Advertisement

Research on Key Algorithms of Segmented Spectrum Retrieval System

  • Jianfeng TangEmail author
  • Jie Huang
Conference paper
  • 2 Downloads
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1143)

Abstract

The quality of characteristic spectra stored in spectral database directly affects the matching calculation of retrieval algorithm. Spectral data is different from mass spectrometry. Mass spectrometry only needs to record the position and intensity of peaks. Spectral data is continuous. There are many useful information concentrated in fingerprint area. Usually, the dimension of spectral data is relatively high. At the beginning of the establishment of spectral database, the amount of data will be relatively small. Therefore, dimensionality reduction and feature extraction of spectral data are needed before building spectral database. The whole dimension reduction feature extraction method can retain less information about the region of interest. In order to solve this problem, this paper proposes a segmentation feature extraction algorithm. The classification verification experiment also proves that the segmentation method is better than the whole method. In addition, on the basis of feature extraction, a feature retrieval algorithm with piecewise weighting is implemented. The time and space complexities of the algorithm are related to the performance of spectral database. The larger the spectral database is, the slower the retrieval is, and the faster the retrieval is.

Keywords

Segmented spectrum retrieval Principal component analysis Dimension reduction 

References

  1. 1.
    Dcebska, B.J., Guzowska-Swider, B.: Spectral databases, infrared. Encyclopedia of Analytical Chemistry. Wiley, New York (2006)Google Scholar
  2. 2.
    Curry, A.S., Read, J.F., Brown, C.: A simple infrared spectrum retrieval system. J. Pharm. Pharmacol. 21(4), 224–231 (2011)CrossRefGoogle Scholar
  3. 3.
    Peschke, K.-D., Haasdonk, B., Ronneber, O., Burkhardt, H., Rösch, P., Harz, M., Popp, J.: Using transformation knowledge for the classification of raman spectra of biological samples. In: Proceedings of the 4th IASTED International Conference on Biomedical Engineering, pp. 288–293 (2006)Google Scholar
  4. 4.
    Ghasemi, Z., Vafaei, A., Monadjemi, S.A.: Comparison of accuracy and time complexity of PCA based algorithms for steel surface segmentation. Int. J. Res. Rev. Comput. Sci. 3(2) (2012)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2021

Authors and Affiliations

  1. 1.School of Software EngineeringTongji UniversityShanghaiChina

Personalised recommendations