Monopoly OCR

An Open-CV project to create a dashboard for the monopoly game. A screenshot of the game results is added to the resources folder and the script runs an OCR using Open-CV and pyTesseract.

Dashboard Site can be viewed here

GSheets Dashboard can be viewed here

import pytesseract
from PIL import Image
import cv2
import sys
import pandas as pd
import csv
import os

1 - Testing for latest files

#Testing conditions
test = False
if test:
    latestCSV = "latest_test.csv"
    dataCSV = "data_test.csv"
    rankCSV = "rank_test.csv"
else:
    latestCSV = "latest.csv"
    dataCSV = "data.csv"
    rankCSV = "rank.csv"

#Check latest file.
check = pd.read_csv(latestCSV)
latest_file = check.iloc[0,0]

def newest(path):
    files = os.listdir(path)
    paths = [os.path.join(path, basename) for basename in files if basename.endswith('.jpg')]
    return max(paths, key=os.path.getctime)
print(newest("resources"))
latest = pd.DataFrame([newest("resources")])
latest.to_csv(latestCSV,index = False)
if latest_file == newest("resources"):
    sys.exit("Item has been appended")

else:
    print("This will be appended")

resources\Screenshot_20200702-173925 - Copy (2).jpg
This will be appended

2 - Read latest file.

file_path= newest("resources")
im = Image.open(file_path)
im.save("raw.jpg", dpi=(300, 300))

im

3 - Pre-process images

#Make image b&w.
im = Image.open("raw.jpg").convert("L")
im.save("bw.jpg")
readImage = cv2.imread("bw.jpg")

#Invert colour chanels.
invertImage = cv2.bitwise_not(readImage)

im

import matplotlib.pyplot as plt
fig, ax = plt.subplots(figsize=(16, 8))
img2 = invertImage[:,:,::-1]
ax.imshow(img2)

<matplotlib.image.AxesImage at 0x1b3623f4d88>

4 - Crop-dimensions to extract the four player names from inverted image.

cropName_1 = invertImage[428:478,683:970]
cropName_2 = invertImage[645:700,683:970]
cropName_3 = invertImage[869:918,683:970]
cropName_4 = invertImage[1086:1136,683:970]

#Global Crop width for wealth except first value.
wStart = 2085
wEnd = 2225

#Global variables for threshold value.
tn = 25
tw = 28

cropList = [cropName_1,cropName_2,cropName_3,cropName_4]
def plotImages(list):
    w=10
    h=10
    fig=plt.figure(figsize=(8, 8))
    columns = 1
    rows = 4
    for i in range(1, columns*rows +1):
        img = list[i-1]
        fig.add_subplot(rows, columns, i)
        plt.imshow(img)
    plt.show()
plotImages(cropList)

5 - Crop dimensions for four player wealths.

cropWealth_1 = invertImage[360:430,2065:wEnd]
cropWealth_2 = invertImage[570:655,wStart:wEnd]
cropWealth_3 = invertImage[790:870,wStart:wEnd]
cropWealth_4 = invertImage[1010:1100,wStart:wEnd]

cropList = [cropWealth_1,cropWealth_2,cropWealth_3,cropWealth_4]
plotImages(cropList)

6 - Change threshhold for names.

retval, thresholdName_1 = cv2.threshold(cropName_1,tn,255,cv2.THRESH_BINARY)
retval, thresholdName_2 = cv2.threshold(cropName_2,tn,255,cv2.THRESH_BINARY)
retval, thresholdName_3 = cv2.threshold(cropName_3,tn,255,cv2.THRESH_BINARY)
retval, thresholdName_4 = cv2.threshold(cropName_4,tn,255,cv2.THRESH_BINARY)

cropList = [thresholdName_1,thresholdName_2,thresholdName_3,thresholdName_4]
plotImages(cropList)

7 - Change threshhold for wealth.

retval, thresholdWealth_1 = cv2.threshold(cropWealth_1,tw,255,cv2.THRESH_BINARY)
retval, thresholdWealth_2 = cv2.threshold(cropWealth_2,tw,255,cv2.THRESH_BINARY)
retval, thresholdWealth_3 = cv2.threshold(cropWealth_3,tw,255,cv2.THRESH_BINARY)
retval, thresholdWealth_4 = cv2.threshold(cropWealth_4,tw,255,cv2.THRESH_BINARY)

cropList = [thresholdWealth_1,thresholdWealth_2,thresholdWealth_3,thresholdWealth_4]
plotImages(cropList)

8 - Write edited images for preview.

"""
cv2.imwrite("w1.jpg",thresholdWealth_1)
cv2.imwrite("w2.jpg",thresholdWealth_2)
cv2.imwrite("w3.jpg",thresholdWealth_3)
cv2.imwrite("w4.jpg",thresholdWealth_4)
"""

'\ncv2.imwrite("w1.jpg",thresholdWealth_1)\ncv2.imwrite("w2.jpg",thresholdWealth_2)\ncv2.imwrite("w3.jpg",thresholdWealth_3)\ncv2.imwrite("w4.jpg",thresholdWealth_4)\n'

name_1 = pytesseract.image_to_string(thresholdName_1)
name_2 = pytesseract.image_to_string(thresholdName_2)
name_3 = pytesseract.image_to_string(thresholdName_3)
name_4 = pytesseract.image_to_string(thresholdName_4)

9 - Force pytesseract to read numbers using the pre-config 'outputbase digits'.

custom_config = r'--oem 3 --psm 6 outputbase digits'

wealth_1 = pytesseract.image_to_string(thresholdWealth_1, config=custom_config)
wealth_2 = pytesseract.image_to_string(thresholdWealth_2, config=custom_config)
wealth_3 = pytesseract.image_to_string(thresholdWealth_3, config=custom_config)
wealth_4 = pytesseract.image_to_string(thresholdWealth_4, config=custom_config)

10 - Manual correction for misread -ve sign.

wealth_1 = int(wealth_1)
wealth_2 = -abs(int(wealth_2))
wealth_3 = -abs(int(wealth_3))
wealth_4 = -abs(int(wealth_4))

11 - Add read data and ranks to csv.

#Add read data to csv.
data = pd.read_csv(dataCSV)
names = ["Game No",name_1,name_2,name_3,name_4]
wealths = [(len(data["Game No"])+1),wealth_1,wealth_2,wealth_3,wealth_4]
ocrData = pd.DataFrame([wealths],columns=names)
mergedData_wealth = pd.concat([data, ocrData], join = "outer")
mergedData_wealth.to_csv(dataCSV, index=False)

#Add ranks to csv.
data_r = pd.read_csv(rankCSV)

rank_1 = 1
rank_2 = 2
rank_3 = 3
rank_4 = 4

ranks = [(len(data_r["Game No"])+1),rank_1,rank_2,rank_3,rank_4]
ocrRank = pd.DataFrame([ranks],columns=names)
mergedData_rank = pd.concat([data_r, ocrRank], join = "outer")
mergedData_rank.to_csv(rankCSV, index=False)

12 - Calculate wins and win-ratio

#Calculate wins and win-ratio
players = ["Sanjay", "thamarmo", "Abhishek", "Shishtaouk", "NipPincher", "Diraj", "SugaDaddy", "Spartan"]
data_r = pd.read_csv(rankCSV)
winRatio = []
wins =[]
played = []
losses = []

13 - Check for new files and update tables

for i in players:
    gamesPlayed= data_r[i].value_counts().sum()
    played.append(gamesPlayed)
    if 1 in data_r[i].unique():
        win = data_r[i].value_counts()[1]
        wins.append(win)
        wr = ((win/gamesPlayed*100).round(2))
        winRatio.append(wr)
    else:
        winRatio.append(0)
        wins.append(0)
for i in range(len(played)):
    losses.append(played[i]-wins[i])
winLoss = pd.DataFrame(list(zip(players, wins, played,losses,winRatio)), 
               columns =['Names', 'Wins', 'Played', 'Lost', 'winRatio'])
winLoss.to_csv(r'winLoss.csv', index = False)

winLoss

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	Names	Wins	Played	Lost	winRatio
0	Sanjay	25	71	46	35.21
1	thamarmo	5	28	23	17.86
2	Abhishek	21	78	57	26.92
3	Shishtaouk	3	7	4	42.86
4	NipPincher	6	22	16	27.27
5	Diraj	3	26	23	11.54
6	SugaDaddy	14	48	34	29.17
7	Spartan	13	47	34	27.66

snjsomnath / Monopoly-2