The challenge of real-time object detection in dynamic environments is complicated by issues like noise, occlusion, changes in illumination, and ambiguity of object boundaries. Non-Bayesian models of deep learning tend to be confident about their predictions. Such approach is dangerous since it renders these models inappropriate for use in safety-critical applications such as self-driving vehicles, surveillance and robotics. In this paper, an uncertainty-aware deep learning method is suggested which can be applied to real-time object detection in dynamic computational vision environments. This method combines a lightweight detector based on YOLO architecture, Monte Carlo dropout and uncertainty estimation via entropy measure to account for both aleatoric and epistemic uncertainties. The approach is expected to increase robustness to challenges including occlusion, motion blur and illumination variation. Experimental results obtained on COCO and KITTI data sets show that the proposed model reaches mAP of 88.9%, that is, 6.8% better compared to baseline YOLO and CNNs models. False positives are reduced by 12.3% and ECE score is increased by 9.5%. The model runs at 38 FPS which ensures its real-time operation. The results confirm that uncertainty-aware reasoning significantly enhances prediction reliability and interpretability in object detection systems.
Nidhi Mishra, Aakansha Soy· 2026 6th International Confe...· 0 citations
Advent of Internet of Things (IoT) sensor networks of scale, the amount and speed of data produced has become a major problem in real-time data processing and analytics. Conventional computing models cannot provide real-time insights because are constrained by scalability, resource management, and computational capabilities. The paper discusses how parallel computing frameworks, including Apache Spark, Apache Flink, and CUDA, can be used to improve the efficiency of processing and analyzing massive sensor data in real time. These models allow distributed data processing across multiple nodes, significantly reducing latency and enhancing throughput. The load balancing, fault tolerance, and data partitioning problems are some of the challenges that proposed approach will address using these frameworks to ensure significant performance gains in large-scale IoT settings. Data processing is found to be significantly faster, and the analytics latency is reduced, which is shown by experimental results, thereby showing the potential of parallel computing in real-time IoT analytics. The contributions made in this work are the design and implementation of an optimized framework of IoT sensor networks and the performance evaluation of a comprehensive framework across different real-world conditions.
Dr.Nidhi Mishra, Aakansha Soy· International Conference Com...· 0 citations