Skip to Main Content
In spite of all the progress known frequently by the computer vision field, the user intervention stays always necessary. In all images processing tasks, it is to the user to adjust the parameters of the vision operators in order to reach the desired result. The manually adjustment of these parameters isn't an easy work, it becomes more tedious with complicated vision applications such as texture detection. This paper comes to help the user by proposing him a scientific method to automatically adjust these parameters. Our method is based on reinforcement learning, the optimal parameters are determined according to the definition of states, actions and reward. Our method takes in account not only the system opportunities, but also the user preferences. And through the learning mechanism it will suggest trustworthy solutions. An example to test our method by detecting an object in a textural image is given.
Date of Conference: 28-30 March 2010