Raw silk or soaked silk appearance quality detection device and detection method based on binocular vision

Abstract

The invention discloses a raw silk or soaked silk appearance quality detection device and detection method based on binocular vision. The device comprises a motion control system, an image collection system and an image processing system, wherein the motion control system is used for driving raw silk or soaked silk to move along the vertical axial direction of the motion control system; the image collecting system comprises two cameras, and the two cameras are used for respectively collecting a first horizontal radial image and a second horizontal radial image of the raw silk or the soaked silk; the image processing system comprises a storage module, an analyzing module and a computing module; the first horizontal radial image, the second horizontal radial image, a theoretical diameter value of the raw silk and two equations are stored in the storage module, the analyzing module is used for obtaining a first horizontal radial average diameter value D1 and a second horizontal radial average diameter value D2 of the raw silk or the soaked silk, and the computing module is used for computing to obtain a roundness value and a bulking value. By means of collecting the first horizontal radial image and the second horizontal radial image of the raw silk or the soaked silk, on one hand, the roundness and a bulking degree of the raw silk or the soaked silk are obtained; on the other hand, appearance quality of the raw silk or the soaked silk can be reflected more accurately.

Description

technical field [0001] The invention relates to the field of textile inspection, in particular to a device and method for detecting the appearance quality of raw silk or soaked silk based on binocular vision. Background technique [0002] my country is a big country in the production and export of raw silk, and with the development of the economy and the improvement of people's living standards, people have put forward higher requirements for the quality of silk fabrics. In order to improve the quality of raw silk in my country and enhance the competitiveness of raw silk in the world, it is extremely important to strengthen the research on raw silk detection methods and equipment. [0003] The current raw silk appearance quality inspection methods are mainly blackboard inspection method and capacitance inspection method. The blackboard inspection method is an artificial inspection method that originated in the early 20th century. It is easily affected by the emotion and expe...

AI Technical Summary

Problems solved by technology

[0003] The current raw silk appearance quality inspection methods are mainly blackboard inspection method and capacitance inspection method. The blackboard inspection method is an artificial inspection method that originated in the early 20th century. It is easily affected by the emotion and experience of the inspectors and has poor repeatability; capacitance inspection The method is represented by the Uster evenness meter. Its obvious disadvantages are: high cost; easily affected by environmental conditions, the test needs to be carried out in a specific environment; it can only reflect the size of the defect but not the type of the defect

[0004] In addition, with the continuous development of photoelectric technology, photoelectric detection methods have been applied to the appearance quality detection of raw silk, such as literature 1 (Liu Fengjiao, Zhou Wang. Non-contact real-time dynamic measurement of raw silk fineness[J]. Textile Journal, 2012,33(11):P127-131.) announced a CCD imaging system using a dark background mode, which uses a laser as a light source and a CCD image sensor as a receiving element, and the image signal is converted into a digital image by an A / D converter. signal, and then through the processing of the digital signal and the analysis of the standard raw silk fineness imaging curve, the relationship between the raw silk fineness and the imaging curve is obtained. This method obtains the relationship between the raw silk fineness and the imaging curve by indirectly processing the collected digital signal, and There is no intuitive information on the appearance of raw silk, which belongs to an indirect processing method of photoelectric method, and the error is large, which is not suitable for the precise detection of raw silk

Document 2 (Fei Wanchun, Zhu Yayi, Cui Xiaoli, et al. Research on photoelectric detection method of raw silk cross-sectional area [J]. Textile Journal, 2002, 23(3): 17-19.), proposed a method using two The method of measuring the cross-sectional area of ​​raw silk by an image sensor is a method that uses the cross-sectional area instead of the fineness to characterize the evenness of raw silk, but its data processing method is the same as that of Document 1. Direct processing of image signals is an indirect processing of photoelectric methods method, and it only gives the evenness index of raw silk, and does not consider other detection indexes such as defects and cohesion of raw silk

Document 3 (Tu Wenxiao. A detection method for the appearance quality of raw silk: China, 103257145 [P]. 2013-08-21.) introduces a system based on CCD camera imaging and image processing. The raw silk image is taken, and then the evenness and defect indicators of the raw silk are obtained by image analysis of the raw silk appearance picture. This is a direct inspection method of the raw silk appearance quality, but it does not consider the actual appearance characteristics of the raw silk, that is, the appearance of the raw silk. The part is elliptical, and the ratio of the short diameter to the long diameter is about 0.72-0.76. The single-direction imaging cannot reflect this structural change of the raw silk, and the method of simply applying pretension cannot solve the problem of raw silk well. jitter problem, so this method cannot accurately characterize the appearance quality of raw silk

Images