Sign In

Communications of the ACM

ACM TechNews

Computers Can Perceive Image Curves Like Artists

View as: Print Mobile App Share:
interest curves

Interest curves perceived by a computer.

Credit: Ume University

Umeå University researchers have developed a breakthrough concept in the field of computer vision using curves and lines to represent image shapes and recognize objects. "With this method, the computer can redraw an image using curve strokes and recognize objects through these curves," says Umeå researcher Bo Li. The concept heralds new dimensions of understanding image features including points, regions, lines, and curves.

The research advances the standard for future studies regarding image features, while providing practical guidance to the field. The most important element in feature extraction is the robustness, according to Li. The results show the new method enables curve and lines to be detected robustly under various image transformations and disturbances.

Historically, curves and lines have not been as popular as points and regions in the field of computer vision because they lack the necessary robustness. However, Li's new theory and algorithms solve this problem. "Curves and lines are naturally more useful than points, because humans use these shapes to describe the world," Li says.

From Umeå University
View Full Article


Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA


No entries found