Face recognition (FR) in unconstrained conditions remains an open research topic and an ongoing challenge. The facial images exhibit diverse expressions, occlusions, variations in illumination, and heterogeneous backgrounds. This work aims to produce an accurate and robust system for enhanced Security and Surveillance. A parallel hybrid deep learning model for feature extraction and classification is proposed. An ensemble of three parallel extraction layer models learns the best representative features using CNN and RNN. 2D LBP and 3D Mesh LBP are computed on face images to extract image features as input to two RNNs. A stacked autoencoder (SAE) merged the feature vectors extracted from the three CNN-RNN parallel layers. We tested the designed 2D/3D CNN-RNN framework on four standard datasets. We achieved an accuracy of $98.9\%$. The hybrid deep learning model significantly improves FR against similar state-of-the-art methods. The proposed model was also tested on an unconstrained conditions human crowd dataset, and the results were very promising with an accuracy of $95\%$. Furthermore, our model shows an 11.5% improvement over similar hybrid CNN-RNN architectures, proving its robustness in complex environments where the face can undergo different transformations.

In recent decades, the global installation of photovoltaic (PV) systems has surged, underscoring the growing significance of solar energy. Automatic inspection of photovoltaic power stations provides a means to reliable operation and performance preservation of the solar plant. The efficiency of solar stations can diminish over time due to several problems such as hotspots, shading, and short-circuited bypass diodes. Human on-site manual inspection is expensive and impractical. In order to reduce time and minimize expenses. To reduce time and cost, unmanned aerial vehicles (UAVs) have been proposed as a more reliable and cost-effective inspection paradigm. However, UAVs have limited energy storage and short flight durations, and the existing UAV inspection methods typically rely on the exhaustive visual/thermal scans and ignore the UAV energy constraints. This article proposes a targeted UAV inspection strategy that uses the maximum power point (MPP) telemetry to prioritize candidate modules and compute energy-aware flight paths. An anomaly detection algorithm analyzes MPP-derived features to identify deviations from the expected patterns. An energy-aware shortest flight path planner then computes UAV routes to inspect the highest-risk modules under battery and mission constraints. The results, compared to full-scan and random-sampling baselines, across varying defect densities demonstrate substantial time savings with high defect detection performance. This work involves the simulation of plant geometry and defect densities. The results illustrate a reduction in the operation time up to 85% (versus full-scan) of a 0.5% module defect rate while achieving an F1-score of 92% in defect detection.