A method and a device for acquiring a face recognition training sample

Abstract

The embodiment of the invention provides a method and a device for acquiring a face recognition training sample, which relate to the technical field of face recognition. When biometric information matching the acquired first biometric information is present in the first biometric information database, A first face image associated with the first biometric information is acquired, and an image acquisition device is controlled to acquire a second face image of a user, and the first face image and the second face image are determined as a face recognition training sample. As that first face imageand the second face image are associated by using the first biometric information, the face recognition train sample is reliably obtained without manual participation, which improves efficiency and accuracy and saves human cost; At the same time, by acquiring the second biometric information associated with the first biometric information, a problem that a human replacement occurs during a periodwhen the image acquisition apparatus loses a photographing target, resulting in an error of a captured face sample is avoided.

Description

technical field [0001] The present invention relates to the technical field of face recognition, in particular to a method and device for acquiring training samples for face recognition. Background technique [0002] Face recognition is a series of related technologies that use cameras or cameras to collect images or video streams containing faces, and automatically detect and track faces in the images, and then perform face recognition on detected faces, usually also called portraits. Recognition, facial recognition. In the field of public security, the research and application of face recognition technology has very important practical significance. Face recognition technology includes the collection of face sample data, preprocessing of sample pictures, training models and sample recognition, among which, the collection of sample data is the premise of face recognition. [0003] In the existing technology, special sample collection workers are usually required to manual...

AI Technical Summary

Problems solved by technology

However, due to the huge amount of image data, a large number of sample collection workers are required, which is time-consuming, labor-intensive, slow, and inefficient, and the cost is also high; at the same time, human work itself has errors, and sample calibration is long-term repeatability Work, sample collection workers will inevitably be fatigued, resulting in errors in the final sample data, which will lead to the unreliability of the trained face recognition system

Images