Generalised Object Detection and Semantic Analysis: Casino Example using Matlab

TL;DR

This paper demonstrates a computer vision application using Matlab 7.1 to detect playing cards on a casino table, identify their suits and ranks, showcasing object detection and semantic analysis techniques.

Contribution

It presents a two-pass image processing approach for detecting rectangular objects and extracting detailed information like suit and rank from playing cards.

Findings

Successfully detected and identified card suits and ranks.

Implemented a two-pass detection and recognition process.

Validated the approach on casino table images.

Abstract

Matlab version 7.1 had been used to detect playing cards on a Casino table and the suits and ranks of these cards had been identified. The process gives an example of an application of computer vision to a problem where rectangular objects are to be detected and the information content of the objects are extracted out. In the case of playing cards, it is the suit and rank of each card. The image processing system is done in two passes. Pass 1 detects rectangular shapes and template matched with a template of the left and right edges of the cards. Pass 2 extracts the suit and rank of the cards by matching the top left portion of the card that contains both rank and suit information, with stored templates of ranks and suits of the playing cards using a series of if-then statements.