Determination the optimal imaging conditions for vision systems working on cloudy days for robotic applications

One of important factors affecting the accuracy of the final response of processing algorithm is the lighting of the scene and its properties with respect to the object of interest. The image quality is affected by the lighting surrounding conditions, the scattering of it and the angle of the light source on the object. In many robotic applications, the use of natural light is the only possible option for imaging. Since reducing the effect of sunlight reflections requires complex and computationally expensive Color Constant algorithms, recognizing the time and angle of imaging that reduce the negative impacts of illumination in natural conditions will be of significant importance. Therefore this study was performed to evaluate the color components of taken images on cloudy day which there was not any direct sunlight. The images were taken with a constantly-positioned camera 6 hours from several different colored cards in 24 rotational angles in January 2014 in Mashhad, Iran (located at 36.20º North latitude and 59.35º East longitude,). Performance parameters such as Root Mean Square Error (RMSE) and Color’s Coefficient of Variation (CCV) were employed to compare the variation of colors from standard controls and to investigate the effect of time and direction on the color components. The results showed that as well as reducing the negative effects of light reflection, values of CCV and RMSE were lower.

Keywords:

vision system , indirect sunlight , evaluation , imaging , color component , illumination