Research On Online Screening System Of Defective Baoyu-Flavor-Slices

Qualified Baoyu-Flavor-Slices have no obvious pores,the fillings and wrapper are complete and the wrapper thickness is uniform,while there are three main defects in the defective Baoyu-Flavor-Slice products: external contour damage,uneven thickness of the wrapper,and large hole defects.At present,enterprises still use manual inspection and elimination methods to complete the screening of product appearance quality,and the screening criteria are difficult to quantify,and it is very easy to miss inspection and error detection.In this paper,an online screening system based on machine vision was developed according to the appearance characteristics of Baoyu-Flavor-Slices,so as to realize the online identification of defective products and the classification of defective products of BaoyuFlavor-Slices.This in-line screening system for defective products provides big data support for process improvements in the pre-sequence process.The research content of this article is as follows:(1)The research of the detection principle mainly includes: through the analysis of the product characteristics of Baoyu-Flavor-Slices,the design points of the material separation,carding and rejecting device before the detection and identification are determined to ensure the design reliability of the system hardware device;according to the type defects and characteristics of Baoyu-Flavor-Slices,Compare commonly used edge extraction operators in the extraction of Baoyu-Flavor-Slices,and find the best Baoyu-Flavor-Slices feature detection method.(2)The design of system hardware mainly includes monolithic separation device,carding device,image acquisition device and rejection device.The monolithic separation device realizes the monolithic Baoyu-Flavor-Slices after slicing,so that the Baoyu-FlavorSlices adhering to each other are separated.The lifting mesh belt transports and organizes the disorderly Baoyu-Flavor-Slices into an orderly,single-layer array arrangement.The image acquisition device collects the image of the Baoyu-Flavor-Slices product on the secondary conveyor belt,and then after the feature detection and identification is completed,the execution element of the reject device completes the corresponding rejection action according to the detection results to achieve the rejection of the unqualified product.(3)The design of the system software mainly includes image acquisition and preprocessing module of Baoyu-Flavor-Slices,feature extraction module,and defective product rejection coding.The image acquisition module collects the image of the Baoyu-FlavorSlices in real time,and the image preprocessing module preprocesses the image of the ROI region to improve the contrast of the target image.By extracting the characteristics of BaoyuFlavor-Slices and comparing them with the characteristics of qualified products,the product testing and defective product classification are completed.(4)Experiments and data analysis,including Baoyu-Flavor-Slices sorting and sorting test,image acquisition system calibration experiment,wrapper thickness statistics experiment,screening experiment,through the analysis of experimental results,to demonstrate the reliability and shortcomings of the system.The online screening system of Baoyu-Flavor-Slices defective products studied in this paper can accurately detect and eliminate Baoyu-Flavor-Slices online,which not only reduces labor costs,improves work efficiency and detection accuracy,but also realizes the quantification of Baoyu-Flavor-Slices testing standards and moves towards intelligent food production.