Image processing device, image processing method, and integrated circuit in which super resolution process data is output as output data for input pixel data that is classified as high-frequency region data and that has noise less than or equal to a threshold

1. An image processing device for generating an output image by performing an image process on an input image, the image processing device comprising:

a super resolution processing unit configured to generate super resolution process data by performing a super resolution process on a plurality of pieces of input pixel data that represent the input image;

a region determining unit configured to classify each of the plurality of pieces of input pixel data into one of high-frequency region data and non-high-frequency region data; and

a generating unit configured to generate, as at least part of output pixel data that represents the output image, one or more pieces of output pixel data corresponding to one or more pieces of input pixel data classified as the high-frequency region data, by using the super resolution process data of the one or more pieces of input pixel data as the one or more pieces of output pixel data based on an amount of noise included in the one or more pieces of input pixel data;

wherein when an amount of noise included in a piece of input pixel data classified as the high-frequency region data is smaller than or equal to a predetermined threshold, the generating unit outputs a piece of super resolution process data corresponding to the piece of input pixel data without change as a piece of output pixel data corresponding to the piece of input pixel data and

wherein when an amount of noise included in the piece of input pixel data classified as the high-frequency region data is larger than the predetermined threshold, the generating unit outputs data other than the piece of super resolution process data corresponding to the piece of input pixel data as the piece of output pixel data corresponding to the piece of input pixel data.

2. The image processing device of claim 1 , wherein

the region determining unit assigns a label to each of the input pixel data that differs depending on whether the input pixel data is classified into the high-frequency region data or the non-high-frequency region data.

3. The image processing device of claim 1 further comprising

a filter processing unit configured to generate filter process data by applying a filter process to the plurality of pieces of input pixel data,

the generating unit including

a first mixing unit configured to generate one or more pieces of first mixture data by mixing the super resolution process data with the filter process data in accordance with the amounts of noise of the one or more pieces of input pixel data,

wherein, as the data other than the piece of super resolution process data corresponding to the piece of input pixel data, the generating unit outputs the one or more pieces of first mixture data as the one or more pieces of output pixel data.

4. The image processing device of claim 3 , wherein

the first mixing unit calculates a mixture ratio α: (1−α) satisfying 0≦α≦1 such that a value of α increases as the amount of noise increases, and generates the one or more pieces of first mixture data by mixing the filter process data with the super resolution process data in accordance with the mixture ratio α: (1−α).

5. The image processing device of claim 3 , wherein

the generating unit further includes

a second mixing unit configured to generate one or more pieces of second mixture data by mixing the one or more pieces of input pixel data with the one or more pieces of first mixture data in accordance with a predetermined mixture ratio, and

the generating unit outputs the one or more pieces of second mixture data instead of the one or more pieces of first mixture data.

6. An image processing method for generating an output image by performing an image process on an input image, the image processing method comprising the steps of:

generating super resolution process data by performing a super resolution process on a plurality of pieces of input pixel data that represent the input image;

classifying each of the plurality of pieces of input pixel data into one of high-frequency region data and non-high-frequency region data; and

generating, as at least part of output pixel data that represents the output image, one or more pieces of output pixel data corresponding to one or more pieces of input pixel data classified as the high-frequency region data, by using the super resolution process data of the one or more pieces of input pixel data as the one or more pieces of output pixel data based on an amount of noise included in the one or more pieces of input pixel data;

wherein when an amount of noise included in a piece of input pixel data classified as the high-frequency region data is smaller than or equal to a predetermined threshold, said generating outputs a piece of super resolution process data corresponding to the piece of input pixel data without change as a piece of output pixel data corresponding to the piece of input pixel data; and

wherein when an amount of noise included in the piece of input pixel data classified as the high-frequency region data is larger than the predetermined threshold, said generating outputs data other than the piece of super resolution process data corresponding to the piece of input pixel data as the piece of output pixel data corresponding to the piece of input pixel data.

7. An integrated circuit for use in an image processing device for generating an output image by performing an image process on an input image, the integrated circuit comprising:

a super resolution processing unit configured to generate super resolution process data by performing a super resolution process on a plurality of pieces of input pixel data that represent the input image;

a region determining unit configured to classify each of the plurality of pieces of input pixel data into one of high-frequency region data and non-high-frequency region data; and

a generating unit configured to generate, as at least part of output pixel data that represents the output image, one or more pieces of output pixel data corresponding to one or more pieces of input pixel data classified as the high-frequency region data, by using the super resolution process data of the one or more pieces of input pixel data as the one or more pieces of output pixel data based on an amount of noise included in the one or more pieces of input pixel data;

wherein when an amount of noise included in a piece of input pixel data classified as the high-frequency region data is smaller than or equal to a predetermined threshold, the generating unit outputs a piece of super resolution process data corresponding to the piece of input pixel data without change as a piece of output pixel data corresponding to the piece of input pixel data; and

wherein when an amount of noise included in the piece of input pixel data classified as the high-frequency region data is larger than the predetermined threshold, the generating unit outputs data other than the piece of super resolution process data corresponding to the piece of input pixel data as the piece of output pixel data corresponding to the piece of input pixel data.

8. A program that is recorded on a computer-readable non-transitory recording medium and executes an image process, the image process comprising the steps of:

generating super resolution process data by performing a super resolution process on a plurality of pieces of input pixel data that represent the input image;

classifying each of the plurality of pieces of input pixel data into one of high-frequency region data and non-high-frequency region data; and

generating, as at least part of output pixel data that represents the output image, one or more pieces of output pixel data corresponding to one or more pieces of input pixel data classified as the high-frequency region data, by using the super resolution process data of the one or more pieces of input pixel data as the one or more pieces of output pixel data based on an amount of noise included in the one or more pieces of input pixel data;

wherein when an amount of noise included in a piece of input pixel data classified as the high-frequency region data is smaller than or equal to a predetermined threshold, said generating outputs a piece of super resolution process data corresponding to the piece of input pixel data without change as a piece of output pixel data corresponding to the piece of input pixel data; and

wherein when an amount of noise included in the piece of input pixel data classified as the high-frequency region data is larger than the predetermined threshold, said generating outputs data other than the piece of super resolution process data corresponding to the piece of input pixel data as the piece of output pixel data corresponding to the piece of input pixel data.