Scene understanding based on heterogeneous data fusion
Artificial intelligence has boosted human’s life; this technology has become something that will totally change people’s life in the future. Scene understanding is one of the most popular research areas under this topic. This project focuses on developing a high-performance deep learning neural netw...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/70755 |
| _version_ | 1824456770347597824 |
|---|---|
| author | Zhu, Lingzhi |
| author2 | Mao Kezhi |
| author_facet | Mao Kezhi Zhu, Lingzhi |
| author_sort | Zhu, Lingzhi |
| collection | NTU |
| description | Artificial intelligence has boosted human’s life; this technology has become something that will totally change people’s life in the future. Scene understanding is one of the most popular research areas under this topic. This project focuses on developing a high-performance deep learning neural network which could help scene understanding model perform well in image classification. This project uses convolutional neural network as the fundamental network architecture. Nearly ten thousand images are collected, and these images are classified into 20 different classes based on image descriptions. With pre-trained Keras VGG-16 model, several sets of features are extracted from different layers. New classifiers are created and trained by passing those features through. This network achieves 0.77 mean average precision without any fine tuning. Moreover, after fine tuning process, the highest mAP score it can reach is 0.804. Experiments on testing different variables are implemented, and the results are elaborated as well. Difference between these tests are discussed as well. |
| first_indexed | 2025-02-19T03:59:23Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/70755 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-02-19T03:59:23Z |
| publishDate | 2017 |
| record_format | dspace |
| spelling | ntu-10356/707552023-07-07T17:21:07Z Scene understanding based on heterogeneous data fusion Zhu, Lingzhi Mao Kezhi School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering Artificial intelligence has boosted human’s life; this technology has become something that will totally change people’s life in the future. Scene understanding is one of the most popular research areas under this topic. This project focuses on developing a high-performance deep learning neural network which could help scene understanding model perform well in image classification. This project uses convolutional neural network as the fundamental network architecture. Nearly ten thousand images are collected, and these images are classified into 20 different classes based on image descriptions. With pre-trained Keras VGG-16 model, several sets of features are extracted from different layers. New classifiers are created and trained by passing those features through. This network achieves 0.77 mean average precision without any fine tuning. Moreover, after fine tuning process, the highest mAP score it can reach is 0.804. Experiments on testing different variables are implemented, and the results are elaborated as well. Difference between these tests are discussed as well. Bachelor of Engineering 2017-05-09T09:11:40Z 2017-05-09T09:11:40Z 2017 Final Year Project (FYP) http://hdl.handle.net/10356/70755 en Nanyang Technological University 57 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Electrical and electronic engineering Zhu, Lingzhi Scene understanding based on heterogeneous data fusion |
| title | Scene understanding based on heterogeneous data fusion |
| title_full | Scene understanding based on heterogeneous data fusion |
| title_fullStr | Scene understanding based on heterogeneous data fusion |
| title_full_unstemmed | Scene understanding based on heterogeneous data fusion |
| title_short | Scene understanding based on heterogeneous data fusion |
| title_sort | scene understanding based on heterogeneous data fusion |
| topic | DRNTU::Engineering::Electrical and electronic engineering |
| url | http://hdl.handle.net/10356/70755 |
| work_keys_str_mv | AT zhulingzhi sceneunderstandingbasedonheterogeneousdatafusion |