US 12,405,167 B2
Active infrared thermography system and computer-implemented method for generating thermal image
Chun Yong Andrew Ngo, Singapore (SG); Vinod Kumar S/O Sivaraja, Singapore (SG); and Wenjiang Guo, Singapore (SG)
Assigned to AGENCY FOR SCIENCE, TECHNOLOGY AND RESEARCH, Singapore (SG)
Appl. No. 17/797,787
Filed by Agency for Science, Technology and Research, Singapore (SG)
PCT Filed Feb. 2, 2021, PCT No. PCT/SG2021/050050
§ 371(c)(1), (2) Date Aug. 5, 2022,
PCT Pub. No. WO2021/158174, PCT Pub. Date Aug. 12, 2021.
Claims priority of application No. 10202001122U (SG), filed on Feb. 7, 2020.
Prior Publication US 2023/0046023 A1, Feb. 16, 2023
Int. Cl. G01J 5/48 (2022.01); G01J 5/00 (2022.01); G01J 5/02 (2022.01); G01J 5/08 (2022.01); G01N 25/72 (2006.01)
CPC G01J 5/48 (2013.01) [G01J 5/025 (2013.01); G01J 5/0896 (2013.01); G01N 25/72 (2013.01); G01J 2005/0077 (2013.01)] 28 Claims
OG exemplary drawing
 
1. An active infrared thermography system, comprising:
one or more excitation sources;
an infrared camera;
one or more portable power sources arranged to power the one or more excitation sources and the infrared camera;
a housing, wherein the one or more excitation sources and the one or more portable power sources are received in the housing;
a non-transitory computer-readable memory storing computer program instructions executable by one or more computer processors to perform operations for generating a thermal image, the operations comprising:
acquiring thermal image data on a subject;
activating one or more excitation sources to radiate the subject;
ending acquisition of the thermal image data on the subject when a cooling pattern is captured;
determining a cooling curve of each pixel of the thermal image data;
calculating a cooling coefficient of each pixel of the thermal image data; and
scaling the cooling coefficient of each pixel of the thermal image data to a colour palette to produce a heat map; and
fusing the heat map to a visual image of the subject;
wherein the operation of fusing the heat map to the visual image of the subject comprises:
identifying coordinates of at least three sets of matching feature points on each of the heat map and the visual image;
calculating a transformation matrix with the coordinates of the at least three sets of matching feature points on each of the heat map and the visual image using least square optimisation; and
applying the transformation matrix onto the heat map to generate a corrected heat map.