Research On Image Enhancement And Software Testing Efficiency Of Vehicle Bottom Detection System

China has become the second largest trading country in the world.Although it still ranks behind the United States in total trade volume,it has surpassed the United States in terms of goods trade volume.Customs plays a key role in foreign trade.It is an important content of Customs daily law enforcement and supervision to supervise the safety and legality of vehicles.It is a common method for the criminals to hide the prohibited goods in the vehicle bottom and try to pass through the border port when they are engaged in illegal transactions.Therefore,vehicle bottom detection becomes an important part of the legitimacy inspection of transit vehicles.Not only that,the vehicle bottom inspection can also be used for various large-scale conference activities,entrance and exit of important state organs,highway toll stations and security inspection stations,prison entrances and exits,airport vehicle entrances and exits,and security inspection work around important places.Vehicle bottom detection technology can quickly find drugs,weapons,smuggled goods,dangerous explosives and suspected contraband goods hidden on the chassis of the car through intuitive image inspection.This is of great significance to transit security,social security and even national security.The traditional vehicle bottom detection work is that the inspector obtains the vehicle bottom information through the vehicle bottom inspection rod endoscope and camera,and the operator directly observes the vehicle bottom status through video or mirror reflection.However,due to the limitations of inspection methods,it is not only difficult and inconvenient to operate,but also difficult to provide high-quality test results.At present,the vehicle bottom detection equipment based on scanning imaging technology can provide better convenience for vehicle bottom detection.The vehicle bottom scanning imaging equipment includes a fixed or mobile vehicle chassis photographing system,which can take a complete image of the vehicle bottom,and provide high-definition vehicle bottom image and high-definition vehicle bottom video for inspectors.This kind of equipment and technology,on the one hand,is easy to operate,on the other hand,can form high-quality test results.Compared with traditional vehicle bottom detection technology,vehicle simultaneous interpreting technology based on scanning imaging has great advantages.However,this detection technology will also be affected by environmental light,component shading,insufficient power of fill light,improper exposure time of linear array camera and other reasons,resulting in deficiencies in brightness,contrast and integrity of the generated vehicle bottom image,which requires the key technologies such as image denoising,image restoration and image enhancement to obtain higher quality vehicle bottom image.In view of the problems of the vehicle bottom detection technology based on scanning imaging,this paper studies the algorithms of denoising,recovery and contrast enhancement of the vehicle bottom image,and tests the stability of the software designed based on the relevant algorithm,and then carries out the experimental verification on the general image set and the bottom line array image.The main research work of this paper as follows:1.Aiming at the problems of dark,high noise and low contrast of vehicle bottom scanning image caused by insufficient power of fill light and unreasonable configuration of exposure parameters,an anisotropic nonlinear Perona-Malik differential equation(PMDE)image enhancement algorithm with initial boundary conditions is proposed.The PMDE image enhancement algorithm adopts nonlinear minimization technology and multiple optimization to remove image noise,which can effectively enhance the image contrast and retain the edge details of the image.The algorithm in this paper can solve the PMDE better,and make the image enhancement process have better stability and convergence.The experimental results show that,no matter in the general image data set or in the denoising and contrast enhancement processing of the vehicle bottom scanning image,the anisotropic nonlinear PMDE enhancement algorithm with initial boundary conditions can achieve ideal image enhancement effect than traditional image enhancement algorithms.2.Aiming at the noise,blur and loss of vehicle bottom image caused by imaging environment,imaging mode and coding transmission,a semi supervised image restoration algorithm based on Fourier transform is proposed,which is used to repair vehicle bottom image in Fourier transform domain.In order to solve the nonsingular system,a truncated singular value decomposition algorithm is proposed.In order to reduce the influence of auxiliary points,a semi-supervised feature selection method using information such as the strength of auxiliary points is proposed.Theoretical analysis proves the convergence of the proposed algorithm,the experimental results verify the effectiveness of the proposed algorithm for the restoration of general image datasets and vehicle bottom scanned images containing blur and noise.3.Aiming at the low efficiency of vehicle bottom image processing software stability testing,a Daikon software testing method based on the optimal reordering of tracking files is proposed.With the aid of program invariant detection tool Daikon,variable attributes are introduced to reduce the number of invariants,and non deterministic heuristic method is used to reorder the data tracking files to improve the test speed.Vehicle bottom image software testing experiments show that the proposed method reduces testing time and improves testing efficiency.The more software files there are,the more obvious the improvement in testing efficiency.This paper proposes a numerical solution method based on the fractional-order nonlinear anisotropic diffusion equation using Grunwald-Letnikov derivative,a semi-supervised image restoration algorithm based on Fourier transform and an improved method of Daikon software testing based on optimal reordering of tracking files.Theses innovative results have achieved significant image denoising and restoration enhancement effects,and significantly improved the efficiency of Daikon software testing.These innovative results can promote the development of image enhancement algorithm,improve the technical level of the under-vehicle inspection system,and is of great significance for supervising vehicle safety and ensuring social safety.