BlackJack, a popular casino game [1], presents players with the challenge of making strategic decisions to beat the dealer without exceeding a total card value of 21.

Players often face inherent disadvantages due to limited information and the statistically optimized strategies employed by casinos. In our final assignment, we developed the Blackjack ‘Super Assistant’ application - designed to mitigate these challenges through technological innovation.

The assistant uses live video feeds to detect and classify cards and chips. It enhances the player's strategy through a combination of statistical analysis, card counting, and bet suggestions, while tracking the player’s profit over multiple sessions.

In the short term, distinguishing the difference may not be feasible due to statistical variability. However, over the long term, as per the law of large numbers, there will be an enhancement in the player's probability of winning. The system operates by analyzing the video input, identifying gaming elements such as the gaming board, cards and chips, and by using a state-machine algorithm, it dynamically updates the game state and recommends optimal betting strategies and moves.

This process increases the likelihood of profitability by adapting to changes in the game environment in real-time.

Project video - https://www.youtube.com/watch?v=so_-Fl3tl98&t=25s