Method for determining heat and reflected heat in thermal image
A method for determining heat and reflected heat in a thermal image, according to the present invention, comprises the steps of: a) photographing a subject through a thermal imaging camera so as to collect a thermal image of the subject; b) analyzing features according to the thermal image temperature distribution of the thermal image; c) detecting the heat and the reflected heat in the thermal image according to the analysis in step b); and d) displaying, as heat and reflected heat regions in the thermal image, the heat and the reflected heat in the thermal image detected in step c).
1 . A method of determining heat and reflected heat in a thermal image, the method comprising:
a) photographing a subject through a thermal imaging camera, and collecting a thermal image of the subject;
b) analyzing features based on a thermal image temperature distribution of the thermal image;
c) detecting heat and reflected heat in the thermal image based on the analysis in step b); and
d) displaying the heat and the reflected heat in the thermal image, detected in step c), as heat and reflected-heat regions in the thermal image,
wherein the step b) comprises:
(i) extracting a region-of-interest of the thermal image through machine learning based on an object extraction neural network;
(ii) extracting raw data for each pixel of the region-of-interest; and
iii) extracting temperature data of the region-of-interest, and separately managing or filing the extracted temperature data,
wherein the step c) comprises analyzing variance in temperature between a pixel to be analyzed in the region-of-interest and surrounding pixels to determine reflected-heat spots corresponding to the pixels between which the variance in temperature is greater than or equal to a preset threshold.
2 . The method of claim 1 , wherein the step a) comprises generating moving image data of the subject through the thermal imaging camera, and extracting and collecting the thermal image for each frame from the generated moving image data.
3 . The method of claim 1 , wherein the temperature data extracted from the region-of-interest is cleansed, the region-of-interest is searched for low-heat distribution regions in units of pixels, and the found low-heat distribution regions are excluded from the region-of-interest.
4 . The method of claim 1 , wherein heat distributions having similar features in the region-of-interest are clustered, and clusters smaller than a certain reference size are excepted.
5 . The method of claim 1 wherein the step c) comprises detecting heat spots, which have a value greater than or equal to a certain threshold in the region-of-interest, in units of pixels.