019, 31(7): 1235-1270.

	KNO₃/g	NH₄H₂PO₄/g	KH₂PO₄/g	K₂SO₄/g	MgSO₄·7H₂O/g	CaCl₂/g	CaNO₃·4H₂O/g
CK	0.0	0.0	27.2	52.3	24.6	44.4	0.0
T₁	13.4	7.7	0.0	34.8	24.6	44.4	0.0
T₂	26.9	7.7	0.0	17.4	24.6	44.4	0.0
T₃	40.4	7.7	0.0	0.0	24.6	44.4	0.0
T₄	40.4	7.7	0.0	0.0	24.6	29.6	11.8
T₅	40.4	7.7	0.0	0.0	24.6	14.8	23.6
T₆	40.4	7.7	0.0	0.0	24.6	0.0	35.4

Molecular	Molecular weight standard concentration/（mg/L）
FeSO₄·7H₂O	13.206
EDTA·H₂O	17.679
EDTA·3H₂O	20.000
MnSO₄·H₂O	1.614
MnSO₄· 4H₂O	2.130
H₃BO₃	2.860
ZnSO₄·7H₂O	0.220
CuSO₄·5H₂O	0.080
Na₂MO₄·2H₂O	0.020

Network Structure	Parameters
Convolutional kernel	3×3
Convolutional stride	1
Convolutional padding	1
Pooling kernel	2×2
Pooling stride	2
Fully connected input	16×300×300
Fully connected output	16

Network structure	Parameters
Input features	34
RNN hidden size	30
RNN layers	2
Learning rate	0.001
Loss function	MSE_Loss
Optimizer	Adam

RNN network layers	Hidden layer size	Prediction scenario
		Short-term prediction		Mid-term prediction		Long-term prediction
		MSE	MAPE/%	MSE	MAPE/%	MSE	MAPE/%
2	15	6.89	2.16	55.21	6.03	113.17	6.74
	20	1.89	1.25	68.13	6.87	288.93	10.81
	25	2.11	1.11	56.51	5.41	297.91	10.85
	30	3.09	1.87	56.78	4.83	236.51	10.12
3	15	6.03	2.01	21.09	2.82	388.68	11.26
	20	1.52	1.07	27.53	4.11	180.16	7.14
	25	3.07	1.74	46.15	6.36	165.09	6.51
	30	2.49	1.50	60.96	6.71	203.26	8.26