Radiography method, trained model, radiography module, radiography program, radiography system, and machine learning method
A control device includes an input unit configured to accept an input of condition information indicating either operating conditions of a source of X-rays when the X-rays are radiated to capture an image of a target object or imaging conditions during capturing an image of the target object, a calculation unit configured to calculate average energy of the X-rays passing through the target object on the basis of the condition information, and a narrowing unit configured to narrow down candidates for a trained model from a plurality of trained models constructed through machine training in advance using image data on the basis of the average energy.
1 . A radiographic image processing method comprising:
inputting condition information indicating either conditions of a source of radiation or imaging conditions when the radiation is radiated to capture an image of a target object;
calculating average energy related to the radiation passing through the target object on the basis of the condition information;
narrowing down candidates for a trained model from a plurality of trained models constructed through machine training in advance using image data on the basis of the average energy; and
executing image processing for removing noise from a radiographic image of the target object using the candidates.
2 . The radiographic image processing method according to claim 1 , wherein narrowing down includes narrowing down the candidates by comparing the average energy with average energy specified from the image data.
3 . The radiographic image processing method according to claim 1 , wherein the condition information includes at least any one of a tube voltage of the source, information on a filter included in a camera used to capture an image of the target object, information on a filter included in the source, information on a scintillator included in the camera, a distance between the source and an imaging device, information relating to an X-ray detection camera used to capture an image of the target object, and information relating to the target object.
4 . The radiographic image processing method according to claim 1 , further comprising acquiring a radiographic image by radiating the radiation to capture an image of a jig and selecting a trained model from the candidates on the basis of image characteristics of the radiographic image.
5 . The radiographic image processing method according to claim 1 , wherein the machine training is deep learning.
6 . A trained model used for the radiographic image processing method according to claim 1 , wherein the trained model is constructed through machine training using image data and causes a processor to execute image processing for removing noise from a radiographic image of the target object.
7 . A radiographic image processing module comprising:
at least one processor,
wherein the at least one processor is configured to
accept an input of condition information indicating either conditions of a source of radiation or imaging conditions when the radiation is radiated to capture an image of a target object,
calculate average energy related to the radiation passing through the target object on the basis of the condition information,
narrow down candidates for a trained model from a plurality of trained models constructed through machine training in advance using image data on the basis of the average energy, and
execute image processing for removing noise from a radiographic image of the target object using the candidates.
8 . The radiographic image processing module according to claim 7 , wherein the at least one processor narrows down the candidates by comparing the average energy with average energy specified from the image data.
9 . The radiographic image processing module according to claim 7 , wherein the condition information includes information relating to a tube voltage of the source, a filter, a distance between the source and an imaging device, an X-ray detection camera, and the target object.
10 . The radiographic image processing module according to claim 7 , wherein the at least one processor acquires a radiographic image by radiating the radiation to capture an image of a jig and selects a trained model from the candidates on the basis of image characteristics of the radiographic image.
11 . The radiographic image processing module according to claim 7 , wherein the machine training is deep learning.
12 . A radiographic image processing program causing a processor to function as:
accepting an input of condition information indicating either conditions of a source of radiation or imaging conditions when the radiation is radiated to capture an image of a target object;
calculating average energy related to the radiation passing through the target object on the basis of the condition information;
narrowing down candidates for a trained model from a plurality of trained models constructed through machine training in advance using image data on the basis of the average energy; and
executing image processing for removing noise from a radiographic image of the target object using the candidates.
13 . A radiographic image processing system comprising:
the radiographic image processing module according to claim 7 ;
the source configured to radiate radiation to the target object; and
an imaging device configured to capture an image of the radiation passing through the target object and acquire the radiographic image.
14 . A machine training method comprising constructing a trained model for outputting image data from which noise has been removed on the basis of a training image using, as training data, the training image that is a radiographic image of a target object corresponding to average energy related to radiation passing through the target object, the average energy being calculated on the basis of condition information indicating either conditions of a source of the radiation or imaging conditions when the radiation is radiated to capture an image of the target object.