Still subtitle detection apparatus and image processing method therefor

A still subtitle detection apparatus configured to detect in pixel units or in a pixel unit a subtitle part included in an image in one video frame including a plurality of pixels, the apparatus comprising:

• a subtitle part determination module configured to determine within the one video frame, as the subtitle part, a pixel or a pixel group
• on a first luminance region side that neighbors a detection position of an image edge where a first luminance level changes, or
• on a second luminance region side that neighbors a detection position of an image edge where a second luminance level changes, the second luminance level being lower than the first luminance level;
• a still subtitle detection module configured to detect the subtitle part as a still subtitle by comparing two video frames and determining that the subtitle part remains unchanged in the two video frames, the two video frames being continuous along a time axis, wherein an input image signal including the subtitle part includes a previous frame and a current frame having a first frame rate;
• and
• a frame interpolation module configured to convert the input image signal into an output image signal having a second frame rate higher than the first frame rate, the frame interpolation module including
• a motion vector detection module configured to detect a motion vector of an input image from the previous frame and the current frame of the input image signal, and
• an interpolating frame insertion module configured to insert an interpolating frame having the second frame rate between the previous frame and the current frame, and the interpolating frame insertion module is configured to generate the interpolating frame based on a motion vector detection result by the motion vector detection module or a still subtitle detection result by the still subtitle detection module,
• wherein the subtitle part determination module includes a current frame edge detection module configured to detect a luminance level change in a luminance level distribution of a group of pixels on a horizontal line in the video frame, and to input a current frame edge detection result indicating the luminance level change to the current frame edge neighboring peak region detection module, and a current frame edge neighboring peak region detection module configured to detect in pixel units or in a pixel unit an edge of a character of the subtitle part in the current frame from the current frame edge detection result.

A still subtitle detection apparatus configured to detect in pixel units or in a pixel unit a subtitle part included in an image in one video frame including a plurality of pixels, the apparatus comprising:

• a subtitle part determination module configured to determine within the one video frame, as the subtitle part, a pixel or a pixel group
• on a first luminance region side that neighbors a detection position of an image edge where a first luminance level changes, or
• on a second luminance region side that neighbors a detection position of an image edge where a second luminance level changes, the second luminance level being lower than the first luminance level;
• and
• a still subtitle detection module configured to detect the subtitle part as a still subtitle by comparing two video frames and determining that the subtitle part remains unchanged in the two video frames, the two video frames being continuous along a time axis,
• wherein the subtitle part is formed of a first character type indicating that the pixels in a region of the first luminance level continue for a given width from an edge of a character with the first luminance level, or a second character type indicating that the pixels in a region of the second luminance level continue for a given width from an edge of a character with the second luminance level.

A still subtitle detection apparatus configured to detect in pixel units or in a pixel unit a subtitle part included in an image in one video frame including a plurality of pixels, the apparatus comprising:

• a subtitle part determination module configured to determine within the one video frame, as the subtitle part, a pixel or a pixel group
• on a first luminance region side that neighbors a detection position of an image edge where a first luminance level changes, or
• on a second luminance region side that neighbors a detection position of an image edge where a second luminance level changes, the second luminance level being lower than the first luminance level;
• and
• a still subtitle detection module configured to detect the subtitle part as a still subtitle by comparing two video frames and determining that the subtitle part remains unchanged in the two video frames, the two video frames being continuous along a time axis,
• wherein the subtitle part determination module comprises
• an edge detecting module configured to detect an edge of the subtitle part based on a difference between the luminance levels within the one video frame;
• and
• a determining module configured to determine a continuous pixels existing region in which same luminance level pixels continuously exist near the edge detected by the edge detecting module.