Visual clutter detection

MIT researchers have designed a software tool that measures "visual clutter." According to the scientists, the system could someday help designers create better displays, maps, and data visualizations and steer our attention in various ways. The prototype tool, written in MATLAB, is freely available here. From the MIT News Office:

"We lack a clear understanding of what clutter is, what features, attributes and factors are relevant, why it presents a problem and how to identify it," said Ruth Rosenholtz, principal research scientist in MIT's Department of Brain and Cognitive Sciences (BCS) and the paper's lead author.

The fact that one person's clutter is the next person's organized workspace makes it hard to come up with a universal measure of clutter. Rosenholtz and colleagues modeled what makes items in a display harder or easier to pick out. They used this model, which incorporates data on color, contrast and orientation, to come up with a software tool to measure visual clutter.

To be useful, such a tool has to capture the effect of clutter on performance. In their paper, Rosenholtz and her colleagues– MIT BCS graduate student Yuanzhen Li and BCS undergraduate Lisa Nakano–tested the influence of clutter on searching for a symbol in a map, like an arrow indicating "you are here." They found good correlation between the time it takes to find a symbol in a map and the amount of clutter according to their measure.

Link