DetectChars.py

# DetectChars.py
import os

import cv2
import numpy as np
import math
import random

import Main
import Preprocess
import PossibleChar

# module level variables ##########################################################################

kNearest = cv2.ml.KNearest_create()

        # constants for checkIfPossibleChar, this checks one possible char only (does not compare to another char)
MIN_PIXEL_WIDTH = 2
MIN_PIXEL_HEIGHT = 8

MIN_ASPECT_RATIO = 0.25
MAX_ASPECT_RATIO = 1.0

MIN_PIXEL_AREA = 80

        # constants for comparing two chars
MIN_DIAG_SIZE_MULTIPLE_AWAY = 0.3
MAX_DIAG_SIZE_MULTIPLE_AWAY = 5.0

MAX_CHANGE_IN_AREA = 0.5

MAX_CHANGE_IN_WIDTH = 0.8
MAX_CHANGE_IN_HEIGHT = 0.2

MAX_ANGLE_BETWEEN_CHARS = 12.0

        # other constants
MIN_NUMBER_OF_MATCHING_CHARS = 3

RESIZED_CHAR_IMAGE_WIDTH = 20
RESIZED_CHAR_IMAGE_HEIGHT = 30

MIN_CONTOUR_AREA = 100

###################################################################################################
def loadKNNDataAndTrainKNN():
    allContoursWithData = []                # declare empty lists,
    validContoursWithData = []              # we will fill these shortly

    try:
        npaClassifications = np.loadtxt("classifications.txt", np.float32)                  # read in training classifications
    except:                                                                                 # if file could not be opened
        print("error, unable to open classifications.txt, exiting program\n")  # show error message
        os.system("pause")
        return False                                                                        # and return False
    # end try

    try:
        npaFlattenedImages = np.loadtxt("flattened_images.txt", np.float32)                 # read in training images
    except:                                                                                 # if file could not be opened
        print("error, unable to open flattened_images.txt, exiting program\n")  # show error message
        os.system("pause")
        return False                                                                        # and return False
    # end try

    npaClassifications = npaClassifications.reshape((npaClassifications.size, 1))       # reshape numpy array to 1d, necessary to pass to call to train

    kNearest.setDefaultK(1)                                                             # set default K to 1

    kNearest.train(npaFlattenedImages, cv2.ml.ROW_SAMPLE, npaClassifications)           # train KNN object

    return True                             # if we got here training was successful so return true
# end function

###################################################################################################
def detectCharsInPlates(listOfPossiblePlates):
    intPlateCounter = 0
    imgContours = None
    contours = []

    if len(listOfPossiblePlates) == 0:          # if list of possible plates is empty
        return listOfPossiblePlates             # return
    # end if

            # at this point we can be sure the list of possible plates has at least one plate

    for possiblePlate in listOfPossiblePlates:          # for each possible plate, this is a big for loop that takes up most of the function

        possiblePlate.imgGrayscale, possiblePlate.imgThresh = Preprocess.preprocess(possiblePlate.imgPlate)     # preprocess to get grayscale and threshold images

        if Main.showSteps == True: # show steps ###################################################
            cv2.imshow("5a", possiblePlate.imgPlate)
            cv2.imshow("5b", possiblePlate.imgGrayscale)
            cv2.imshow("5c", possiblePlate.imgThresh)
        # end if # show steps #####################################################################

                # increase size of plate image for easier viewing and char detection
        possiblePlate.imgThresh = cv2.resize(possiblePlate.imgThresh, (0, 0), fx = 1.6, fy = 1.6)

                # threshold again to eliminate any gray areas
        thresholdValue, possiblePlate.imgThresh = cv2.threshold(possiblePlate.imgThresh, 0.0, 255.0, cv2.THRESH_BINARY | cv2.THRESH_OTSU)

        if Main.showSteps == True: # show steps ###################################################
            cv2.imshow("5d", possiblePlate.imgThresh)
        # end if # show steps #####################################################################

                # find all possible chars in the plate,
                # this function first finds all contours, then only includes contours that could be chars (without comparison to other chars yet)
        listOfPossibleCharsInPlate = findPossibleCharsInPlate(possiblePlate.imgGrayscale, possiblePlate.imgThresh)

        if Main.showSteps == True: # show steps ###################################################
            height, width, numChannels = possiblePlate.imgPlate.shape
            imgContours = np.zeros((height, width, 3), np.uint8)
            del contours[:]                                         # clear the contours list

            for possibleChar in listOfPossibleCharsInPlate:
                contours.append(possibleChar.contour)
            # end for

            cv2.drawContours(imgContours, contours, -1, Main.SCALAR_WHITE)

            cv2.imshow("6", imgContours)
        # end if # show steps #####################################################################

                # given a list of all possible chars, find groups of matching chars within the plate
        listOfListsOfMatchingCharsInPlate = findListOfListsOfMatchingChars(listOfPossibleCharsInPlate)

        if Main.showSteps == True: # show steps ###################################################
            imgContours = np.zeros((height, width, 3), np.uint8)
            del contours[:]

            for listOfMatchingChars in listOfListsOfMatchingCharsInPlate:
                intRandomBlue = random.randint(0, 255)
                intRandomGreen = random.randint(0, 255)
                intRandomRed = random.randint(0, 255)

                for matchingChar in listOfMatchingChars:
                    contours.append(matchingChar.contour)
                # end for
                cv2.drawContours(imgContours, contours, -1, (intRandomBlue, intRandomGreen, intRandomRed))
            # end for
            cv2.imshow("7", imgContours)
        # end if # show steps #####################################################################

        if (len(listOfListsOfMatchingCharsInPlate) == 0):			# if no groups of matching chars were found in the plate

            if Main.showSteps == True: # show steps ###############################################
                print("chars found in plate number " + str(
                    intPlateCounter) + " = (none), click on any image and press a key to continue . . .")
                intPlateCounter = intPlateCounter + 1
                cv2.destroyWindow("8")
                cv2.destroyWindow("9")
                cv2.destroyWindow("10")
                cv2.waitKey(0)
            # end if # show steps #################################################################

            possiblePlate.strChars = ""
            continue						# go back to top of for loop
        # end if

        for i in range(0, len(listOfListsOfMatchingCharsInPlate)):                              # within each list of matching chars
            listOfListsOfMatchingCharsInPlate[i].sort(key = lambda matchingChar: matchingChar.intCenterX)        # sort chars from left to right
            listOfListsOfMatchingCharsInPlate[i] = removeInnerOverlappingChars(listOfListsOfMatchingCharsInPlate[i])              # and remove inner overlapping chars
        # end for

        if Main.showSteps == True: # show steps ###################################################
            imgContours = np.zeros((height, width, 3), np.uint8)

            for listOfMatchingChars in listOfListsOfMatchingCharsInPlate:
                intRandomBlue = random.randint(0, 255)
                intRandomGreen = random.randint(0, 255)
                intRandomRed = random.randint(0, 255)

                del contours[:]

                for matchingChar in listOfMatchingChars:
                    contours.append(matchingChar.contour)
                # end for

                cv2.drawContours(imgContours, contours, -1, (intRandomBlue, intRandomGreen, intRandomRed))
            # end for
            cv2.imshow("8", imgContours)
        # end if # show steps #####################################################################

                # within each possible plate, suppose the longest list of potential matching chars is the actual list of chars
        intLenOfLongestListOfChars = 0
        intIndexOfLongestListOfChars = 0

                # loop through all the vectors of matching chars, get the index of the one with the most chars
        for i in range(0, len(listOfListsOfMatchingCharsInPlate)):
            if len(listOfListsOfMatchingCharsInPlate[i]) > intLenOfLongestListOfChars:
                intLenOfLongestListOfChars = len(listOfListsOfMatchingCharsInPlate[i])
                intIndexOfLongestListOfChars = i
            # end if
        # end for

                # suppose that the longest list of matching chars within the plate is the actual list of chars
        longestListOfMatchingCharsInPlate = listOfListsOfMatchingCharsInPlate[intIndexOfLongestListOfChars]

        if Main.showSteps == True: # show steps ###################################################
            imgContours = np.zeros((height, width, 3), np.uint8)
            del contours[:]

            for matchingChar in longestListOfMatchingCharsInPlate:
                contours.append(matchingChar.contour)
            # end for

            cv2.drawContours(imgContours, contours, -1, Main.SCALAR_WHITE)

            cv2.imshow("9", imgContours)
        # end if # show steps #####################################################################

        possiblePlate.strChars = recognizeCharsInPlate(possiblePlate.imgThresh, longestListOfMatchingCharsInPlate)

        if Main.showSteps == True: # show steps ###################################################
            print("chars found in plate number " + str(
                intPlateCounter) + " = " + possiblePlate.strChars + ", click on any image and press a key to continue . . .")
            intPlateCounter = intPlateCounter + 1
            cv2.waitKey(0)
        # end if # show steps #####################################################################

    # end of big for loop that takes up most of the function

    if Main.showSteps == True:
        print("\nchar detection complete, click on any image and press a key to continue . . .\n")
        cv2.waitKey(0)
    # end if

    return listOfPossiblePlates
# end function

###################################################################################################
def findPossibleCharsInPlate(imgGrayscale, imgThresh):
    listOfPossibleChars = []                        # this will be the return value
    contours = []
    imgThreshCopy = imgThresh.copy()

            # find all contours in plate
    (_, contours, _) = cv2.findContours(imgThreshCopy, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

    for contour in contours:                        # for each contour
        possibleChar = PossibleChar.PossibleChar(contour)

        if checkIfPossibleChar(possibleChar):              # if contour is a possible char, note this does not compare to other chars (yet) . . .
            listOfPossibleChars.append(possibleChar)       # add to list of possible chars
        # end if
    # end if

    return listOfPossibleChars
# end function

###################################################################################################
def checkIfPossibleChar(possibleChar):
            # this function is a 'first pass' that does a rough check on a contour to see if it could be a char,
            # note that we are not (yet) comparing the char to other chars to look for a group
    if (possibleChar.intBoundingRectArea > MIN_PIXEL_AREA and
        possibleChar.intBoundingRectWidth > MIN_PIXEL_WIDTH and possibleChar.intBoundingRectHeight > MIN_PIXEL_HEIGHT and
        MIN_ASPECT_RATIO < possibleChar.fltAspectRatio and possibleChar.fltAspectRatio < MAX_ASPECT_RATIO):
        return True
    else:
        return False
    # end if
# end function

###################################################################################################
def findListOfListsOfMatchingChars(listOfPossibleChars):
            # with this function, we start off with all the possible chars in one big list
            # the purpose of this function is to re-arrange the one big list of chars into a list of lists of matching chars,
            # note that chars that are not found to be in a group of matches do not need to be considered further
    listOfListsOfMatchingChars = []                  # this will be the return value

    for possibleChar in listOfPossibleChars:                        # for each possible char in the one big list of chars
        listOfMatchingChars = findListOfMatchingChars(possibleChar, listOfPossibleChars)        # find all chars in the big list that match the current char

        listOfMatchingChars.append(possibleChar)                # also add the current char to current possible list of matching chars

        if len(listOfMatchingChars) < MIN_NUMBER_OF_MATCHING_CHARS:     # if current possible list of matching chars is not long enough to constitute a possible plate
            continue                            # jump back to the top of the for loop and try again with next char, note that it's not necessary
                                                # to save the list in any way since it did not have enough chars to be a possible plate
        # end if

                                                # if we get here, the current list passed test as a "group" or "cluster" of matching chars
        listOfListsOfMatchingChars.append(listOfMatchingChars)      # so add to our list of lists of matching chars

        listOfPossibleCharsWithCurrentMatchesRemoved = []

                                                # remove the current list of matching chars from the big list so we don't use those same chars twice,
                                                # make sure to make a new big list for this since we don't want to change the original big list
        listOfPossibleCharsWithCurrentMatchesRemoved = list(set(listOfPossibleChars) - set(listOfMatchingChars))

        recursiveListOfListsOfMatchingChars = findListOfListsOfMatchingChars(listOfPossibleCharsWithCurrentMatchesRemoved)      # recursive call

        for recursiveListOfMatchingChars in recursiveListOfListsOfMatchingChars:        # for each list of matching chars found by recursive call
            listOfListsOfMatchingChars.append(recursiveListOfMatchingChars)             # add to our original list of lists of matching chars
        # end for

        break       # exit for

    # end for

    return listOfListsOfMatchingChars
# end function

###################################################################################################
def findListOfMatchingChars(possibleChar, listOfChars):
            # the purpose of this function is, given a possible char and a big list of possible chars,
            # find all chars in the big list that are a match for the single possible char, and return those matching chars as a list
    listOfMatchingChars = []                # this will be the return value

    for possibleMatchingChar in listOfChars:                # for each char in big list
        if possibleMatchingChar == possibleChar:    # if the char we attempting to find matches for is the exact same char as the char in the big list we are currently checking
                                                    # then we should not include it in the list of matches b/c that would end up double including the current char
            continue                                # so do not add to list of matches and jump back to top of for loop
        # end if
                    # compute stuff to see if chars are a match
        fltDistanceBetweenChars = distanceBetweenChars(possibleChar, possibleMatchingChar)

        fltAngleBetweenChars = angleBetweenChars(possibleChar, possibleMatchingChar)

        fltChangeInArea = float(abs(possibleMatchingChar.intBoundingRectArea - possibleChar.intBoundingRectArea)) / float(possibleChar.intBoundingRectArea)

        fltChangeInWidth = float(abs(possibleMatchingChar.intBoundingRectWidth - possibleChar.intBoundingRectWidth)) / float(possibleChar.intBoundingRectWidth)
        fltChangeInHeight = float(abs(possibleMatchingChar.intBoundingRectHeight - possibleChar.intBoundingRectHeight)) / float(possibleChar.intBoundingRectHeight)

                # check if chars match
        if (fltDistanceBetweenChars < (possibleChar.fltDiagonalSize * MAX_DIAG_SIZE_MULTIPLE_AWAY) and
            fltAngleBetweenChars < MAX_ANGLE_BETWEEN_CHARS and
            fltChangeInArea < MAX_CHANGE_IN_AREA and
            fltChangeInWidth < MAX_CHANGE_IN_WIDTH and
            fltChangeInHeight < MAX_CHANGE_IN_HEIGHT):

            listOfMatchingChars.append(possibleMatchingChar)        # if the chars are a match, add the current char to list of matching chars
        # end if
    # end for

    return listOfMatchingChars                  # return result
# end function

###################################################################################################
# use Pythagorean theorem to calculate distance between two chars
def distanceBetweenChars(firstChar, secondChar):
    intX = abs(firstChar.intCenterX - secondChar.intCenterX)
    intY = abs(firstChar.intCenterY - secondChar.intCenterY)

    return math.sqrt((intX ** 2) + (intY ** 2))
# end function

###################################################################################################
# use basic trigonometry (SOH CAH TOA) to calculate angle between chars
def angleBetweenChars(firstChar, secondChar):
    fltAdj = float(abs(firstChar.intCenterX - secondChar.intCenterX))
    fltOpp = float(abs(firstChar.intCenterY - secondChar.intCenterY))

    if fltAdj != 0.0:                           # check to make sure we do not divide by zero if the center X positions are equal, float division by zero will cause a crash in Python
        fltAngleInRad = math.atan(fltOpp / fltAdj)      # if adjacent is not zero, calculate angle
    else:
        fltAngleInRad = 1.5708                          # if adjacent is zero, use this as the angle, this is to be consistent with the C++ version of this program
    # end if

    fltAngleInDeg = fltAngleInRad * (180.0 / math.pi)       # calculate angle in degrees

    return fltAngleInDeg
# end function

###################################################################################################
# if we have two chars overlapping or to close to each other to possibly be separate chars, remove the inner (smaller) char,
# this is to prevent including the same char twice if two contours are found for the same char,
# for example for the letter 'O' both the inner ring and the outer ring may be found as contours, but we should only include the char once
def removeInnerOverlappingChars(listOfMatchingChars):
    listOfMatchingCharsWithInnerCharRemoved = list(listOfMatchingChars)                # this will be the return value

    for currentChar in listOfMatchingChars:
        for otherChar in listOfMatchingChars:
            if currentChar != otherChar:        # if current char and other char are not the same char . . .
                                                                            # if current char and other char have center points at almost the same location . . .
                if distanceBetweenChars(currentChar, otherChar) < (currentChar.fltDiagonalSize * MIN_DIAG_SIZE_MULTIPLE_AWAY):
                                # if we get in here we have found overlapping chars
                                # next we identify which char is smaller, then if that char was not already removed on a previous pass, remove it
                    if currentChar.intBoundingRectArea < otherChar.intBoundingRectArea:         # if current char is smaller than other char
                        if currentChar in listOfMatchingCharsWithInnerCharRemoved:              # if current char was not already removed on a previous pass . . .
                            listOfMatchingCharsWithInnerCharRemoved.remove(currentChar)         # then remove current char
                        # end if
                    else:                                                                       # else if other char is smaller than current char
                        if otherChar in listOfMatchingCharsWithInnerCharRemoved:                # if other char was not already removed on a previous pass . . .
                            listOfMatchingCharsWithInnerCharRemoved.remove(otherChar)           # then remove other char
                        # end if
                    # end if
                # end if
            # end if
        # end for
    # end for

    return listOfMatchingCharsWithInnerCharRemoved
# end function

###################################################################################################
# this is where we apply the actual char recognition
def recognizeCharsInPlate(imgThresh, listOfMatchingChars):
    strChars = ""               # this will be the return value, the chars in the lic plate

    height, width = imgThresh.shape

    imgThreshColor = np.zeros((height, width, 3), np.uint8)

    listOfMatchingChars.sort(key = lambda matchingChar: matchingChar.intCenterX)        # sort chars from left to right

    cv2.cvtColor(imgThresh, cv2.COLOR_GRAY2BGR, imgThreshColor)                     # make color version of threshold image so we can draw contours in color on it

    for currentChar in listOfMatchingChars:                                         # for each char in plate
        pt1 = (currentChar.intBoundingRectX, currentChar.intBoundingRectY)
        pt2 = ((currentChar.intBoundingRectX + currentChar.intBoundingRectWidth), (currentChar.intBoundingRectY + currentChar.intBoundingRectHeight))

        cv2.rectangle(imgThreshColor, pt1, pt2, Main.SCALAR_GREEN, 2)           # draw green box around the char

                # crop char out of threshold image
        imgROI = imgThresh[currentChar.intBoundingRectY : currentChar.intBoundingRectY + currentChar.intBoundingRectHeight,
                           currentChar.intBoundingRectX : currentChar.intBoundingRectX + currentChar.intBoundingRectWidth]

        imgROIResized = cv2.resize(imgROI, (RESIZED_CHAR_IMAGE_WIDTH, RESIZED_CHAR_IMAGE_HEIGHT))           # resize image, this is necessary for char recognition

        npaROIResized = imgROIResized.reshape((1, RESIZED_CHAR_IMAGE_WIDTH * RESIZED_CHAR_IMAGE_HEIGHT))        # flatten image into 1d numpy array

        npaROIResized = np.float32(npaROIResized)               # convert from 1d numpy array of ints to 1d numpy array of floats

        retval, npaResults, neigh_resp, dists = kNearest.findNearest(npaROIResized, k = 1)              # finally we can call findNearest !!!

        strCurrentChar = str(chr(int(npaResults[0][0])))            # get character from results

        strChars = strChars + strCurrentChar                        # append current char to full string

    # end for

    if Main.showSteps == True: # show steps #######################################################
        cv2.imshow("10", imgThreshColor)
    # end if # show steps #########################################################################

    return strChars
# end function