A review on various methodologies used for vehicle classification, helmet detection and number plate recognition
- PDF / 640,647 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 91 Downloads / 179 Views
SPECIAL ISSUE
A review on various methodologies used for vehicle classification, helmet detection and number plate recognition S. Sanjana1 · S. Sanjana1 · V. R. Shriya1 · Gururaj Vaishnavi1 · K. Ashwini1 Received: 12 June 2020 / Revised: 16 August 2020 / Accepted: 11 September 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Vehicle detection and classification has been an area of application of image processing and machine learning which is being researched extensively in accordance with its importance due to increasing number of vehicles, traffic rule defaulters and accidents. This paper aims to review various methodologies used and how it has evolved to give better results in the past years, closely moving towards usage of machine learning. This has resulted in advancing the problem statement towards helmet detection followed by number plate detection of defaulters. Object detection and Text recognition that are available in various frameworks offer built-in models which are easy to use or offer easy methods to build and train customized models. Keywords Object detection · Text recognition · Classification · Image processing · Machine learning
1 Introduction Object classification and recognition was previously accomplished through image processing (IP) techniques alone. Hand-crafted feature extraction algorithms were constructed for specific applications. Occlusion, illumination, weather conditions, poor video quality, static object changes in background etc., are some of the various challenges faced by it. Accuracy of these models depends mainly on factors like angle of inclination of the camera and shadow removal [1]. With the rapid advancements in machine learning and deep learning, IP is being used in conjecture with CNNs to arrive at better results [2].
1.1 Image processing Computer Vision is a major field of interest in practical applications like autonomous vehicles [3], video surveillance that uses object tracking [4], robotics. etc. Image * K. Ashwini [email protected] S. Sanjana [email protected] S. Sanjana [email protected] 1
Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru, India
Processing portrays a significant role in enabling Computer Vision. Digital Image Processing is the domain which helps a machine to view its surroundings. The images that are captured by the machine’s camera are processed by the image processing algorithms, making it easier for the machine to detect, track and classify objects etc. Image processing also helps in enhancing, restoring existing images. Filtering, transforms, morphological operations, clustering, segmentation, edge detection, feature extraction etc. are the basic concepts used [5]. Videos can be broken down into frames and can be processed using Image Processing in the same way as images. With increase in availability of resources like computational power, storage etc. using IP algorithms for video surveillance has also increased [6]. OpenCV [7] is an open source com
Data Loading...
