forked from Deevoluation/ALPR
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Control_flow.txt
25 lines (20 loc) · 2.04 KB
/
Control_flow.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
The Work flow of the Program:
Main.main ------> DetectChars.loadKNNDataAndTrainKNN() --> Load classification.txt and flattened_image.txt to train a knn classifier for handwritten digit. return the classifier object
(create an object of the OpenCV knn classifier)
-------> DetectPlates.detectPlatesInScene() ----> Detecting all the possible plates in the image
-----> Preprocess.preprocess() ---> convert the image to grayscale,increase the visibility of plate_area(by blurring and thresholding) ,return the grayscale and thresholded object.
-----> findPossibleCharsInScene() ---> extract all the contours from the image,
---> PossibleChar.PossibleChar() ---> create a class object of each conti=our that contain physical informations of that particular contour
---> DetectChars.checkIfPossibleChar() ---> A small check that if the current contour object can be a character or not, if yes than append the contour to listOfPossibleChars
---> return listOfPossibleChars
-----> re-arrange the one big list of chars into a list of lists of matching chars,where each list can be a license plate in itself.
---> return listOfListsOfMatchingChars
-----> extractPlate ---> Attempt to extract plates.
---> PossiblePlate.PossiblePlate() --> Make a class object of the Possible Plate class to store the desires informations of the detected plates
---> DetectChars.distanceBetweenChars
---> return possiblePlate
-----> return listOfPossiblePlates
------> DetectChars.detectCharsInPlates -----> detect chars in plates
-----> return listOfPossiblePlates
-------> drawRedRectangleAroundPlate ----> Draw a box around the prabable position of the name plates
-------> writeLicensePlateCharsOnImage ----> write license plate text on the image