Research On The Electrical Inspection System Of Condoms

With the development of medical and health standards and the use of new materials,the condom using polyurethane material is emerging.However,there is a problem that the accuracy of electric inspection is low for both the polyurethane condom and the latex condom in the current production process of condoms.For the polyurethane condom,it is in the form of curled ring coming from production line,so it needs to be unfolded for electric inspection.Now,the unfolding work is down manually.In order to realize automatically spreading condoms,it is necessary to solve the problem of recognizing the positive and negative sides of the polyurethane condom by machine vision.The main content of this paper is to deal with the above problems by designing a new electric inspection circuit and introducing the machine vision.First of all,this paper described the working principle of two mainstream electric inspection devices for condoms,which are chain-type and rotary desk-type.Aiming at the problems of electric coupling of condoms and easy damage existing in chain-type electric inspection machines,as well as the problem of the low detection efficiency of rotary desk-type electric inspection machines,this paper designed an improved electric inspection circuit to solve the problem in the current electric inspection process.Secondly,after observing the positive and negative sides of the ringlike condom,it was found that the biggest difference between the the positive and the negative was the performance of the inner ring edge.Extracting and describing the inner ring accurately was the key to study in this paper.Through the current mainstream image edge extraction methods such as Sobel,Laplace,Prewitt,Canny operator etc.to test the images of the same one ringlike condom,the edge images extracted by different algorithms were obtained.Comparing these images,it was found that Canny operator had stronger anti-interference ability and more accurate single-edge characteristics than other algorithms in extracting condom edge.Therefore,Canny operator was selected as the basic idea of the algorithm in this paper.Thirdly,on the basis of Canny operator,the gradient amplitude calculation link was improved for the edge extraction of the inner and outer rings of the ringlike condom.The twice calculations of the gradient amplitudes of Canny operator in horizontal and vertical directions changed the once calculation of the sub-region in a specified direction.This paper gave the specific method for identifying the positive and negative sides of the ringlike condom,including the concept of regional division,the construction of the directional gradient convolution kernel method and its mathematical description,and designed a method for quantitatively describing the integrity of the inner ring and a method for determining the positive and negative sides of ringlike condoms.Finally,this paper briefly described the experimental environment required for the algorithm verification and illustrated the samples of the specific value of the 3-6 order convolution kernel and the edge detection images with 3~6 orders.Through observation and analysis of the results,it was determined that when the convolution kernel size was greater than or equal to 5th order,the purpose of retaining ring information and filtering out non-cyclical clutter could be achieved.Taking into account the efficiency of recognition,this paper finally used a 5th order convolution kernel to recognize the positive and negative sides of 700 ringlike condoms.The experimental result shows that the algorithm has a good detection accuracy and the correct rate can be up to 100%.