Novel techniques for sparse representation problems
Sparse representations have been used in solving many problems in computer science. Two issues that need to be addressed in formulating such a representation are: the problem design; and the optimization technique. Many optimization problems contain one/multiple non-smooth terms in the objective f...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Doctor of Philosophy |
| Language: | English |
| Published: |
Nanyang Technological University
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/146392 |
| _version_ | 1824454234470350848 |
|---|---|
| author | Chai, Woon Huei |
| author2 | Quek Hiok Chai |
| author_facet | Quek Hiok Chai Chai, Woon Huei |
| author_sort | Chai, Woon Huei |
| collection | NTU |
| description | Sparse representations have been used in solving many problems in computer science. Two issues that need to be addressed in formulating such a representation are:
the problem design; and the optimization technique.
Many optimization problems contain one/multiple non-smooth terms in the objective function. Besides, the feasibility of an optimization problem depends on the availability of adequate computational resources. In this thesis, a new parallelizable optimization technique that uses more information and has better convergence than state-of-the-art counterparts is presented. Theoretical derivation of the bound of the recovery probability of using sparse representation based on a L_1-minimization is also shown. A clustering-based technique for dictionary and signal dimension reduction to replace the traditional naïve downsampling technique is introduced to address computational resource constraints. Finally, an anomaly detection and localization technique using a sparse representation problem and used in a case study for an important and challenging field; namely automated visual inspection (AVI) is presented. The experimental results are encouraging. |
| first_indexed | 2025-02-19T03:19:04Z |
| format | Thesis-Doctor of Philosophy |
| id | ntu-10356/146392 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-02-19T03:19:04Z |
| publishDate | 2021 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1463922023-03-05T16:38:02Z Novel techniques for sparse representation problems Chai, Woon Huei Quek Hiok Chai Interdisciplinary Graduate School (IGS) Energy Research Institute @ NTU (ERI@N) ASHCQUEK@ntu.edu.sg Engineering::Computer science and engineering Science::Mathematics Sparse representations have been used in solving many problems in computer science. Two issues that need to be addressed in formulating such a representation are: the problem design; and the optimization technique. Many optimization problems contain one/multiple non-smooth terms in the objective function. Besides, the feasibility of an optimization problem depends on the availability of adequate computational resources. In this thesis, a new parallelizable optimization technique that uses more information and has better convergence than state-of-the-art counterparts is presented. Theoretical derivation of the bound of the recovery probability of using sparse representation based on a L_1-minimization is also shown. A clustering-based technique for dictionary and signal dimension reduction to replace the traditional naïve downsampling technique is introduced to address computational resource constraints. Finally, an anomaly detection and localization technique using a sparse representation problem and used in a case study for an important and challenging field; namely automated visual inspection (AVI) is presented. The experimental results are encouraging. Doctor of Philosophy 2021-02-16T01:29:59Z 2021-02-16T01:29:59Z 2021 Thesis-Doctor of Philosophy Chai, W. H. (2021). Novel techniques for sparse representation problems. Doctoral thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/146392 10.32657/10356/146392 en This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering Science::Mathematics Chai, Woon Huei Novel techniques for sparse representation problems |
| title | Novel techniques for sparse representation problems |
| title_full | Novel techniques for sparse representation problems |
| title_fullStr | Novel techniques for sparse representation problems |
| title_full_unstemmed | Novel techniques for sparse representation problems |
| title_short | Novel techniques for sparse representation problems |
| title_sort | novel techniques for sparse representation problems |
| topic | Engineering::Computer science and engineering Science::Mathematics |
| url | https://hdl.handle.net/10356/146392 |
| work_keys_str_mv | AT chaiwoonhuei noveltechniquesforsparserepresentationproblems |