Agricultural productivity and food security are heavily impacted by plant diseases, and thus there is a high demand for accurate and automated plant disease detection that can be achieved by applying deep learning techniques. This research proposes a Multi-Model Ensemble Method Based on Deep Learning for multi-plant disease detection using ResNet50 to improve classification performance across multiple crop varieties. The proposed framework takes five important categories of plants into consideration including tomato, potato, grape, apple and maize, and 10 classes of healthy and diseased plants are generated from the PlantSeg dataset. The Anaconda platform was used along with Python to create a development environment that allows data preprocessing, augmentation, training and testing to be implemented efficiently. The proposed ensemble framework combines the feature extraction power of ResNet50 with several deep learning classifiers so as to obtain a good identification performance at different resolutions and environments. The proposed model performance is tested with the following metrics Accuracy, Precision, Inference Time, and Resolution quality and compared with MobileNetV2, YOLOv8 and the baseline CNN models. Experimental results show that the proposed ensemble ResNet50 framework achieves an accuracy of 98.7% and precision of 98.3%, which is about 6.4%, 4.8%, and 9.2% higher than that of MobileNetV2, YOLOv8, and CNN respectively. Moreover, the proposed method achieves high resolution disease localization capability with an inference time improvement of almost 18% compared with YOLOv8. The proposed system greatly improves the detection accuracy of the early stage and the calculation speed of the system, which is very suitable for smart agriculture applications and real-time monitoring of the health status of crops.
This paper proposes an Intelligent Link Failure Prediction and QoS-Aware Routing framework for Mobile Ad Hoc Networks (MANETs) using the Harris Hawk Optimization (HHO) algorithm to achieve reliable and efficient data transmission under highly dynamic network conditions. The proposed model integrates proactive link failure prediction with HHO-based multi-objective route optimization to select stable, energy-efficient, and QoS-compliant paths. The framework is implemented and evaluated using the SimPy simulation environment under a realistic node mobility and traffic workload generated using Random Waypoint mobility with CBR and VBR traffic patterns, which is widely adopted for MANET performance evaluation. The proposed method is compared with Multi-Agent Deep Learning (MADL), Multi-Agent Deep Reinforcement Learning (MADRL), and the Communication-Aware Hierarchical Routing Framework (CAHRF). Experimental results demonstrate that the HHO-based approach significantly improves network performance by increasing the Packet Delivery Ratio (PDR) by 9.8-15.6%, reducing Link Failure Recovery Time by 21.4-34.7%, extending Network Lifetime by 18.2-27.9%, and decreasing Control Packet Cost by 16.5-25.3% compared to the benchmark methods. These improvements confirm that the proposed HHO-driven intelligent routing framework provides a robust, scalable, and QoS-aware solution for reliable communication in highly dynamic MANET environments.
K. Helini, Malleswari Lakkapogu, Suryateja Kothuru et al.· 2026 7th International Confe...· 0 citations
The rapid growth of Internet of Things (IoT) applications has introduced significant challenges in efficient task scheduling within distributed cloud environments, particularly in meeting Quality of Service (QoS) requirements while minimizing operational costs and SLA violations. To address this issue, this paper proposes an AI-Based Hybrid Detective Behavior Optimization (DBA) technique integrated with fuzzy systems for intelligent task scheduling of IoT workloads. The proposed approach leverages the exploration–exploitation capabilities of DBA along with fuzzy logic-based decision-making to dynamically prioritize and allocate tasks under uncertain and heterogeneous cloud conditions. The model is evaluated using the DigitalOcean cloud workload in the WorkflowSim simulation environment and compared against traditional methods including Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Deep Reinforcement Learning (DRL). Experimental results demonstrate that the proposed DBA-Fuzzy approach significantly outperforms baseline methods by reducing SLA violations by 24.6%, improving QoS by 21.3%, minimizing execution cost by 18.9%, and enhancing throughput by 26.7%. These improvements highlight the robustness and adaptability of the proposed model in handling dynamic IoT workloads. The findings suggest that integrating metaheuristic optimization with fuzzy reasoning provides an effective solution for multi-objective task scheduling, making it highly suitable for next-generation distributed cloud environments supporting large-scale IoT applications.
Suryateja Kothuru, Sudipta Priyadarshini, Sai Mounika Chintalapudi et al.· International Conference Com...· 0 citations