Human face recognition method based on artificial Slime Mould

Abstract

The invention discloses a human face recognition method based on artificial Slime Mould, and the method is characterized by simulating a Slime Mould foraging behavior, carrying out the matching and recognition of a human face image, and extracting a human face feature vector through the expansion and contraction processes of the artificial Slime Mould in the human face image. The specific algorithm comprises the following steps: step 1, initializing a face image and the artificial Slime Mould; step 2, carrying out feature extraction on the face image by using artificial Slime Mould; and step 3, matching and identifying the face image by using artificial Slime Mould. The myxamoeba of the artificial Slime Mould expands around on the face image, face feature pixel points are searched, and the searched face feature pixel points are added into a face feature pixel point set of the myxamoeba; afterwards, the artificial Slime Mould starts to contract continuously, namely, the external food source or face feature pixel points are digested, and the face matching and recognition functions are achieved. According to the method, optimization solution of face recognition can be efficiently completed.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a face recognition method based on artificial slime mold. Background technique [0002] Face recognition technology has always been a high-profile technology in the field of computer vision. In recent years, a series of sub-algorithms have emerged around this technology, such as face recognition algorithms based on convolutional neural networks, deep learning-based Face recognition algorithm and so on. [0003] Although the traditional face recognition algorithm can play a role in recognition, there are still the following problems: First, the extraction algorithm of the face image feature value used by the traditional face recognition algorithm increases rapidly with the increase of computing nodes, It is difficult to adapt to the needs of rapidly increasing pictures, and it is not convenient to obtain deeper and more semantic depth features from the original image. T...

AI Technical Summary

Problems solved by technology

[0003] Although the traditional face recognition algorithm can play a role in recognition, there are still the following problems: First, the extraction algorithm of the face image feature value used by the traditional face recognition algorithm increases rapidly with the increase of computing nodes, It is difficult to adapt to the needs of rapidly increasing images, and it is not convenient to obtain deeper and more semantic depth features from the original image

The second is that in order to obtain better recognition results, the traditional face recognition algorithm must be combined with artificial features, but in this process, artificial features often bring interference and errors that do not meet expectations, reducing the accuracy of human faces. Accuracy of recognition

Third, in the case of reducing human interference, traditional face recognition often shows difficulties and shortcomings of low algorithm performance in the face of large data sets

Images