论文信息 - DEVELOPMENT OF AN AUTOMATIC SYSTEM FOR COUNTING ASBESTOS FIBERS USING IMAGE PROCESSING

DEVELOPMENT OF AN AUTOMATIC SYSTEM FOR COUNTING ASBESTOS FIBERS USING IMAGE PROCESSING

ABSTRACT Atmospheric concentrations of carcinogenic asbestos fibers, have traditionally been measured visually using phase contrast microscopy. However, because this measurement method requires great skill, and has poor reproducibility and objectivity, the development of automatic counting methods has been long anticipated. We have developed the Asbestos Fibers Automatic Counting System (AFACS), and verified its accuracy via application to illustrations obtained from the Asbestos Internationa) Association (AIA). Important features of the AFACS system are as follows: (1) Counting criteria of the automatic counting system conform to those of visual counting (2) The cost of the whole system is low because it is microcomputer-based.

Katsuhito Yamaguchi | Yoshio Inoue | Akikazu Kaga | S. Kamoi

[1] Ching Y. Suen,et al. A fast parallel algorithm for thinning digital patterns , 1984, CACM.

[2] C. J. Hilditch,et al. Linear Skeletons From Square Cupboards , 1969 .

[3] Jack Sklansky,et al. Fast polygonal approximation of digitized curves , 1980, Pattern Recognit..

[4] P.K Sahoo,et al. A survey of thresholding techniques , 1988, Comput. Vis. Graph. Image Process..

[5] N. Otsu. A threshold selection method from gray level histograms , 1979 .

[6] Josef Kittler,et al. Minimum error thresholding , 1986, Pattern Recognit..

[7] L. C. Kenny,et al. Asbestos fibre counting by image analysis , 1984 .

[8] P. Iles,et al. Comparison of a fibrous aerosol monitor (FAM) with the membrane filter method for measuring airborne asbestos concentrations. , 1986, The Annals of occupational hygiene.