• MUTHU DAYALAN

The computer vision technology has continued to evolve through advancement in researches. This has continually widened the computer vision and imaging processing to extended applications. This paper analyzes the technological advancement in these aspects. Machine vision system is among the emergent technologies, which digitizes the video images or information from a camera, processes the information to inspect the concerned object, and communicates with the system or the motion control[ ]. 3D and virtual environments help to mimic the real life situations. The abnormality detection for the video parsing helps to solve the inter-class variability, viewpoint dependence and the object articulation, which makes the practice of object detection quite difficult. A video surveillance where foreground pixels are detected through the background subtraction. The curvature self-similarity visual recognition is a feature descriptor, which increase dimensionality such as the self-similarity and the co-occurrence statistics. It extends beyond the current common approximation of the objects by the use of straight lines. The result on the self-similarity curvature proves that there is provision of the accurate information on the straight lines, while the performance has improved based on the feature selection algorithm

virtual environments, vision system, 3D, video parsing, virtual recognition, self-similarity

"New techniques improving accuracy of computer vision technologies", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.4, Issue 9, page no.98-102, September-2017, Available :http://www.jetir.org/papers/JETIR1709020.pdf

"New techniques improving accuracy of computer vision technologies", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.4, Issue 9, page no. pp98-102, September-2017, Available at : http://www.jetir.org/papers/JETIR1709020.pdf