FGSC: Fuzzy Guided Scale Choice SSD Model for Edge AI Design on Real-Time Vehicle Detection and Class Counting
The aim of this paper is to distinguish the vehicle detection and count the class number in each classification from the inputs. We proposed the use of Fuzzy Guided Scale Choice (FGSC)-based SSD deep neural network architecture for vehicle detection and class counting with parameter optimization. Th...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/21/7399 |
| _version_ | 1797511749956534272 |
|---|---|
| author | Ming-Hwa Sheu S. M. Salahuddin Morsalin Jia-Xiang Zheng Shih-Chang Hsia Cheng-Jian Lin Chuan-Yu Chang |
| author_facet | Ming-Hwa Sheu S. M. Salahuddin Morsalin Jia-Xiang Zheng Shih-Chang Hsia Cheng-Jian Lin Chuan-Yu Chang |
| author_sort | Ming-Hwa Sheu |
| collection | DOAJ |
| description | The aim of this paper is to distinguish the vehicle detection and count the class number in each classification from the inputs. We proposed the use of Fuzzy Guided Scale Choice (FGSC)-based SSD deep neural network architecture for vehicle detection and class counting with parameter optimization. The ‘FGSC’ blocks are integrated into the convolutional layers of the model, which emphasize essential features while ignoring less important ones that are not significant for the operation. We created the passing detection lines and class counting windows and connected them with the proposed FGSC-SSD deep neural network model. The ‘FGSC’ blocks in the convolution layer emphasize essential features and find out unnecessary features by using the scale choice method at the training stage and eliminate that significant speedup of the model. In addition, FGSC blocks avoided many unusable parameters in the saturation interval and improved the performance efficiency. In addition, the Fuzzy Sigmoid Function (FSF) increases the activation interval through fuzzy logic. While performing operations, the FGSC-SSD model reduces the computational complexity of convolutional layers and their parameters. As a result, the model tested Frames Per Second (FPS) on edge artificial intelligence (AI) and reached a real-time processing speed of 38.4 and an accuracy rate of more than 94%. Therefore, this work might be considered an improvement to the traffic monitoring approach by using edge AI applications. |
| first_indexed | 2024-03-10T05:52:27Z |
| format | Article |
| id | doaj.art-ecbaca4a0aa54a0cabbbf4ddcbf7b907 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T05:52:27Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-ecbaca4a0aa54a0cabbbf4ddcbf7b9072023-11-22T21:41:01ZengMDPI AGSensors1424-82202021-11-012121739910.3390/s21217399FGSC: Fuzzy Guided Scale Choice SSD Model for Edge AI Design on Real-Time Vehicle Detection and Class CountingMing-Hwa Sheu0S. M. Salahuddin Morsalin1Jia-Xiang Zheng2Shih-Chang Hsia3Cheng-Jian Lin4Chuan-Yu Chang5Department of Electronic Engineering, National Yunlin University of Science and Technology, Douliu 64002, TaiwanDepartment of Electronic Engineering, National Yunlin University of Science and Technology, Douliu 64002, TaiwanDepartment of Electronic Engineering, National Yunlin University of Science and Technology, Douliu 64002, TaiwanDepartment of Electronic Engineering, National Yunlin University of Science and Technology, Douliu 64002, TaiwanDepartment of Computer Science and Information Engineering, National Chin-Yi University of Technology, Taichung 411030, TaiwanComputer Science and Information Engineering, National Yunlin University of Science and Technology, Douliu 64002, TaiwanThe aim of this paper is to distinguish the vehicle detection and count the class number in each classification from the inputs. We proposed the use of Fuzzy Guided Scale Choice (FGSC)-based SSD deep neural network architecture for vehicle detection and class counting with parameter optimization. The ‘FGSC’ blocks are integrated into the convolutional layers of the model, which emphasize essential features while ignoring less important ones that are not significant for the operation. We created the passing detection lines and class counting windows and connected them with the proposed FGSC-SSD deep neural network model. The ‘FGSC’ blocks in the convolution layer emphasize essential features and find out unnecessary features by using the scale choice method at the training stage and eliminate that significant speedup of the model. In addition, FGSC blocks avoided many unusable parameters in the saturation interval and improved the performance efficiency. In addition, the Fuzzy Sigmoid Function (FSF) increases the activation interval through fuzzy logic. While performing operations, the FGSC-SSD model reduces the computational complexity of convolutional layers and their parameters. As a result, the model tested Frames Per Second (FPS) on edge artificial intelligence (AI) and reached a real-time processing speed of 38.4 and an accuracy rate of more than 94%. Therefore, this work might be considered an improvement to the traffic monitoring approach by using edge AI applications.https://www.mdpi.com/1424-8220/21/21/7399fuzzy guided scale choicefuzzy sigmoid functionvehicle detectionfuzzy logicvehicle class countingand intelligent AIoT vehicles application |
| spellingShingle | Ming-Hwa Sheu S. M. Salahuddin Morsalin Jia-Xiang Zheng Shih-Chang Hsia Cheng-Jian Lin Chuan-Yu Chang FGSC: Fuzzy Guided Scale Choice SSD Model for Edge AI Design on Real-Time Vehicle Detection and Class Counting Sensors fuzzy guided scale choice fuzzy sigmoid function vehicle detection fuzzy logic vehicle class counting and intelligent AIoT vehicles application |
| title | FGSC: Fuzzy Guided Scale Choice SSD Model for Edge AI Design on Real-Time Vehicle Detection and Class Counting |
| title_full | FGSC: Fuzzy Guided Scale Choice SSD Model for Edge AI Design on Real-Time Vehicle Detection and Class Counting |
| title_fullStr | FGSC: Fuzzy Guided Scale Choice SSD Model for Edge AI Design on Real-Time Vehicle Detection and Class Counting |
| title_full_unstemmed | FGSC: Fuzzy Guided Scale Choice SSD Model for Edge AI Design on Real-Time Vehicle Detection and Class Counting |
| title_short | FGSC: Fuzzy Guided Scale Choice SSD Model for Edge AI Design on Real-Time Vehicle Detection and Class Counting |
| title_sort | fgsc fuzzy guided scale choice ssd model for edge ai design on real time vehicle detection and class counting |
| topic | fuzzy guided scale choice fuzzy sigmoid function vehicle detection fuzzy logic vehicle class counting and intelligent AIoT vehicles application |
| url | https://www.mdpi.com/1424-8220/21/21/7399 |
| work_keys_str_mv | AT minghwasheu fgscfuzzyguidedscalechoicessdmodelforedgeaidesignonrealtimevehicledetectionandclasscounting AT smsalahuddinmorsalin fgscfuzzyguidedscalechoicessdmodelforedgeaidesignonrealtimevehicledetectionandclasscounting AT jiaxiangzheng fgscfuzzyguidedscalechoicessdmodelforedgeaidesignonrealtimevehicledetectionandclasscounting AT shihchanghsia fgscfuzzyguidedscalechoicessdmodelforedgeaidesignonrealtimevehicledetectionandclasscounting AT chengjianlin fgscfuzzyguidedscalechoicessdmodelforedgeaidesignonrealtimevehicledetectionandclasscounting AT chuanyuchang fgscfuzzyguidedscalechoicessdmodelforedgeaidesignonrealtimevehicledetectionandclasscounting |