Detection of carbon fiber plate edge based on grey relational fusion Hough transform
Wind power generation is an important part of energy structure,and carbon fiber sheet is the key composite material for manufacturing high-performance wind power blades.The edge of carbon fiber sheet is easy to appear defects because it is not covered with release cloth.In order to ensure the quality of carbon fiber sheet,it is necessary to strictly implement quality inspection standards in production inspection.Machine vision has the characteristics of real-time and high efficiency in quality inspection,which can meet the needs of real-time inspection in production line.The edge detection algorithm of carbon fiber sheet is based on industrial control computer,which realizes machine vision detection of carbon fiber sheet edge.Grey relational fusion Hough trans-form was used to detect the bonding distance between the edge of the plate and the bonding accuracy of the release cloth.In the process of edge detection of carbon fiber sheet,the image was preprocessed firstly,and the noise reduction algorithm and fuzzy edge correlation degree were combined in gray correlation to reduce the amount of data to be processed.Secondly,the edge layer and invalid layer were distinguished by gray correlation adaptive filtering,and the effective data of the edge layer were filtered.Finally,the effective edge layer gray correlation degree data was combined with Hough transform for edge parameter space operation to realize edge detection of carbon fiber sheet.The experimental data show that the number of frames per second,average error and detection accuracy meet the requirements of carbon fiber plate quality inspection.
NETL
NSTL
万方数据
