Research On The Recognition Algorithm For Concentration And Diameter Distribution Of Indoor Suspended Particulate

Suspended particles in the atmosphere, especially small particles damage on human health greatly, and suspended particles have serious impact on the environment, therefore, people concern about the particles in the air increasingly. Particle parameter and concentration are important for the characters of particles; they are the base of the other characters of particles. The research of morphologic eigenvalue is also basic, so it is necessary to measure the shape factor and fractal of particles.At present the routine methods, for example, sedimentation and fractionation technique are chiefly adopted in measurement of granularity. There are some disadvantages in these routine methods, for example, they take many steps and cast much time, and the accuracy of results is affected greatly by subjective factors of human inspectors. Along with development of the image processing techniques and improvement of micro-observation, particle detection based on micro-observation and image processing techniques has been brought forward. This thesis applies the detection method to analysis of suspended particles, which can overcome shortage of routine methods, and so it is one of the most effective methods measurements of particle parameter.The sample of particles is gotten by using sampler in this experiment. Images can be gotten by microscope with camera and computer.The system of particle measurement based on micro-image method is built to study the image processing with the software of VC++. The process includes preprocessing, binarization, area-filling, morphologic, edge detection and so on. Morphological binary algorithm, morphological edge detection algorithm and morphological filter algorithm are proposed in the image processing. It has adopted the technology, such as image region selecting, image demarcating, particle alternative filtration of the image, picture single analyzing, statistical analysis of the image. The micro-images of particle morphologic and fractal are applied to process of image processing. Morphologic parameters have been brought forward: area, girth, diameter, shape factor, and fractal. The result of these parameters is shown by table. The result of concentration and diameter distribution is shown by distribution figures.Experiments proved that method of image processing techniques of granularity could determine concentration and parameter distribution clearly and directly. The process of image recognition is speedy and stable.If the steps of taking samples, collecting photos and software working can be auto completed, particle sample will reflect actually the truth. Particle image auto analysis system can be built if the algorithms are gathered into software, and the measurement system can be equipped with auto image collection system.