The Study and Implementation of the Methods for Solving Nonogram Puzzles

• Main Authors: Chen, Shiuan-Yi, 陳璿伊
• Other Authors: Lin, Shun-Shii
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/16439960176196982971

碩士 === 國立臺灣師範大學 === 資訊工程學系 === 104 === Nonogram is a logic puzzle game, invented in 1987 by a Japanese named Tetsuya Nishio. The basic method of playing Nonogram puzzle is that, at the beginning a fixed-sized blank board will be given, where each row and each column will be given a set of numbers (clues) as a reminder to the players who need these clues to set or unset the squares. Nonogram is a Constraint satisfaction problem, which has also been proven to be an NP-complete problem. We are uncertain whether it can be solved in polynomial time, but the execution time and the correctness of the answer are the most important indications in the study of efficiency and reliability for designing its algorithms. At present, some relevant papers have proposed different methods to solve the problem. Most of them tend to find the answer based on the prompted clues from a blank board using the logic rules to deduce the set/unset states to part of the squares, and finally using the Backtracking method to solve all the squares exhaustively. Problem-solving approach proposed in this study is to follow the example of n queens problem using Random-restart hill climbing method. First of all, we randomly arrange all lines so that they meet the requirements of all the row clues. Then we randomly select a line and horizontally move it to a new position in order to further meet the requirements of the column clues. This process will be repeated many times. If it finally falls into local optimal solution, we will again restart filling the board randomly, and then move the lines to new positions, until we find out a solution that meets the requirements of all the column clues.