Research On Automatic Window Adjustment Algorithm And X-ray Detection System

Automobile wheels are the main stress-bearing parts in automobile operation,and qualified wheels are the guarantee of driving stability and safety.At present,X-ray inspection is widely used due to its high imaging accuracy,wide detection depth,and high real-time performance.Compared with 8-bit ray grayscale images,16-bit ray grayscale images have higher grayscale representation and higher accuracy.Therefore,16-bit grayscale images are often used in wheel inspection.However,16-bit X-ray grayscale images have many problems such as redundant information and low contrast.Window adjustment is a good way to deal with the above-mentioned problems,but the traditional manual adjustment of windows by hand is time-consuming and laborious,and requires higher prior knowledge of the inspectors.This paper is dedicated to realizing a self-adaptive window adjustment method with good real-time performance,strong robustness and high accuracy to solve the above problems.This article first introduces some existing adaptive window adjustment methods,mainly including:1 The adaptive window adjustment algorithm based on neural network is complicated,time-consuming,and requires a large number of well-adjusted pictures as a training set.2 The improved extreme value method,area method,cross-section method and other algorithms are simple and easy to implement and run fast.But the robustness is too poor.In order to overcome the above shortcomings,this paper proposes an algorithm based on voice endpoint detection.Compared with the adaptive windowing algorithm based on neural network,this method has a simpler algorithm principle,fewer process steps,and does not require a lot of manual work.The windowed pictures are used as training samples,so the problem can be handled efficiently and quickly.At the same time,experiments show that the algorithm proposed in this paper can handle new types of pictures very well,has good robustness,overcomes the problem of poor adaptability of simple algorithms,and takes into account both efficiency and performance..Another work of this article is based on the above algorithm to design and implement a set of defect detection system with beautiful interface and rich functions.Combined with the characteristics of this algorithm,it can process pictures quickly and efficiently,and at the same time increase the function of batch processing pictures.A large number of pictures can be processed at once.