The majority of current licence plate (LP) detection systems have advanced dramatically in image processing, but with limitations due to environmental factors and plate variations. Different lighting, temperature, and background conditions are among the environmental factors. Plate variations include plates mounted anywhere on the vehicle, multiple plates in a single picture, numerous combinations of vehicles with different plate orientations, various sizes of plates, background colour of plates, dirt on plates, rotated plates, LPs with two lines of characters, and tilted plates. With increased mobility and internationalisation comes the need for a universal LP detection system that can handle LPs from any country and any vehicle, including motorcycles, in any weather condition. This paper introduces a novel LP detection method based on geometrical properties of the LP characters, as well as a proposed new character extraction method for LP character components that are skipped due to noise between the LP characters and the LP boundary. Because of the geometrical properties of the set of characters in LP, the proposed method detects the number plate of any type of vehicle (including buses, cars, trucks, bikes, and so on) with different plate variations, under various environmental and weather conditions. The proposed approach is not influenced by LP colour, rotation, or scale variances. Normal media-lab and Application Oriented License Plate (AOLP) benchmark LP recognition databases are used to test the principle. The proposed solution for LP detection has a success rate of 97.3 percent when using the media-lab database and 93.7 percent when using the AOLP database. The findings show that the proposed method is comparable to previously published papers that compared their output to publicly accessible benchmark databases.
Author (s) Details
Dr. Narasimha Reddy Soora
Department of Computer Science and Engineering, Kakatiya Institute of Technology and Science, Warangal, Telangana, India.
View Book :- https://stm.bookpi.org/AAER-V3/issue/view/42