Study on Data Processing and Compression of Cobalt-60 Digital Radiographic System

The Co-60 Train Cargo Inspection System (TCT-SCAN) inherits and expands the advantages of the Co-60 Container Inspection System (TC-SCAN), which uses the radioisotope Co-60 as radiation source and gas-pressurized array ion-chamber as detector to inspect large-scale objects with digital radiographic methods. TCT-SCAN that uses DR scanning mode can obtain the images of vehicle in high speed, and it is the first on-line train inspection equipment in practice all over the world. The system provides a more powerful weapon of beating smuggle and drug activities. Based on the work in developing TCT-SCAN system, the paper discusses the design method of high-speed data acquisition system, presents the data correction algorithm in integral mode, analyzes the characteristics of noise in radiant images and corresponding de-noising methods. At last, the paper investigates the compression methods for radiant images. After describing the development and present situation of non-destructive technology and container inspection technology, the paper introduces the principle and the components of TCT-SCAN. Then a brief comparison between TCT-SCAN and general TC-SCAN is given. The high-speed data acquisition system is an important part of TCT-SCAN. The influence of speed to data acquisition is concerned and a distributed-cooperation scheme is presented. Then the paper analyzes and explicates the design and realization in detail. At the end of this chapter, the paper puts forward a data correction algorithm in integral mode, which can improve the image quality distinctly. The noise is one of the important factors to affect the quality of radiant images. The paper analyzes the specialty of the noise and puts forward the "whiten" algorithm. The wavelet transform is introduced as the method of filtering noise. The experiment results show that not only the statistical noise is wiped off, but also the details of the images are preserved commendably. In the last chapter, the paper introduces the compression method for radiant images. At first, the paper analyzes the characteristics of radiant images, the limitation of the existing method is discussed here. Then, the wavelet transform is adopted and the quantization and encoding algorithm are discussed in detail. The paper also analyzes the influence factors to restored images. The methods can compress the radiant images efficiently and preserve the quality of the images at the same time. At last, the paper does an elementary research of lossless compression method on radiant images. The scheme is proved to be better than traditional entropy encoding method.

Alternate abstract:

Co-60铁路货运列车检测系统(TCT-SCAN)是Co-60集装箱检测技术的继承和发展。它是利用Co-60放射性同位素射线源，以高压充气电离室探测器阵列对货运列车进行DR成像的无损检测系统，可在高速下实现对列车车厢的实时成像，是世界首套投入实际运行的列车在线检测设备。它的研制成功，为反恐、反走私提供了更强有力的武器。本论文在参与研制该系统的基础上，研讨了高速数据采集系统的设计方法、提出了积分模式下数据校正算法，分析了辐射图像的噪声特点和去除办法，并对辐射图像的压缩方法进行研究。 在概述无损检测技术和集装箱检测技术的历史和发展现状后，论文阐述了TCT-SCAN的成像原理，对系统的各部分组成和技术特点进行了介绍，并将TCT-SCAN系统与常规集装箱检测系统作了简单的比较。 高速数据采集系统是TCT-SCAN的重要组成部分。论文首先分析了速度问题对数据采集的影响，提出了分布式协同处理的解决方案，并对各部分的设计思路和实现细节进行了详细的分析和说明。针对系统特点，提出了一种积分模式下的数据校正算法，极大地提高了成像质量。 辐射图像的噪声是影响图像质量的一大因素。论文对此进行了详细的分析，并对噪声的白化算法进行了理论推导并给出实现方法，在滤波过程中引入小波变换。滤波后的图像表明，该方法能在有效去除统计噪声的同时，较好地保持图像细节，使图像质量得到很大改善。 论文最后研究了辐射图像的压缩方法。首先分析了辐射图像的特点和现有方法的不足之处，并据此提出采用小波变换作为变换编码，详细讨论了量化和编码算法，并分析了边界拓展方式等因素对复原图像的影响。在不降低图像质量的前提下有效地压缩了图像数据。另外，论文也对辐射图像的无损压缩方法进行了初步研究，取得了优于传统熵编码的压缩效果。