Research Of Droplet Size Measurement System Based On Computer Vision

Droplets size is an important index for estimating the performance of equipment for crop protection. In the paper, on the basis of computer vision and gray system theory, the droplet size automatic measurement and optimization system were established. The main work included in the thesis was shown as follows:(1) According to the droplet image characteristic, an adaptive mean filter algorithm and a segmentation method based on the optimum threshold were proposed. Droplet image edge detection algorithm of mathematical morphology and operator for edge detection were studied. The experiment results showed that these methods are effective improved algorithms, and the droplets were detected and labeled and the droplet parameter were measured according to the above results.(2) The conglutinated droplets were segmented based on the watershed algorithm of distance transform aiming at the conglutinated droplets in the droplet image. A distance transform optimization algorithm was proposed, and the algorithm efficiency was greatly improved. Grayscale reconstruction algorithm was applied to avoid image over-segmentation phenomenon. The experiments showed that these methods were effective.(3) In establishing the model of forecasting droplet size, the gray theory was applied. An example showed this method was proved to be precise by the droplet size forecasts and was helpful to choose the other parameters of spray system within a certain droplet size scope.