Jump to content

User:UJNLHN/sandbox

From Wikipedia, the free encyclopedia
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Vibration Perceptual Hashing

Perceptual Vibration Hashing is to perceive/extract and hash the machine condition information from vibration signals from a perceivable perspective. It is relevant to perceptual hashing, but with vibration signals as the processing objects. Vibration signals contain a lot of information about machine condition, which makes it widely adopted for condition monitoring. But the transmission, storage and computation of vibration signals unusually require a strong information infrastructure. Because the acquired vibration signals would increase exponentially with the timespan and number of sensors deployed. So, why not just transmit, store and compute the machine condition information, instead of the original vibration signals? Perceptual vibration hashing is giving an answer to it. Two computing advantages of perceptual hashing make it a perfect choice for vibration signal processing, especially in remote machine condition monitoring, which are perceptual information extraction and compact information representation.

Hashing

Hashing is a commonly technique for Information retrieval, Cryptography, File verification, [Authentication], and so on. Any bit-level change in the data would make completely different hash code. Commonly used hash function includes MD5, SHA-1, SHA256, etc.

Perceptual Hashing

For similar objects, taking two images with a dog and a wolf as an example, similar hash code should be generated. Here, the similarity is justified with some defined distance metrics, such as the Hamming distance, Mahalanobis distance, etc. Various algorithm are also proposing, which includes feature point based methods[1], clustering based methods[2], and so on.

Condition Monitoring with Perceptual Hashing

Conceptual architecture is proposed[3] based on perceptual hashing for machine condition monitoring. With this architecture, not only the data dimension can be reduced, but also the diagnostic modeling can benefit from the geometric structure of the machine condition hash.

It is still an open research about Vibration Perceptual Hashing.

See Also

References

  1. ^ Monga, Vishal; Evans, Brian L. (2006). "Perceptual image hashing via feature points: Performance evaluation and tradeoffs". IEEE. {{cite journal}}: Cite journal requires |journal= (help)
  2. ^ Vishal Monga; Arindam Banerjee; Brian L. Evans (2006). "A Clustering Based Approach to Perceptual Image Hashing". IEEE. {{cite journal}}: Cite journal requires |journal= (help)
  3. ^ Haining Liu and Xiuhua Men and Fajia Li and Jinkai Zhang and Xiaohong Wang and Chengliang Liu (2018). A new methodology for condition monitoring based on perceptual hashing. Vol. 1. IEEE. pp. 919–923. {{cite book}}: |journal= ignored (help)