No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Using AI Approaches for Predicting Tomato Growth in Hydroponic Systems
Abstract
In any hydroponic system, effective plant growth and yield prediction are important in precision farming. Improving model growth systems can improve plant growth conditions for enhanced crop production and better quality food and increase resource use efficiency (e.g. light use efficiency, fertiliser use efficiency, water use efficiency) that meets the market demand with lower costs. Recently, different Artificial Intelligence (AI) techniques include machine learning and, deep learning methods are employed for providing powerful analytical tools. The proposed research in this study uses different machine/deep learning techniques to predict yield and plant growth variation for tomato crop (Tiny Tim mini tomato) under three different experimental conditions. The three tomato plants were grown under 3 different light treatments. The number of yielded fruits is predicted depends on the environmental conditions in each light treatment. This research deploys deep learning techniques includes Bidirectional Long short-term Memory (Bi-LSTM) with an attention mechanism for punctuation restoration and a standard Long Short-Term Memory (LSTM) recurrent neuron network. As well as, some other machine learning techniques including Support Vector Machine (SVM) and Random Forest (RF) for predicting the tomato yield. In the three different treatments, the tomato fruits yield growth and number yielded fruits values are used by the employed AI techniques to model the targeted growth parameters. The comparative results obtained from each technique are presented and discussed utilising the cross-validation process, to evaluate the performance achieved by the different methods. Very promising results, based on the data generated from the tomato growth and development with three light treatments are presented. The results that have been achieved are () and () achieved by employing the Bi-LSTM and LSTM algorithms, respectively.
... In this section, some related literatures are examined. Amongst those literatures is the study of Mohmed et al. [10]. In their study, Artificial Intelligence was used to forecast tomato growth in hydroponic systems. ...
... Currently, most CSG climate solutions are controlled manually based on experience, resulting in poor performance on greenhouse production. Many research works have been conducted to analyse and predict plant growth performance using many different Artificial Intelligence (AI) approaches to predict the environmental conditions, mainly ambient temperature 13 . Hence, there is currently no data model for lettuce growth in CSGs which incorporates climate management control including the effects of air temperature, humidity, CO 2 concentration, and radiation under CSG scenarios, where environmental conditions are usually not within the optimal range for crop growth and development due to the limited climate control ability. ...
Understanding how plants respond to environmental conditions such as temperature, CO2, humidity, and light radiation is essential for plant growth. This paper proposes an Artificial Neural Network (ANN) model to predict plant response to environmental conditions to enhance crop production systems that improve plant performance and resource use efficiency (e.g. light, fertiliser and water) in a Chinese Solar Greenhouse. Comprehensive data collection has been conducted in a greenhouse environment to validate the proposed prediction model. Specifically, the data has been collected from the CSG in warm and cold weather. This paper confirms that CSG’s passive insulation and heating system was effective in providing adequate protection during the winter. In particular, the CSG average indoor temperature was 18 ∘C higher than the outdoor temperature. The difference in environmental conditions led to a yield of 320.8g per head in the winter after 60 growing days compared to 258.9g in the spring experiment after just 35 days. Three different architectures of Bayesian Neural Networks (BNN) models have been evaluated to predict plant response to environmental conditions. The results show that the BNN network is accurate in modelling and predicting crop performance.
... The data can be transmitted through a wireless solution, and then be presented in a cloudhosted monitoring and control platform that offers real-time data, remote management of robotics (to ensure optimum conditions are maintained), an enduser dashboard and alerts. All data can be stored in the wireless sensor and actuator networks, to be utilised by data scientists/vertical farming specialists to develop an algorithm including deep learning and AI approaches to calculate optimum growing recipes for nutritionally dense crops and ensure optimum energy/ resource efficiency (Kozai et al., 2018;Mohmed et al., 2022). An automated system would be an excellent alternative for indoor vertical farming in combination with real-time data from an IoT system allowing the enhancement of cost-efficacy monitoring and control systems and the reduction of human interaction, along with smartphone applications (Adhau et al., 2018). ...
Global food security has been significantly threatened by the Covid-19 pandemic and several prolonged challenges such as climate change, population increases, shortage of natural resources, energy crisis, and rapid urbanisation worldwide. Although numerous attempts have been made to secure resilience in the food system, many countries are suffering from hunger and malnutrition, particularly in African and some Asian countries. This review paper presents one of the sustainable farming practices – vertical farming that could play a key role in mitigating global food security in the current uncertain world. It addresses the recent development of vertical farming with advanced precision monitoring and controlling system by the Internet of Things (IoT) applications. It also provides information about the opportunities and challenges of vertical-urban agriculture and how urban agriculture meets economic, social and educational needs.
The commonly used greenhouse crop yield prediction models today have their specific application scenarios, which may not ensure the accuracy of the results if the greenhouse environment changes. This greatly restricts their use in the greenhouse environment. To solve this problem, two widely used tomato growth models were compared in the study: TOMGRO and Vanthoor, and then an integrated model was obtained. Through the extended Fourier amplitude sensitivity test (EFAST), the model parameters were divided into three categories: optimized, fixed and ignored. In addition, Bayesian optimization was used as an optimization algorithm, through which the parameters applicable to the greenhouse can be optimized based on the greenhouse data. Compared with TOMGRO and Vanthoor, the output of the integrated model was more reasonable and universal, and the RMSE in the integrated model was 2.5974 while that in TOMGRO and Vanthoor both were over 17, reflecting the fact that the model output was closer to the actual value. According to the verification results of four-year greenhouse data, the model had high performance in predicting yield.
Effective plant growth and yield prediction is an essential task for greenhouse growers and for agriculture in general. Developing models which can effectively modelgrowth and yield can help growers improve the environmental control for better production, match supply and market demand and lower costs. Recent developmentsin Machine Learning (ML) and, in particular, Deep Learning (DL) can provide powerfulnew analytical tools. The proposed study utilises ML and DL techniques to predict yield and plant growth variation across two different scenarios, tomato yield forecasting and Ficus benjamina stem growth, in controlled greenhouse environments. We deploy anew deep recurrent neural network (RNN), using the Long Short-Term Memory (LSTM) neuron model, in the prediction formulations. Both the former yield, growth and stem diameter values, as well as the microclimate conditions, are used by the RNN architecture to model the targeted growth parameters. A comparative study ispresented, using ML methods, such as support vector regression and random forest regression, utilising the mean square error criterion, in order to evaluate the performance achieved by the different methods. Very promising results, based on data that have been obtained from two greenhouses, in Belgium and the UK, in the framework of the EU Interreg SMARTGREEN project (2017-2021), are presented.
Biomass and yield are key variables for assessing the production and performance of agricultural systems. Modeling and predicting the biomass and yield of individual plants at the farm scale represents a major challenge in precision agriculture, particularly when salinity and other abiotic stresses may play a role. Here, we evaluate a diversity panel of the wild tomato species (Solanum pimpinellifolium) through both field and unmanned aerial vehicle (UAV)-based phenotyping of 600 control and 600 salt-treated plants. The study objective was to predict fresh shoot mass, tomato fruit numbers, and yield mass at harvest based on a range of variables derived from the UAV imagery. UAV-based red–green–blue (RGB) imageries collected 1, 2, 4, 6, 7, and 8 weeks before harvest were also used to determine if prediction accuracies varied between control and salt-treated plants. Multispectral UAV-based imagery was also collected 1 and 2 weeks prior to harvest to further explore predictive insights. In order to estimate the end of season biomass and yield, a random forest machine learning approach was implemented using UAV-imagery-derived predictors as input variables. Shape features derived from the UAV, such as plant area, border length, width, and length, were found to have the highest importance in the predictions, followed by vegetation indices and the entropy texture measure. The multispectral UAV imagery collected 2 weeks prior to harvest produced the highest explained variances for fresh shoot mass (87.95%), fruit numbers (63.88%), and yield mass per plant (66.51%). The RGB UAV imagery produced very similar results to those of the multispectral UAV dataset, with the explained variance reducing as a function of increasing time to harvest. The results showed that predicting the yield of salt-stressed plants produced higher accuracies when the models excluded control plants, whereas predicting the yield of control plants was not affected by the inclusion of salt-stressed plants within the models. This research demonstrates that it is possible to predict the average biomass and yield up to 8 weeks prior to harvest within 4.23% of field-based measurements and up to 4 weeks prior to harvest at the individual plant level. Results from this work may be useful in providing guidance for yield forecasting of healthy and salt-stressed tomato plants, which in turn may inform growing practices, logistical planning, and sales operations.
Crop yield is a highly complex trait determined by multiple factors such as genotype, environment, and their interactions. Accurate yield prediction requires fundamental understanding of the functional relationship between yield and these interactive factors, and to reveal such relationship requires both comprehensive datasets and powerful algorithms. In the 2018 Syngenta Crop Challenge, Syngenta released several large datasets that recorded the genotype and yield performances of 2,267 maize hybrids planted in 2,247 locations between 2008 and 2016 and asked participants to predict the yield performance in 2017. As one of the winning teams, we designed a deep neural network (DNN) approach that took advantage of state-of-the-art modeling and solution techniques. Our model was found to have a superior prediction accuracy, with a root-mean-square-error (RMSE) being 12% of the average yield and 50% of the standard deviation for the validation dataset using predicted weather data. With perfect weather data, the RMSE would be reduced to 11% of the average yield and 46% of the standard deviation. We also performed feature selection based on the trained DNN model, which successfully decreased the dimension of the input space without significant drop in the prediction accuracy. Our computational results suggested that this model significantly outperformed other popular methods such as Lasso, shallow neural networks (SNN), and regression tree (RT). The results also revealed that environmental factors had a greater effect on the crop yield than genotype.
Human activity recognition (HAR) has become a popular topic in research because of its wide application. With the development of deep learning, new ideas have appeared to address HAR problems. Here, a deep network architecture using residual bidirectional long short-term memory (LSTM) is proposed. The advantages of the new network include that a bidirectional connection can concatenate the positive time direction (forward state) and the negative time direction (backward state). Second, residual connections between stacked cells act as shortcut for gradients, effectively avoiding the gradient vanishing problem. Generally, the proposed network shows improvements on both the temporal (using bidirectional cells) and the spatial (residual connections stacked) dimensions, aiming to enhance the recognition rate. When testing with the Opportunity dataset and the public domain UCI dataset, the accuracy is significantly improved compared with previous results.
Genomic selection is revolutionizing plant breeding. However, still lacking are better statistical models for ordinal phenotypes to improve the accuracy of the selection of candidate genotypes. For this reason, in this paper we explore the genomic based prediction performance of two popular machine learning methods: the Multi Layer Perceptron (MLP) and support vector machine (SVM) methods vs. the Bayesian threshold genomic best linear unbiased prediction (TGBLUP) model. We used the percentage of cases correctly classified (PCCC) as a metric to measure the prediction performance, and seven real data sets to evaluate the prediction accuracy, and found that the best predictions (in four out of the seven data sets) in terms of PCCC occurred under the TGLBUP model, while the worst occurred under the SVM method. Also, in general we found no statistical differences between using 1, 2 and 3 layers under the MLP models, which means that many times the conventional neuronal network model with only one layer is enough. However, although even that the TGBLUP model was better, we found that the predictions of MLP and SVM were very competitive with the advantage that the SVM was the most efficient in terms of the computational time required.
Global climate change imposes increasing impacts on our environments and crop production. To decipher environmental impacts on crop plants, the concept “envirotyping” is proposed, as a third “typing” technology, complementing with genotyping and phenotyping. Environmental factors can be collected through multiple environmental trials, geographic and soil information systems, measurement of soil and canopy properties, and evaluation of companion organisms. Envirotyping contributes to crop modeling and phenotype prediction through its functional components, including genotype-by-environment interaction (GEI), genes responsive to environmental signals, biotic and abiotic stresses, and integrative phenotyping. Envirotyping, driven by information and support systems, has a wide range of applications, including environmental characterization, GEI analysis, phenotype prediction, near-iso-environment construction, agronomic genomics, precision agriculture and breeding, and development of a four-dimensional profile of crop science involving genotype (G), phenotype (P), envirotype (E) and time (T) (developmental stage). In the future, envirotyping needs to zoom into specific experimental plots and individual plants, along with the development of high-throughput and precision envirotyping platforms, to integrate genotypic, phenotypic and envirotypic information for establishing a high-efficient precision breeding and sustainable crop production system based on deciphered environmental impacts.
Continuing population and consumption growth will mean that the global demand for food will increase for at least another 40 years. Growing competition for land, water, and energy, in addition to the overexploitation of fisheries, will affect our ability to produce food, as will the urgent requirement to reduce the impact of the food system on the environment. The effects of climate change are a further threat. But the world can produce more food and can ensure that it is used more efficiently and equitably. A multifaceted and linked global strategy is needed to ensure sustainable and equitable food security, different components of which are explored here.
There has been an increased interest in the development of models to identify and predict human activities. However, the sparsity of the data gathered from the sensory devices in an ambient living environment creates the challenge of representing activities accurately. Also, such data usually comprise arbitrary lengths of dimensions. Recurrent Neural Networks (RNNs) are one of the widely used algorithms in sequential modelling due to their ability to handle the arbitrary lengths of data. In an attempt to address the above challenges, this paper proposes a method of fuzzy feature representation with Bidirectional Long Short-Term Memory (Bi-LSTM) for human activities modelling and recognition. To obtain optimal feature representation, sensor data are fuzzified and the membership degrees represent the selected features which are then applied to the Bi-LSTM model for activity modelling and recognition. The learning capability of the Bi-LSTM allows the model to learn the temporal relationship in sequential data which is used to identify human activities pattern. The learned pattern is then utilised in the prediction of further activities. The proposed method is tested and evaluated using dataset representing Activity of Daily Living (ADL) for a single user in a smart home environment. The obtained results are also compared with existing approaches that are used for modelling and recognising human activities.
A challenging key aspect of recognising and modelling human activity is to design a model that can deal with the uncertainty in human behaviour. Several machine learning and deep learning techniques are employed to model the Activity of Daily Living (ADL). This paper provides a new method based on Fuzzy Finite State Machine (FFSM) and Long Short-Term Memory (LSTM) neural network for modelling and recognising human activities. The learning capability in the LSTM allows the system to learn the relations in the temporal data to identify the parameters of the rule-based system through time steps in the learning mode. The learned parameters are then used for generating the fuzzy rules that govern the transitions between the system's states. Experimental results are presented to demonstrate the effectiveness of the proposed approach.
A field experiment was conducted at Geregera (11° 45’ N latitude and 38° 45’ E longitude) in north Wollo, Ethiopia during the 2003 cropping season with the objectives of estimating genotypic variability, heritability, genetic advance and associations among yield and yield-related characters of durum wheat. Forty four randomly taken Ethiopian durum wheat accessions were evaluated using randomized complete block design in three replications. Analysis of variance showed significant differences (p < 0.05) among the durum wheat accessions for all the characters considered. Genotypic coefficients of variation (GCV %) were medium for spike length (12.5%) and kernel yield plant-1 (12.3%). Broad sense heritability were high for spike length (89.2%), plant height (87.1%) and thousand kernels weight (80.2%), indicating that these characters were predominantly controlled by genetic factors. Maximum heritability in broad sense coupled with high genetic advance were exhibited for spike length, plant height and thousand kernels weight, implying that phenotypic selection could identify superior genotypes for these traits. Genotypic correlation analysis revealed that grain yield had strong positive associations (p < 0.01) with kernel yield plant-1 (rg = 0.89), plant height (rg= 0.84), thousand kernels weight (rg = 0.82), biomass yield (rg = 0.80) and number of kernels spike-1(rg = 0.78), where biomass yield and harvest index exerted maximum positive direct effects on grain yield. On the other hand, days to heading had a significant negative correlation (p < 0.01) with grain yield (rg = -0.75) which exerted maximum negative direct effect on grain yield. Therefore, earliness coupled with high biomass yield could be considered as an indirect selection criterion for durum wheat grain yield improvement in moisture stressed areas.
Keywords: Genetic; Genetic Advance; Harvest Index; Heritability; Variability
Random Forests are considered for classification of multisource remote sensing and geographic data. Various ensemble classification methods have been proposed in recent years. These methods have been proven to improve classification accuracy considerably. The most widely used ensemble methods are boosting and bagging. Boosting is based on sample re-weighting but bagging uses bootstrapping. The Random Forest classifier uses bagging, or bootstrap aggregating, to form an ensemble of classification and regression tree (CART)-like classifiers. In addition, it searches only a random subset of the variables for a split at each CART node, in order to minimize the correlation between the classifiers in the ensemble. This method is not sensitive to noise or overtraining, as the resampling is not based on weighting. Furthermore, it is computationally much lighter than methods based on boosting and somewhat lighter than simple bagging. In the paper, the use of the Random Forest classifier for land cover classification is explored. We compare the accuracy of the Random Forest classifier to other better-known ensemble methods on multisource remote sensing and geographic data.