Research On Infrared Detection Method Of Container Leakage Based On Thermal Radiation Characteristics Of Materials

With the rapid development of industrial technology,the requirements for the containment of containers in industrial production are becoming higher and higher.However,the leakage detection of containers is currently a major problem in industrial production,and the traditional leakage detection method cannot meet the requirements of industrial automation detection and has high false detection.This paper proposes an infrared detection method for container leakage based on the thermal radiation properties of the material.The method is based on an integrated computer-controlled operation process,where the location of the leakage point of the container to be tested is obtained through infrared image acquisition and algorithm analysis.The method is simple to operate,cost saving,safe and reliable,and meets some of the current requirements of industrial production.Due to the infrared camera in the process of data acquisition is susceptible to the influence of environmental noise and thus produce false detection of leak points,and for smaller size leak points general detection methods are less effective,the leakage rate is high.To address the above problems,this paper investigates the physical changes at the leak point of the container based on the theory of material heat radiation and heat transfer heat convection,and experimentally verifies that the temperature at the leak point will change according to the pressure change when the container leaks,i.e.the grey value at the leak point changes in the infrared image captured by the infrared camera.This paper firstly builds a complete container leak detection platform,firstly by filling the container with dry air,pressurising it and simultaneously filling and deflating it to increase the temperature change at the leak point.During this period,an infrared camera is used for image acquisition,and the collected infrared images are firstly filtered to remove a large amount of noise,and secondly to determine the change in grey scale value at the leak point.Accordingly,this paper proposes a multi-frame based leak detection algorithm(improved bilateral filtering and frame difference variance method)to verify the detected leaks and further reduce the false detection rate,and compare with other algorithms to verify the superiority of this paper’s algorithm.This paper proposes a new super-resolution algorithm based on image super-resolution reconstruction for smaller leaks.The new super-resolution algorithm is constructed with an improved multi-scale residual and attention module,so that smaller leaks are amplified,i.e.it is easier to capture changes in the grey value of the leak,and then the container leak is detected by the multi-frame detection algorithm,which increases the detection accuracy of smaller leaks and further reduces the The superiority of this paper’s algorithm is verified by comparing it with various super-resolution reconstruction algorithms.After extensive experiments,the improved bilateral filtering and frame difference variance algorithm and the improved super-resolution reconstruction algorithm have achieved better detection results and meet the requirements for leak detection in industrial production.It can provide technical services for the subsequent leak detection in industrial production.