Polarization Image Processing And Effect Evaluation For Machine Vision Inspectio

With the development of industrial production speed,the quality of products is also improved.In order to meet the needs of the continuous development of the industry,many equipments for machine vision inspection systems have been developed,and their advantages lie in non-contact inspection methods.Because sterile syringes can avoid cross-infection,easy to use,economical,and help reduce the workload of medical workers,they are favored by the majority of medical workers.However,defects such as cracks and incomplete screen printing may occur in the manufacturing process of the syringe,which seriously affects the safety of use.At present,the commonly used digital image processing technologies in machine vision are mainly grayscale,binarization and morphological processing,but the traditional image processing technology will lead to the lack of effective information for the treatment of surface cracks in the syringe barrel or foreign bodies in the tube,such as the crack information will be covered by background noise.In response to these problems,this paper combines the machine vision imaging system with the polarization imaging technology,designs a non-intuitive optical imaging device for machine vision inspection.The device includes main structure design,hardware design and assembly,polarization device design and software design.It mainly includes solving the driving and control of the image acquisition system,the design and control of the liquid crystal phase retarder,the design of the lighting system,and the processing of polarized images.The article combines dark field lighting and oblique lighting to design and complete the lighting system,In addition,the article also introduces the related digital image processing technology in detail.In the image preprocessing,the median filter algorithm is used to smooth the image,and after the histogram correction,binarization and morphological processing are used,the effective information of the syringe barrel crack and the foreign body in the pipe is missing.This paper introduces polarization imaging technology and uses the theoretical model to obtain a set of polarization parameter maps including depolarization map_{（9）Stokes parameter maps0,1,2,3 and sample parameter maps sin,,and linear polarization parameter maps0,7}.The results of the contrast between the original image and the polarization parameter show that:compared with the original image and the preprocessed image,the contrast of cracks and foreign bodies in the syringe is significantly improved.different defect types have different sensitity to different polarization parameters,including the polarization parameters sinandimageing effect of syringe surface scale and number is better.