Face portrait synthesis method based on coupled nearest neighbor index

Abstract

The invention discloses a face figure synthetic method based on coupling neighbor indexes, mainly solving the problem that a present method does not consider the training figure effect in neighbor selection. The method comprises the steps of: 1, diving a training figure sample set, a training photograph sample set and a test sample set; 2, diving all images into image blocks and forming block sets; 3, diving training photograph block sets and corresponding image blocks sets into a plurality of subsets; 4, establishing the indexes of each pair of photograph-image block in the training photograph block subsets and training image block subsets to K neighbor pairs thereof; 5, searching for a nearest neighbor of each test photograph block, and furthermore obtaining K neighbors thereof; 6, utilizing a Markov weight network to solve a reconstruction weight; 7, solving an image block to be synthesized; and 8, iteratively executing steps 5-7 for N times, and obtaining a finally synthesized image through fusion. Compared with the prior art, the synthesized image has higher definition and more complete details, and can be used for face retrieval and identification.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a face image synthesis method, which can be used for face retrieval and recognition in criminal investigation and case solving. Background technique [0002] Face portrait synthesis plays a huge role in digital entertainment and criminal investigation. For example, in 3D chocolate printing, it is usually necessary to use the face portrait synthesis algorithm to synthesize a black-and-white portrait as the printing track of the printer; in the process of criminal investigation and case detection, it is often not always possible to obtain photos of criminal suspects, but often witnesses Some descriptions or video image data, in order to quickly solve the case, a feasible solution is to draw a portrait based on the witness description and the clues provided by the video image, and then use the portrait to confirm the identity. However, due to the large differences in...

AI Technical Summary

Problems solved by technology

However, the shortcomings of this method are: because only the information of the photo block is used to find the neighbors and the role of the training image block is ignored, resulting in noise and blur in the synthesis result.

However, the disadvantage of this method is that only one training image block is selected for image synthesis for each photo block position, resulting in deformation of the synthesis result.

However, because this method does not consider the role of training image blocks, the searched neighbor blocks are not accurate, resulting in a lot of noise in the synthesis results.

Images