Playing Robot Soccer under Natural Light: A Case Study

@inproceedings{Mayer2003PlayingRS,
  title={Playing Robot Soccer under Natural Light: A Case Study},
  author={Gerd Mayer and Hans Utz and Gerhard K. Kraetzschmar},
  booktitle={RoboCup},
  year={2003}
}
The recent debate in the RoboCup middle-size community about natural light conditions shows that a more in-depth analysis of the problems incurred by this is necessary in order to draft out a focused and realistic roadmap for research. Based on real-world images taken under varying lighting conditions, we performed descriptive and statistical analysis of the effects on color-based vision routines. The results show that pure color-based image processing is not likely to perform well under… 
Adaptive Field Detection and Localization in Robot Soccer
TLDR
An efficient approach for adaptive soccer field detection model utilizing NAO’s two-camera system is proposed, building upon real-time image histogram analysis between top and bottom camera frames, that is robust under inconsistent lighting conditions, and can be further processed to generate field boundaries.
Perception and Strategy Learning in Robot Soccer
TLDR
This work explored some novel techniques in lighting-invariant perception and high-level team strategies which allowed the team UPennalizers to stay competitive in the Standard Platform League (SPL).
Line Based Robot Localization under Natural Light Conditions
TLDR
This framework comprises an algorithm for robust selflocalization and a set of benchmarks which can be used offline to test other algorithms and to compare their outcomes with the authors' results.
RoboCup Small-Size League: Using Neural Networks to Learn Color Segmentation during Visual Processing
TLDR
This paper describes the use of a neural network to classify camera image pixels to a discrete set of color classes that is robust under different light conditions and shows how the recall time of the neural network can be improved to achieve vision processing of over 30 fps using high resolution images.
Neural Robot Detection in RoboCup
TLDR
The described multilevel approach documents, that the combination of a simple color based attention control and a subsequent neural object classification can be applied successfully in real world scenarios, indicate a very good overall performance regarding robustness, flexibility and computational needs.
A Survey on Visual Perception for RoboCup MSL Soccer Robots
TLDR
The paper reviews the advancement of visual perception in RoboCup MSL soccer robots from several points of view including the design and calibration of the vision system, the visual object recognition, the estimation of the object's motion, robot visual self-localization and multi-robot cooperative sensing.
An Automatic Approach to Online Color Training in RoboCup Environments
TLDR
This paper presents an algorithm for automatic online color training, which is able to robustly adapt the mapping of colors to color classes onto different lighting situations online.
Improving Visual Processing in RoboCup SSL by Learning Color Segmentation with Neural Networks
TLDR
This paper describes the use of a neural network to classify camera image pixels to a discrete set of color classes that is robust under different light conditions and shows how the recall time of the neural network can be improved to achieve vision processing of over 30 fps using high resolution images.
A robust object recognition method for soccer robots
TLDR
A robust object recognition method based on the omni-directional vision system for RoboCup Middle Size League robot-NuBot that can be adaptive to light conditions when the illumination is not changed very suddenly and greatly.
...
1
2
3
4
...

References

SHOWING 1-10 OF 10 REFERENCES
Toward RoboCup without color labeling
TLDR
An algorithm, called the CONTRACTING CURVE DENSITY (CCD) algorithm, for fitting parametric curves to image data and shows that the CCD algorithm reliably localizes the ball even in the presence of heavily changing illumination, strong clutter, specularity, partial occlusion, and texture.
Toward RoboCup without Color Labeling
TLDR
An algorithm named the Contracting Curve Density (CCD) algorithm for fitting parametric curves to image data and it is shown that the CCD algorithm reliably localizes the ball even in the presence of heavily changing illumination, strong clutter, specularity, partial occlusion, and texture.
A novel algorithm for color constancy
  • D. Forsyth
  • Computer Science
    International Journal of Computer Vision
  • 2004
TLDR
The experimental work shows that for good constancy, a color constancy system will need to adjust the gain of the receptors it employs in a fashion analagous to adaptation in humans.
Towards autonomous vision self-calibration for soccer robots
TLDR
A method for autonomous calibration of a color classificator used for color blob-based image segmentation and landmark and object recognition is presented and the experimental results demonstrate significantly improved robustness of visual processing.
Sampling, resampling and colour constancy
  • D. Forsyth
  • Computer Science
    Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149)
  • 1999
TLDR
This work forms colour constancy as a problem of Bayesian inference, where one is trying to represent the posterior on possible interpretations given image data using a Markov chain Monte Carlo method.
Properties and performance of a center/surround retinex
TLDR
A practical implementation of the retinex is defined without particular concern for its validity as a model for human lightness and color perception, and the trade-off between rendition and dynamic range compression that is governed by the surround space constant is described.
The contracting curve density algorithm and its application to model-based image segmentation
  • Robert Hanek
  • Mathematics
    Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001
  • 2001
TLDR
The article addresses the problem of model-based image segmentation by fitting deformable models to the image data by adapting local statistics to distinguish the two sides of the edge, even in the presence of spatially changing properties such as texture, shading, or color.
Multi-scale retinex for color image enhancement
TLDR
A multi-scale retinex (MSR) which overcomes this limitation for most scenes and both color rendition and dynamic range compression are successfully accomplished except for some "pathological" scenes that have very strong spectral characteristics in a single band.
Towards robocup without color labeling. RoboCup International Symposium
  • Towards robocup without color labeling. RoboCup International Symposium
  • 2002
A novel approach for color constancy
  • International Journal of Computer Vision 5
  • 1990