IMPROVING ALGORITHM FOR VEHICLE MODEL USING IMAGE PROCESSING | ||||
| Aswan Science and Technology Bulletin | ||||
| Volume 2, Issue 2, December 2024, Page 58-74 PDF (679.21 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/astb.2024.310538.1003 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
IEHAB ABDULJABBAR KAMIL   1; MOHANAD Abdulsallam AL-ASKARI 2
| ||||
| 1UNIVERSITY OF ANBAR | ||||
| 2Department Of Information Systems College Of COMPUTER SCIENCES AND INFORMATION TECHNOLOGY, UNIVERSITY OF ANBAR, IRAQ | ||||
| Abstract | ||||
| The research will employ contemporary image processing with convolutional neural network technology to enhance automobile model recognition and categorisation. Image processing improves the accuracy and efficiency of car identification and categorization. Control of traffic and safety surveillance might benefit significantly from this technology. CNNs improve vehicle recognition accuracy and efficiency. It examines complex image processing techniques, including bilateral filtrating and directional diffusion. PCA, or principal component analysis, reduces the number of parameters in models with multiple dimensions. This is essential to minimising computational difficulty and limiting over-fitting while preserving system quality. This strategy improves model efficiency and accuracy by targeting the most critical data discrepancies. Monocular vision, along with infrared sensors, are essential for vehicle detection. The CNN algorithm, trained on two-dimensional images and three-dimensional Bezier curves, reduces restoration errors and accurately recognizes automobile models. The results showed fewer mistakes, greater precision, recall, and F1 rating scores. In order to enhance car recognition and classification, further research is needed to expand databases and examine hybrid solutions. | ||||
| Keywords | ||||
| Image Processing; Convolutional Neural Networks (CNNs); Deep Learning; Principal Component Analysis (PCA); Traffic | ||||
|
Statistics Article View: 119 PDF Download: 182 |
||||
