algorithm - Fastest method to search for a specified item on an image? -

imagine have simple 2d drawing, filled lots of non-overlapping circles , few stars.

if find stars among these circles, can think of few methods. brute force 1 of them. 1 possibly reduce image size (to optimal point can still distinguish objects apart) , apply brute force , map original image. drawback of brute force of course, time consuming. looking faster methods, possibly fastest one.

what fastest image processing method search specified item on simple 2d image?

one typical way of looking object in image through cross correlation. basically, position cross-correlation between mask (the object you're attempting find) , image highest. position location of object you're trying find.

for sake of simplicity, refer object you're attempting find star, in general can shape.

some problems above approach:

• the size of mask has match size of star. if don't know size of star, have try different size masks. image pyramids more effective iteratively trying different size masks, still require effort.
• similarly, orientations of mask , star have match. if don't, cross-correlation won't work.

for these reasons, more know problem, simpler becomes. reason why people have asked more information in comments. general purpose solution doesn't exist, best of knowledge. maybe more knowledgeable can correct me on this.

as you've mentioned, reducing size of image reduce computational time of approach. in opinion, it's hardly core element of solution -- it's optional optimization step.