Font Size: a A A

Design Of Intelligent Cloth Pattern Cutting System Based On Machine Vision

Posted on:2019-02-18Degree:MasterType:Thesis
Country:ChinaCandidate:H M TaoFull Text:PDF
GTID:2428330566472238Subject:Control engineering
Abstract/Summary:PDF Full Text Request
With the improvement of people's living standards and the pursuit of fashionable cloth,the demand for trendy clothing is increasing day by day,and the types of clothing patterns are becoming more and more abundant.In clothing production,many factories still rely on manual cutting for the pattern of fabrics,resulting in single varieties,poor technology,low efficiency,and high costs,which cannot adapt to the rapid changes in the market.The advancement of machine vision and image processing technology has promoted the development and demand for smart cloth cutting equipment.Its application has reduced production costs,improved the production efficiency and the quality of product.According to the user's requirements,the machine vision system is installed on the laser cutting machine,the image processing technology and the new algorithm are used to realize intelligent extraction of fabric pattern cutting lines and automatic cutting of fabric patterns.The specific research content is as follows:System design and hardware system setup.Based on the research of related technologies in China and abroad,based on the design requirements put forward by users,the article designed the system hardware structure,image processing flow and software architecture,also selected the hardware devices and built the hardware system.Camera calibration.The imaging principle and the world coordinate system,the camera coordinate system,the pixel coordinate system and the image coordinate system as well as the conversion relationship between them are studied in the article.The camera calibration of the system is completed,and the camera's internal and external parameters are obtained.The distortion of the cloth image is corrected,the image coordinates are matched with the laser cutting device,the extracted cutting line data can be accurately transmitted to the laser cutting machine,and the laser cutting is completed on the platform.Research on extraction algorithm of cloth cutting lines.1.Combining the advantages of mean and median filtering,a hybrid filtering method is proposed,which can reduce the image noise while protecting the edge pixel information.Compared with the median filtering algorithm,it greatly reduces the time-consuming of the algorithm and satisfies the efficient of cutting machine.2.Aiming at the image of long-width fabrics,this article study the image mosaic algorithm.The SIFT registration and spatial domain image fusion are used to realize the image mosaic of widewidth fabrics.3.Apply a new color cone and circular table segmentation method,through the use of hyperplanes in the RGB color space on the fabric pattern for fine segmentation,and achieved a good segmentation effect.4.Two cutting line extraction methods are designed: edge detection and template matching.The edge detection method uses the morphology and the Sobel method to detect the edge,and then performs the maze search and approximates the polygon algorithm to optimize the edge.In the template matching method,first establish a pattern template to match,and then extract and optimize the matched pattern cutting line,the deformed template are used to match the pattern when the fabric has a wrinkle.The two cutting line extraction methods can be used individually and together.System integration and analysis.The interface display and control program was designed.Through the integration of hardware and software functions,a complete cloth pattern intelligent cutting system was developed.Since the system was put into use,it has performed well.Compared with the manual cutting cost analysis,equipment investment recovery is fast and has a good promotion and application value.
Keywords/Search Tags:machine vision, image processing, cloth cutting machine, cutting line extraction, image segmentation
PDF Full Text Request
Related items