US 12,105,033 B2
Inspection system, determination processing apparatus, and inspection method
Teruaki Sano, Kanagawa (JP); Kiminori Shigetomi, Kanagawa (JP); Shigeo Otsuki, Kanagawa (JP); and Nobutaka Tanaka, Kanagawa (JP)
Assigned to JGC CORPORATION, Kanagawa (JP)
Appl. No. 17/638,176
Filed by JGC CORPORATION, Kanagawa (JP)
PCT Filed Jul. 16, 2020, PCT No. PCT/JP2020/027719
§ 371(c)(1), (2) Date Feb. 24, 2022,
PCT Pub. No. WO2022/014019, PCT Pub. Date Jan. 20, 2022.
Prior Publication US 2022/0299451 A1, Sep. 22, 2022
Int. Cl. G01N 23/04 (2018.01); G01N 23/083 (2018.01); G01N 33/2045 (2019.01); G01N 33/207 (2019.01); G01T 7/00 (2006.01); G06T 7/00 (2017.01)
CPC G01N 23/04 (2013.01) [G01N 23/083 (2013.01); G01N 33/2045 (2019.01); G01N 33/207 (2019.01); G06T 7/001 (2013.01); G01N 2223/04 (2013.01); G01N 2223/3303 (2013.01); G01N 2223/401 (2013.01); G01N 2223/628 (2013.01); G01N 2223/629 (2013.01); G01N 2223/646 (2013.01); G06T 2207/10116 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20212 (2013.01); G06T 2207/30136 (2013.01)] 11 Claims
OG exemplary drawing
 
1. An inspection system relating to a nondestructive inspection of a defect of a welded portion of a pipe or a pipe member, the inspection system comprising:
an imaging data acquisition unit configured to acquire imaging data obtained by transmitting radiation through the welded portion of the pipe or pipe member to be inspected, wherein the imaging data acquisition unit comprises a radiation irradiation unit configured to irradiate radiation and a detector configured to detect the radiation;
a first processor configured to perform processing of associating, with image data showing a distribution of a transmission intensity of the radiation obtained from the imaging data, determination data which indicates a result of determining a defect of the welded portion to be inspected based on the distribution of the transmission intensity of the radiation, wherein the determination data is associated with information with which a position of a flaw in a welding line included in the image data, a size of the flaw, and a shape feature of the flaw are identifiable, and includes information indicating a type of a defect which is a harmful flaw for which a result of welding of the welded portion is determined to be a failure; and
a second processor configured to create, based on results obtained by performing machine learning of flaws included in past image data with which the determination data is associated, preliminarily determined image data by associating preliminary determination data obtained by automatically determining a type of a predicted defect by identifying the position, size, and shape feature of the flaw in the welding line, with new image data showing a distribution of the transmission intensity of the radiation,
wherein the second processor is configured to transmit the preliminarily determined image data to the first processor, and
wherein the first processor is configured to perform processing of receiving a worker's determination for the preliminary determination data associated with the preliminarily determined image data that the preliminary determination data is correct and associating the resultant preliminary determination data with the new image data as proper determination data.