The invention discloses an efficient
forehead temperature
estimation method based on wide-spectrum human
forehead imaging and distance sensing, and provides the scheme for solving a problem of complexcalculation of
forehead temperature measurement in the prior art. The method comprises a neural network training step, a step of calculating an optimal value of a coefficient in a
temperature measurement model, a step of performing forehead region image interception and energy value calculation by using a neural network, a step of obtaining a distance of an object
human body, and a step of calculating the temperature of the object
human body by using the
temperature measurement model. The method is suitable for a whole mid-
infrared spectrum range, efficient in calculation and low in hardwarecost. Compared with other areas of a human face, the mid-
infrared imaging of the forehead region of the
human body not only contains stable temperature information, but also can effectively avoid thermal induction interference caused by shielding of glasses and masks possibly worn on the human face. The method can be used for
forehead temperature estimation of multiple persons, which means that when multiple persons exist in a scene, after
infrared imaging and imaging distances in face regions of different persons are obtained, a
forehead temperature value of each person can be rapidly calculated.