aiqbench analysis
This research presents a significant efficiency gain for autonomous systems by shifting sonar data processing from image-based representations to raw CSV formats, directly enhancing machine learning object detection and signal quality. By reducing processing time by over 90%, the study addresses critical latency constraints in real-time acoustic sensing applications without relying on heavy computational overhead. The approach validates that optimizing data preprocessing pipelines can substantially improve the performance of ML models in remote sensing tasks.
Signal-Centric Remote Sensing via Alternative Preprocessing and Acoustic Processing for ML-Driven Applications
arXiv:2609.21123v1 Announce Type: cross Abstract: The dominant method of processing sonar data is using image-based representations, requiring the preprocessing of image data on autonomous systems. We propose an alternative data processing method for remote sensing applications via the use of data in Comma-Seperated Value format. Experimentation on our alternative approach shows a reduction of processing time by 91.18%, an improvement in accurate object detection by Machine Learning, and an increase in SNR (Signal-to-noise ratio), PSNR (Peak signal-to-noise ratio), and other evaluation metrics.
Abstract:The dominant method of processing sonar data is using image-based representations, requiring the preprocessing of image data on autonomous systems. We propose an alternative data processing method for remote sensing applications via the use of data in Comma-Seperated Value format. Experimentation on our alternative approach shows a reduction of processing time by 91.18%, an improvement in accurate object detection by Machine Learning, and an increase in SNR (Signal-to-noise ratio), PSNR (Peak signal-to-noise ratio), and other evaluation metrics.
7 pages, 9 figures, 3 tables. Published in Proc. IEEE SoutheastCon 2025, pp. 1078-1084, doi: https://doi.org/10.1109/SoutheastCon56624.2025.10971547
Machine Learning (cs.LG); Robotics (cs.RO)
arXiv:2609.21123 [cs.LG]
(or arXiv:2609.21123v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.21123 arXiv-issued DOI via DataCite (pending registration)
L. Luna, S. Jansen-Sánchez, I. Demirkiran, and L. Ghelarducci, "Signal-Centric Remote Sensing via Alternative Preprocessing and Acoustic Processing for ML-Driven Applications," in Proc. IEEE SoutheastCon 2025, pp. 1078-1084, 2025
https://doi.org/10.1109/SOUTHEASTCON56624.2025.10971547 DOI(s) linking to related resources
Full text under the publisher's open licence · 148 words
