Research On The Water Level And Velocity Measurement System Based On Image Recognition

China is a country suffering from drought,water shortage and uneven distribution of water resources.Therefore,the detection of water level and flow velocity is of great significance.Through these two indicators,the water resources can be strictly controlled and reasonably distributed in areas where water resources are scarce.Therefore,the utilization rate of water resources should be improved.In areas with excess water resources,the security alerts can be set to remind people from danger in special circumstances.So the people's life and property can be protected.In order to solve the problems of turbid water quality,backward communication mode and equipment maintenance difficulties in the remote areas of northeast China,the image recognition technology is used to measure water level and flow velocity.And a new way to measuring water level and flow velocity based on image recognition is proposed.This way not only can measure the water level based on the new water gauge,but also can measure the flow velocity based on the floats of surface at part of the time.At the same time,the measurement results and the image data can be sent to the remote control center through the GPRS wireless transmission technology.So the data transmission can be completed.The main contents of this thesis include the following aspects:(1)The algorithm based on image recognition is researched to measure water level.According to the improved binaryzation algorithm,the location of water level can be realized.Through the positioning and identification of the new scale,the water level can be calculated and error compensation are realized.So the system's measurement accuracy is improved.The measurement error of the water level is less than 4mm.(2)The algorithm based on image recognition is researched to measure flow velocity.It is mainly aimed at the situation of floating objects on the surface of the water.The recognition area of floating objects based on the result of water level measurement is extracted to reduce the running time of the algorithm and enhance its practicability.Combining the frame difference method and the tracking algorithm based on MeanShift,the recognition and localization of the target are realized.So the flow velocity can be calculated.(3)The application of GPRS module is researched.The data transmission rules is formulated.The communication is achieved based on the TCP protocol.To solve the problem of sending image data,the data compression and format conversion are used to achieve image transmission rapidly.