Learning support device, endoscope system, and method for supporting learning
A learning support device includes a processor. The processor is configured to: form a foreground image containing at least one treatment instrument by placing an image of the at least one treatment instrument within an image region; form a superimposed image by superimposing the foreground image on a background image; and form a training image by adjusting at least one of hue, saturation, or brightness of the superimposed image.
1 . A learning support device that supports formation of a learning model that recognizes a treatment instrument within an endoscopic image, the learning support device comprising a processor, wherein the processor is configured to:
form a foreground image containing at least one treatment instrument by placing an image of the at least one treatment instrument within an image region;
form a superimposed image by superimposing the foreground image on a background image; and
form a training image by:
forming a gamma-corrected image from the superimposed image;
forming a hue-corrected image from the superimposed image; and
forming the training image by synthesizing the superimposed image, the gamma-corrected image, and the hue-corrected image, wherein:
the gamma-corrected image is an image in which a gamma value of at least one of a hue, a saturation and a brightness of the superimposed image is corrected, and
the hue-corrected image is an image in which all hue values of the superimposed image are converted to zero.
2 . The learning support device according to claim 1 , wherein the processor is further configured to normalize the brightness of the superimposed image to reduce a difference in brightness between the superimposed image.
3 . The learning support device according to claim 1 , wherein the superimposed image, the gamma-corrected image, and the hue-corrected image are synthesized at a ratio of 0.125:0.5:0.375.
4 . The learning support device according to claim 1 , further comprising a storage unit configured to store a learning-use model, wherein the processor is further configured to cause the learning-use model to learn the training image to form the learning model that recognizes the treatment instrument within the endoscopic image.
5 . An endoscope system comprising:
the learning support device according to claim 4 ;
an endoscope configured to acquire at least one endoscopic image; and
an image processing apparatus including a processor and a storage unit configured to store the learning model, wherein the processor of the image processing apparatus is configured to input the endoscopic image to the learning model to obtain, from the learning model, a recognition result with respect to the treatment instrument within the endoscopic image.
6 . The endoscope system according to claim 5 , further comprising a display device, wherein the processor of the image processing apparatus is further configured to display the recognition result on the display device.
7 . A learning support device that supports formation of a learning model that recognizes a treatment instrument within an endoscopic image, the learning support device comprising a processor, wherein the processor is configured to:
form a foreground image containing at least one treatment instrument by placing an image of the at least one treatment instrument within an image region; and
form a training image by:
forming a gamma-corrected image from the foreground image;
forming a hue-corrected image from the foreground image; and
forming the training image by synthesizing the foreground image, the gamma-corrected image, and the hue-corrected image, wherein:
the gamma-corrected image is an image in which a gamma value of at least one of a hue, a saturation and a brightness of the foreground image is corrected, and
the hue-corrected image is an image in which all hue values of the foreground image are converted to zero.
8 . The learning support device according to claim 7 , wherein adjusting the at least one of the hue, the saturation, or the brightness of the foreground image includes converting a value of each pixel of the foreground image based on a LUT (lookup table).
9 . A method for supporting learning, the method supporting formation of a learning model that recognizes a treatment instrument within an endoscopic image, the method comprising:
forming a foreground image containing at least one treatment instrument by placing an image of the at least one treatment instrument within an image region;
forming a superimposed image by superimposing the foreground image on a background image; and
forming a training image by;
forming a gamma-corrected image from the superimposed image;
forming a hue-corrected image from the superimposed image; and
forming the training image by synthesizing the superimposed image, the gamma-corrected image, and the hue-corrected image, wherein:
the gamma-corrected image is an image in which a gamma value of at least one of a hue, a saturation and a brightness of the superimposed image is corrected, and the hue-corrected image is an image in which all hue values of the superimposed image are converted to zero.