To estimate the correct dimension of the target on the underwater, we can use CCD camera and cast laser light strip onto the target, and then observe the displacement of laser light to get the dimension. Since the laser light will show on different situation, it's due to the surface of the target is not smooth. When we get the image from CCD camera, we need to calibrate the displacement of the laser light and return to the actual dimension of the target on the underwater. We know the optical distortion and non-linearity of the CCD camera will influence to get the correct image, also the location of camera is. That's the reason we need to calibrate the camera first. It was a mathematical way to explain the calibration of CCD camera non-linearity before. On this subject, we lay vertical and horizontal grid lines of 50 mm span on an acrylic plate. These grid lines are same as the longitudes and latitudes of the map. We estimate the target with the pair of interpolated longitude and latitude same as to be used to estimate the location of the point in the world coordinate system. And choose some targets with different size and form to use to verify the approach. By the way, we also test if there is any influence for the clear of water. The results indicate the error is under 3 \% when we catch the image on the underwater by a calibrated CCD camera.