The Study On Image Processing Of Video Liquid Level Detection System

Based on image processing of video level detection system combines the image processing, pattern recognition, ARM embedded system, Linux operating system and other advanced technologies in one. The level images was collected in real-time and detect high liquid level for the field of industrial control information needed to provide accurate data to control.Level of production and testing throughout all areas of life,current detection methods have a lot of liquid, such as float, pressure, radar, ultrasonic, capacitance etc. Of level meters, they usually have certain requirements on the environment affected by environment relatively large and unstable. Therefore, the design of a high precision, stability, and the detection speed of the liquid level detection system become a top priority, and has great development prospects.This design has developed a non-contact liquid level based on the video image detection system. This paper introduces the hardware components, the transfer and the last describes the application of algorithm design level, and ARM-linux kernel and file system cross compiler. Main tasks are as follows:(1) Embedded ARM hardware was Selecteed, including the core plate and description of the underlying template, AV collection plate presentation.(2) According to the actual liquid level detection in the environment, shooting Level prepared a series of image processing algorithms finalized the design of the entire application algorithm, including image filtering, image segmentation, edge detection, morphological line detection, level positioning algorithm design.(3) Embedded Bootloader was instrduced, ARM-Linux kernel and file system of cross-compiling and transplants.Build embedded ARM application environment, to prepare for the application.Finally, in this paper, the application design is completed the simulation algorithm and the stability of the system was tested on, are good experimental results achieved the expected goal.