Research Of Yarn Hairiness Parameters Measurement Technique Based On Image Analysis

In yarn quality inspection and control of the yarn,the yarn hairiness is an important index to evaluate the quality of the yarn.The detection of hairiness can avoid the influence of poor hairiness in the post-processing process and also improve the quality of the yarn and the yarn fabric.In order to meet the control and testing of hairiness,this paper designs a suitable yarn image acquisition device and its hairiness parameter calculation algorithm for the characteristics of the yarn in motion.Firstly,the image acquisition device of moving yarn is designed,which can be divided into yarn transmission system and imaging system.It can meet the speed requirement of 30m/min for the transmission system,and realize uniform and steady movement of the yarn.For the imaging system,the linear CCD camera with high scanning frequency and high resolution is selected as the image sensor.Combined with the telecentric lens and the parallel light source,the image acquisition of the motion yarn is realized,and the image quality is superior to the general imaging system.Secondly,the background difference is selected to remove the fixed noise in the background,which highlights the gray characteristics of the yarn.Through the image gray scale analysis,the 10%-35% of maximum gray value is determined as the threshold range of image segmentation,and the binary operation is performed on the image.The positive direction structure with size 13 operators is used for morphological open operation,the hairiness image is indirectly extracted.The refinement algorithm in Matlab is chosen,the line of hairiness is refined,and the topology structure of hairiness is saved.Finally,the yarn image acquisition method with 1/3 matching line frequency is proposed in the hairiness parameter algorithm.By analyzing the hairiness pixel numbers matching the line frequency image and 1/3 matching line frequency yarn image,the value of a in hairiness value calculation is put forward.Based on the center line of yarn,the equal interval cutting line of hairiness on both sides is made,the number of feather roots on the cutting line is counted,and the feather index is given.The test results show that the a of the same experimental environment is a fixed value,and the hairiness values calculated from the yarn image with 16667 Hz of the row frequency are consistent with those of the yarn images with 50000 Hz of the row frequency.At the same time,in the calculation of hairiness index,it can accurately identify the length of hairiness root number greater than or equal to 1.25 mm,which can effectively identify the hazards of long hairiness.