Salient object in an image is that part of the image where all the human attention goes and rest part is mostly ignored by the vision of humans.
i wonder why many people researching about saliency. i can't get enough illustration how saliency would be any of use to human in the world. could u give me any example that describe usefullness of saliency?

The obvious uses are all around -- I will list a few:

Control room of a complex system (e.g. a nuclear reactor). In a typcially control room, you want the imformation that matters to be salient. That doesn't just include alarms but information that leads up to an alarm, e.g. water pressures or cooling flow rates, or temperature asymmetry in the core etc.

Second is dager type signals/signs. We have all tripped over the stupid "floor is wet" signs. They are NOT salient and represent a bigger hazard than the hazard that is actually signed about. But now consider that someone is using a radioactive source in a room that typically doesn't have one in. The sign should be salient, you really want to know. Same is true to signs on chemistry labs. There is lots of dangerous stuff in a typcial chemistry lab, but most chemists are aware of the normal stuff, but when you add something different/unusual even if it is not particularly dangerous, THAT saftey sign should to be more salient than the normal stuff -- but how should you achieve that?

Further what about reaction time incidents, anything were you want a rapid response from humans, and in particular crowds. The saliency of the signage can greatly influence the rate at which a human or a crowd of humans react to a change in situation.

The is also the salience relative to risk aspect. It seems [disputed and a subject to research] that the saliency of the sign telling someone about a risk is a high factor in the self-determination of the risk. [I understand but can't find -- reference for this -- but recall that there was a question if it is the actual saliency of a sign but the contrast in saliency to recently experienced danger signs] But it seems a large factor in determination of risk. Classic examples is : how big should a sign by a sea-side pier be about risk of being swept away by waves. Too big and everyone laughs and a lot of people go onto the pier to have their photo taken with a wave breaking over the pier, too small and people assume that the 3 foot railing will protect them from the 20foot waves breaking over the pier [... situational analysis by humans seems to have been lost over the last 30 years or so !]

Further there is work on saliency of presentations, i.e. what did you actually remember from the last presentation you saw? What did you remember and can actually apply?

You drive a car -- road signs should be salient -- how many do you need -- too many is just as bad as too few.

In the computational world, the amount of information available is huge, but how do we present multi-dimensional data saliently, i.e. so it is acted on by a human. I normally think of the human as a relativey long latency statistical sampling algorithm, saliency has significant [or could have] impact on that response time, and choise.

.. finally .. advertising but I am not sure that a benefit.

that's surprisingly many use of saliency for human, yet something new to me >_

thank's StuXYZ, now i see

Saliency detection is considered to be a key attentional mechanism that facilitates learning and survival by enabling organisms to focus their limited perceptual and cognitive resources on the most pertinent subset of the available sensory data.
Examples:

Suppose you are driving and suddenly a truck appears. You react immediatelly because of the saliency generated by the moving truck.

Suppose you are driving and suddenly a cyclist appears (wearing yellow clothes). The saliency focus your attention in the cyclist and you slow down your car.

for more info you can go here.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.