The paper is about the idea of a method, with which lightweight camera systems can be extended to get more accurate color meta-information for still images and for frames gained from streamed videos. This meta-information can give more information about the lighting conditions and about the colors of objects in the picture. By having more accurate colors in the picture, many typical in situ and post process visual tasks can be done with greater reliability. This extension could enhance color identification of images taken by low budget camera systems to measurement devices.
Authors:Tamás Storcz, Zsolt Ercsey and Géza Várady
This paper corresponds to the solution of some problems realized during ragweed identification experiments, namely the samples collected on the field by botanical experts did not match the initial conditions expected. Reflections and shadows appeared on the image, which made the segmentation more difficult, therefore also the classification was not efficient in previous study. In this work, unlike those solutions, which try to remove the shadow by restoring the illumination of image parts, the focus is on separating leaf and background points based on chromatic information, basically by examining the histograms of the full image and the border. This proposed solution filters these noises in the subspaces of hue, saturation and value space and their combination. It also describes a qualitative technique to select the appropriate values from the filtered outputs. With this method, the results of segmentation improved a lot.
Authors:Ildiko Jancskar, Zoltan Sári, Adam Schiffer and Geza Várady
The paper suggests a design procedure for fuzzy cascade control. The expert knowledge based tuning procedure is supported by investigation the phase plane of the reference tracking. The plant is a one degree of freedom helicopter model with highly non-linear dynamics in a hardware-in-the-loop arrangement. Starting from a common fuzzy-PI controller in the outer loop and classic PI-controller in the inner loop, the presented tuning procedure enables to obtain fuzzy-adaptive parameter settings of the inner-loop and a heuristic modified rule base in the outer loop. The developed control strategy proves to be very successful: both the settling times and overshoots have become well regulated.