Semantic Segmentation and YOLO Detector over Aerial Vehicle Images

作     者：Asifa Mehmood Qureshi Abdul Haleem Butt Abdulwahab Alazeb Naif Al Mudawi Mohammad Alonazi Nouf Abdullah Almujally Ahmad Jalal Hui Liu 

作者机构：Faculty of Computing and AIAir UniversityIslamabad44000Pakistan Department of Computer ScienceCollege of Computer Science and Information SystemNajran UniversityNajran55461Saudi Arabia Department of Information SystemsCollege of Computer Engineering and SciencesPrince Sattam bin Abdulaziz UniversityAl-Kharj16273Saudi Arabia Department of Information TechnologyCollege of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Cognitive Systems LabUniversity of BremenBremen28359Germany 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2024年第80卷第8期

页      面：3315-3332页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 

基　　金：This researchwas supported by the Deanship of ScientificResearch at Najran University,under the Research Group Funding Program Grant Code(NU/RG/SERC/12/30) This research is supported and funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project Number(PNURSP2024R410) Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia This study is supported via funding from Prince Sattam bin Abdulaziz University Project Number(PSAU/2024/R/1445) 

主　　题：Semantic segmentation YOLOv5 vehicle detection and tracking Kalman filter SURF 

摘      要：Intelligent vehicle tracking and detection are crucial tasks in the realm of highway ***,vehicles come in a range of sizes,which is challenging to detect,affecting the traffic monitoring system’s overall *** learning is considered to be an efficient method for object detection in vision-based *** this paper,we proposed a vision-based vehicle detection and tracking system based on a You Look Only Once version 5(YOLOv5)detector combined with a segmentation *** model consists of six *** the first step,all the extracted traffic sequence images are subjected to pre-processing to remove noise and enhance the contrast level of the *** pre-processed images are segmented by labelling each pixel to extract the uniform regions to aid the detection phase.A single-stage detector YOLOv5 is used to detect and locate vehicles in *** detection was exposed to Speeded Up Robust Feature(SURF)feature extraction to track multiple *** on this,a unique number is assigned to each vehicle to easily locate them in the succeeding image frames by extracting them using the feature-matching ***,we implemented a Kalman filter to track multiple *** the end,the vehicle path is estimated by using the centroid points of the rectangular bounding box predicted by the tracking *** experimental results and comparison reveal that our proposed vehicle detection and tracking system outperformed other state-of-the-art *** proposed implemented system provided 94.1%detection precision for Roundabout and 96.1%detection precision for Vehicle Aerial Imaging from Drone(VAID)datasets,respectively.