Image detection apparatus and operation method thereof

1. An image detection apparatus comprising:

a display;

a memory storing one or more instructions; and

one or more processors configured to execute the one or more instructions stored in the memory to:

detect, by using a neural network, a first additional information area in a first image that is output on the display;

obtain first style information of the first additional information area from the first additional information area;

obtain, from an external computing device, an artificial intelligence (AI) model that has learned to detect a new additional information area having new style information, based on the first style information being transmitted to the external computing device,

update the neural network by using the AI model that has learned to detect the new additional information area having the new style information; and

detect, in a second image that is output on the display, a second additional information area having second style information different from the first style information by using the updated AI model.

2. The image detection apparatus of claim 1 , further comprising a neural network processor configured to generate the second additional information area having the second style information based on the first style information, and obtain the AI model for outputting a new image by learning the second additional information area having the second style information.

3. The image detection apparatus of claim 1 , wherein any one or any combination of the one or more processors is further configured to execute the one or more instructions to obtain text information and color information from the first additional information area, and obtain the first style information from the text information and the color information.

4. The image detection apparatus of claim 3 , wherein the first style information comprises at least one of a location of the first additional information area in the first image, a background color of the first additional information area, a background texture of the first additional information area, a layout of texts included in the first additional information area, a text type, a text font, a text color, or a text texture.

5. The image detection apparatus of claim 3 , wherein any one or any combination of the one or more processors is further configured to execute the one or more instructions to recognize the first image based on the text information.

6. The image detection apparatus of claim 3 , wherein any one or any combination of the one or more processors is further configured to execute the one or more instructions to extract a text region from the first additional information area, recognize a text in the text region, and classify the recognized text into different classes to obtain the text information.

7. The image detection apparatus of claim 6 , wherein any one or any combination of the one or more processors is further configured to execute the one or more instructions to extract a feature map from the first additional information area and analyze the feature map to extract the text region.

8. The image detection apparatus of claim 6 , wherein any one or any combination of the one or more processors is further configured to execute the one or more instructions to classify the text into at least one class from among a channel name, a channel number, a title, a playback time, and other information.

9. The image detection apparatus of claim 6 , wherein any one or any combination of the one or more processors is further configured to execute the one or more instructions to recognize, in the text region, at least one of a number, a word, a language of the word, or a font in order to recognize the text.

10. An image detection method comprising:

detecting, by using a neural network, a first additional information area in a first image;

obtaining first style information of the first additional information area from the first additional information area; and

obtaining, from an external computing device, an artificial intelligence (AI) model that has learned to detect a new additional information area having new style information based on the first style information being transmitted to the external computing device,

updating the neural network by using the AI model that has learned to detect the new additional information area having the new style information, and

detecting, in a second image, a second additional information area having second style information different from the first style information, by using the updated AI model.

11. The image detection method of claim 10 , further comprising:

generating the second additional information area having the second style information based on the first style information; and

obtaining the AI model for outputting a new image by learning the second additional information area having the second style information.

12. The image detection method of claim 10 , wherein the obtaining of the first style information comprises:

obtaining text information and color information from the first additional information area; and

obtaining the style information from the text information and the color information.

13. The image detection method of claim 12 , wherein the style information comprises at least one of a location of the first additional information area in the first image, a layout of the first additional information area, a location of a text, a type of the text, a font of the text, a background color of the first additional information area, or a color of the text.

14. The image detection method of claim 12 , further comprising recognizing the first image based on the text information.

15. The image detection method of claim 12 , wherein the obtaining of the text information comprises:

extracting a text region from the first additional information area;

recognizing a text in the text region; and

classifying the recognized text into different classes.

16. A non-transitory computer-readable recording medium having recorded thereon a program for performing an image detection method, the image detection method comprising:

detecting, by using a neural network, a first additional information area in a first image;

obtaining first style information of the first additional information area from the first additional information area;

obtaining, from an external computing device, an artificial intelligence (AI) model that has learned to detect a new additional information area having new style information, based on the first style information being transmitted to the external computing device;

updating the neural network by using the AI model, and

detecting, in a second image, a second additional information area having second style information different from the first style information by using the updated AI model.