Accurate and robust object classification is an unsolved problem in the field of digital radiographic image (DRI) analysis. The problem is confounded by variations in objects’ scale, rotation, translation, point-of-view, partial obscuration, and noise. In addition, DRIs present a unique set of challenges, since they reveal the internal structures and features of objects rather than the surface features revealed in most other types of images. The objective of this research is to design, implement, and test a general classifier for DRIs that is robust in the presence of occlusions and noise. To achieve this objective, the proposed ALISA Component Module will use a generalized two-tier analysis token to extract feature vectors associated with the pixel at the center of the analysis token. The feature vectors are accumulated in histograms during training, and classification is performed by comparing feature vectors generated from test images to trained histograms. Formal experiments conducted with the Component Module operating on both canonical shapes and real-world applications using DRIs demonstrated robust classification performance. The Experiments have also demonstrated that the Component Module can learn to classify components of objects from small training sets, as well as effectively classify similar components independent of their position and some variation in their orientations. The Component Module is also robust to uniform and non-uniform occlusions, and noise.
|Advisor:||Bock, Peter S.|
|Commitee:||Armstrong, Alice J., Choi, Hyeong-Ah, Happel, Mark D., Martin, C. Dianne, Youssef, Abdou|
|School:||The George Washington University|
|School Location:||United States -- District of Columbia|
|Source:||DAI-B 69/12, Dissertation Abstracts International|
|Subjects:||Artificial intelligence, Computer science|
|Keywords:||Adaptive systems, Machine learning, Pattern recognition, Radiographic images, Shape classification|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be