Automatic Cartridge Case Image Mosaic Using SIFT and Graph Transformation Matching

Characteristic marks on the cartridge can be viewed as a "fingerprint" for identification of a firearm. Sometimes, however, not all information can be obtained from just one image due to the limitations of microscope and the unsmoothed specimen surface in the cartridge case image detection. Image mosaic that refers to the combination of two or more images into a single composite image is precisely to solve this problem. This paper proposes a new cartridge case image mosaic approach by using automatic image registration and fusion techniques. In the registration stage, the scale invariant feature transform (SIFT) is employed to obtain initial matching. Then the graph transformation matching (GTM) is used to remove incorrect matches effectively. In the fusion stage, histogram matching is applied to smooth seams which exist in the stitched image. Finally, a stitched image is obtained, which contains more information than each of the two original images. Experimental results demonstrate that the proposed approach achieves good performance for cartridge case image mosaic in both subjective and objective evaluations.