Title :
Fast Image segmentation of gold immunochromatographic strip based on FCM clustering algorithm in HSV color space
Author :
Jie Zhang ; Min Du
Author_Institution :
Coll. of Electr. Eng. & Autom., Fuzhou Univ., Fuzhou, China
Abstract :
Gold immunochromatographic strip (GICS) quantitative detective can provide more information than the qualitative or semiquantitative testing. In this paper, a fast color image segmentation method is presented to develop the quantitative detective of GICS. The image of GICS was acquired by Charge-coupled Device (CCD) image sensor, and segmented by fuzzy c-means (FCM) clustering algorithm based on color histogram in HSV color space. Maximin-distance algorithm was adopted to get the initial positions of centroids and cluster number to overcome the shortcoming that the FCM algorithm may produce local optimal results. For the segmented target image, a special characteristic parameter was constructed and calculated in HSV color space to achieve the quantitative interpretation of the GICS.
Keywords :
CCD image sensors; image colour analysis; image segmentation; medical image processing; pattern clustering; statistical analysis; CCD image sensor; FCM clustering algorithm; GICS quantitative detection; HSV color space; charge-coupled device; color histogram; fast color image segmentation; fuzzy c-means clustering; gold immunochromatographic strip; hue saturation value; maximin-distance algorithm; medicine clinical diagnosis; qualitative testing; semiquantitative testing; Clustering algorithms; Gold; Histograms; Image color analysis; Image segmentation; Immune system; Strips; FCM; HSV; gold immunochromatographic strip; maximin-distance algorithm; quantitative interpretation;
Conference_Titel :
Image and Signal Processing (CISP), 2012 5th International Congress on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4673-0965-3
DOI :
10.1109/CISP.2012.6470010