A Systematic Review Of Current Methods And Challenges In Automatic Number Plate Recognition

Recent years have witnessed an unprecedented increase in the number of automobiles that have been put to use thereby leading to the design and development of a monitoring system like Automated Number Plate Recognition (ANPR) to deal with several tasks like law and order, supervision, and highway kiosk related operations. The functioning pattern as well as the requirement specification of these structures is varied due to multiple disparities encountered in the proposed systems. To substantiate the case in point, the majority of these applications are executed on smartphones or any other handheld gadgets on cloud servers, maneuvered in poorly illuminated, unfavourable weather situations. There have been numerous approaches that have been designed to implement ANPR that can efficiently suit the aforementioned real-time requirements. Generally, these systems have a pre-defined framework whose major tasks include the detection of license plates, character segmentation, and recognition of individual characters. Owing to the diversity of license plate formats and constraints with respect to environmental conditions, efficiency and preciseness of recognition highly rely on an accurate detection process. This research investigates several methods and techniques utilized in ANPR systems and presents a detailed review of them as a literature survey. Our objective is thus to facilitate towards beneficial analysis, and categorization of associated studies in the domain of ANPR and recognize the critical challenges encountered by scholars and application developers. Moreover, we also intend to present relevant research directions and a comprehensive analysis of modern real-time systems that can be employed in future ventures along with commendations to optimize the existing solutions to enable them to work under intense situations.