Person recognition method based on nose image features in breathing process

Abstract

The invention discloses a person recognition method based on nose image features in a breathing process, and the method comprises the steps: collecting nose breathing features in real time through an infrared thermal imager, comparing the nose breathing features collected in real time with the nose breathing features which is collected in advance in the database, and determining that two sides are the same person when the comparison result reaches a se similarity threshold value, wherein the comparison of the nose breathing features is the comparison of nose static breathing parameter features in a plurality of continuous breathing periods and the comparison of noise dynamic breathing parameter features. The breathing feature parameters based on shape and position clustering reflect the substantive characteristics of the breathing of a person, and the substantive characteristics are independent of a specific breathing mode and do not change along with the external environment and the temporary physiological reaction of the person. Therefore, the method can recognize the person which breathes in any mode. The method can recognize the person which breathes in any mode, and supplement for the current human face recognition.

Description

technical field [0001] The invention relates to face recognition, in particular to a person recognition method based on nose image features during breathing. Background technique [0002] The automatic identification of people is usually based on their physiological and behavioral characteristics, such as the voice or speaking style, face, iris, retina, fingerprint, palm shape and other characteristics of a specific person. When choosing an automatic identification system, the following factors are usually considered: ability to resist fraud, ease of use, interference to identified people, applicability to special groups, speed of identification, size of model (feature information) , Long-term stability of biological characteristics, cost of use, etc. No one identification method can win in all the above aspects, so it is necessary to choose a more suitable identification method according to the specific requirements of an application. For example, the price of equipment f...

AI Technical Summary

Problems solved by technology

However, there are the following problems in face recognition: 1. The effect of recognition is greatly affected by lighting conditions. When suitable lighting conditions are not available (for example, when the size of the device is limited and the power supply capacity is limited), it is impossible to implement reliable facial recognition. Face recognition; 2. Different facial expressions will affect data collection and extraction; 3. Ornament occlusion and damage problems: If the recognized object has accessories such as sunglasses, hats, or there is damage or dirt on the face, It will cause the lack of signal data; 4. The problem of twins; 5. The maturity and use of facial makeup and plastic surgery technology will bring greater variability to the face, and the current recognition system is almost difficult to solve