This paper presents a novel microscope spray droplets detecting technique based on image processing method, which is applied to an experimental measurement performed on a flat fan atomizer nozzle. The droplets images are detected by an image acquisition system. The raw images are pre-processed to enhance the images and avoid the illumination unevenness effect. An image restoration method based on iterative blind deconvolution is applied to improve the quality of the blurred image due to droplets movement and defocus. Circle Hough transform and the least square method are applied to find the droplet contour in the image, and the droplets diameters are calibrated and counted. A criterion is presented to avoid repeated statistic problem. The measurements of water spray droplet size distribution from a small flat fan nozzle atomizers are taken and the experiment results are analyzed.
A novel method for measuring droplet size and complex refractive index simultaneously using a CCD camera by rainbow detecting is proposed. A new mathematic model for rainbow pattern of absorbing droplet is built. Based on this model, a series of new formulas to measure droplet imaginary part of refractive index are derived. Then a new method for simultaneously measure droplet size and the complex refractive index is presented, which is verified by simulation experiments under different conditions. The experiment is performed to measure water with different dye concentrations. To avoid the effect of non-sphericity on refractive index detecting, a long and stable water cylinder instead of droplet is measured. Both the diameter and the complex refractive index of the liquid cylinder are detected by measuring the scattering rainbow light which is received by a linear CCD camera placed in the focus of the lens system. The result shows a satisfactory agreement with the theoretical analysis.
Monochromatic rainbow measurement, as a laser based interference detecting technique, is always used to measure particle refractive index. However, the accuracy of this method is effected by particle non-sphericity. In this paper, the non-sphericity effect is analyzed from both simulation and experiment. The non-sphericity effect for both prolate and oblate spheroid is studied. The simulation result shows that the axial ratio greatly effects the rainbow angle position, and eventually effect the refractive index evaluation. An experiment is also taken. The refractive indexes of ethanol solutions with different concentration are measured by monochromatic rainbow method. A comparison is taken between the results measured from the rainbows formed by a liquid cylinder and rainbows of free falling droplets.
The characterization of spray behavior especially the droplets parameters are very important for many industrial
applications. In this paper, the droplets parameters of a flat fan water spray were measured using a new developed global
rainbow refractometry. the global rainbow mechanism of spray is studied in detail. The effects of discrete parameter,
distribution function, mean diameter and disperse coefficient on global rainbow of spray are concerned. a new rainbow
parameter inversion algorithm of particles is proposed. This algorithm can inverse both particle diameter distribution and
refractive index simultaneously, without pre-knowledge of droplets diameter distribution function and range. The
algorithm removes the noised using empirical mode decomposition (EMD) method, and inverses the data based on
regularized NNLS and regularization method. The experimental measurements are performed on a flat fan air-blast
atomizer spray, and the droplets size distribution and the refractive index have been measured under five different gasliquid