Research On Sand Wave Macro-Motion Characteristics Based On Digital Image Processing Technology

The granules in close contact with the bed surface become the main way of sediment movement near the bed surface;a large number of granules tend to aggregate in a specific form,forming a rich and diverse bed surface form on the bed surface-sand waves.The sand wave changes the boundary conditions at the bottom,and the mechanism of water and sand interaction is very complicated.In recent years,the rapid development and wide application of image processing technology has laid the foundation for high-frequency,highprecision shifting motion.At present,in the field of river dynamics,the development of corresponding detection systems based on digital image processing technology has become a convenient and efficient means to study the movement of the bed.This paper has carried out the test of the load-sand wave movement under the conditions of the outflow of the sink.Image processing technology acquires parameters at different moments in the evolution of sand wave motion,including morphological parameters(wavelength,wave width),motion parameters(displacement,velocity),and statistical analysis.In this paper,flume test is the main method,and theoretical research is supplementary.Quartz sand with particle size of 0.2-0.4 mm is selected as the test sediment,and the test flow condition is constant flow.The experimental study on the evolution law of sand wave movement under the condition of bend outflow is carried out.During the shooting process,according to the actual sediment transport rate in the first part of the flume in Gaza,the flume bed can achieve scouring and silting balance as a whole.The flume test monitors the fluctuation of water level.A highdefinition camera connected with a hard disk video recorder is set up above the flume to take video pictures of the test process.A larger shooting range is selected to obtain the Macro-motion information of sand wave,i.e.the change of sand ridge along the course after the formation of sand wave.(2)Sand Wave Detection System Based on Digital Image Processing TechnologyAiming at the moving image of bed load sand wave in complex background and around the key scientific problems,the existing research results of bed load transport are widely collected.The existing methods of tracking and identifying bed load are improved by means of data analysis and theoretical analysis,so as to meet the analysis requirements of image sequence in the study of bed load sand wave motion in this paper.Image recognition technology and bed load sand wave transport are applied.Based on the experimental study of dynamic evolution law,a bed load sand wave motion research system based on digital image processing technology is developed.Finally,the identification or extraction of sand wave shape parameters(wavelength,wave width,etc.)and sand wave motion parameters(displacement,velocity,etc.)are realized.(3)Research and Analysis of Sand Wave Movement under Bend Outflow ConditionsThe cross section of sand wave bed is divided into three velocity regions: A,B and C.Using digital image processing technology,the profiles of sand wave edges at different times(different beds)are extracted,and the relevant parameters of sand wave are obtained.Statistical analysis is used to study the physical process of sand wave development with time under the condition of bend outflow.1.The longitudinal and transverse deformation characteristics of sand wave development with time in different regions.2.The fluctuation of the overall velocity of bed load sand wave in different regions with time.In the experiment,the image data of sand wave motion evolution in a period of time were obtained by shooting,and the dynamic sand wave bed surface at different times was extracted by a series of static images.Then,the contour line of moving sand wave was extracted by digital image processing technology,and the continuous macroscopic motion characteristics of sand wave under the condition of non-stop water were studied.The main results and conclusions are as follows:(1)In the experiment of sand wave movement evolution in a large Ushaped flume,the method of top photography is proposed.In the past,the side shooting method only studied the wave height and wavelength of sand wave,but ignored the development and change of wave width.The top shooting method made up for this defect and could obtain a large range of complete bed shape simultaneously.Especially suitable for the study of the three-dimensional development law of sand wave movement,which has certain innovative significance compared with previous studies.(2)Aiming at the problem of sand wave edge contour extraction,a complete sand wave edge detection system is developed.Several commonly used edge detection operators(Prewitt,Sobel,Log,Canny)are used to detect sand wave edge.Each operator tries different thresholds.When the threshold is large,the edge information will be missing;when the threshold is small,a large amount of background noise will be extracted,which will interfere with the results;therefore,the key to extract sand wave edges is to precisely set the threshold.According to the comparative analysis of the results,it is shown that Canny operator is more suitable for the detection and extraction of sand wave edge in complex background.(3)Statistical analysis of a large number of parameters extracted from the sand wave motion sequence images shows that for the dynamic sand wave bed at different time,the probability distribution of wave width in different flow velocity regions obeys the normal function,and the wavelength is positively correlated with the flow velocity.(4)Under the action of current,the average velocity of sand wave fluctuates in a certain range when it moves forward in a whole way.In the region with fast velocity,the variation of wave velocity is larger than that in the region with small velocity.