Yj. Han et al., IDENTIFICATION AND MEASUREMENT OF CONVOLUTIONS IN COTTON FIBER USING IMAGE-ANALYSIS, Artificial intelligence review, 12(1-3), 1998, pp. 201-211
An image analysis procedure was developed to quantify morphological ch
aracteristics of convolutions in individual cotton fibers without pre-
tensioning or orientation requirements. The image of each fiber was ca
ptured by a PC-based color imaging system using a conventional microsc
ope. Ends of individual cotton fibers were glued on a microscope slide
without any tension or straightening. A modified watershed technique
was implemented to identify individual convolution segments, which wer
e defined as sections of the fiber bordered by two neighboring convolu
tions. Length, area and perimeter of each convolution segment were mea
sured directly from the image. Average width, shape factor and number
of convolution segments in mm were calculated from the measured parame
ters. Performance of the image analysis algorithm was compared with vi
sual inspection for number and position of convolution segments in thr
ee different varieties of cotton. Image analysis results agreed with v
isual inspection in 89.6% of the tested images.