Click on “Download PDF” for the PDF version or on the title for the HTML version.


If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Detection of Human and Pig Hair on the Surface of Ginned Cotton by Hyper-Spectral Imaging

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  2010 Pittsburgh, Pennsylvania, June 20 - June 23, 2010  1008622.(doi:10.13031/2013.31938)
Authors:   Junxian Guo, Xiuqin Rao, Fang Cheng, Yibin Ying
Keywords:   Hyper-spectral imaging, Reflection model, Human and pig hair, Detection, Recognition

Many foreign objects and materials issued from hand picking is reduced in size and entangled with the lint during harvesting and ginning processes. The current techniques cannot rapidly and reliably detect white & colorless contaminants, such as polypropylene fibers, white hair and plastic film. In this paper, the hyper-spectral imaging technique, capable of collecting spectral information at each pixel of the image, was investigated for its detection potential of human and pig hair on the surface of carded cotton. Human and pig hair with different lengths and colors were inspected by hyper-spectral reflection model over the visible and short-wave near-infrared region. The rate of identification for black human hair, black and white pig hair respectively reached up to 97.56%, 96.43% and 44.44 % by using a series well-planned analytical procedure for hyper-spectral images. In contrast, the identification rate for white pig hair using RGB image with the same resolution (i.e. 188 dpi) of hyper-spectral images was zero. Furthermore, the potential wave band for recognition of black hair was to distribute in a wider band while that for recognition of white hair concentrated in the narrow range. The range from 562.46-632.84 nm was finally identified as key wave band for segmentation and recognition of white pig hair according to quantitative evaluation of equal-interval band images.

(Download PDF)    (Export to EndNotes)