The Study Of On-Site Detection Technology On Asbestos Fibers Suspending In The Air

Asbestos is used in Construction, Metallurgy, Machinery, Petroleum, Electricity, Chemical industry or even Military fields. There is no toxic for asbestos itself. It's greatest harm is the very small fibers suspended in the air after differentiation. When these tiny fibers are inhaled in the human body, they will attach and deposite in the lungs, causing lung disease, even lung cancer. People understand the hazards of asbestos fibers growing in recent years. So the developed countries have introduced restrictions on the use of asbestos products and many bills and a variety of the relevant standards on environmental asbestos fibers content. Asbestos fibers on-site monitoring technology has also become an ongoing solution problem.At present, domestic and international testing for asbestos fibers use of the combination of mineralogical and physical methods which is completed in the laboratory after sampling. For the on-site air monitoring of asbestos fibers, there has not a good way. Therefore, to find a automatic detection method of asbestos fibers in air is necessary.The author has spended a lot of time reading a lot of literature and found a few literatures on asbestos and asbestos fibers. On base of studying various properties of the relevant standards of asbestos and asbestos sample preparation, the author proposes a kind of structure of automatic identification system for asbestos fiber, and its overall design structure is described.The paper proposes for image recognition measurement of asbestos fiber counting by the first time. The system has a polarized light microscopy with image acquisition using function, and a unique auto-sampling structure. The system samples continuously with a micro-pump. The system deals with and recognizes the images collected of asbestos fibers collected, and designs several recognition algorithm according to the characteristics of asbestos fibers. The paper strengthens the image through the wavelet-based histogram equalization algorithm; denoises the image through a kind of adaptive hybrid filtering algorithm for the mixed noise; detects the image's edge through the Canny edge detection operator; extracts features from the image after the operations of expansion through the technology of bounding;and finally counts automatically from the final image through the labeling approach based on the eight neighborhood search. In the paper, the autor also makes Matlab simulation for the image.The automated counting method of asbestos fibers is still theoretical study and the initial pilot phase. The paper is the initial exploration of introducing image recognition and processing into the field of asbestos fibers detection. It will play the role of early predictions of the in-depth study as well as the practical application.