Technologies and services aimed at autonomous vehicles and Intelligent Transportation Systems (ITS) are transforming different aspects of everyday life. Automatic Vehicle License Plate Detection (LPD) and License Plate Recognition (LPR) systems for vehicles have various potential uses, discovering a stolen vehicle, monitoring the flow of traffic, automatic car parking systems, surveillance solicitations, toll systems, etc. This paper presents a comprehensive performance comparison of many Automatic Number Plate Recognition (ANPR) algorithms, including that incorporating computer vision, as well as a comprehensive analysis of current methodologies and advancements in ANPR. With the improvement of sophisticated machine learning algorithms, numerous techniques have been developed by researchers in the past decades for efficient LPD and LPR. Even with the greatest algorithms, ANPR system implementation may require multiple methods for optimum accuracy. However, the detection of License Plate (LP) is a more challenging task concerning the nature of the number plate, camera quality, non-standard formats, scene complexity, contrast problems, indoor/outdoor or day/night images, distortion tolerance, camera mount position, software tools may affect its effectiveness. This paper not only describes the numerous strategies used for LPD and LPR but also compares their performance and gives suggestions to help researchers choose the optimal technique for their work.