Color learning and illumination invariance on mobile robots: A survey

Abstract

Recent developments in sensor technology have made it feasible to use mobile robots in several fields, but robots still lack the ability to accurately sense the environment. A major challenge to the widespread deployment of mobile robots is the ability to function autonomously, learning useful models of environmental features, recognizing environmental changes, and adapting the learned models in response to such changes. This article focuses on such learning and adaptation in the context of color segmentation on mobile robots in the presence of illumination changes. The main contribution of this article is a survey of vision algorithms that are potentially applicable to color-based mobile robot vision. We therefore look at algorithms for color segmentation, color learning and illumination invariance on mobile robot platforms, including approaches that tackle just the underlying vision problems. Furthermore, we investigate how the interdependencies between these modules and high-level action planning can be exploited to achieve autonomous learning and adaptation. The goal is to determine the suitability of the state-of-the-art vision algorithms for mobile robot domains, and to identify the challenges that still need to be addressed to enable mobile robots to learn and adapt models for color, so as to operate autonomously in natural conditions.

Using Saliency-Based Visual Attention Methods for Achieving Illumination Invariance in Robot Soccer

“…As a workaround, either restricted configurations in structured domains are considered, or specific models of segmentation and recognition that do not provide generalized solutions [2] are used. Various approaches to address this challenge include describing the problem in terms of illumination [3], surface reflectance [4], and sensor sensitivity [5].…”

“…Bayesian decision theory and hierarchical model based approaches also exist in the literature [6,7]. Methods that do not require domain specific tuning are shown to be computationally less complex and more adaptive, whereas usually the opposite is shown to be true for the classical and model based methods [2]. Illumination invariance has been studied in the robot soccer domain as well since the overall performance of the teams heavily depend on the successful perception of the environment [7,8].…”

Learning and Vision-Based Obstacle Avoidance and Navigation

“…In their work on color learning on legged robots, Sridharan and Stone [25,27] model illuminations using autonomously collected image statistics. Mohan Sridharan and Peter Stone [26] gave a good survey of color learning and illumination invariance on mobile robots.…”

Bootstrap Learning and Visual Processing Management on Mobile Robots

“…The learned models are used to detect and adapt to a range of illuminations. See [45] for a recent survey on color learning and illumination invariance.…”