| In recent years,vein imaging and vein recognition technology have become a hot research topic,more and more researchers threw themselves into the study of vein related technology.With the research project developing,the quality of the dorsal hand vein images that are used for imaging or recognition is expected with a higher standard.Projecting methods and matching methods can’t work well without basing a high quality vein image despite these methods are excellent.Therefore,enhancing the dorsal hand vein image to get more character information will provide a significant condition for the follow-up job to work on.In this paper,the dorsal hand vein image enhancement is seen as the research object,and the specific work is as follows:1.A dorsal hand vein image acquisition system was set up after choosing suitable devices.A normalization method and a denoise method were implemented as the preprocessing to prepare for the follow-up job.2.A dorsal hand vein image enhancement method based on gradient domain was proposed.With this method,the original vein image was firstly mapped to the gradient domain by the sobel operator,then operating the gradient field of the original image by a modified function aiming at the characters of dorsal hand vein images to enlarge the smaller gradients as well as decrease the bigger gradients and putting different weight on partial derivatives of different directions to construct a new gradient field,lastly a new enhanced image was reconstructed from the new gradient field by the discrete sine transform.3.The proposed gradient domain method was tested in the experiment and the processing results were compared with the effects of the other two common vein enhancement methods.The experimental results demonstrated that the proposed gradient domain method had a better evaluation than the other two methods,this method was proved to be a fast and effective new enhancement algorithm for the infrared dorsal hand vein images. |
