Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.487071
Title: Wavelet-based processing for acquisition and transmission of large image data sets using climbing robot to deploy NDT sensors
Author: Shang, Jianzhong
Awarding Body: London South Bank University
Current Institution: London South Bank University
Date of Award: 2007
Availability of Full Text:
Access through EThOS:
Abstract:
The thesis describes original work that has developed a novel climbing robot system for the Non-Destructive Testing (NDT) of aircraft fuselage and wings, and the processing and compression ofNDT image data based on the Wavelet Transform. The work establishes the functional requirements for a wall climbing robotic NDT system for aircraft inspection. A climbing robot has been designed and prototyped that can climb on all surface curvatures presented by the fuselage and wings. The novelty in the design is the use of a hierarchy of universal joints to allow flexible adaptation of the robot to varying surface curvatures while at the same time making the structure rigid when suction cups are activated. The structure of the flexible climbing robot has been analysed to establish its intrinsic stability. The hybrid electric and pneumatic control system obtains X-Y movement of the robot and is able to correct off-course errors. The control system tests cylinder rod positions and vacuum pressure before allowing movement of the robot. The climbing robot has been tested successfully on a real aircraft fuselage section. The robot can climb reliably and deploy wet or dry ultrasound transducers with a four axes Cartesian scanner mounted on the climbing robot. The development is reported of a new invention called a regional NDT image compression tool based on the Wavelet Transform. The development includes feature extraction of NDT images and Regional Set Partitioning in Hierarchical Trees (RSPIHT) coding algorithm. With this tool, the NDT image compression showed that the tested images could be compressed at ratios of up to 1000:1, and when the compressed images were reconstructed, the defects could be still observed clearly. This proves that the initial idea of using the regional Wavelet Transform based image compression technique is correct. The results of the work can be used as a general guideline for design and development of other new types of wall-climbing robotic NDT instrumentation and image data processing.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.487071  DOI: Not available
Share: