{"id":2915,"date":"2011-07-01T11:23:28","date_gmt":"2011-07-01T11:23:28","guid":{"rendered":"https:\/\/blogs.mathworks.com\/pick\/2011\/07\/01\/video-sudoku\/"},"modified":"2017-02-04T14:29:51","modified_gmt":"2017-02-04T19:29:51","slug":"video-sudoku","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/pick\/2011\/07\/01\/video-sudoku\/","title":{"rendered":"Video Sudoku"},"content":{"rendered":"
\n

Michael<\/a>'s pick this week is Video Sudoku Solver<\/a> by Teja Muppirala<\/a>.<\/p>\n

About once a year, the Application Engineers here at MathWorks gather from across the globe to share ideas, strategies, best
\npractices, horror stories, and more. One of the highlights of this time is our demo showcase, where we show off our most
\nuseful, most sophisticated, or most jaw-droppingly neat demos. One of this year's standouts happens to be on the File Exchange.<\/p>\n

The algorithm that solves Teja's Sudoku demo is nothing new: Cleve Moler discussed the recursive technique back in 2009. What is<\/i> new, though, is that Teja has wrapped some video acquisition and number recognition around this algorithm to create a simple
\nversion of augmented reality. Hold an unsolved Sudoku puzzle in front of your webcam, and this solver can recognize what it's
\nseeing and write the solution into the video feed.<\/p>\n

Teja defines each step in this process with concise functions that are demonstrated in a script acting on a static image.
\nBy the time we move to the live video analysis, we understand exactly what's happening. This makes the demo not just entertaining
\nbut educational as well!<\/p>\n

Teja has furthermore provided a video<\/a> explaining how this algorithm works on our Image and Video Processing demo page.<\/p>\n

\"\"<\/p>\n

Comments<\/b><\/p>\n

Give it a try and let us know what you think here<\/a> or leave a comment<\/a> for Teja.<\/p>\n