Image processing method that performs gamma correction to update neural network parameter, image processing apparatus, and storage medium

1. An image processing method comprising the steps of:

acquiring a training image and a correct image;

inputting the training image into a multilayer neural network to generate an output image;

performing a gamma correction for each of the correct image and the output image and calculating an error between the correct image after the gamma correction and the output image after the gamma correction; and

updating a network parameter of the neural network using the error.

2. The image processing method according to claim 1 ,

wherein each of the training image and the correct image has a plurality of color components periodically arranged,

wherein the image processing method further comprises:

a separating step for the training image configured to separate each of the plurality of color components of the training image; and

a separating step for the correct image configured to separate each of the plurality of color components of the correct image,

wherein the separating step for the training image is performed before the training image is input into the neural network, and

wherein the separating step for the correct image is performed before the error is calculated with the correct image.

3. The image processing method according to claim 1 , further comprising the step of clipping a signal value of the output image in a predetermined range before the gamma correction is performed.

4. The image processing method according to claim 1 , wherein the output image is an image having a resolution or contrast higher than that of the training image.

5. The image processing method according to claim 1 , further comprising the steps of:

acquiring a lower limit value and an upper limit value of a signal value of each of the training image and the correct image; and

normalizing the signal value of each of the training image and the correct image using the lower limit value and the upper limit value.

6. The image processing method according to claim 1 , further comprising the step of designating the gamma correction,

wherein the step of calculating the error performs the gamma correction designated in the step of designating the gamma correction, for each of the output image and the correct image.

7. The image processing method according to claim 6 , further comprising the step of storing information on the gamma correction designated in the step of designating the gamma correction and the network parameter updated in the step of updating the network parameter.

8. The image processing method according to claim 1 , wherein the updating of the network parameter updates the network parameter using a plurality of errors calculated by performing a plurality of gamma corrections for each of the correct image and the output image.

9. An image processing apparatus comprising:

at least one memory configured to store instructions; and

at least one processor communicatively connected to the at least one memory and configured to execute the stored instructions to:

acquire a training image and a correct image;

input the training image into a multilayer neural network and generate an output image;

perform a gamma correction for each of the correct image and the output image and calculate an error between the correct image after the gamma correction and the output image after the gamma correction; and

update a network parameter of the neural network using the error.

10. The image processing apparatus according to claim 9 , wherein the network parameter is stored in the memory.

11. A non-transitory computer-readable storage medium configured to store a program for causing a computer to execute an image processing method comprising the steps of:

acquiring a training image and a correct image;

inputting the training image into a multilayer neural network to generate an output image;

performing a gamma correction for each of the correct image and the output image and calculating an error between the correct image after the gamma correction and the output image after the gamma correction; and

updating a network parameter of the neural network using the error.