In this project, the Classic Snake Game (which uses a keypad-control) has been recreated to a Stylus-controlled Snake Game, where the movement of the snake is controlled by a Stylus in front of the computer camera.

The Stylus Snake Game was constructed using the concepts of Computer Vision from OpenCV and the game environment from Pygame.

• Created a window where the user will place the Stylus within a rectangular region to capture it.
• Take the average of the HSV values of the pixels in that Region of Interest.

• Take each frame captured by the computer camera.
• Convert it from BGR to HSV.
• Thresholded the HSV image within a range of the captured average.
• Reduced considerable amount of noise by using Morphological Transformations of Closing and Opening.
• Applied Gaussian Blur.
• Using Contour features, we obtained the coordinates of the centroid of the object.

• The camera window is divided into 4 triangles- marking the region of UP, DOWN, LEFT, RIGHT.
• Used a function to find out if the centroid coordinates of the Stylus lies within these regions, using the area of triangles concept.

• Created the main game window, along with the snake body and also added fruits at random positions.
• Used a function to add 5 different obstacle layouts randomly to the game screen.
• Directed the motion of the snake according to the movements of the Stylus.
• Added the following 'Game Over' constraints:
  • Getting out-of-bounds of the game window.
  • Touching the snake body.
  • Touching the obstacles.
• Used a Game Over function to display the final score and end the game.

Click here to view the demonstration of the Stylus Snake Game.