Mixed YOLOv3-LITE: A Lightweight Real-Time Object Detection Method
Embedded and mobile smart devices face problems related to limited computing power and excessive power consumption. To address these problems, we propose Mixed YOLOv3-LITE, a lightweight real-time object detection network that can be used with non-graphics processing unit (GPU) and mobile devices. B...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/7/1861 |
| _version_ | 1827761340850110464 |
|---|---|
| author | Haipeng Zhao Yang Zhou Long Zhang Yangzhao Peng Xiaofei Hu Haojie Peng Xinyue Cai |
| author_facet | Haipeng Zhao Yang Zhou Long Zhang Yangzhao Peng Xiaofei Hu Haojie Peng Xinyue Cai |
| author_sort | Haipeng Zhao |
| collection | DOAJ |
| description | Embedded and mobile smart devices face problems related to limited computing power and excessive power consumption. To address these problems, we propose Mixed YOLOv3-LITE, a lightweight real-time object detection network that can be used with non-graphics processing unit (GPU) and mobile devices. Based on YOLO-LITE as the backbone network, Mixed YOLOv3-LITE supplements residual block (ResBlocks) and parallel high-to-low resolution subnetworks, fully utilizes shallow network characteristics while increasing network depth, and uses a “shallow and narrow” convolution layer to build a detector, thereby achieving an optimal balance between detection precision and speed when used with non-GPU based computers and portable terminal devices. The experimental results obtained in this study reveal that the size of the proposed Mixed YOLOv3-LITE network model is 20.5 MB, which is 91.70%, 38.07%, and 74.25% smaller than YOLOv3, tiny-YOLOv3, and SlimYOLOv3-spp3-50, respectively. The mean average precision (mAP) achieved using the PASCAL VOC 2007 dataset is 48.25%, which is 14.48% higher than that of YOLO-LITE. When the VisDrone 2018-Det dataset is used, the mAP achieved with the Mixed YOLOv3-LITE network model is 28.50%, which is 18.50% and 2.70% higher than tiny-YOLOv3 and SlimYOLOv3-spp3-50, respectively. The results prove that Mixed YOLOv3-LITE can achieve higher efficiency and better performance on mobile terminals and other devices. |
| first_indexed | 2024-03-11T10:11:31Z |
| format | Article |
| id | doaj.art-1eee1ab85fb8463c9c6923fafa7414ca |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T10:11:31Z |
| publishDate | 2020-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-1eee1ab85fb8463c9c6923fafa7414ca2023-11-16T14:28:20ZengMDPI AGSensors1424-82202020-03-01207186110.3390/s20071861Mixed YOLOv3-LITE: A Lightweight Real-Time Object Detection MethodHaipeng Zhao0Yang Zhou1Long Zhang2Yangzhao Peng3Xiaofei Hu4Haojie Peng5Xinyue Cai6The Institute of Geospatial Information, Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaThe Institute of Geospatial Information, Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaBeijing Institute of Remote Sensing Information, Beijing 100192, ChinaThe Institute of Geospatial Information, Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaThe Institute of Geospatial Information, Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaThe Institute of Geospatial Information, Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaThe Institute of Geospatial Information, Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaEmbedded and mobile smart devices face problems related to limited computing power and excessive power consumption. To address these problems, we propose Mixed YOLOv3-LITE, a lightweight real-time object detection network that can be used with non-graphics processing unit (GPU) and mobile devices. Based on YOLO-LITE as the backbone network, Mixed YOLOv3-LITE supplements residual block (ResBlocks) and parallel high-to-low resolution subnetworks, fully utilizes shallow network characteristics while increasing network depth, and uses a “shallow and narrow” convolution layer to build a detector, thereby achieving an optimal balance between detection precision and speed when used with non-GPU based computers and portable terminal devices. The experimental results obtained in this study reveal that the size of the proposed Mixed YOLOv3-LITE network model is 20.5 MB, which is 91.70%, 38.07%, and 74.25% smaller than YOLOv3, tiny-YOLOv3, and SlimYOLOv3-spp3-50, respectively. The mean average precision (mAP) achieved using the PASCAL VOC 2007 dataset is 48.25%, which is 14.48% higher than that of YOLO-LITE. When the VisDrone 2018-Det dataset is used, the mAP achieved with the Mixed YOLOv3-LITE network model is 28.50%, which is 18.50% and 2.70% higher than tiny-YOLOv3 and SlimYOLOv3-spp3-50, respectively. The results prove that Mixed YOLOv3-LITE can achieve higher efficiency and better performance on mobile terminals and other devices.https://www.mdpi.com/1424-8220/20/7/1861object detectioncomputer visionconvolutional neural networkembedded systemreal-time performance |
| spellingShingle | Haipeng Zhao Yang Zhou Long Zhang Yangzhao Peng Xiaofei Hu Haojie Peng Xinyue Cai Mixed YOLOv3-LITE: A Lightweight Real-Time Object Detection Method Sensors object detection computer vision convolutional neural network embedded system real-time performance |
| title | Mixed YOLOv3-LITE: A Lightweight Real-Time Object Detection Method |
| title_full | Mixed YOLOv3-LITE: A Lightweight Real-Time Object Detection Method |
| title_fullStr | Mixed YOLOv3-LITE: A Lightweight Real-Time Object Detection Method |
| title_full_unstemmed | Mixed YOLOv3-LITE: A Lightweight Real-Time Object Detection Method |
| title_short | Mixed YOLOv3-LITE: A Lightweight Real-Time Object Detection Method |
| title_sort | mixed yolov3 lite a lightweight real time object detection method |
| topic | object detection computer vision convolutional neural network embedded system real-time performance |
| url | https://www.mdpi.com/1424-8220/20/7/1861 |
| work_keys_str_mv | AT haipengzhao mixedyolov3litealightweightrealtimeobjectdetectionmethod AT yangzhou mixedyolov3litealightweightrealtimeobjectdetectionmethod AT longzhang mixedyolov3litealightweightrealtimeobjectdetectionmethod AT yangzhaopeng mixedyolov3litealightweightrealtimeobjectdetectionmethod AT xiaofeihu mixedyolov3litealightweightrealtimeobjectdetectionmethod AT haojiepeng mixedyolov3litealightweightrealtimeobjectdetectionmethod AT xinyuecai mixedyolov3litealightweightrealtimeobjectdetectionmethod |