The size-frequency distribution of asteroids with diameters < 10 km is essentially unknown. We propose to use ISOCAM (LW3, CAM01) in microscanning mode to test current predictions of the number of asteroids per square degree down to sizes below 1 km. Our objective is to measure the number of asteroids per unit area of sky observed as a function of IR brightness. (A 1 km main-belt asteroid has a flux density at 15 microns of about 1 mJy.) From these data we will be able to constrain, and perhaps decide among, current estimates for the numbers of asteroids with diameters greater than 1 km. These estimates differ by up to a factor of six. Any region of sky along the ecliptic is suitable for this purpose and so the target regions are independent of the launch date. In addition to the solar system science resulting from this study, we will provide data useful for assessing the "contamination" rate from asteroids in all ISOCAM observations.