Study on Identification Method of Foreign Fibers of Seed Cotton in Hyper-spectral Images Based on Minimum Noise Fraction - Computer and Computing Technologies in Agriculture VI - Part I
Conference Papers Year : 2013

Study on Identification Method of Foreign Fibers of Seed Cotton in Hyper-spectral Images Based on Minimum Noise Fraction

Laiqi Xu
  • Function : Author
Xinhua Wei
  • Function : Author
  • PersonId : 986503
Xinyun Zhou
  • Function : Author
Dazhi Yu
  • Function : Author
Jinmin Zhang
  • Function : Author

Abstract

In order to improve the recognition accuracy of seed cotton foreign fibers, a study on identification method in hyper-spectral images based on Minimum Noise Fraction (MNF) was proposed ,which was applied to feature extraction to reduce the dimension of hyper-spectral images. This method reduced the numbers of hyper-spectral data, lessened the images noise to the minimum , but also decreased the computational requirements for subsequent processing. The white foreign fibers and cotton which were in small discrimination were selected in this paper as the research object. The hyper-spectral images were displayed in software ENVI with 256 bands in the wavelenghth range of 871.60nm-1766.32nm. Afterwards, the images would be processed with the iteration threshold segmentation method, inflation and corrosion. Meanwhile, the correlation of template images and destination images were calculated to find the spectral peaks so that to make template matching to eliminate the images of the cotton seeds. Results of experiments show that the above methods is suitable for identifying foreign fibers of seed cotton which achieved 84.09% rate of recognition.
Fichier principal
Vignette du fichier
978-3-642-36124-1_21_Chapter.pdf (554.76 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-01348096 , version 1 (22-07-2016)

Licence

Identifiers

Cite

Laiqi Xu, Xinhua Wei, Xinyun Zhou, Dazhi Yu, Jinmin Zhang. Study on Identification Method of Foreign Fibers of Seed Cotton in Hyper-spectral Images Based on Minimum Noise Fraction. 6th Computer and Computing Technologies in Agriculture (CCTA), Oct 2012, Zhangjiajie, China. pp.166-176, ⟨10.1007/978-3-642-36124-1_21⟩. ⟨hal-01348096⟩
67 View
153 Download

Altmetric

Share

More