We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

eess.SP

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Electrical Engineering and Systems Science > Signal Processing

Title: Incremental learning with online SVMs on LiDAR sensory data

Abstract: The pipelines transmission system is one of the growing aspects, which has existed for a long time in the energy industry. The cost of in-pipe exploration for maintaining service always draws lots of attention in this industry. Normally exploration methods (e.g. Magnetic flux leakage and eddy current) will establish the sensors stationary for each pipe milestone or carry sensors to travel inside the pipe. It makes the maintenance process very difficult due to the massive amount of sensors. One of the solutions is to implement machine learning techniques for the analysis of sensory data. Although SVMs can resolve this issue with kernel trick, the problem is that computing the kernel depends on the data size too. It is because the process can be exaggerated quickly if the number of support vectors becomes really large. Particularly LiDAR spins with an extremely rapid rate and the flow of input data might eventually lead to massive expansion. In our proposed approach, each sample is learned in an instant way and the supported kernel is computed simultaneously. In this research, incremental learning approach with online support vector machines (SVMs) is presented, which aims to deal with LiDAR sensory data only.
Comments: This paper has been published at the International Conference on Digital Image and Signal Processing (DISP 2019)At: Oxford, United Kingdom
Subjects: Signal Processing (eess.SP); Machine Learning (cs.LG)
Cite as: arXiv:2101.01667 [eess.SP]
  (or arXiv:2101.01667v1 [eess.SP] for this version)

Submission history

From: Zhiyuan Chen Dr [view email]
[v1] Wed, 30 Dec 2020 03:54:35 GMT (420kb)

Link back to: arXiv, form interface, contact.