人工智能-图像处理-基于图像处理技术的油管识别系统的设计与实现.pdf

Pipeline is the most important industrial transportation product in petroleum

industry, it is used to transport and mining oil, the amount of it is very large, so it is

particularly important to use and manage effectively. According to statistics, the

proportion of oil leakage or explosion accidents is not small that caused by blowout,

corrosion or service extended and so on, which has caused huge economic losses. To

improve this situation, many industry use the different mark on the pipeline to

distinguish and manage them, through the study of the recognition, we can realize to

sort the pipeline effectively based on the mark that stands for pipeline’s attribute.

With the development of science and technology, the realization of automation

identification system to replace the human eyes to finish the work of identifying the

pipeline. Based on the reference of relevant data and image processing technology, we

use Java programming to achieve the identity recognition system. Through the image

processing of the pipeline mark, we can get the information, and the system can

identify the effective part, at last, the system exports the correct coding of the mark,

we imported a picture of the mark, the system output the correct coding and has a

better result.

In this paper, the design and implementation of recognition system is mainly

completed the following parts:

(1) The sample picture of pipeline is completed the digitization and the computer

technology of image segmentation to detect the edge of signal, at last, the

background and target become smooth, the edges become sharp.

(4) After binarization processing on the segmentation area, the system gets the

coding of “1” and “0”, through the “回”type locator position, the system

output the binary code in the middle of locator area.

(5) Based on the research of original system, this paper also uses another

method to design the identification system, at first, the system uses Sobel

operator of image processing for the poorer picture, from which we can see

the result after removing the noise is obviously, it lays a foundation for

subsequent identification.