Detection Of Defects In Steelwire Ropes For Oil Well Based On Image Processing

Abstract:The steelwire rope is really important in the national economy. It is widely used in mining, machinery, construction, petroleum, coal and some other areas. The steelrope is not only be widely used in the main industry area, but also play an important role in people’s daily life.All the time, people express concern about the security issues of using the steelwire ropes. In the oil and gas drilling platforms, people commonly use artificial detection to decide whether to replace the rope or not. The ropes working condition does not be taken into consideration so that there will be a great waste of resource. In order to confirm the span of the replacement and prolong the rope life, the scientists are trying different ways to detect the defects of the steelwire ropes. This paper is using a digital image processing method which is based on texture feature of steel cable to detect the fracture of steel wire. At first, using Retinex filtering method to eliminate environment heterogeneous shining. Then to pick up the steel line body image by the edge detecting and section counting filtering method. Then by dilate and erode the edge image and Hough line detecting method to pick up the steel wire bunch part. By using an improved Radon transformation method to suggest if those steel wire are in good condition or not. Finally, by using BP neural network model to judge the final result. Test result shows that this method is easy to use and fulfill real time request.Finally, by using BP neural network model to judge the final result. Test results show that this method based on image processing is easy to use and fulfill real time request with high accuracy and good reliability.