Terminal and image processing method thereof

Patent metadata

Field	Value
Publication number	US-10855933-B2
Application number	US-201916259401-A
Country	US
Kind code	B2
Filing date	Jan 28, 2019
Priority date	Jan 31, 2018
Publication date	Dec 1, 2020
Grant date	Dec 1, 2020

Abstract

A terminal according to the present disclosure comprises an inputter for receiving input of a preview image converted from light reflected from surroundings of the terminal; a controller for identifying a type of an object included in the preview image and selecting two or more image filters using a recommended algorithm regarding an image filter to be applied to the preview image; and a display for displaying the preview image to which the selected two or more image filters are applied, wherein the controller may apply a first image filter of the two or more filters to a first image area of the preview image and apply a second image filter that is different from the first image filter to a second image area that is an area excluding the first image area of the preview image.

First claim

What is claimed is:

1. An image processing method being performed in a terminal, the method comprising:
receiving an input of a preview image converted from light reflected from surroundings of the terminal;
identifying a type of an object included in the preview image;
obtaining a recommended algorithm regarding an image filter to be applied to the preview image;
selecting two or more different image filters corresponding to the type of the object using the recommended algorithm; and
outputting the preview image to which the selected two or more image filters are applied,
wherein the selecting comprises:
obtaining a first learning model for recommending the image filter based on an artificial neural network; and
selecting a first image filter, which is predicted by using the first learning model to have a highest probability to be selected by a user of the terminal according to the type of the object included in the preview image, and a second image filter, which is predicted by using the first learning model to have a second highest probability to be selected by the user of the terminal according to the type of the object included in the preview image.

2. The image processing method according to claim 1 , further comprising:
obtaining a second learning model for identifying the type of the object included in the preview image, the second learning model being learned regarding correlation between image data of numerous photographs and types of objects included in the numerous photographs prior to identifying the type of the object included in the preview image.

3. The image processing method according to claim 1 ,
wherein the obtaining the first learning model comprises obtaining the first learning model that is learned regarding a correlation between types of objects included in numerous images and types of image filters selected in order to correct the numerous images.

4. The image processing method according to claim 1 ,
wherein the outputting comprises:
applying the first image filter to a first image area of the preview image; and
applying the second image filter to a second image area that is an area excluding the first image area of the preview image.

5. The image processing method according to claim 4 ,
wherein the outputting further comprises:
receiving a touch drag input that is input on a first display area of the terminal corresponding to the first image area or a second display area of the terminal corresponding to the second image area; and
changing an image filter applied to an image area corresponding to a display area to which the touch drag input is received to a third image filter that is different from the first image filter and the second image filter.

6. The image processing method according to claim 5 ,
wherein the third image filter is predicted by using the first learning model to have a third highest probability to be selected by the user of the terminal according to the type of the object included in the preview image.

7. The image processing method according to claim 5 ,
wherein the receiving the touch drag input comprises receiving the touch drag input that is input on the first display area or the second display area, the touch drag input being towards a first direction that is perpendicular to a direction from the first display area towards the second display area, or towards a second direction that is opposite to the first direction.

8. The image processing method according to claim 5 ,
wherein the outputting further comprises:
receiving a touch drag input that starts from an arbitrary point on a boundary of the first display area and the second display area;
adjusting a ratio of the first image area and a ratio of the second image area based on the touch drag input;
determining whether one of the ratio of the first image area and the ratio of the second image area becomes a reference ratio or above; and
applying the image filter applied to one of the first image area and the second image area having the reference ratio or above to an entire area of the preview image, based on a result of the determining.

9. The image processing method according to claim 8 ,
wherein the determining comprises determining whether one of a size of the first image area and a size of the second image area becomes a reference size or above.

10. The image processing method according to claim 8 ,
wherein the determining comprises determining whether a width of the first image area and a width of the second image area becomes a reference width or above.

11. The image processing method according to claim 8 , further comprising:
after the applying the image filter applied to one of the first image area and the second image area having the reference ratio or above to the entire area of the preview image, outputting a photograph icon; and
in response to receiving a touch input on the photograph icon, storing the preview image to which the image filter is applied, the image filter being that applied to the image area having the ratio that became the reference ratio or above.

12. The image processing method according to claim 8 ,
wherein the adjusting comprises, in response to receiving the touch drag input from the arbitrary point on the boundary of the first display area and the second display area in a direction from the second display area towards the first display area, increasing the ratio of the second image area in the preview image to be proportionate to a distance between a starting point of the touch drag input and an end point of the touch drag input.

13. The image processing method according to claim 8 ,
wherein the adjusting comprises, in response to receiving the touch drag input from the arbitrary point on the boundary of the first display area and the second display area in a direction from the first display area towards the second display area, increasing the ratio of the first image area in the preview image to be proportionate to a distance between a starting point of the touch drag input and an end point of the touch drag input.

14. A non-transitory computer-readable record medium having a program for executing the image processing method according to claim 1 in at least one processor.

15. A terminal comprising:
an inputter configured to receive an input of a preview image converted from light reflected from surroundings of the terminal;
a controller configured to identify a type of an object included in the preview image and select two or more image filters using a recommended algorithm regarding an image filter to be applied to the preview image; and
a display configured to display the preview image to which the selected two or more image filters are applied,
wherein the controller is further configured to:
obtain a learning model for recommending the image filter based on an artificial neural network; and
apply a first image filter, which is predicted by using the learning model to have a highest probability to be selected by a user of the terminal according to the type of the object included in the preview image, to a first image area of the preview image and apply a second image filter, which is predicted by using the learning model to have a second highest probability to be selected by the user of the terminal according to the type of the object included in the preview image, to a second image area that is an area excluding the first image area of the preview image.

16. The terminal according to claim 15 ,
wherein the display is further configured to receive a touch drag input that is input on a first display are

Assignees

• Hyperconnect Inc

Inventors

• Ahn Sangil
• Shin Beomjun
• Oh Minsik
• Choi Eunhee

Classifications

• H04N23/632

  for displaying or modifying preview images prior to image capturing, e.g. variety of image resolutions or capturing parameters · CPC title

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• H04N5/2621 Primary

  Cameras specially adapted for the electronic generation of special effects during image pickup, e.g. digital cameras, camcorders, video cameras having integrated special effects capability · CPC title

• G06T5/60 Primary

  using machine learning, e.g. neural networks · CPC title