Visualization is one of the important aspects for data representation. A proper artificial stimulus can produce the same effect as natural objects, with visual stimuli being extremely effective. In general, visualization helps human to handle vast amount of data easily. The object of visualization is to increase human understanding and readability of complex data by taking advantage of the high-bandwidth human visual channel. In addition, the techniques used in visualization are majority from the field of computer graphics.
A comprehensive definition for visualization is given below ,[2], [1]:
‘Visualization is a cognitive process using the powerful information processing and analytical functions of the human vision system. It has always been a major factor in scientific progress, and now, with the assistance of computer graphics, it extends our vision system from sub-atomic to interstellar dimensions and allows geometric representation and simulations of any multidimensional data set. The fundamental objective is to acquiring new knowledge rather than generating images’.
A modern technique used for visualization is tone mapping. Tone mapping is only dealing with scalar data that came from difference input sources. In addition, tone mapping can extend data range significantly. Furthermore, the scale that the raw data are represented in is not compulsory compatible with sensory response curve of human photoreceptors, and therefore a direct linear mapping of the input data to light intensity does not have the desired visual effect. The majorities of tone mapping techniques work on the spatial domain and are therefore categorized under either local or global process, depending on the nature of the compression.
Transfer function is commonly used in tone mapping. A transfer function is a special algorithm that mapping a space to another space based on some interest. A common adopt transfer function is based on luminance-chrominance HDR (High Dynamic Range).
In designing a visualization, technical issues is not the one must take into consideration but also the characteristics of the human visual system, that is, matters of perception. As known, human eye are sensitive to variation in intensity as well as chromaticity. A contrast sensitivity research, [3], showing human sensitivity to luminance contrast is very different from human sensitivity to chrominance contrast. [4] found out that red-green and blue-yellow contrast sensitivity functions have similar spatial bandwidth.
Several color models can serve as intermediate level for conversion. RGB color model generally suffers from high channel correlation and mixing on luminance and chrominance. YUV and YCrCb are used to solve to problem from RGB model as Y channel being the luminance value and the other two channels is a combination of red-green and blue-yellow. These models do not solve entirely the correlation problem because the chrominance channel still correlated. Thus, other color model is required to decouple the channels. XYZ color model and LAB color model offer such requirement. In addition to LAB color model, it offer a nice property that the L, A, and B channel have a Euclidean relationship. Other color model exists for special proposed such as HSV color model and CMYK color model.
In conclusion, visualization is the most effective way for human to analyze and manipulate data. Various visualization principles need to consider for achieving highest possible way for conveying information. Once again, a picture is worth that a thousand words !
References:
[1] T. Theoharis, G. Papaioannou, N. Platis, N.M. Patrikalakis. Graphics & Visualization: Principles and Algorihthm. pp. 231 - 365. A K Peters, Ltd. 2008.
[2] University of Edinburgh. Visualization, 2005.
[3] G.J.C. van der Horst, C.M.M. de Weert, and M.A. Bouman, ‘Transfer of spatial chromaticity-contrast at threshold in human eye’, Journal of the Optical Society of America, vol. 57, no. 10, pp. 1260-1266, October 1967.
[4] K.T. Mullen, ‘The contrast sensitivity
of human colour vision to red-green and blue-yellow chromatic gratings’,
Journal of Physiology, vol. 359, pp. 381-400, February 1985.
No comments:
Post a Comment