Achieving Of Image Acquisition And Processing Based On X-ray

Trucks which carry fruit and vegetable carry something else to cross traffic levels will cause economic loss, the development of X-ray imaging technology allows us to penetrate the spacer to get the images we need.so the main work of this paper is to study and implement a system based on scanning truck with X-ray to get data, then process the X-ray image with the knowledge of image processing.Image acquisition is a course of collecting simulate image from the real world through the channel of image processing, and then converting it into digital image. This paper analysis the principle of imaging by the X-ray image acquisition equipment, it collects the data we need by the X-ray detector, then transmits the data of trucks getting by X-ray scanning to the computer through Ethernet cable, the computer platform capture the raw data from the Ethernet cable by the tool named wire-shark, then it stores the scanning data of the ray strength in a TXT file using the form of two-dimensional matrix. The ray strength in this paper is divided into65536grades, including eight high effective, just corresponding256gray scale of a gray image.However, the course of image acquisition from real world is always being interfered by many kinds of noise, so the digital image needs to be preprocessing before it has enough practicability, what’s more, the X-ray images are always overall darker than normal images, much outline can not be distinguished by our naked eyes, so we need to use the image processing techniques to extract the edge and segment the image.. The focus of this paper is to de-noise the data, then get the edge of the image, and then segment the image to help people easier to judge whether the truck has entrainment by the X-ray scanning image. In this course, we use several kinds of filter to filter the noise of digital image, such as mean filter, median filter and Gaussian filter, and compare the actual effect of these filters and the template size influence on filtering effect, finally we select the3*3median filter for de-nosing.This paper introduces the common image segmentation based on edge, region, threshold, wavelet transform, neural network and so on, emphatically analyzes and implements the image segmentation based on genetic, the image segmentation based on traditional operator of edge, the algorithm of image segmentation based on Otsu, and the image segmentation base on maximum entropy. For some images, we need to do the image gray level histogram equalization to make the edge more complete and continuous. Achieving these image segmentation algorithms help us understanding the image segmentation better, compare the effect of these algorithms on X-ray image segmentation, so we choose the image segmentation based on genetic with maximum entropy threshold segmentation.