Long N. Le

Long N. Le, Douglas L. Jones

Traditional distributed detection systems are often designed for a single target application. However, with the emergence of the Internet of Things (IoT) paradigm, next-generation systems are expected to be a shared infrastructure for multiple applications. To this end, we propose a modular, cascade design for resource-efficient, multi-task detection systems. Two (classes of) applications are considered in the system, a primary and a secondary one. The primary application has universal features that can be shared with other applications, to reduce the overall feature extraction cost, while the secondary application does not. In this setting, the two applications can collaborate via feature sharing. We provide a method to optimize the operation of the multi-application cascade system based on an accurate resource consumption model. In addition, the inherent uncertainties in feature models are articulated and taken into account. For evaluation, the twin-comparison argument is invoked, and it is shown that, with the optimal feature sharing strategy, a system can achieve 9$\times$ resource saving and 1.43$\times$ improvement in detection performance.

Long N. Le, Douglas L. Jones

Energy-efficiency is highly desirable for sensing systems in the Internet of Things (IoT). A common approach to achieve low-power systems is duty-cycling, where components in a system are turned off periodically to meet an energy budget. However, this work shows that such an approach is not necessarily optimal in energy-efficiency, and proposes \textit{guided-processing} as a fundamentally better alternative. The proposed approach offers 1) explicit modeling of performance uncertainties in system internals, 2) a realistic resource consumption model, and 3) a key insight into the superiority of guided-processing over duty-cycling. Generalization from the cascade structure to the more general graph-based one is also presented. Once applied to optimize a large-scale audio sensing system with a practical detection application, empirical results show that the proposed approach significantly improves the detection performance (up to $1.7\times$ and $4\times$ reduction in false-alarm and miss rate, respectively) for the same energy consumption, when compared to the duty-cycling approach.

Fadeel Sher Khan, Long N. Le, Abhinau K. Venkataramanan et al.

Traditional image quality assessment (IQA) methods rely on mean opinion scores (MOS), which are resource-intensive to collect and fail to provide interpretable, localized feedback on specific image distortions. We overcome these limitations by shifting from absolute quality prediction to a relational and directional assessment. Our approach utilizes a self-supervised synthetic distortion engine to generate training data, eliminating the need for manual annotation. A distortion prediction network is trained with an anti-symmetric objective to produce spatially-aware, disentangled maps that identify the type, intensity, and direction of distortions relative to a reference image. Subsequently, a scoring network is trained via contrastive learning on ordinally ranked image sets to predict a relational quality score. Our method provides a more granular and interpretable approach to IQA for the targeted optimization of image processing algorithms without requiring any human-labeled quality scores.