Communications of the ACM
The Machine-Vision Algorithm For Analyzing Children's Drawings
A newly developed machine-vision algorithm can analyze children's drawings objectively.
Credit: Technology Review
Researchers at the Swiss Federal Institute of Technology in Lausanne (EPFL) have developed a machine-vision algorithm that can objectively analyze children's drawings.
The researchers used the algorithm to create an "average" of children's drawings, which enabled them to examine which parts of the page children prefer to draw on, the colors they use, the color intensity, and the complexity of the drawings. In addition, the new method enables the team to see how these aspects of children's drawings vary by age and location.
The researchers focused on the way children draw God. The database consists of 2,389 drawings of God by children between the ages of 5 and 15, and from a variety of cultural backgrounds and religions.
The researchers note the most impressive result is a clear demonstration that the complexity of the drawing changes as children get older. The researchers also found, using the machine-vision analysis, the average images from some parts of the world tend to be above the midline of a piece of paper, indicating the children consider God to be unworldly, while images from other regions are more centered. The researchers also found children tend to use yellow in the center of the images while green appears at the bottom, and they suggest green represents earthly objects and yellow represents supernatural objects.
From Technology Review
View Full Article
Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA
No entries found