Pictures at an Exhibition

This project is to recognize an input image through the database which are learned from a set of training images. The first thing we have done is to extract only the main image. This step requires an adaptive threshold to decide whether the selected region is a painting we want or not. The morphological operation follows the filtering process. The edges of the painting extracted in the source image are dilated and eroded. As a result, the four corner points of centered images are searched by using a radon transform. Inverse projective spatial transform leads us to obtain eigenimages, which enable to form a database. The test input is calculated the correlation values with eigenimage in database. The result shows that learning process with three images recognize the test input with one hundred probability.