Data augmentation for improving few-shot learning on ResNet50
With the onset of rapid climate change and declining biodiversity, forest recovery management is becoming increasingly important. Tree inventory keeping and species identification are two necessary aspects to this, which can be very labour intensive. To alleviate this, a way to automate these tasks...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/176270 |
| _version_ | 1811683848625324032 |
|---|---|
| author | Chan, Jia Ler |
| author2 | Ji-Jon Sit |
| author_facet | Ji-Jon Sit Chan, Jia Ler |
| author_sort | Chan, Jia Ler |
| collection | NTU |
| description | With the onset of rapid climate change and declining biodiversity, forest recovery management is becoming increasingly important. Tree inventory keeping and species identification are two necessary aspects to this, which can be very labour intensive. To alleviate this, a way to automate these tasks using machine learning models can be very helpful. However, due to the nature of how tree data is captured, there is very little usable data to train an image learning model, leading to very low classification accuracy. Because of this, finding a few-shot learning method that can yield high accuracies is of utmost importance.
Data augmentation is a machine learning technique that can help in cases where data is scarce. In the case of image-based learning, small modifications would be made to the image, presenting a wider array of variations of the data during training and can help the model to generalise better. |
| first_indexed | 2024-10-01T04:19:15Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/176270 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:19:15Z |
| publishDate | 2024 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1762702024-05-17T15:44:59Z Data augmentation for improving few-shot learning on ResNet50 Chan, Jia Ler Ji-Jon Sit School of Electrical and Electronic Engineering jijon@ntu.edu.sg Engineering Electrical and electronic engineering With the onset of rapid climate change and declining biodiversity, forest recovery management is becoming increasingly important. Tree inventory keeping and species identification are two necessary aspects to this, which can be very labour intensive. To alleviate this, a way to automate these tasks using machine learning models can be very helpful. However, due to the nature of how tree data is captured, there is very little usable data to train an image learning model, leading to very low classification accuracy. Because of this, finding a few-shot learning method that can yield high accuracies is of utmost importance. Data augmentation is a machine learning technique that can help in cases where data is scarce. In the case of image-based learning, small modifications would be made to the image, presenting a wider array of variations of the data during training and can help the model to generalise better. Bachelor's degree 2024-05-15T05:37:19Z 2024-05-15T05:37:19Z 2024 Final Year Project (FYP) Chan, J. L. (2024). Data augmentation for improving few-shot learning on ResNet50. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/176270 https://hdl.handle.net/10356/176270 en A2088-231 application/pdf Nanyang Technological University |
| spellingShingle | Engineering Electrical and electronic engineering Chan, Jia Ler Data augmentation for improving few-shot learning on ResNet50 |
| title | Data augmentation for improving few-shot learning on ResNet50 |
| title_full | Data augmentation for improving few-shot learning on ResNet50 |
| title_fullStr | Data augmentation for improving few-shot learning on ResNet50 |
| title_full_unstemmed | Data augmentation for improving few-shot learning on ResNet50 |
| title_short | Data augmentation for improving few-shot learning on ResNet50 |
| title_sort | data augmentation for improving few shot learning on resnet50 |
| topic | Engineering Electrical and electronic engineering |
| url | https://hdl.handle.net/10356/176270 |
| work_keys_str_mv | AT chanjialer dataaugmentationforimprovingfewshotlearningonresnet50 |