Image Processing in Automatic License Plate Recognition Using Combined Methods
Serdica Journal of Computing, Tome 16 (2022) no. 1, pp. 1-23
Cet article a éte moissonné depuis la source Bulgarian Digital Mathematics Library
There are many existing studies released in the field of Computer Vision, especially the field of Automatic License Plate Recognition. However, most of them are focused on using one method at the time, such as Thresholding algorithms, Edge Detections or Morphological transformations. This research paper proposes to automate the License plate recognition process, by combining four algorithms from the three methods mentioned above: Adaptive Thresholding, Otsu's Thresholding, Canny Edge Detection and Morphological Gradient applied to Edge Detection. The Goal achieved is to obtain the best binary image from those methods, and the statistical technique used in, is the median of pixel's intensity of all output images obtained by the four methods. Additionally, this research offers a comparative study on thresholding techniques to choose the best method for binarizing an image, which is the first and crucial step of Automatic License Plate Recognition Process.
Keywords:
Image Pre-Processing, Deep Learning, Automatic License Plate Recognition, Thresholding Image, Canny Edge Detection, Morphological Transformations, Otsu Algorithm, Adaptive Thresholding, Global Thresholding
@article{SJC_2022_16_1_a1,
author = {Hamdoun, Nabila and Mentagui, Driss},
title = {Image {Processing} in {Automatic} {License} {Plate} {Recognition} {Using} {Combined} {Methods}},
journal = {Serdica Journal of Computing},
pages = {1--23},
year = {2022},
volume = {16},
number = {1},
language = {en},
url = {http://geodesic.mathdoc.fr/item/SJC_2022_16_1_a1/}
}
Hamdoun, Nabila; Mentagui, Driss. Image Processing in Automatic License Plate Recognition Using Combined Methods. Serdica Journal of Computing, Tome 16 (2022) no. 1, pp. 1-23. http://geodesic.mathdoc.fr/item/SJC_2022_16_1_a1/