The Study And Application For A Liquid Level Detection System Based On Image Edge Extraction

The topic of this paper achieves a level detection the actual system combining image processing technologies and embedded technologies, through a variety of testing environments, a variety of image processing technologies and embedded technologies research, pointing shortcoming of the traditional level detection technologies. This paper proposes a suitable method to locate the target of this system, based on the existing targeting method, through studying the principles of level detection by image processing techniques, pointing to the relevant requirements of image processing techniques and goals (level) for the orientation. This paper improves the corresponding algorithm based on the actual problem through researching and analyzing the target location of the existing algorithms in different stages. The focal point of this paper is the regional level goal orientation and precise positioning. Thispaper mainly includes:1. First of all, paper introduces the related background, outlining the edge image processing technology and basics of detection algorithm. Around the importance of testing the level, development status, paper introduces liquid level detection device based on image processing technology, and one of the key technology edge extraction. Then around the image edge extraction techniques paper discusses and analyzes the image boundary extraction problems.2. In the level of regional positioning stage, paper proposes an improved Prewitt operator on the initial image edge detection because of traditional Prewitt edge detection operator for image edge extraction is not strong vertical defects. Paper raises a binary image improved mathematical morphology algorithm regional orientation. The improvement of the morphological double structure element is used for the system design characteristics of the image. Simulation results show that the proposed algorithm for the environment, adaptation to different light well and fast computing algorithm.3. In Level of precision positioning stage, according to the characteristics of the target image, and in order to achieve the purpose of reducing computation, based on the traditional Hough transform curve, this paper uses standardized objective method to reduce the complexity, to achieve accurate detection of liquid level. The resulting target curve often be intermittent, not continuous, it needs to get continuous curve fitting target curve. Simulation results show that the algorithm to detect speed, high accuracy, reliability, real time to meet the system requirements for hardware implementation.4. A more detailed description of the proposed algorithm is applied to detect the actual level of embedded systems software and hardware design and implementation, the actual operation of the system test also proved that the improved algorithm. This detection and tracking algorithm is successfully achieved in the embedded system platform level.