Research On Finding Missing Pieces In The Square Of Jigsaw Puzzle

An image can be represent as a thousand words.We often used images to replace long description of information became more simpler,as well as images on traffic signs like is prohibited to stop can be symbolized by the letter S in a red circle accompanied by a red cross line.In addition,the use of images is also often used in buildings as a means of safety instructions.Jigsaw Puzzles is one game that utilizes transformed images into a game,where some of the objects in the image are cut into small pieces and randomly positioned.To play this Jigsaw puzzle,players simply rearrange the pieces according to the order until the last piece.The end of this game,the pieces that are arranged will form an image object.The image used in this puzzle game is usually an interesting object taken from a collection of other images,as well as a landscape image of a zoo,where there are many animal objects in the picture,so one of the animal object is used as an image for Jigsaw Puzzle games.Many studies have been conducted to be able to recognize the pattern of objects in the image.Both from the single and multi objects in the image,the results of the process are used as information for other processes such as human or other objects recognition.Blob analysis and morphological reconstruction is one of the methods that used to carry out this process.Basically blob analysis separates the object in the image based on the interconnection between points on the object.The interconnections found in blob analysis are 4 points and 8 points.While morphological reconstruction is a method that analyzes the geometrical structure of an image,if there is noise in the center part of the image,this method will eliminate the noise.The present study was design to examines two images,where the first image is a cluttered scene or collection of objects contained in one image,while the second image is part of the first image which is slice in square shape into several pieces like puzzle with several missing parts.The purpose of this study is how computers can find the location and area of the second image in the first image and cut off the missing part on the first image and put into the second image that is used to fill the missing part on it,so that the second image can become a complete image.This study is divided into two parts,first is jigsaw puzzle with border,and the second one is jigsaw puzzle without borders.The blob analysis method combined with morphological reconstruction were applied in the first section and the result showed that some processes were not perfectly obtained.Therefore,intersection line segment was added to cover the deficiencies in the previous method to get the piece intact,so that information about size and location of each piece can be well obtained.For the borderless section,the methods in the first section were combined with new algorithm for line search based on column called BALSEM.The BALSEM algorithm was able to separate each piece of the puzzle,so the information about size and position of each piece could be well obtained.In addition,the Blob analysis method combined with Genetic Algorithm(BADPIG)accompanied by an Auto Switch System(ASS)were applied to get optimal time in terms of finding information.From the results of our experiments,the BALSEM method is more effective in finding size,location and missing part in the second image in section one.Whereas the BADPIG method is very effective to finding size,location and missing part in the second image in section two with many missing parts on it.