Python Open-CV T-Rex Run(No Hardware needed)

In this project, I demonstrate the use of open cv and python for Gaming bot. So you think what is different in this bot?

• No Hardware Needed
• Direct Screen as Video input in python Code

You think how it's possible to make this code without any hardware.

The following libraries are used in this code.

1. Numpy Version = 1.18.0 (Link for Installation)2.CV2 Version = 3.4.1(Link for Installation)3.mss Version = 2.0.22(Link for Installation)4.PIL Version = 5.4.1(Link for Installation)

After installation you can run this code on jupyter notebook.

import numpy as np
import cv2
from mss import mss
from PIL import Image
import keyboard
from time import sleep 
mon1 = {'top': 200, 'left': 160, 'width': 800, 'height': 200}
sct1 = mss()
font                   = cv2.FONT_HERSHEY_SIMPLEX
bottomLeftCornerOfText = (50,50)
fontScale              = 1
fontColor              = (255,0,0)
lineType               = 2
xcnts = []
count = 0
last_count = 0
while 1:
    #sleep(0.0001)
    sct1.get_pixels(mon1)
    img1 = Image.frombytes('RGB', (sct1.width, sct1.height), sct1.image)
    img_arry1 = np.array(img1)
    obj2 = img_arry1[125:155 , 110:155]
    gray = cv2.cvtColor(obj2, cv2.COLOR_BGR2GRAY)  
    th, threshed = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV|cv2.THRESH_OTSU) 
    cnts = cv2.findContours(threshed, cv2.RETR_LIST,cv2.CHAIN_APPROX_SIMPLE)[-2] 
    count = len(cnts)
    if count >= 1:
        keyboard.press_and_release('space')
        keyboard.press_and_release('space')
    cv2.putText(img_arry1,str(count), bottomLeftCornerOfText, font, fontScale,fontColo    r, lineType)
    cv2.imshow('test1', np.array(img_arry1))
    cv2.imshow('Obj2', np.array(gray))
    cv2.imshow('cnts', np.array(threshed))
    if cv2.waitKey(25) & 0xFF == ord('q'):
        cv2.destroyAllWindows()
        break

Now see the video first and set both Windows of the application according to video. You can see two little windows on the screen. This little window is for black object detection on the screen.