SSDLiteX: Enhancing SSDLite for Small Object Detection
Object detection in many real applications requires the capability of detecting small objects in a system with limited resources. Convolutional neural networks (CNNs) show high performance in object detection, but they are not adequate to resource-limited environments. The combination of MobileNet V...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/21/12001 |
| _version_ | 1797632155732410368 |
|---|---|
| author | Hyeong-Ju Kang |
| author_facet | Hyeong-Ju Kang |
| author_sort | Hyeong-Ju Kang |
| collection | DOAJ |
| description | Object detection in many real applications requires the capability of detecting small objects in a system with limited resources. Convolutional neural networks (CNNs) show high performance in object detection, but they are not adequate to resource-limited environments. The combination of MobileNet V2 and SSDLite is one of the common choices in such environments, but it has a problem in detecting small objects. This paper analyzes the structure of SSDLite and proposes variations leading to small object detection improvement. The feature maps with the higher resolution are utilized more, and the base CNN is modified to have more layers in the high resolution. Experiments have been performed for the various configurations and the results show the proposed CNN, SSDLiteX, improves the detection accuracy AP of small objects by 1.5 percent points in the MS COCO data set. |
| first_indexed | 2024-03-11T11:33:54Z |
| format | Article |
| id | doaj.art-af3723e904a544ba8022bdb0ce0d9f2c |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T11:33:54Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-af3723e904a544ba8022bdb0ce0d9f2c2023-11-10T14:59:26ZengMDPI AGApplied Sciences2076-34172023-11-0113211200110.3390/app132112001SSDLiteX: Enhancing SSDLite for Small Object DetectionHyeong-Ju Kang0School of Computer Science and Engineering, Korea University of Technology and Education, Cheonan 31253, Republic of KoreaObject detection in many real applications requires the capability of detecting small objects in a system with limited resources. Convolutional neural networks (CNNs) show high performance in object detection, but they are not adequate to resource-limited environments. The combination of MobileNet V2 and SSDLite is one of the common choices in such environments, but it has a problem in detecting small objects. This paper analyzes the structure of SSDLite and proposes variations leading to small object detection improvement. The feature maps with the higher resolution are utilized more, and the base CNN is modified to have more layers in the high resolution. Experiments have been performed for the various configurations and the results show the proposed CNN, SSDLiteX, improves the detection accuracy AP of small objects by 1.5 percent points in the MS COCO data set.https://www.mdpi.com/2076-3417/13/21/12001convolutional neural networksobject detectionsingle shot multibox detector |
| spellingShingle | Hyeong-Ju Kang SSDLiteX: Enhancing SSDLite for Small Object Detection Applied Sciences convolutional neural networks object detection single shot multibox detector |
| title | SSDLiteX: Enhancing SSDLite for Small Object Detection |
| title_full | SSDLiteX: Enhancing SSDLite for Small Object Detection |
| title_fullStr | SSDLiteX: Enhancing SSDLite for Small Object Detection |
| title_full_unstemmed | SSDLiteX: Enhancing SSDLite for Small Object Detection |
| title_short | SSDLiteX: Enhancing SSDLite for Small Object Detection |
| title_sort | ssdlitex enhancing ssdlite for small object detection |
| topic | convolutional neural networks object detection single shot multibox detector |
| url | https://www.mdpi.com/2076-3417/13/21/12001 |
| work_keys_str_mv | AT hyeongjukang ssdlitexenhancingssdliteforsmallobjectdetection |