Design and implementation of real time image acquisition and processing systems

Liu, Junhong
Major Professor
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Electrical and Computer Engineering

Nondestructive evaluation (NDE) is a way to characterize a material or a structure without compromising its usability. Generally, the inspection methods of NDE testing may be based on acoustics, penetrating radiation, light, electric and magnetic fields, or more special possibilities. Whatever methods are used in NDE, imaging technology is one of the important components for these systems. The rapid growth of sophisticated and low priced image acquisition and processing devices has opened up the possibility of applying imaging analysis to more NDE areas. Imaging technology is becoming a very powerful tool in NDE for material properties. The objective of this thesis is to develop a robust, open, easily extendable software platform for real time imaging acquisition and processing. This platform can support image format transform, histogram based look up table, real time image/slice display and device control integration. Three applications were implemented based on this platform. For Rapid Whole-Kernel Single-Seed Analyzer project, the special requirements for the CCD camera and Liquid Crystal Tunable Filter (LCTF) control were met. Multi-thread synchronization was used to cooperate between the CCD camera and the LCTF control. In order to speed up the whole image acquisition procedure, a predefined palette was used. The overlapping between the LCTF tuning time and image storing time made the whole data acquisition as fast as possible. This thesis also used the 14 bit cooled CCD camera to do radiographic digitization. Calibration, focusing, and distance measurement were implemented. The test showed the system could meet the basic requirements for radiographic digitization. In new X-ray Vision system, real time image/slice displaying under multi-video systems were developed. Image integration, averaging and subtracting were finished. It also provided a friendly user interface to motion control. Based on the integration of image acquisition and motion control, the automation of real-time scans was achieved. It is very flexible and can be used in complicated automatic scanning. The tests for the above three applications showed this platform has high stability and powerful functionality.