The fast optimization of digital holography system

In digital holography applications, the position of the object, which is in the system, substantially affects the quality and clarity of three dimensional reconstructed image obtained from hologram. Therefore, obtaining a good three dimensional image by locating the image to be reconstructed in the correct position in the system, can sometimes take a long time. The usage of artificial intelligence algorithms instead of capturing the best images by recording the image several times and reduction of this time would be wise. For the first time in this study by using the classification method approach used in artificial neural networks, which is an artificial intelligence methods, the best image is obtained with optimization via software, not on the digital holographic system. The three important features that affected the system, while the holographic setup is constructing, are examined for 142 different situations and the results of these situations are classified into 5 different classes. After classification by using artificial neural networks, the random values are taken for the test process of the trained system and the performance of the system is determined.