Analyzing Gas Data Using Deep Learning and 2-D Gramian Angular Fields

The notion of employing deep learning (DL) for gas classification has kindled a revolution that improved both data collection measures and performance. Yet, the current literature, with its vast contributions, potential in enhancing state art by DL novel visualization methods to boost performance speed. Therefore, this article presents dual system high-performance classification: on 1-D time series 2-D Gramian Angular Field (GAF) data. For GAF case study, are converted into counterparts means normalization, segmentation, averaging, color coding. sensor array (GSA) dataset is used evaluating implemented AlexNet model classifying an version GasNet time-based Using cloud-based architecture, two models evaluated benchmarked art. Evaluation results modified signify state-of-the-art accuracy 96.5%, while achieved 81.0% test near real-time edge computing platforms.