Research On The Detection Technology Of Collectible Coin's Face Quality Based On Machine Vision

The detection result of the face quality of the collectible coin can well reflect the collection value of the collectible coin,and is an important reference for the graders to evaluate the value of the collectible coin.The surface quality inspection of collectible coins includes the following objectives.The first is to complete the collection of high-quality collectible currency images;the other is to qualitatively determine the characteristics and types of currency defects;and the third is to perform statistical analysis on these defects.To determine the extent of the impact of these defects on the quality of the currency.In order to accomplish the above goals,this article mainly does the following work.In terms of image acquisition,starting from the machine vision system hardware selection,acquisition background and acquisition environment requirements,in this paper,the selection of light source,camera and lens was completed,and construction of a closed acquisition environment and the compilation of image acquisition software was designed.high quality collection coin images collection and storage are realized.In the image preprocessing and collectible coin type recognition stage,in view of the uncertainty and variability of the features of different collectible coins,a correlation-based multi-objective method of identifying the coin type of collectible coins was adopted.After the same preprocessing,the target image and reference image undergoes a series of image transformation and matching.Finally,the type recognition and registration of the collectible coin was realized.In terms of the extraction,classification,and recognition of defects on the coin surface of collectible coins,considering the difference between the pixel values of the same position whether defected or no defected,the target collection coin image and the reference collection coin image are first registered and then differentiated by the image difference method and realize the extraction of the defect area of the collectible coin.At the same time,in order to remove the noise and other interference in the image,more effectively screen out the defect area of the coin and more accurately determine the type of the coin defect,multiple thresholds and image segmentation methods was used,according to different defects in the target image,reference the different pixel values in the image and the difference image,and the different geometric and shape parameters of the different defect areas after segmentation the accurate classification of the defect types of the collectible coin was achieved.In the aspect of currency quality inspection,taking the three-year Yuan Shih-kai silver dollar of the Republic of China as an example,divides the currency area into currency circles,important features.There are three areas of the currency background.According to these three areas,the types of defects of the collectible currency are subdivided.The different categories reflect the different degrees of influence of the defects on the appearance of the collectible currency.Based on the above research content,a machine vision-based inspection system for the face quality of collectible coins was designed and implemented,the defect areas of collectible coins can identified and proved through the Experiments,and the inspection results stable and reliable.