Apparatus and method for image region detection of object based on seed regions and region growing

1. An image processing apparatus, comprising:

a memory configured to store a content; and

a processor configured to:

identify a first seed region and a second seed region from an image frame included in the content based on pixel value information,

identify a first object region and a second object region from the image frame based on the first seed region and the second seed region, respectively, and

obtain an output image frame by performing contrast adjustment for the first object region and the second object region, respectively.

2. The image processing apparatus as claimed in claim 1 , wherein the pixel value information includes at least one of color information, brightness information, variation information, standard deviation information, edge information or texture information.

3. The image processing apparatus as claimed in claim 1 , wherein the processor is further configured to:

identify a pixel block having a first pixel value from the image frame as the first seed region, and

identify a pixel block having a second pixel value different from the first pixel value from the image frame as the second seed region.

4. The image processing apparatus as claimed in claim 3 , wherein the processor is further configured to:

identify the second pixel value based on a predetermined threshold grayscale range.

5. The image processing apparatus as claimed in claim 1 , wherein the processor is further configured to:

identify a first expansion region and a second expansion region from the image frame based on the first seed region and the second seed region, respectively, and

identify the first object region and the second object region from the image frame based on the first expansion region and the second expansion region, respectively.

6. The image processing apparatus as claimed in claim 1 , wherein the processor is further configured to:

perform the contrast adjustment by generating a contrast curve for the first object region and the second object region, respectively.

7. The image processing apparatus as claimed in claim 1 , wherein the processor is further configured to:

perform the contrast adjustment based on a High Dynamic Range (HDR) function.

8. The image processing apparatus as claimed in claim 7 , wherein the HDR function is a function that performs tone mapping on the image frame based on metadata corresponding to the content, and

wherein the metadata includes at least one of color space information of the content, bit number information of the content, peak luminance information of the content, tone mapping information of the content, and mastering monitor information.

9. The image processing apparatus as claimed in claim 1 , wherein the processor is further configured to:

obtain the output image frame by inputting the image frame into a training network model, and

wherein the training network model is a model trained to perform the contrast adjustment for a plurality of object regions, respectively.

10. The image processing apparatus as claimed in claim 1 , further comprising:

a display including a liquid crystal display (LCD) panel using a back light;

wherein the processor is further configured to:

perform the contrast adjustment for the first object region and the second object region, respectively, by using a local dimming method.

11. An image processing method for an image processing apparatus storing a content, the image processing method comprising:

identifying a first seed region and a second seed region from an image frame included in the content based on pixel value information,

identifying a first object region and a second object region from the image frame based on the first seed region and the second seed region, respectively, and

obtaining an output image frame by performing contrast adjustment for the first object region and the second object region, respectively.

12. The image processing method as claimed in claim 11 , wherein the pixel value information includes at least one of color information, brightness information, variation information, standard deviation information, edge information or texture information.

13. The image processing method as claimed in claim 11 , wherein the identifying the first seed region and the second seed region comprises:

identifying a pixel block having a first pixel value from the image frame as the first seed region, and

identifying a pixel block having a second pixel value different from the first pixel value from the image frame as the second seed region.

14. The image processing method as claimed in claim 13 , wherein the identifying the pixel block having the second pixel value comprises:

identifying the second pixel value based on a predetermined threshold grayscale range.

15. The image processing method as claimed in claim 11 , wherein the identifying the first object region and the second object region comprises:

identifying a first expansion region and a second expansion region from the image frame based on the first seed region and the second seed region, respectively, and

identifying the first object region and the second object region from the image frame based on the first expansion region and the second expansion region, respectively.

16. The image processing method as claimed in claim 11 , wherein the obtaining the output image frame comprises:

performing the contrast adjustment by generating a contrast curve for the first object region and the second object region, respectively.

17. The image processing method as claimed in claim 11 , wherein the obtaining the output image frame comprises:

performing the contrast adjustment based on a High Dynamic Range (HDR) function.

18. The image processing method as claimed in claim 17 , wherein the HDR function is a function that performs tone mapping on the image frame based on metadata corresponding to the content, and

wherein the metadata includes at least one of color space information of the content, bit number information of the content, peak luminance information of the content, tone mapping information of the content, and mastering monitor information.

19. The image processing method as claimed in claim 11 , wherein the obtaining the output image frame comprises:

obtaining the output image frame by inputting the image frame into a training network model, and

wherein the training network model is a model trained to perform the contrast adjustment for a plurality of object regions, respectively.

20. The image processing method as claimed in claim 11 , wherein the image processing apparatus includes a display including a liquid crystal display (LCD) panel using a back light, and

wherein the obtaining the output image frame comprises:

performing the contrast adjustment for the first object region and the second object region, respectively, by using a local dimming method.