Field finger and palm print image enhancement method and device and electronic equipment

Abstract

The invention provides an on-site finger and palm print image enhancement method and device and electronic equipment, and the method comprises the steps: carrying out the on-site finger and palm print simulation of a stamped finger and palm print image, and obtaining an on-site sample image of the simulated on-site finger and palm print image; the stamped finger and palm print image is clear, so that the target image corresponding to the stamped finger and palm print image can be used as a real image for model training, and the image enhancement model obtained by training based on the target image and a plurality of sample images including the field sample image can be well suitable for image enhancement of the field finger and palm print image. According to the field finger and palm print image enhancement method and apparatus, and the electronic device provided by the invention, the sample image-real image pair used for training the image enhancement model is generated from the stamped finger and palm print image in an automatic manner, and the field finger and palm print image enhancement is performed by using the image enhancement model. The problems of insufficient sample image data volume and difficulty in labeling are avoided, and the enhancement accuracy of the field finger and palm print image is improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a method, device and electronic equipment for enhancing an on-site finger and palmprint image. Background technique [0002] Finger and palm print comparison The on-site finger and palm print image (referring to the finger and palm print image collected at the scene of the accident) is often compared with the printed finger and palm print image contained in the base database to determine the identity of the person corresponding to the on-site finger and palm print image. The on-site finger and palm print images are often collected from the media, and there are many noises. It is usually difficult to obtain high accuracy by directly comparing the on-site finger and palm print images with the printed finger and palm print images contained in the bottom library. [0003] In order to improve the comparison accuracy, it is necessary to perform image enhancement on the ...

AI Technical Summary

Problems solved by technology

However, this calculation method will cause false enhancements to images with complex backgrounds, and is not sensitive to some detailed features (such as sticks or dots) in the palmprint.

If the method of deep learning is used, on the one hand, the number of on-site finger and palm print images is far less than the number of imprinted finger and palm print images, resulting in insufficient training data; The real image corresponding to the palmprint image

Images