Pattern recognition through brushstroke analysis: an inspection to classify van Gogh from others

This thesis report is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2017.Cataloged from PDF version of thesis report.Includes bibliographical references (page 38-41).What makes a painter unique? Does every painter leave a “fingerprint”? How can we determine whether a painting is an original or a fake? Or how can we determine if two different pictures are painted by the same artist or not? The main objective of this project is to find the unique attributes of a painter, and to use those attributes to find similarities amongst other paintings to determine whether that painting is done by the same painter or not. The process includes to check if a painting is done by that painter or not by looking towards the similarities in the set of individual data collected from paintings of various painters, and the thorough analyzation of those data to find similarities through convenient methods. From the movement of brushstrokes [1] and the thorough mathematical analysis amongst those patterns incorporate many aspects of a painter's unique style. The key idea to attain the goal is by using the concept of how do we see art [2], by using the strokes and artistic patterns used by a painter in his/her works to find matches, and using appropriate algorithms to match the brushstroke patterns in other paintings to determine whether the artwork is done by the artist or not. In this experiment, the main focus was on Vincent van Gogh and his works to analyze his paintings, and to analyze his work patterns with his contemporaries and others.Souradeep DasguptaMd. Ayaz HossainB. Computer Science and Engineerin