An Beverage Closure Defect Detection Design Using Reconstruction Of Algorithms Based On Machine Vision Technology

In beverage bottle production process, if the location of capping machine adjusted error, there will be a variety of styles of cap defects, it will seriously affect the quality of the product and brand benefits. But the traditional method of bottle quality inspection is inefficient, so the method should be renovated urgently.According to the feature of cap defects, a function modules can be reconstructed bottle defect detection system is designed based on machine vision and digital image processing after considering the precision, efficiency and cost.The major contributions in this dissertation are as follows:1) In order to make the thread of image acquisition, image processing, defect detection and results of real-time display can be concurrent execution, improves system efficiency and reduces hardware requirements, the system software develops using multi-threading technology.2) For the sake of overcoming those disadvantages which exist in traditional visual inspection software, for example, the reconstructing performance of software is poor and functional structure is single, the system using reconstructing development method. Users can quickly construct software features according to the new requirements by using the graphical tool modules in the interactive interface.3) The paper designs three regional extraction methods to remove a mass of useless image information and reduce amount of process data to speed up the detection speed. They include fixed region of interest extraction, manual region of interest extraction and adaptive region of interest extraction.4) By researching the edge features of cap after pretreatment, the dissertation design and implement a defect detection algorithm, the experiments show that it has good results and high efficiency of real-time processing;Finally, constructs a complete set of detection process using the inspection system platform after analysis and researching the process of bottle defect detection and the external physical conditions carefully. Comparing with a large number of measurement data of bottle samples, the vision inspection system is proved practically. And it has the advantage of high efficiency and accuracy rating.