Closed-system greenhouses have gained increasing popularity due to their ability to precisely control environmental conditions, support vertical farming, and enhance crop yield in limited spaces. However, maintaining optimal temperature control under continuously changing outdoor weather conditions while improving energy efficiency remains a significant challenge. This study presents the development of an Artificial Intelligence (AI) and Internet of Things (IoT)based architecture incorporating energy management for temperature regulation in a closed-system green-house using a machine learning-based classification control decision approach. IoT sensors continuously collect environmental and electrical parameters, including indoor and outdoor temperature, humidity, voltage, current, and power consumption, and transmit the data to a cloud platform for monitoring and analysis. Machine learning classification models were trained offline in MATLAB/Simulink and employed to determine air-conditioning activation for energy-efficient greenhouse operation. Three modelsQuadratic Support Vector Machine (QSVM), Cubic Support Vector Machine (CSVM), and Ensemble Bagged Trees (BT)were trained and evaluated. Experimental results show that the BT model is suitable for real greenhouse deployment, achieving 98.46% accuracy comparable to QSVM with lower computational complexity, while its lightweight nature suits cloud platforms with limited communication resources. Furthermore, real-world implementation testing conducted over a two-day period demonstrates that the BT-based control system reduces energy consumption by approximately 20.54% compared with a conventional system without AI-based control. This approach has the potential to reduce energy consumption in closed-system greenhouses; however, it would require adjustments in data collection and model configuration to account for differing plant physiological characteristics.
Learning-enabled optimization in dynamic energy systems demands more than high-quality solutions; it requires sustained adaptation under non-stationary and shock-driven environments. This paper proposes Adaptive Genetic Algorithm with Dynamic Operator Scoring and Operator Elimination (AGA-DO), an evolutionary learning framework that explicitly models operator-level learning to achieve long-term robustness. In contrast to conventional genetic algorithms with static or heuristically adaptive operators, AGA-DO continuously evaluates operator contributions, applies temporal smoothing to stabilize learning dynamics, and eliminates persistently ineffective operators to prevent negative knowledge accumulation. Unlike conventional adaptive genetic algorithms that primarily focus on parameter adaptation and peak optimization performance, AGA-DO emphasizes operator-level learning, learning stability, and recovery capability under dynamic environments. A ve-phase validation framework is employed to progressively evaluate effectiveness, learning behavior, robustness, and real-world recovery capability. Beyond instantaneous performance metrics, Cumulative Regret Loss (CRL) is adopted as an observational indicator to quantify the accumulated cost of delayed adaptation and repeated suboptimal decisions over time. Results show that benchmark algorithms accumulate substantially higher regret under repeated environmental shocks, whereas AGA-DO achieves the lowest cumulative regret (CRL = 0.81) and the best overall Friedman rank (1.2). These findings suggest that AGA-DO can serve as an effective learning-driven evolutionary optimizer for real-world dynamic energy systems. The proposed framework emphasizes learning stability and retention over aggressive convergence, aligning with practical requirements of real-world adaptive optimization.
Multimodal learning that integrates Color Fundus Photography (CFP) and Optical Coherence Tomography (OCT) can enhance retinal disease recognition by combining surface appearance with depth-resolved structural cues. However, the limited availability of synchronized 2D-3D CFP-OCT datasets hinders the deployment of multimodal AI in real-world screening settings where OCT devices are often unavailable. This study investigates whether synthetic 3D OCT volumes can serve as a practical auxiliary modality to support multimodal classification in fundus-only scenarios. We synthesize 3D OCT from fundus images using three generative paradigms-a 3D GAN (DCGAN/WGAN-GP style), a 3D VAEGAN, and a fundus-conditioned latent diffusion model (3D-DDPM with classifier-free guidance)-and integrate the generated volumes into an EyeMoST- based uncertainty-aware fusion framework. We evaluate the pipeline on RFMiD and ODIR-5K across multiple backbone configurations. Empirically, diffusion-generated OCT yields the most stable fusion performance. In contrast, synthetic OCT does not consistently outperform strong fundus-only models due to domain gap and imperfect biomarker preservation.
Edge-vision systems on resource-constrained platforms require low-latency and energy-efficient front-end processing. Conventional gradient-based edge operators continue to rely on multiply-accumulate (MAC) operations, which can increase logic utilization, switching activity, and power consumption in eld-programmable gate array (FPGA) implementations. This study introduces a kernel-aware arithmetic-selection framework, supported by synthesized evidence, for energy-efficient FPGA-based edge detection. The framework combines XNORpopcount-based arithmetic mapping with selective approximate computation. Its central design principle is to replace MAC operations with exclusive-NOR-population count (XNORpopcount) when the kernel structure is suitable for binary-friendly arithmetic, and to apply approximation only to the remaining adder-dominated stages when a complete XNOR-popcount mapping is not practical. Under this rule, Prewitt-like operators are mapped to an XNORpopcount datapath over their active nonzero taps. In contrast, Sobel-like operators are realized using a hybrid datapath that combines binary matching, shift-add processing, and approximate accumulation. The resulting framework shows that multiplier removal is the dominant source of hardware savings, while approximate arithmetic provides a controlled secondary optimization. Overall, the proposed approach establishes a structured design methodology for low-power FPGA edge-detection architectures on embedded platforms.
An intelligent honeypot system designed to mimic legitimate websites using Sequence-to-Sequence (Seq2Seq) learning and Deep Q-Learning. The system generates realistic, contextually appropriate responses to attacker queries, prolonging interactions and providing insights into malicious behaviors while safeguarding actual systems. The Seq2Seq model, trained on HTTP request-response pairs, enables the honeypot to produce responses that closely resemble those of real servers, enhancing its ability to deceive attackers. Deep Q-Learning optimizes engagement by selecting the most effective responses through a custom reward function, balancing realism and interactivity to maximize session length. Performance was evaluated using metrics such as Response Realism Rate (RRR), Semantic Consistency Accuracy (SCA), and Average Session Length (ASL). The honeypot achieved an RRR of 92.3%, an SCA of 89.7%, and a 94.5% Optimal Response Selection Rate (ORSR). These advancements increased ASL by 143.5%, from 3.2 to 7.8 exchanges, reflecting prolonged attacker engagement. By integrating Seq2Seq and Deep Q-Learning, this honeypot demonstrates significant improvements in generating convincing responses and sustaining interactions. These results contribute to modern cybersecurity by providing a practical and theoretical framework for developing next-generation honeypots capable of deceiving attackers and gathering actionable intelligence.
