X-ray image processing apparatus and X-ray image processing method
This X-ray image processing apparatus includes: an image acquisition unit; a frequency resolution processing unit configured to perform frequency resolution processing for resolving an X-ray image into a high-frequency component image and a low-frequency component image; a high-resolution image generation unit configured to generate a high-resolution high-frequency component image from the high-frequency component image by a trained learning model, the high-resolution high-frequency component image being an image higher in resolution than the high-frequency component image; and an image synthesis unit configured to synthesize an image based on the low-frequency component image and an image based on the high-resolution high-frequency component image to generate a high-resolution X-ray image.
1 . An X-ray image processing apparatus comprising:
an image acquisition unit configured to acquire an X-ray image;
a frequency resolution processing unit configured to perform frequency resolution processing for resolving the X-ray image into a high-frequency component image and a low-frequency component image;
a high-resolution image generation unit configured to generate a high-resolution high-frequency component image from the high-frequency component image by a trained learning model that has been trained to enhance a resolution of an image without performing processing for generating an image higher in resolution than the low-frequency component image from the low-frequency component image, the high-resolution high-frequency component image being an image higher in resolution than the high-frequency component image; and
an image synthesis unit configured to synthesize an image based on the low-frequency component image without being made high definition and an image based on the high-resolution high-frequency component image that is high definition to generate a high-resolution X-ray image.
2 . The X-ray image processing apparatus as recited in claim 1 ,
wherein the frequency resolution processing unit is configured to perform frequency resolution processing in which a frequency band to be resolved is set such that frequency components of a predetermined target site in the X-ray image become high-frequency side components.
3 . The X-ray image processing apparatus as recited in claim 2 ,
wherein the X-ray image is an image reflecting a blood vessel of a subject and a device introduced into the blood vessel, and
wherein the target site includes an edge of the blood vessel and an edge of the device.
4 . The X-ray image processing apparatus as recited in claim 1 ,
wherein the image acquisition unit is configured to acquire the X-ray image as a moving image,
wherein the frequency resolution processing unit is configured to perform frequency resolution processing on an acquired frame each time a frame of the X-ray image as a moving image is acquired to acquire the low-frequency component image and the high-frequency component image for each frame,
wherein the high-resolution image generation unit is configured to generate the high-resolution high-frequency component image from the high-frequency component image acquired for each frame, and
wherein the image synthesis unit is configured to synthesize the low-frequency component image for each frame and the high-resolution high-frequency component image for each frame to generate the high-resolution X-ray image as a moving image.
5 . The X-ray image processing apparatus as recited in claim 1 ,
wherein the frequency resolution processing unit is configured to perform smoothing filter processing on the X-ray image to acquire the low-frequency component image and subtract the low-frequency component image from the X-ray image to acquire the high-frequency component image.
6 . The X-ray image processing apparatus as recited in claim 1 , further comprising:
an image enlargement unit configured to enlarge an image,
wherein the image enlargement unit is configured to enlarge the image either before performing the frequency resolution processing by the frequency resolution processing unit or after performing the frequency resolution processing.
7 . The X-ray image processing apparatus as recited in claim 6 ,
wherein the image enlargement unit is configured to enlarge the low-frequency component image acquired by performing the frequency resolution processing by the frequency resolution processing unit to generate a low-frequency component enlarged image,
wherein the learning model has further trained to enlarge the image to be generated when increasing the resolution of the image,
wherein the high-resolution image generation unit is configured to generate a high-resolution high-frequency component enlarged image improved in resolution than the high-frequency component image and enlarged in size, from the high-frequency component image by the learning model, and
wherein the image synthesis unit is configured to synthesize the low-frequency component enlarged image and the high-resolution high-frequency component enlarged image to generate the high-resolution X-ray image.
8 . The X-ray image processing apparatus as recited in claim 6 , further comprising:
an image reduction unit configured to reduce an image,
wherein the image enlargement unit is configured to enlarge the X-ray image before performing the frequency resolution processing by the frequency resolution processing unit,
wherein the learning model has further trained to enlarge the image to be generated when enhancing the resolution of the image,
wherein the frequency resolution processing unit is configured to generate a low-frequency component enlarged image and a high-frequency component enlarged image, from the enlarged X-ray image,
wherein the image reduction unit is configured to reduce the high-frequency component enlarged image to acquire a high-frequency component reduced image,
wherein the high-resolution image generation unit is configured to generate a high-resolution high-frequency component enlarged image from the high-frequency component reduced image, by the learning model, and
wherein the image synthesis unit is configured to synthesize the low-frequency component enlarged image and the high-resolution high-frequency component enlarged image to generate the high-resolution X-ray image.
9 . The X-ray image processing apparatus as recited in claim 6 ,
wherein the image enlargement unit is configured to enlarge the X-ray image before performing the frequency resolution processing by the frequency resolution processing unit,
wherein the frequency resolution processing unit is configured to generate a low-frequency component enlarged image and a high-frequency component enlarged image from the enlarged X-ray image,
wherein the high-resolution image generation unit is configured to generate the high-resolution high-frequency component enlarged image from the high-frequency component enlarged image, by the learning model, and
wherein the image synthesis unit is configured to synthesize the low-frequency component enlarged image and the high-resolution high-frequency component enlarged image to generate the high-resolution X-ray image.
10 . The X-ray image processing apparatus as recited in claim 6 ,
wherein the image enlargement unit is configured to enlarge the image by a reversibly convertible interpolation algorithm.
11 . The X-ray image processing apparatus as recited in claim 8 ,
wherein the image reduction unit is configured to reduce the image by a reversibly convertible interpolation algorithm.
12 . An X-ray image processing method comprising the steps of:
acquiring an X-ray image;
performing frequency resolution processing for resolving the X-ray image into a high-frequency component image and a low-frequency component image;
producing a high-resolution high-frequency component image from the high-frequency component image by a trained learning model that has been trained to enhance a resolution of an image without generating an image higher in resolution than the low-frequency component image from the low-frequency component image, the high-resolution high-frequency component image being an image higher in resolution than the high-frequency component image; and
synthesizing an image based on the low-frequency component image without being made high definition and an image based on the high-resolution high-frequency component image being made high definition to generate a high-resolution X-ray image.
13 . An X-ray image processing apparatus comprising:
an image acquisition unit configured to acquire an X-ray image of a blood vessel of a subject and a device introduced into the blood vessel;
a frequency resolution processing unit configured to perform frequency resolution processing for resolving the X-ray image into a high-frequency component image and a low-frequency component image, wherein the high-frequency component image corresponds to an edge of the blood vessel and an edge of the device;
a high-resolution image generation unit configured to generate a high-resolution high-frequency component image from the high-frequency component image by a trained learning model that has been trained to enhance a resolution of an image without performing processing for generating an image higher in resolution than the low-frequency component image from the low-frequency component image, the high-resolution high-frequency component image being an image higher in resolution than the high-frequency component image; and
an image synthesis unit configured to synthesize an image based on the low-frequency component image without being made high definition and an image based on the high-resolution high-frequency component image being made high definition to generate a high-resolution X-ray image.
14 . The X-ray image processing apparatus of claim 13 , wherein the low-frequency component image corresponds to a background.
15 . The X-ray image processing apparatus as recited in claim 13 ,
wherein the image acquisition unit is configured to acquire the X-ray image as a moving image,
wherein the frequency resolution processing unit is configured to perform frequency resolution processing on an acquired frame each time a frame of the X-ray image as a moving image is acquired to acquire the low-frequency component image and the high-frequency component image for each frame,
wherein the high-resolution image generation unit is configured to generate the high-resolution high-frequency component image from the high-frequency component image acquired for each frame, and
wherein the image synthesis unit is configured to synthesize the low-frequency component image for each frame and the high-resolution high-frequency component image for each frame to generate the high-resolution X-ray image as a moving image.
16 . The X-ray image processing apparatus as recited in claim 13 ,
wherein the frequency resolution processing unit is configured to perform smoothing filter processing on the X-ray image to acquire the low-frequency component image and subtract the low-frequency component image from the X-ray image to acquire the high-frequency component image.
17 . The X-ray image processing apparatus as recited in claim 13 , further comprising:
an image enlargement unit configured to enlarge the low-frequency component image acquired by performing the frequency resolution processing by the frequency resolution processing unit to generate a low-frequency component enlarged image,
wherein the learning model has been further trained to enlarge the image to be generated when increasing the resolution of the image,
wherein the high-resolution image generation unit is configured to generate a high-resolution high-frequency component enlarged image improved in resolution than the high-frequency component image and enlarged in size, from the high-frequency component image by the learning model, and
wherein the image synthesis unit is configured to synthesize the low-frequency component enlarged image and the high-resolution high-frequency component enlarged image to generate the high-resolution X-ray image.
18 . The X-ray image processing apparatus as recited in claim 17 , further comprising:
an image reduction unit configured to reduce an image,
wherein the image enlargement unit is configured to enlarge the X-ray image before performing the frequency resolution processing by the frequency resolution processing unit,
wherein the learning model has been further trained to enlarge the image to be generated when enhancing the resolution of the image,
wherein the frequency resolution processing unit is configured to generate a low-frequency component enlarged image and a high-frequency component enlarged image, from the enlarged X-ray image,
wherein the image reduction unit is configured to reduce the high-frequency component enlarged image to acquire a high-frequency component reduced image,
wherein the high-resolution image generation unit is configured to generate a high-resolution high-frequency component enlarged image from the high-frequency component reduced image, by the learning model, and
wherein the image synthesis unit is configured to synthesize the low-frequency component enlarged image and the high-resolution high-frequency component enlarged image to generate the high-resolution X-ray image.