Dual-band self-adaptive intelligent temperature measurement face recognition method

Abstract

The invention discloses a dual-band self-adaptive intelligent temperature measurement face recognition system, which accurately matches a visible light image with an infrared image through an image fusion technology, optimizes a recognition algorithm for a wearing state of a mask, and can intelligently detect the temperature of an effective region. When the body temperature of a person is abnormal, an alarm can be given automatically; if the person is in a face database, assistance in identifying the person and recording the name of the person is carried out. The system is an intelligent safety epidemic prevention monitoring system integrating a visible light wave band, an infrared wave band, a temperature measurement function, a face detection function, a face recognition function and a face recognition function under the condition that a mask is worn, and therefore non-contact, low-error and high-efficiency blackbody-free ultra-precision body temperature detection application is achieved.

Description

technical field [0001] The invention belongs to the field of artificial intelligence, and relates to an intelligent image face recognition and interaction hardware device and software processing intelligent monitoring equipment. Background technique [0002] How to use stricter prevention and control measures in public places on a large scale to ensure accurate temperature measurement while improving temperature measurement efficiency and avoiding cross-infection will be an effective prevention and control preparation for the double pressure of population movement. Now there are a variety of erection infrared thermometers emerging on the market, ranging from thousands of yuan to hundreds of thousands of yuan, which can be described as "a mixed bag". The recognition function causes a very high false alarm rate, which cannot be effectively applied, and the temperature measurement value has a large error. The calibration temperature accuracy of ±0.5°C will actually bring a temp...

AI Technical Summary

Problems solved by technology

Now there are a variety of erection infrared thermometers emerging on the market, ranging from thousands of yuan to hundreds of thousands of yuan, which can be described as "a mixed bag". The recognition function causes a very high false alarm rate, which cannot be effectively applied, and the temperature measurement value has a large error. The calibration temperature accuracy of ±0.5°C will actually bring a temperature span of 1°C, resulting in extremely inaccurate body temperature measurement, and the field of view Very small, it is necessary to regulate the action route of personnel, and it is impossible to realize a wide range of non-sensing temperature measurement capabilities

The relatively high-end is based on the "infrared + visible light" binocular camera, which can realize simultaneous detection by multiple people by displaying true-color images in real time. The recognition error can be between ±0.5°C and ±0.3°C. The temperature measurement results will also change accordingly. On the one hand, the reason is that the infrared thermal imager is too cheap, the temperature measurement accuracy is low, and the effective temperature measurement position is not accurate. On the other hand, the temperature measurement result of the infrared thermal imager will change with the ambient temperature Changes and changes, so accurate results cannot be obtained in different temperature environments

Images