Open In App
Related Articles

Google Chrome Dino Bot using Image Recognition | Python

Improve Article
Save Article
Like Article

What would you see in your Chrome browser when there is no internet connection? Yes, everybody knows that dinosaur game that comes on screen. So, in this article, we are going to build a simple python bot that plays Chrome Dino Game without user interaction. Here we are not using any machine learning or artificial intelligence to counter this problem but we will use simple image/screen processing.

Google Chrome Dino Bot

We will work with Pyautogui and PIL (Python Imaging Library) for implementation. This project is very basic and consists of only about 50 lines of code but its result will make you surprise. 
Some libraries used are: 

  1. PIL: Python Imaging Library (PIL) is a free library for the Python programming language that adds support for opening, manipulating, and saving many different image file formats. 
  2. Pyautogui: PyAutoGUI is a Python module for programmatically controlling the mouse and keyboard without any user interaction.
  3. Time: Python “Time” Module which allows us to handle various operations regarding time, its conversions and representations, which find its use in various applications in life.
  4. Numpy:NumPy is a library for the Python programming language, adding support for large, multi-dimensional arrays and matrices, along with a large collection of high-level mathematical functions to operate on these arrays.



  1. Click on the restart button using Pyautogui library using “replaybutton” coordinates.
  2. Calculate the sum of all white pixels values present in the box in front of Dinosaur.
  3. If the sum of pixels values present at any time in the box becomes less than the sum of white pixels values, it means either “bush” or “bird” is coming. So either we have to make our Dino jump or bend down.
  4. In order to protect Dino from “Bush”, we make a jump.
  5. In order to protect Dino from “Bird”, we always keep our Dino down.

Below is the Python implementation:


# importing above defined libraries to
# implement the functionalities 
from PIL import ImageGrab, ImageOps
import pyautogui
import time
import numpy as np  
class coordinates():
    # coordinates of replay button to start the game
    replaybutton =(360, 214)
    # this coordinates represent the top-right coordinates
    # that will be used to define the front box
    dinasaur = (149, 239 )
def restartGame():
    # using pyautogui library, we are clicking on the 
    # replay button without any user interaction
    # we will keep our Bot always down that 
    # will prevent him to get hit by bird 
def press_space():
    # releasing the Down Key 
    # pressing Space to overcome Bush
    # so that Space Key will be recognized easily
    # printing the "Jump" statement on the
    # terminal to see the current output 
    # releasing the Space Key 
    # again pressing the Down Key to keep my Bot always down 
def imageGrab(): 
    # defining the coordinates of box in front of dinosaur 
    box = (coordinates.dinasaur[0]+30, coordinates.dinasaur[1],
           coordinates.dinasaur[0]+120, coordinates.dinasaur[1]+2)
    # grabbing all the pixels values in form of RGB tuples   
    image = ImageGrab.grab(box)
    # converting RGB to Grayscale to
    # make processing easy and result faster 
    grayImage = ImageOps.grayscale(image)
    # using numpy to get sum of all grayscale pixels 
    a = np.array(grayImage.getcolors())
    # returning the sum
    return a.sum()
# function to restart the game
while True
     # 435 is the sum of white pixels values of box. 
     # You may get different value is you are taking bigger 
     # or smaller box than the box taken in this article. 
     # if value returned by "imageGrab" function is not equal to 435, 
     # it means either bird or bush is coming towards dinosaur 
     if(imageGrab()!= 435):   
        # time to recognize the operation performed by above function 

Output : 

Improvements : Over a period of time, the Dino Bot Game becomes fast. The Birds and Bushes start coming very fast. So we are not making our Bot to learn all these things, changing its speed based on past learning. So our bot will function for around 2000 score. In order to score more, we have to apply machine learning and artificial intelligence. 

Whether you're preparing for your first job interview or aiming to upskill in this ever-evolving tech landscape, GeeksforGeeks Courses are your key to success. We provide top-quality content at affordable prices, all geared towards accelerating your growth in a time-bound manner. Join the millions we've already empowered, and we're here to do the same for you. Don't miss out - check it out now!

Last Updated : 26 Apr, 2023
Like Article
Save Article
Similar Reads
Complete Tutorials