Skip to content

Latest commit

 

History

History
38 lines (25 loc) · 2.04 KB

README.md

File metadata and controls

38 lines (25 loc) · 2.04 KB

#SudokuSolver

============

Android application that captures image of Sudoku puzzle using camera and processes solution using image processing and OCR technology.

#Screenshots sudoku-before sudoku-solved

#Libraries Used

#Future Optimizations

  • Sodoku solving algorithm

    1. currently using greedy backtracking algorithm- could use different heuristics to improve performance of algorithm.
  • Image processing

    1. Noise reduction currently using connected component labelling algorithm to remove parts of the image which aren't numbers which requires a passthrough of all the pixels in the image.
      Improvement: Could possibly use a more efficient approach to reduce image noise.

    2. Puzzle detection: currently scanning through image from center across all four directions until outer gridlines of puzzle is reached, which is really inefficient.
      Improvement: Could use an algorithm to detect the largest rectangle/square in the image and assume that to be the puzzle.

    3. Continuous feed detection feature: currently only executes image processing when button is pressed.
      Improvement: Could have a feature that continuously detects a sudoku puzzle and overlays the solution onto the preview.

    4. Rotation detection: currently can detect a puzzle with a few degrees of rotation.
      Improvement: Could have a pre-processing feature that detects the angle of the rotation and rotates the image before using the image processing algorithms.

Analytics