Parameters’ Measurement Of Micro-bubbles In Water Based On Digital Image Processing

In the applications of wastewaters treatment technology, the micro-bubbles are oftenused to remediate the volatile pollutants in the aqueous phase and separate the particulatematerials from the aqueous phase. The efficiency of the treatment of wastewaters hasdirect relationship with the diameter, speed and other parameters of the micro-bubbles inwater. Therefore, parameters’ measurement of micro-bubbles in water is particularlynecessary. Image method as a non-contact measurement method of intuitive, instantaneousand high precision was more and more applied to the parameters’ measurement of themicro-bubbles in water. On the needs of the practical applications of wastewaterstreatment technology, the following work has mainly to be done in this paper:Firstly, the bubble image acquisition device is designed in laboratory conditions, andthen the detailed analysis about the factors of affecting the quality of the image acquisitionand the measurement error in the processing of experiment were carried out.Secondly, the principle and implementation of digital image processing algorithmswere detailed elaborated. Different digital image processing methods were used in sparsebubbles image taken in laboratory, whose effects of processing were contrasted. For thecharacteristics of the image in this paper, first,the image was removed the background bySubtraction. Then the enhanced image was obtained by reducing image grayscale andmedian filtering.The iterative method, Otsu method and two-dimensional histogrammaximum entropy method were applied to threshold segmentation. Relatively betterbinary image was obtained by Contrasting the effects of the three threshold segmentationmethods.Thirdly, the preprocessing methods of the sparse bubbles image were summarized,horizontal scanning method was applied to measure the diameter of the bubbles for theimage of pre-treatment, and then the area and perimeter of the bubble were obtained by thediameter. The concentration of the bubbles was obtained by the statistical ratio of bubblepixel attributable to the total number of pixels of the image. An improved nearest neighbormethod and particle tracking velocimetry (PTV) were applied to measurements of thebubbles’ speed. Finally, bubbles images of a high density shot in laboratory were preprocessed. Thespeed of the high density of bubbles were measured by particle image velocimetry (PIV)based on the fast Fourier transform and an improved method proposed in this paper,whichcombined the PIV method based on wavelet multi-scale analysis with sub-pixel fittingmethod based on the3×3Gaussian method.