Aran Hansuebsai

Posture recognition invariant to background, cloth textures, body size, and camera distance using morphological geometry

The human posture estimation in surveillance caring application can improves the people everyday life, In this paper, we propose a method that is invariant to background, distance of camera location, size and cloths of people in the frames. A silhouette is projected to the horizontal and vertical histograms for features extraction. The important features are based on the length and width of body parts of human. The proposed features are more suitable for classifying human posture into four main categories such as standing, lying, sitting, and bending, obviously appeared with the high percentage of recognition when compared with the traditional features in the ANFIS model. The increase of accuracy comes from the robustness of various environments such as the complicated posture of a changed body position and camera distance.

Failure of Color Constancy for High Luminance of a Test Patch that Appears Unnatural as an Object in a Space

The theory of the recognized visual space of illumination (RVSI) is that the color appearance of objects in a space is determined in relation to its recognition axis RX whose direction is determined by the brain action to adapt to the illumination in the space. Thus the color constancy holds. RX is applicable to objects in the space but not to an object or a portion that does not belong in the space in terms of illumination, that is to say, the luminance of which is too high based on the illumination for that space. In that case the color appearance would be determined in relation to the fundamental axis FX and the color constancy would not hold. In the present paper the chromaticity points were measured for a test patch that appeared achromatic for various luminance of the patch. The points were close to the color of illumination to indicate the color constancy when the luminance was low enough to assure its appearance as the object color, but they departed from the color of illumination and approached the colorimetrically achromatic color. The color constancy gradually failed for the test patch with the high luminance when the color mode became an unnatural object color and then a light color.

Integrated LCA for Printing Service in an emerging country

In most of emerging countries in Asia, such as Thailand, Printing Service LCA is not organized, so schematization and localization of practical methodology is transferred thanks to the joint research project by Shimizu Printing Inc. and Chulalongkorn University. Printing Service LCA developed in Japan is focusing not only on LCCO2 but also on Integrated LCA. It has strong possibility to take root even in emerging countries for environment-conscious Printing Services.

Color image processing using sRGB subdivided space technique

A new method to improve the accuracy of color transformation, using sub-divided sRGB color space, is proposed. This combines tetrahedral partition technique and linear regression model. It is found that the 3rd order coefficient matrices give acceptable result with average color difference, (Delta) Erms equals 3.31. This process is quick and simple for general use of Internets application.