Abstract:In order to eliminate the drawbacks in grain reserve management, a grain reserve intelligent audit method was designed. It was achieved by using the edge detection technology towards image samples to determine the edge of each object in the photograph, including the grain, the wall and benchmarks, and then separately holding pattern recognition to ascertain the identity of each object. Then according to the basic theories of 3-D reconstruction technology, combining with the location information of these objects in the original photographs, the underlying quantitative information could be dug out. At last, using grain weight measuring algorithm, which had been combined with the perspective error correction method, achieve the real-time and precise audit to the grain in stock and each installment.