YOLO-VEHICLE: REALTIME VEHICLE LICENCE PLATE DETECTION AND CHARACTER RECOGNITION USING YOLOV7 NETWORK
Journal Title: International Journal of Data Science and Artificial Intelligence - Year 2024, Vol 2, Issue 01
Abstract
The demand for a secure lifestyle and travel is increasing due to the rapid development of technology. Since the turn of the century, the number of road vehicles has risen dramatically. The rapid growth of the vehicular sector makes tracking individual vehicles increasingly difficult. In this work, a novel proposed YOLO-VEHICLE has been introduced to detect the licence plate in the highway using Yolov7 network. Initially, a CCTV camera captures the input highway traffic video. The collected video is converted into frames. The frames are detected the license plate using the YOLOv7 network. The detected Licence plates (LP) are segmented for partitions a digital image into discrete groups of pixels using U-Net. Finally, the segmented LP recognizing the character for clear view. The simulation outcomes show the performance is assessed by using the accuracy reached by the proposed YOLO-VEHICLE method, as well as its accuracy (ACU), precision (PRE), recall (RCL), and F1 score (F1S). According to the results, the proposed network accuracy was 99.59 %. In the comparison, the YOLOv7 network improves the overall accuracy of the YOLOv3, YOLOv4, and YOLOv5 is 95.14%, 96.32%, and 97.36% respectively. The YOLO-VEHICLE approach improves the overall accuracy of 13.37%, 2.13%, 14.03% better than edge intelligence-based enhanced YOLOv4, Faster R-CNN, and recognition system respectively.
Authors and Affiliations
R. A. Mabel Rose, J. Vasuki and N. Bhavana
Keywords
Related Articles
- EP ID EP734435
- DOI -
- Views 85
- Downloads 0
DEEP REINFORCEMENT LEARNING BASED REAL TIME VIOLENCE DETECTION
In recent days violence is rampant, and the violence that occurs in public places is considered trembling violence in this study. To avert these anomalous activities, a real-time violence identification model is required...
A NOVEL INTERNET OF THINGS-BASED ELECTROCARDIOGRAM DENOISING METHOD USING MEDIAN MODIFIED WEINER AND EXTENDED KALMAN FILTERS
The Internet of Things (IoT) offers healthcare applications that benefit customers, physicians, hospitals, and insurance companies. Wearable technology like fitness bands and other wirelessly connected gadgets like blood...
DINGO OPTIMIZED FUZZY CNN TECHNIQUE FOR EFFICIENT PROTEIN STRUCTURE PREDICTION
Protein is made up of a variety of molecules that are required by living organisms, such as enzymes, hormones, and antibodies. In step 2, the max-pooling layer and the convolutional layer evaluate the input data to creat...
CLASSIFICATION OF LIVER CANCER VIA DEEP LEARNING BASED DILATED ATTENTION CONVOLUTIONAL NEURAL NETWORK
Liver cancer occur when normal cells develop aberrant DNA alterations and reproduce uncontrollably. Patients with cirrhosis, hepatitis B or C, or both have an increased risk of developing the progressing stage of cancer....
CHICKEN SWARM OPTIMIZATION BASED ENSEMBLED LEARNING CLASSIFIER FOR BLACK HOLE ATTACK IN WIRELESS SENSOR NETWORK
Wireless Sensor Networks (WSNs) are an inevitable technology prevalently used in various critical and remote monitoring applications. The security of WSNs is compromised by various attacks in wireless mediums. Even thoug...