| With the continuous progress of weaving technology and improvement of people's living standards,people have put forward higher requirements for the quality of raw silk and silk fabrics.As one of raw silk import-export powers,it is of great significance to strengthen further study on the method of raw silk inspection in order to ensure the leading position of China's raw silk in the international market.Based on analyzing and summarizing the existing methods of raw silk inspection,We present a method for measuring the evenness and structure tightness of raw silk based on binocular vision.The detection method has the advantages of high objectivity,intuition and accuracy,which can provide better guarantee for the inspection of raw silk appearance quality.An binocular Vision-based raw silk evenness and structure tightness detection system is proposed in this paper.The system is comprised of an image acquisition system with two CCD image sensors,two telecentric lens,two image capture cards.a over feeding device,a light source,a signal generator and a graphic workstation,plus an image processing system using Opencv programming method based on c++ and tasked with image smoothing,threshold segmentation,morphology,Pixel access and other image processing operations.Based on the establishment of the system,the overall performance of the system is verified through experiments.The results show that the detection system has good repeatability,stability and accuracy.Firstly,the images of two orthogonal directions of raw silk are obtained by image acquisition system,and the diameters of two directions are extracted by image processing system.In this paper,the cross-section shape of raw silk is regarded as an ellipse.The two diameters are regarded as long and short diameter of the ellipse.The section areas of raw silk can be obtained through the method of calculating the areas of ellipse,the coefficient of variation of section areas(CV value)is used to characterize the evenness of raw silk(In the following article,the evenness index calculated by section areas is called "bi-directional evenness" for short and the CV value of cross section areas is called "bi-directional CV" for short).Next,an experiment which refers to current international standard ISO 15625 and takes "bi-directional CV" as an indicator is conducted to determine the sampling capacity of the evenness test system.Then,we compare the detection system with Uster ME100 in evenness test,the results show that the difference among ME100 evenness test result,uni-directional CV value of our detection system,bi-directional CV value of our detection system is large and the value increases in turn,but there is a significant correlation among three of them under the significance level at 0.01.On the basis of the above system,a method for detecting the tightness of raw silk structure is put forward,and characterizing the structural tightness of raw silk with the "bulkiness" indexes in this paper.First of all,we describe the definition of three "bulkiness" indexes and its specific implementation methods in detail.Secondly,in order to analyze the occasional low cohesion phenomenon of the raw silk,This paper makes a study on the raw silk "bulkiness" of short segment.Finally,we explore the relationship between " bulkiness" and raw silk cohesion.The experimental results show that three "bulkiness" indexes have a significant correlation with raw silk cohesion.In addition,the regression equation of three "bulkiness" indexes and the score of raw silk cohesion(detected by Duplan tester)is determined respectively by further data analysis.The experiment not only verify the scientificity of the "bulkiness" indexes,but also provide a new way of thinking for the inspection of raw silk cohesion. |
PDF Full Text Request