Liquid Level Detection Based On Digital Image Processing

Currently, there are various ways to detect liquid level automatically, such as Differential Pressure Level Measurement, Float style level measurement, Sound level measurement, capacitive liquid level measurement, there are different applications in different environments. In order to simulate tidal movement in the lab environment, A research unit need to design a automatic liquid level detection system with high-precision. This dissertation proposes a non-contact liquid level auto-detection program based on digital image processing for the special application. The technology, based on digital image processing, provides a new way for automatically detecting the water level and enriches the content of the image processing.In this dissertation, the solved key technology problems and the proposed methods are as follows:(1)The overall structure of the level automatic detection system is designed according to specific application objects and planning.(2)The dissertation proposes a new way for automatically detecting the water level based on digital image processing. In order to get a high-precision automatic liquid level detection system, the water level video captured by CCD will have a series of processing, including Video frame extraction, Noise Filtering, Objective edge detection, Level Line Orientation, Template matching, Curve fitting.The method proposed in the dissertation has been programmed and implemented on Visual Studio 2005, and the experiments show that the system can detect the water level in real time, detection error is less than 0.2mm, and the system runs stable, and can be able to overcome the light change impact in a certain range.