Analysis of deep learning models on low-light pest detection
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2022.
| Autors principals: | , , , , |
|---|---|
| Altres autors: | |
| Format: | Thesis |
| Idioma: | English |
| Publicat: |
Brac University
2023
|
| Matèries: | |
| Accés en línia: | http://hdl.handle.net/10361/21848 |
| id |
10361-21848 |
|---|---|
| record_format |
dspace |
| spelling |
10361-218482023-10-16T21:05:57Z Analysis of deep learning models on low-light pest detection Irtiza, Md. Samin Ahmed, Fattah Haque, Md. Tahmidul Tamim, Arifur Rahman Sultana, Samia Chakrabarty, Amitabha Department of Computer Science and Engineering, Brac University Low-light images Enhancement network Deep learning Object detection Insect Machine learning Computer vision This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2022. Cataloged from PDF version of thesis. Includes bibliographical references (pages 31-33). It is undeniable that in recent years, exceptional progress has been made toward building the most accurate and efficient object detectors. However, existing low- light object detectors still require a substantial amount of resources to perform at their best. Our main goal in this research is to train and evaluate recently developed deep learning object detection models on low-light images and see if they can show decent performance without any additional enhancement networks. Furthermore, we aim to achieve those results with minimum computational cost. In this research, we have created our own custom dataset from a publicly available insect image dataset called ‘IP102’. The new dataset now named ‘IP013’ consists of 13 classes of insects and approximately 8k annotated images. Moreover, we chose recently developed YOLOv7 and DETR object detectors and compared their performance against now older state-of-the-art RetinaNet and EfficientDet deep learning models. YOLOv7, EfficientDet, and RetinaNet are purely CNN-based models whereas DETR uses a Transformer as both encoder and decoder and a CNN as the backbone. Our research shows that YOLOv7 outperforms all of the other models with a mAP0.5:.95 of 45.9 while using the lowest training time and the model that used the least computational resources was EfficientDet which admittedly showed lackluster mAP0.5:.95 of 33.2 with only 3.9M parameters and using 2.5 GFLOPs. Md. Samin Irtiza Fattah Ahmed Md. Tahmidul Haque Arifur Rahman Tamim Samia Sultana B.Sc. in Computer Science 2023-10-16T07:43:04Z 2023-10-16T07:43:04Z ©2022 2022-09-28 Thesis ID 18101429 ID 18101442 ID 18101570 ID 18101510 ID 18101446 http://hdl.handle.net/10361/21848 en Brac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. 43 pages application/pdf Brac University |
| institution |
Brac University |
| collection |
Institutional Repository |
| language |
English |
| topic |
Low-light images Enhancement network Deep learning Object detection Insect Machine learning Computer vision |
| spellingShingle |
Low-light images Enhancement network Deep learning Object detection Insect Machine learning Computer vision Irtiza, Md. Samin Ahmed, Fattah Haque, Md. Tahmidul Tamim, Arifur Rahman Sultana, Samia Analysis of deep learning models on low-light pest detection |
| description |
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2022. |
| author2 |
Chakrabarty, Amitabha |
| author_facet |
Chakrabarty, Amitabha Irtiza, Md. Samin Ahmed, Fattah Haque, Md. Tahmidul Tamim, Arifur Rahman Sultana, Samia |
| format |
Thesis |
| author |
Irtiza, Md. Samin Ahmed, Fattah Haque, Md. Tahmidul Tamim, Arifur Rahman Sultana, Samia |
| author_sort |
Irtiza, Md. Samin |
| title |
Analysis of deep learning models on low-light pest detection |
| title_short |
Analysis of deep learning models on low-light pest detection |
| title_full |
Analysis of deep learning models on low-light pest detection |
| title_fullStr |
Analysis of deep learning models on low-light pest detection |
| title_full_unstemmed |
Analysis of deep learning models on low-light pest detection |
| title_sort |
analysis of deep learning models on low-light pest detection |
| publisher |
Brac University |
| publishDate |
2023 |
| url |
http://hdl.handle.net/10361/21848 |
| work_keys_str_mv |
AT irtizamdsamin analysisofdeeplearningmodelsonlowlightpestdetection AT ahmedfattah analysisofdeeplearningmodelsonlowlightpestdetection AT haquemdtahmidul analysisofdeeplearningmodelsonlowlightpestdetection AT tamimarifurrahman analysisofdeeplearningmodelsonlowlightpestdetection AT sultanasamia analysisofdeeplearningmodelsonlowlightpestdetection |
| _version_ |
1814309520903503872 |