Research On The Image Measurement For Vortex Parameters In River Model

With the coming of the era about information,the interdisciplinary development have becomeincreasingly common,so that people use the computer technology to improve the technical content of waterindustry becomes an important trend under the background of technology convergence for waterindustry.Now the techology of computer image measurement has be more and more applicated in the rivermodel test,but more applications focuse on the tests of surface flow field and river regime,to monitor theharmful vortex on the surface of the water inlet in the reservoir model by computer image measurement isstill rare.This paper summarizes the algorithms of image measurements for river model test,and proposes aset of solutions for measuring the vortex flow field parameters based on the computer image processing.Itcan be used to judge the hazardous about the vortex flow field generated in the river model experiment.In this paper,a monitor platform for observating the vortex flow field would be built,it is a part ofthe monitoring system for the model of Yellow River and the paper in detail introduced the vortex flowfield image measurement system,and the method of measuring the range and radius of the vortex and ofextrcating the tangetial velocity about the vortex surface.First of all, through the digital video camera which has the network routing functions captured aset of vortex images,and transmitted them through the network to the main processing computer.Thenprecessed the vortex images by denosing and compressing.At last,according to different purpose usingdifferent image processing methods to extract the relevant vortex parameters.In the part of vortex imageacquisition and transmission,this paper put forward a compression algorithm based on the EZW imagecoding algorithm improved by the rule of “entropy”.It could satisfy the requirement between the imageresolution and network bandwidth.At the same time,it also makes the output codes have the characteristicof sparse signal,provided a possibility for achieving the lastest compressed sensing technology and greatlyimproving the efficiency of image transmission to satisfy the demands of real-time in river model test.When measuring the vortex radius,according to the characteristics of the vortex image whichafter pretreatment,firstly stretched the image contrast,then through comparing the typical edge detectionoperator,this paper choosed the LOG operator to detect the vortex image.At the same time, put forward aeliminating false contour method based on a spiral texture feature to realize the accurate vortex boundary extraction.Finally,used the plane figure coordinate formula to calculate the vortex center and according thenumber of pixels from the vortex center to the boundary to calculate the vortex radius.In order to measure the vortex surface tangential velocity,this paper used the PTV techniquewhich can match a single tracer particle track.Throw selected tracer particles on the surface of the vortexand measure the displacement of the reference particle in the time interval between two adjacent images,sothat we can get the tangential velocity by the particle’s velocity.Before measuring,we compared the typicalmatching algorithm of the PTV technique by the MATLAB simulation.Through comparing the efficiencyof the four frames method,PCSS,SPG and VGT matching algorithm used in vortex flow field,it wasverified that VGT method had the ideal particle matching results and choose it as the PTV matchingmethod.