The majority of countries rely largely on agriculture for employment. Irrigation accounts for a sizable amount of water use. Crop irrigation is an important step in crop yield prediction. Field harvesting is very reliant on human supervision and experience. It is critical to safeguard the field’s water supply. The shortage of fresh water is a major challenge for the world, and the situation will deteriorate further in the next years. As a result of the aforementioned challenges, smart irrigation and precision farming are the only viable solutions. Only with the emergence of the Internet of Things and machine learning have smart irrigation and precision agriculture become economically viable. Increased efficiency, expense optimization, energy maximization, forecasting, and general public convenience are all benefits of the Internet of Things (IoT). As systems and data processing become more diversified, security issues arise. Security and privacy concerns are impeding the growth of the Internet of Things. This article establishes a framework for detecting and classifying intrusions into IoT networks used in agriculture. Security and privacy are major concerns not only in agriculture-related IoT networks but in all applications of the Internet of Things as well. In this framework, the NSL KDD data set is used as an input data set. In the preprocessing of the NSL-KDD data set, first all symbolic features are converted to numeric features. Feature extraction is performed using principal component analysis. Then, machine learning algorithms such as support vector machine, linear regression, and random forest are used to classify preprocessed data set. Performance comparisons of machine learning algorithms are evaluated on the basis of accuracy, precision, and recall parameters.

The rapid expansion of the Internet of Things (IoT) has introduced new vulnerabilities that traditional security mechanisms often fail to address effectively. Signature-based intrusion detection systems cannot adapt to zero-day attacks, while rule-based solutions lack scalability for the diverse and high-volume traffic in IoT environments. To strengthen the security framework for IoT, this paper proposes a deep learning-based anomaly detection approach that integrates Convolutional Neural Networks (CNNs) and Bidirectional Gated Recurrent Units (BiGRUs). The model is further optimized using the Moth–Flame Optimization (MFO) algorithm for automated hyperparameter tuning. To mitigate class imbalance in benchmark datasets, we employ Generative Adversarial Networks (GANs) for synthetic sample generation alongside Z-score normalization. The proposed CNN–BiGRU + MFO framework is evaluated on two widely used datasets, UNSW-NB15 and UCI SECOM. Experimental results demonstrate superior performance compared to several baseline deep learning models, achieving improvements across accuracy, precision, recall, F1-score, and ROC–AUC. These findings highlight the potential of combining hybrid deep learning architectures with evolutionary optimization for effective and generalizable intrusion detection in IoT systems.