Performance analysis of object detection algorithms using small training datasets
Object detection using machine learning approach has seen wide adoption in virtually all known industries in the past decade. Much investment and research has been put into building the most accurate object detection algorithm. However, implementation of these algorithms is only accessible to organi...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/157247 |
| _version_ | 1811691177862234112 |
|---|---|
| author | Anthony, Benedict |
| author2 | Kedar Hippalgaonkar |
| author_facet | Kedar Hippalgaonkar Anthony, Benedict |
| author_sort | Anthony, Benedict |
| collection | NTU |
| description | Object detection using machine learning approach has seen wide adoption in virtually all known industries in the past decade. Much investment and research has been put into building the most accurate object detection algorithm. However, implementation of these algorithms is only accessible to organizations with vast amount of computing and data procurement resources. In this study, the correlation of overall detection rate, training time and training sample size will be explored. In addition, threshold for minimum effective training sample size will be investigated in order to aid implementation of object detection in environments where annotated training samples are difficult to obtain. The experiment revealed that models trained using the LBP feature type performed significantly better in the 50-100 sample size range in terms of effectiveness compared to the HAAR feature type. |
| first_indexed | 2024-10-01T06:15:45Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/157247 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:15:45Z |
| publishDate | 2022 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1572472022-06-03T22:09:25Z Performance analysis of object detection algorithms using small training datasets Anthony, Benedict Kedar Hippalgaonkar School of Materials Science and Engineering Institute of Materials Research and Engineering, A*STAR Jayce Cheng Jian Wei kedar@ntu.edu.sg Engineering::Materials::Material testing and characterization Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Object detection using machine learning approach has seen wide adoption in virtually all known industries in the past decade. Much investment and research has been put into building the most accurate object detection algorithm. However, implementation of these algorithms is only accessible to organizations with vast amount of computing and data procurement resources. In this study, the correlation of overall detection rate, training time and training sample size will be explored. In addition, threshold for minimum effective training sample size will be investigated in order to aid implementation of object detection in environments where annotated training samples are difficult to obtain. The experiment revealed that models trained using the LBP feature type performed significantly better in the 50-100 sample size range in terms of effectiveness compared to the HAAR feature type. Bachelor of Engineering (Materials Engineering) 2022-05-12T05:43:38Z 2022-05-12T05:43:38Z 2022 Final Year Project (FYP) Anthony, B. (2022). Performance analysis of object detection algorithms using small training datasets. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/157247 https://hdl.handle.net/10356/157247 en application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Materials::Material testing and characterization Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Anthony, Benedict Performance analysis of object detection algorithms using small training datasets |
| title | Performance analysis of object detection algorithms using small training datasets |
| title_full | Performance analysis of object detection algorithms using small training datasets |
| title_fullStr | Performance analysis of object detection algorithms using small training datasets |
| title_full_unstemmed | Performance analysis of object detection algorithms using small training datasets |
| title_short | Performance analysis of object detection algorithms using small training datasets |
| title_sort | performance analysis of object detection algorithms using small training datasets |
| topic | Engineering::Materials::Material testing and characterization Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence |
| url | https://hdl.handle.net/10356/157247 |
| work_keys_str_mv | AT anthonybenedict performanceanalysisofobjectdetectionalgorithmsusingsmalltrainingdatasets |