Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN model
With the growing demand for geospatial data, challenging aerial images with high spatial, spectral, and temporal resolution achieve excellent development. Currently, deep Convolutional Neural Network (CNN) structures are applied widely for object detection. Nevertheless, existing deep CNN-based mode...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-08-01
|
| Series: | Alexandria Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1110016821007602 |
| _version_ | 1818474004979122176 |
|---|---|
| author | Mohamad Haniff Junos Anis Salwa Mohd Khairuddin Mahidzal Dahari |
| author_facet | Mohamad Haniff Junos Anis Salwa Mohd Khairuddin Mahidzal Dahari |
| author_sort | Mohamad Haniff Junos |
| collection | DOAJ |
| description | With the growing demand for geospatial data, challenging aerial images with high spatial, spectral, and temporal resolution achieve excellent development. Currently, deep Convolutional Neural Network (CNN) structures are applied widely for object detection. Nevertheless, existing deep CNN-based models consist of complex network structures and require immense amounts of graphics processing unit (GPU) computation power with high energy consumption. Thus, achieving efficient real-time object detection for limited memory and processing capacity embedded device is a major challenge. This paper proposes a feasible and lightweight object detection model based on deep CNN where a mobile inverted bottleneck module is adopted in the backbone structure. Moreover, an enhanced spatial pyramid pooling is adopted to increase the receptive field in the network by concatenating the multi-scale local region features. The experimental results demonstrated that the proposed model achieved higher average precision and required the smallest memory storage compared to previous works. Moreover, the proposed model offers the best trade-offs in terms of detection accuracy, model size, and detection time which has excellent potential to be deployed on limited capacity embedded device. |
| first_indexed | 2024-04-14T04:30:44Z |
| format | Article |
| id | doaj.art-852c4868bd1144699dc17e4db06d0ad0 |
| institution | Directory Open Access Journal |
| issn | 1110-0168 |
| language | English |
| last_indexed | 2024-04-14T04:30:44Z |
| publishDate | 2022-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Alexandria Engineering Journal |
| spelling | doaj.art-852c4868bd1144699dc17e4db06d0ad02022-12-22T02:12:03ZengElsevierAlexandria Engineering Journal1110-01682022-08-0161860236041Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN modelMohamad Haniff Junos0Anis Salwa Mohd Khairuddin1Mahidzal Dahari2Department of Electrical Engineering, Faculty of Engineering, Universiti Malaya, 50603 Kuala Lumpur, MalaysiaCorresponding author.; Department of Electrical Engineering, Faculty of Engineering, Universiti Malaya, 50603 Kuala Lumpur, MalaysiaDepartment of Electrical Engineering, Faculty of Engineering, Universiti Malaya, 50603 Kuala Lumpur, MalaysiaWith the growing demand for geospatial data, challenging aerial images with high spatial, spectral, and temporal resolution achieve excellent development. Currently, deep Convolutional Neural Network (CNN) structures are applied widely for object detection. Nevertheless, existing deep CNN-based models consist of complex network structures and require immense amounts of graphics processing unit (GPU) computation power with high energy consumption. Thus, achieving efficient real-time object detection for limited memory and processing capacity embedded device is a major challenge. This paper proposes a feasible and lightweight object detection model based on deep CNN where a mobile inverted bottleneck module is adopted in the backbone structure. Moreover, an enhanced spatial pyramid pooling is adopted to increase the receptive field in the network by concatenating the multi-scale local region features. The experimental results demonstrated that the proposed model achieved higher average precision and required the smallest memory storage compared to previous works. Moreover, the proposed model offers the best trade-offs in terms of detection accuracy, model size, and detection time which has excellent potential to be deployed on limited capacity embedded device.http://www.sciencedirect.com/science/article/pii/S1110016821007602Computer visionDeep learningImage processingObject detectionAerial imaging |
| spellingShingle | Mohamad Haniff Junos Anis Salwa Mohd Khairuddin Mahidzal Dahari Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN model Alexandria Engineering Journal Computer vision Deep learning Image processing Object detection Aerial imaging |
| title | Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN model |
| title_full | Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN model |
| title_fullStr | Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN model |
| title_full_unstemmed | Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN model |
| title_short | Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN model |
| title_sort | automated object detection on aerial images for limited capacity embedded device using a lightweight cnn model |
| topic | Computer vision Deep learning Image processing Object detection Aerial imaging |
| url | http://www.sciencedirect.com/science/article/pii/S1110016821007602 |
| work_keys_str_mv | AT mohamadhaniffjunos automatedobjectdetectiononaerialimagesforlimitedcapacityembeddeddeviceusingalightweightcnnmodel AT anissalwamohdkhairuddin automatedobjectdetectiononaerialimagesforlimitedcapacityembeddeddeviceusingalightweightcnnmodel AT mahidzaldahari automatedobjectdetectiononaerialimagesforlimitedcapacityembeddeddeviceusingalightweightcnnmodel |