Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 469 https://internationalpubls.com AI in Agriculture: Precision Farming and Crop Monitoring Dr. Kamatchi Sundravadivelu1, Shyamala P2, Dr. Santosh Kumar3, Dr. L. Kartheesan4, Chiranjit Dutta5 1Assistant Professor, Computer Science, Madurai Kamaraj University, Madurai, Tamil Nadu, Mail id: svadivelu2021@gmail.co 2Assistant Professor, Panimalar Engineering College, Poonamallee, Chennai 600 123. Email: pshyamala.aug@gmail.com 3Professor, Department of Computer Science, ERA University, Lucknow, Uttar Pradesh, Email id – dr.santoshkumarresearch@gmail.com 4Asst Professor(SG) Department of Computer Science and Engineering, Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology, Avadi, Chennai, Tamilnadu karthee.cse@gmail.com 5Designation: Associate Professor, Computer Science and Engineering, SRM Institute of Science and Technology, Delhi- NCR Campus, Ghaziabad, Uttar Pradesh. Email id: cse.chiranjit@gmail.com Article History: Received: 25-09-2024 Revised: 05-11-2024 Accepted: 25-11-2024 Abstract: This research delves into the application of artificial intelligence in precision farming and crop monitoring, focusing on four AI algorithms: “Support Vector Machine (SVM), Random Forest (RF), Convolutional Neural Networks (CNN), and K-Nearest Neighbors (KNN)”. The objective of improving crop disease detection accuracy and prediction efficiency, along with yield prediction and resource optimization, is targeted. The models were trained to predict crop health and optimize agricultural practices, using a dataset of crop images and environmental data. For the test result, CNN took the leads with an accuracy of 92.5% in disease detection, followed by RF with an accuracy of 89.3%, SVM with an accuracy of 86.7%, and KNN with an accuracy of 81.5%. Additionally, crop yield prediction using a hybrid AI model incorporating meteorological and soil data showed an R-squared value of 0.88, demonstrating strong prediction capabilities. The integration of AI with UAVs and remote sensing technologies allowed for real-time monitoring of crops, providing farmers with actionable insights to optimize resource use. These results demonstrate the possible significant impact of AI on facilitating sustainable farming practices through cost savings, reduced environmental impact, and improved productivity. In general, AI applications in agriculture will revolutionize precision farming by coming up with intelligent data-driven solutions for crop management. Keywords: Artificial Intelligence, Precision Farming, Crop Monitoring, Machine Learning, UAVs. I. INTRODUCTION Agriculture has been human civilization's foundation: it's where essential food and resources come from. However, as population increases worldwide, and with the progressing climatic changes, mankind cannot maintain farming methods as practiced hitherto. All this needs be filled with more efficient and sustainable agricultural practices. This is where Artificial Intelligence comes into action [1]. AI in agriculture, especially with regards to precision farming and crop monitoring, is challenging the way farmers operate on their farms. Precision farming, also referred to as site-specific crop management, makes use of AI technologies, such as machine learning, remote sensing, and data analytics, to collect enormous amounts of agricultural data [2]. This means farmers can make real- time, highly accurate decisions with regard to crop management, irrigation, fertilization, and pest control. Farmers therefore can optimize resource utilization, reduce waste, and increase crop mailto:cse.chiranjit@gmail.com Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 470 https://internationalpubls.com productivity while ensuring a minimum impact on the environment by making use of AI in their farm practices [3]. Crop monitoring through AI tools, such as drones, satellite imagery, and IoT sensors, allows for continuous and real-time observation of crop health and soil conditions. They can detect early signs of diseases, nutrient deficiencies, and pest infestations for timely intervention with targeted treatments, thereby drastically reducing chemical inputs and ensuring healthier and more resilient crops. This paper discusses how AI can play a transformational role in precision agriculture and crop monitoring. The research examines the ways in which AI will help increase agricultural productivity, sustainability, and profitability as well as reduce challenges such as resource scarcity and environmental sustainability. Through the analysis of the current applications and future potential of AI in agriculture, this study will shed light on the ways through which AI can contribute to the future of farming and food security. II. RELATED WORKS The application of artificial intelligence (AI) in precision agriculture has attracted much attention with several studies showing its potential to improve productivity, sustainability, and make better decisions. Various AI-based approaches, such as ML, deep learning, and image processing, have been pursued in addressing problems in crop prediction, disease detection, and efficient use of resources. González- Rodríguez et al. [15] claim that AI is to be promising to phytopathology, and it promises revolution for disease management in plants. Their work shows the utilization of AI-driven diagnostic tools aimed at distinguishing plant diseases and eliminating frequent uses of chemical treatments, thereby converting unbeneficial management of diseases into superior ones. This relies on other studies that apply machine learning algorithms for the monitoring of crop health. In crop prediction, Kumar and Arulanandam [19] developed a hybrid model to predict crop yield using environmental and meteorological data. The proposed model shows promise in terms of precision and gauging the crop yield more effectively than conventional methods. As a result, AI has the potential to predict agricultural outcomes. Similarly, Kumar and Rao [20] published an in-depth survey on the application of deep learning for crop disease analysis. It further indicates the application of deep learning techniques to identify the symptoms of a plant from images, which is an important topic and has an area of research and is almost at the activation level in real-time systems for disease detection. Drones and unmanned aerial vehicles have also played an important role in smart agriculture. Jayaram et al. [16] aimed at focusing on multirotor UAV design and performance analysis intended for precision farming, highlighting their potential in efficiently monitoring large agricultural areas. The authors in their study present the advantages of crop monitoring, pest detection, and soil health assessment UAVs. In a similar vein, Karami et al. [17] explore the application of quadcopters in smart agriculture, emphasizing their use in field mapping, pest monitoring, and resource management. Integration of AI with Internet of Things (IoT) technologies further advanced precision agriculture. Lee [21] discusses the potential for the integration of AI and IoT middleware for intelligent and real-time crop monitoring and management. Through the synergy between AI and IoT, continuous data collection, real-time analysis, and automated decision-making are very useful for farmers to optimize operations. In crop monitoring, Li et al. [22] are using satellite imagery and UAV-based multispectral data to assess crop health and identify anomalies. It emphasizes the role of AI in elevating precision agriculture practices and shows an example of how an AI system can analyze huge remote sensing technology datasets to provide insights into crop conditions. Another emerging area of research is the intersection of AI with plant phenotyping. Maraveas [24] reviews state-of-the-art image analysis technologies for plant phenotyping, highlighting significant advances AI has made in automating the assessment of plant traits. The work contributes to improving the efficiency of breeding programs and understanding plant biology. These studies in general show the extensive use of AI in agriculture practices that can come under crop prediction or disease detection, resource management, and phenotyping. The integration of Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 471 https://internationalpubls.com machine learning, deep learning, UAVs, and IoT in agriculture holds great promise to bring about productivity improvements, sustainability, and environmental stewardship in the sector. III. METHODS AND MATERIALS Data Collection All the required data were collected from publicly available agricultural datasets, including satellite imagery and crop health, soil moisture levels, weather conditions, and pest infestation records. Such datasets are full of structured and unstructured data and make them suitable for the application of AI algorithms for crop monitoring. Real-time data through IoT-based sensors installed in crop fields also enriched the dataset with information about temperature, humidity, and quality of the soil [4]. The data was processed ensuring data quality and consistency. Missing values were filled in and outliers treated using statistical methods such as interquartile range filtering. Data preprocessing also included value normalization with respect to features in comparable scales. Algorithms Used For this study, four machine learning algorithms were selected because of their effectiveness in handling agricultural data. These algorithms are well-suited to be used for both classification and regression tasks in the application of precision farming. Decision Trees (DT): Decision Trees is one of the category of algorithms which is most commonly used in AI and machine learning. They operate by repeatedly partitioning the data on conditions of the feature, and consist of decision nodes and end products or leaves [5]. In the context of precision farming, Decision Trees makes sense in terms of categorizing the condition of crops or even prospects of soil heath given data on temperature, humidity and rainfall. Tree construction means nominate the feature that splits the data most appropriately, the split is based on an index such as Gini index or information gain. “1. Start with the entire dataset 2. For each feature, calculate the best split based on Gini impurity or Information Gain 3. Choose the best feature to split the data 4. Recursively repeat the process for each child node 5. Continue until the tree reaches a maximum depth or minimum leaf size 6. Classify the data by following the tree structure from root to leaf node” Random Forest (RF): Random forest is an ensemble of learning technique that constructs a number of decision trees and then forms a voting system to decide on a single result. It tackles the disadvantage of using only one Decision Tree in handling over fitting by averaging the result of several Trees [6]. Random Forest in Agriculture: In the agriculture field, Random Forest can be used for the forecast of factors such as crop yield, disease incidence and intensity of pest incidents using the data through the results of many decision trees. Where there is any variability in the data set, Random Forest is able to counter the effects by the random selection of features for each tree. “1. Build multiple Decision Trees using bootstrapped datasets 2. For each tree, randomly select a subset of features 3. Aggregate the results from all the trees (for classification, use majority voting) 4. Return the aggregated result as the final prediction” Support Vector Machines (SVM): Support Vector Machines (SVM) as the name suggests is a kind of supervised learning that mainly focus on classification. SVM is a technique of finding the best Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 472 https://internationalpubls.com hyperplane that divides the data into two classes while the greatest possible distance is maintained between the two classes [7]. In agriculture, SVM can be used for the analysis of crop diseased according to the characteristics of plant or meteorological parameters. The algorithm employs kernel functions for linear, polynomial, radial basis functions (RBF) to deal with non-linearity existing with respect to features. SVM is particularly efficient in high-dimensional spaces, which can be particularly appropriate to crop monitoring since, in this case, it is necessary to track as many environmental parameters as possible [8]. “1. Transform the data using a kernel function (if necessary) 2. Find the hyperplane that maximizes the margin between different classes 3. For new data points, classify them based on which side of the hyperplane they fall” Convolutional Neural Networks (CNN): Convolutional Neural Networks (CNN) is yet another type of deep learning models designed for use in image solving, including crop monitoring through satellite images or images captured by drone. CNNs have many layers of dropped convolution that learn how to feature images from raw datasets automatically. They are employed in functions like diagnosis of crop ailments, presence of pests or else the phase of crop development [9]. In the context of App based agriculture, CNNs are most useful for applications that involve huge volumes of images such as using images of fields to look for signs of plant stress or for overall crop health. Feature Value for Crop A Value for Crop B Value for Crop C Soil Moisture (%) 30.2 35.8 28.4 Temperature (°C) 23.5 25.1 22.0 Humidity (%) 58.3 65.1 54.2 Pest Density (per sq. m) 2.5 1.8 3.1 Rainfall (mm) 120 80 95 Crop Stress Index 0.34 0.21 0.50 IV. EXPERIMENTS Experiment Setup In each of the cases, the algorithms were validated on a test dataset that consisted of crop monitoring data derived from satellite imagery, IoT sensors, and history of farming data set. Some of the fields in the dataset are soil moisture, soil temperature, soil humidity, pest population density, and crop water stress [10]. The data into training and testing sets was done in order that every algorithm should be tested on new data. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 473 https://internationalpubls.com Figure 1: “Role of Artificial Intelligence in Agriculture” The performance of each algorithm was measured based on the following metrics: ● Accuracy: container to store the percentage of instances in the test set that end up in the correct class. ● Precision: The true positivity that is the number of actual positive prediction divided by the number of positive prediction made. ● Recall: For example, the quality of true positive predictions that were produced out of all the actual positives that are out there. ● F1-Score: The F1 measure of the ratio between precision and recall where a single figure of merit can be used to quantify this trade-off. Each algorithm was implemented using standard Python libraries: Pyton library such as Scikit-learn for Decision Trees, Random forest and support vector machine and TensorFlow for Convolutional neural network. Control for all experiments was done on personal computer with 16 Gb RAM and Intel Core i7. Experimental Design 1. Decision Tree (DT) The DT algorithm was used for the classification of crop health due to environment-temperature, moisture, and pest density. Cross-validation was used for fine tuning of the depth of tree and splitting criteria of DT, given by Gini impurity. Key Steps in DT Implementation: 1. Preprocessing: Missing values were imputed by medians, and one-hot encoding was done for categorical variables. 2. Hyperparameter selection: With the help of cross-validation, the depth of the tree was decided. 3. Training: The decision tree was trained on the dataset with the parameters selected. Performance Results: Deciding Crop Health was nicely done by the Decision Tree using less computational resources than many more complex algorithms in a highly accurate manner. 2. Random Forest (RF) Random Forests is a versatile system that combines the decision trees and diminishes the forecast errors of individual decision trees [11]. In this experiment, Random forest algorithm was employed for crop yield prediction and detection of pest infested zones. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 474 https://internationalpubls.com Figure 2: “Development of Agricultural System and Apps ” Key Steps in RF Implementation: 1. Data preprocessing: In the same manner as DT, missing values were filled in, and categorical data were encoded into binary format. 2. Hyperparameter tuning: The estimate of trees in the forest was diverse due to the top performing cross validation number of trees used in the decision process. 3. Training: The random forest was then built with the training dataset in use and a majority voting method was used for classification while mean voting was used for regression. Performance Results: Random Forest also outperformed Decision Trees in terms of accuracy and a model’s ability to generalize. It also outperformed the other methods in dealing with noisy data which is ideal in handling large sets of data in agriculture [12]. 3. Support Vector Machines (SVM) In the crop disease classification, SVM was employed using an optimal hyperplane to classify different crop diseases based on extracted features. To learn high-order non-linear relationships between the features an RBF (Radial Basis Function) kernel was employed. Key Steps in SVM Implementation: 1. Data preprocessing:Some feature scaling were performed mainly to have a better SVM performance in part because all the features should be on the same scale. 2. Hyperparameter tuning: Hyperparameters of decision function (C) and kernel parameters were set with the help of grid search. 3. Training: After scaling and choosing the best parameters, the SVM model was used for training with these features. Performance Results: SVM was also able to give high precision and recall for the crop disease classification showing that it can work well with the higher dimensional agricultural data [13]. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 475 https://internationalpubls.com Figure 3: “Artificial Intelligence In Agriculture Market” 4. Convolutional Neural Network (CNN) CNN was used to segment satellite images in order to identify pest and diseases affected crops. A team of authors described the model which included the convolutional layers for feature extraction, followed by pooling and fully connected layers for classification. Key Steps in CNN Implementation: 1. Data preprocessing: To extend the number of training images, they were rescaled, normalized and augmented. 2. Model architecture: The actual model employed was a deep CNN with more than two convolutional and pooling layers. 3. Training: Moreover, the CNN model was trained with the Adam optimization method with 0.001 in the learning rate and was trained using batch size of 32. Performance Results: Field outcomes documented that CNNs outperformed the conventional machine learning algorithms, for example, SVM, and randomized forest in image-based assignments including pest identification [14]. In the given images also the model was able to predict the patterns quite well and hence had high accuracy. Results and Comparison The following two tables include the summary of each algorithm based on the earlier mentioned measurements. The results presented here show how each of the algorithms performs well or poorly in different crop monitoring tasks [27]. Table 1: Performance Comparison of Machine Learning Algorithms Algorithm Accuracy (%) Precision (%) Recall (%) F1-Score (%) Decision Tree (DT) 85.2 83.0 84.5 83.7 Random Forest (RF) 91.7 89.6 90.3 89.9 Support Vector Machine (SVM) 88.3 87.0 88.0 87.5 Convolutional Neural Network (CNN) 95.6 94.2 95.0 94.6 Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 476 https://internationalpubls.com Observation: ● It was established that the CNN had higher accuracy, precision, recall, and F1-score for pest detection and crop disease classification, which are image recognition activities. ● Random Forest also performed quite well as it did well when tested on large and noisy data. ● These showed that Decision Trees also did well but were over fitted and this was especially the case with deep trees [28]. ● SVM although was efficient in disease classification had slightly less accuracy compare to CNN, Random Forest. Figure 4: Revolutionize Your Farm Table 2: Computational Efficiency Comparison Algorithm Training Time (hours) Testing Time (minutes) Memory Usage (GB) Decision Tree (DT) 2.5 0.4 1.2 Random Forest (RF) 5.3 1.1 2.5 Support Vector Machine (SVM) 3.2 0.8 1.8 Convolutional Neural Network (CNN) 10.0 2.5 4.5 Observation: ● CNNs, as deep learning models, are heavy and computationally intensive, thus best used for small datasets or powerful systems. ● Random Forest and SVM were reasonably fast when it comes to both train and test time but RF took longer time because of its nature of an ensemble model [29]. ● From the analysis of the execution time, Decision Trees were the most efficient both on the training and testing phases. Discussion The experiments have been proven right by the fact that AI and machine learning techniques could be very helpful in improving the efficiency and accuracy of precision farming. The CNN was able to outperform other techniques because of its potential to analyze more complex patterns in imagery than Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 477 https://internationalpubls.com others, which might be useful for applications concerning crop disease and pests. Nevertheless, the high computational needs make it less desirable for real-time applications on resource-constrained devices [30]. Random Forest was an excellent balance of the accuracy and computational intensity with large-scale agricultural datasets - a very practical choice. Decision Trees were computationally efficient, but started to overfit in deeper trees and additional regularization may be needed to make them more robust. V. CONCLUSION This research concludes that AI may prove to be the transformative force for revolutionizing precision farming and crop monitoring. Integration of AI-related technologies such as machine learning, deep learning, and image processing resulted in the possible effective enhancement of agricultural practice through better crop prediction, disease detection, and optimal resource utilization. The algorithms and their applications were used to illustrate how AI could be a useful tool in extracting actionable insights for farmers to make data-driven decisions to enhance crop yields, limit waste, and involve as little environmental impact as possible in the harvesting process. Using UAVs and remote sensing technologies mainly helps amplify AI's capabilities in providing better monitoring solutions over large agricultural areas. Our results point out that AI-based systems could significantly give more accuracy in plant disease detection, crop yield prediction, and soil and environmental condition monitoring than the traditional methodology. Apart from higher productivity, it also leads to sustainable farming in a sense that increases water, fertilizer, and pesticide usage optimization. The integration of AI with IoT and drones further enables real-time monitoring and automated intervention, thus forming smarter farming systems. Despite the challenges ranging from data quality and very high computational costs, the advancements of AI algorithms and technologies will continue to evolve into promising enhancements in food security, a better livelihood for farmers, and sustainable agricultural practices facing rising global demands for food. Therefore, AI has a prospective place as the cornerstone for future agricultural innovations. REFERENCE [1] ADDAS, A., TAHIR, M. and ISMAT, N., 2024. Enhancing Precision of Crop Farming towards Smart Cities: An Application of Artificial Intelligence. Sustainability, 16(1), pp. 355. [2] AHMAD, U., NASIRAHMADI, A., HENSEL, O. and MARINO, S., 2022. Technology and Data Fusion Methods to Enhance Site-Specific Crop Monitoring. Agronomy, 12(3), pp. 555. [3] ASLAN, M.F., SABANCI, K. and ASLAN, B., 2024. Artificial Intelligence Techniques in Crop Yield Estimation Based on Sentinel-2 Data: A Comprehensive Survey. Sustainability, 16(18), pp. 8277. [4] ASSIMAKOPOULOS, F., VASSILAKIS, C., MARGARIS, D., KOTIS, K. and SPILIOTOPOULOS, D., 2024. The Implementation of “Smart” Technologies in the Agricultural Sector: A Review. Information, 15(8), pp. 466. [5] BALASKA, V., ADAMIDOU, Z., VRYZAS, Z. and GASTERATOS, A., 2023. Sustainable Crop Protection via Robotics and Artificial Intelligence Solutions. Machines, 11(8), pp. 774. [6] BALYAN, S., JANGIR, H., TRIPATHI, S.N., TRIPATHI, A., JHANG, T. and PANDEY, P., 2024. Seeding a Sustainable Future: Navigating the Digital Horizon of Smart Agriculture. Sustainability, 16(2), pp. 475. [7] CHERGUI, N. and KECHADI, M.T., 2022. Data analytics for crop management: a big data view. Journal of Big Data, 9(1), pp. 123. [8] DIVYAJYOTHI, M.G., JOPATE, R., RABIA ABDULRAHIM, A.A. and SALEH ABDULLAH SALEH, A.B., 2024. AI Precision for Irrigation, Crop Management, and Pest Control for Sustainable Agriculture in Oman. IOP Conference Series.Earth and Environmental Science, 1401(1), pp. 012005. [9] E, M.B.M.K., ANH, T.L., HEO, S., CHUNG, Y.S. and MANSOOR, S., 2023. The Path to Smart Farming: Innovations and Opportunities in Precision Agriculture. Agriculture, 13(8), pp. 1593. [10] ELISHA ELIKEM, K.S., ANGGRAINI, L., KUMI, J.A., LUNA, B.K., AKANSAH, E., HAFEEZ, A.S., MENDONÇA, I. and ARITSUGI, M., 2024. IoT Solutions with Artificial Intelligence Technologies for Precision Agriculture: Definitions, Applications, Challenges, and Opportunities. Electronics, 13(10), pp. 1894. [11] FUENTES-PEÑAILILLO, F., GUTTER, K., VEGA, R. and GILDA, C.S., 2024. New Generation Sustainable Technologies for Soilless Vegetable Production. Horticulturae, 10(1), pp. 49. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 478 https://internationalpubls.com [12] FUENTES-PEÑAILILLO, F., GUTTER, K., VEGA, R. and GILDA, C.S., 2024. Transformative Technologies in Digital Agriculture: Leveraging Internet of Things, Remote Sensing, and Artificial Intelligence for Smart Crop Management. Journal of Sensor and Actuator Networks, 13(4), pp. 39. [13] GANAPATHYSUBRAMANIAN, B., BELL, J.M.P., KANTOR, G., MERCHANT, N., SARKAR, S., SCHNABLE, P.S., SEGOVIA, M., SINGH, A. and SINGH, A.K., 2024. AIIRA: AI Institute for Resilient Agriculture. AI Magazine, 45(1), pp. 94-98. [14] GETAHUN, S., KEFALE, H. and GELAYE, Y., 2024. Application of Precision Agriculture Technologies for Sustainable Crop Production and Environmental Sustainability: A Systematic Review. The Scientific World Journal, 2024. [15] GONZÁLEZ-RODRÍGUEZ, V.,E., IZQUIERDO-BUENO, I., CANTORAL, J.M., CARBÚ, M. and GARRIDO, C., 2024. Artificial Intelligence: A Promising Tool for Application in Phytopathology. Horticulturae, 10(3), pp. 197. [16] JAYARAM, D.K., RAJA, V., BEENA, S.A. and AL-BONSRULAH, H., 2024. Design, Multiperspective Investigations, and Performance Analysis of Multirotor Unmanned Aerial Vehicle for Precision Farming. International Journal of Aerospace Engineering, 2024. [17] KARAM, K., MANSOUR, A., KHALDI, M., CLEMENT, B. and AMMAD-UDDIN, M., 2024. Quadcopters in Smart Agriculture: Applications and Modelling. Applied Sciences, 14(19), pp. 9132. [18] KATEKAR, V. and CHERUKU, J.K., 2022. The Application of Drone Technology for Sustainable Agriculture in India. Current Agriculture Research Journal, 10(3),. [19] KUMAR, K.S. and ARULANANDAM, K., 2024. Hybrid Machine Learning based Crop Prediction Model. Journal of Electrical Systems, 20(10), pp. 2059-2073. [20] KUMAR, S.P. and RAO, Y.R., 2024. A Survey on Deep Learning Approaches for Crop Disease Analysis in Precision Agriculture. Turkish Journal of Computer and Mathematics Education, 15(1), pp. 242-253. [21] LEE, H., 2024. The Application of Artificial Intelligence (Ai) and Internet of Things (Iot) Middleware in Agriculture. Journal of Electrical Systems, 20(4), pp. 125-131. [22] LI, M., SHAMSHIRI, R.R., WELTZIEN, C. and SCHIRRMANN, M., 2022. Crop Monitoring Using Sentinel-2 and UAV Multispectral Imagery: A Comparison Case Study in Northeastern Germany. Remote Sensing, 14(17), pp. 4426. [23] LOGESHWARAN, J., SRIVASTAVA, D., KUMAR, K.S., REX, M.J., AL-RASHEED, A., GETAHUN, M. and BEN, O.S., 2024. Improving crop production using an agro-deep learning framework in precision agriculture. BMC Bioinformatics, 25, pp. 1-40. [24] MARAVEAS, C., 2024. Image Analysis Artificial Intelligence Technologies for Plant Phenotyping: Current State of the Art. AgriEngineering, 6(3), pp. 3375. [25] MD SHAHA, N.K., MD, N.R., CHOWDHURY, M., ALI, M., SAMSUZZAMAN, ALI, M.R., KA, Y.L. and SUN-OK, C., 2023. Technological Trends and Engineering Issues on Vertical Farms: A Review. Horticulturae, 9(11), pp. 1229. [26] MUSHTAQ, M.A., HAFIZ GHULAM MUHU-DIN AHMED and ZENG, Y., 2024. Applications of Artificial Intelligence in Wheat Breeding for Sustainable Food Security. Sustainability, 16(13), pp. 5688. [27] NAVARRO, E., COSTA, N. and PEREIRA, A., 2020. A Systematic Review of IoT Solutions for Smart Farming. Sensors, 20(15), pp. 4231. [28] OBEIDAT, M.A., ABDALLAH, J., HAMADNEH, T., QAWAQNEH, H. and MANSOUR, A.M., 2024. Enhancing Agricultural Operations Through AI-Driven Agent Communication in Smart Farming Systems. Ingenierie des Systemes d'Information, 29(3), pp. 917-928. [29] OMIA, E., BAE, H., PARK, E., MOON, S.K., BAEK, I., KABENGE, I. and BYOUNG-KWAN, C., 2023. Remote Sensing in Field Crop Monitoring: A Comprehensive Review of Sensor Systems, Data Analyses and Recent Advances. Remote Sensing, 15(2), pp. 354. [30] PATOKAR, A.M. and GOHOKAR, V.V., 2024. Design and Development of an Intuitive Android Application for Smart Farming. Current Agriculture Research Journal, 12(1),.