A reinforcement learning framework for parameter control in computer vision applications

Abstract

We propose a framework for solving the parameter selection problem for computer vision applications using reinforcement learning agents. Connectionist-based function approximation is employed to reduce the state space. Automatic determination of fuzzy membership functions is stated as a specific case of the parameter selection problem. Entropy of a fuzzy event is used as a reinforcement. We have carried out experiments to generate brightness membership functions for several images. The results show that the reinforcement learning approach is superior to an existing simulated annealing-based approach.

Novel adaptive Discrete Cuckoo Search Algorithm for parameter optimization in computer vision

“…Most of the proposed techniques for parameter tuning have not been widely adopted. The problem has been solved using learning models [1] and optimization models [2][3][4] [5]. This can be partly awarded to few application examples in the real image analysis.…”

“…As first works, we can cite Taylor [1] who proposed a method based on reinforcement learning to monitor parameters in vision applications. B.Nikolay and al.…”

A reinforcement agent for object segmentation in ultrasound images

“…Recently, some other approaches have been introduced which show the application of RL on some image-based problems (Sahba & Tizhoosh, 2003;Sahba, Tizhoosh, & Salama, 2005a, Sahba, Tizhoosh, & Salama, 2006a, Sahba, Tizhoosh, & Salama, 2006bShokri & Tizhoosh, 2004;Taylor, 2006;Tizhoosh & Taylor, 2006;Yin, 2002).…”

Automatic Selection of Object Recognition Methods Using Reinforcement Learning

“…Results were presented for both indoor and outdoor color images, showing a performance improvement over time for both image segmentation and object recognition using RL (Bhanu and Peng, 2000). Taylor (2004) also followed this line of research, applying RL algorithms to learn parameters of an existing image segmentation algorithm. Using the Fuzzy ARTMAP artificial neural network, he was able to optimize ten parameters of Wolf and Jolion (2003) algorithm for text detection in still images.…”