Step 1: Input image of a car.
Step 2: Detecting the License Plate. (output as confidence and coordinates)
Step 3: Detecting the characters.
Step 4: Recognizing the characters.
In some cases there will be problems in accuracy for certain groups as they look similar like: [0, O, D] [I, 1] (Like in the 3rd image, I was wrongly recognized as 1) [S, 5] [2, Z] [B, 8]