Pipe Counting Method Research Based On Digital Image Processing

Pipe automatic counting system which is gathered with optics, mechanology and electronics is developed to solve the record and statistics problem of pipes during the transportation. Its main working principle is to use an industry digital camera to acquire pipe end face image, and then processing the image and recognize pipes, achieve the purpose of pipe automatic counting. Based on the specific project of oilfield pipe entry counting system for practical application background, this paper research relevant digital image processing and application problem.First, with analyzing the entire pipe counting process, design the overall realizing program. Including the use of image stitching technology to splice multiple camera images collected complete pipe end image, the pipe end face image preprocessing to get binary image, binary image processing object to the use of the identification method to identify the pipe region three stages.Then, for each stage research, propose specific methods in the paper. In the splicing stage, the first use of the SURF algorithm to extract the feature points and matching, the right image perspective transformation is completed to splice through a combination of the RANSAC algorithm to obtain an accurate projection matrix; according to the characteristics of the pipe end face image, in the image pre-processing stage, median filter to remove interference noise points, homomorphic filtering method to improve the uneven illumination image, and then use the Otsu binary separation of the need to identify the target area and background area, and finally dilation and closing operation to further enhance the quality of the binary image; through the pipe end face of the binary image analysis, pipe identification method using a combination of fuzzy means clustering and least squares ellipse fitting, the first area cluster analysis excluded smaller interference area, then use the ellipse fitting method to identify the target area on the image after the clustering process.In addition, this paper discusses that the latest.NET algorithm parallel programming technology used in pipe recognition counting algorithm is proposed to improve the program in the running speed of nuclear computer. For real-time demand pipe automatic counting system, it certainly has very vital significance.Based on the above research and design to the preparation of computer program tested in the laboratory prototype system and a large number of tests to verify the feasibility of the algorithm...