looking for advice on manually labeling portions of unlabled images

I have a set of images that are somewhat complicated - they have many examples of what I am looking for on each image and they are currently unlabeled.

I am looking for a tool that would let me select rectangles on the image using my mouse/pointer to build up a training data set. Do you all know of any tools or simple code examples that might already do this?

code preference (best first): python, octave/matlab, R, java, C

EDIT: thanks all, here is an implementation in python if anyone is interested

from pylab import * fish= imread('./stevesFish/1b2.tif') figsrc = figure() figzoom = figure() axsrc = figsrc.add_subplot(111) axzoom = figzoom.add_subplot(111, xlim=(0,20), ylim=(0, 20), autoscale_on=False) axsrc.set_title('Click to zoom') axzoom.set_title('zoom window') axsrc.imshow(fish) axzoom.imshow(fish) def onClick(event): if event.button!=1: return x,y = event.xdata, event.ydata axzoom.set_xlim(x-25, x+25) axzoom.set_ylim(y-25, y+25) figzoom.canvas.draw() def onKeyPress(event): if event.key!='d': return print axzoom.get_xlim(), axzoom.get_ylim() figsrc.canvas.mpl_connect('button_press_event', onClick) figsrc.canvas.mpl_connect('key_press_event', onKeyPress) show()