KR102014443B1 - Method and apparatus for determining content type of video content - Google Patents

Abstract

Receives a frame of video content, detects pixel component color characteristics of the received frame, and determines a content type indicating whether the received frame includes content for reproducing a scene of a predetermined genre according to the detected pixel component color characteristics. A method of determining a content type of video content is disclosed.

Description

Method and apparatus for determining content type of video content

The present invention relates to a method and apparatus for determining a content type of video content, and more particularly, to a method and apparatus for determining whether to include a field game for each frame of video content.

The video content may undergo processing such as brightness, contrast, or sharpening before being displayed on the display. In this case, the video content may be processed in consideration of the genre or type of video content.

There exists a method of detecting the type of video content by using the audio feature of the video content, but there is a problem that it cannot be used when the video and audio tracks are stored separately.

In addition, there is a method of detecting the type of video content for each segment, but there is a problem that it takes too long because the type of the video content must be detected in consideration of one or more frames included in the segment.

Accordingly, there is a need for a method of quickly detecting a type of video content frame by frame.

The present invention relates to a method for determining a type of video content for each frame of video content, and more particularly, to a method for determining whether a field game is included for each frame of video content.

In one embodiment of the present invention for solving the above problems, in the method of determining the content type of the video content, the step of receiving the frame of the video content detecting the color component characteristics of each pixel of the received frame And determining a content type indicating whether the received frame includes content for reproducing a scene of a predetermined genre according to the detected color component characteristic of each pixel.

In addition, when the received frame and the previous frame is the same scene, determining the content type of the received frame according to the determined content type of the previous frame.

In addition, detecting a color component characteristic of each pixel of the received frame may include detecting luminance and saturation of a plurality of pixels included in the received frame using the detected luminance and saturation and RGB channel values of each pixel. Detecting color component characteristics of each pixel by using the detected luminance value and detecting the degree of change of luminance of each pixel using the detected luminance value and the reception using the variation of luminance of each detected pixel and the color component characteristics of each pixel. Detecting a statistical analysis value of one frame.

In addition, detecting a statistical analysis value of the received frame may include detecting a statistical analysis value of a pixel included in the received frame, and statistics of a pixel of which the color component characteristic of each pixel among the received frames is green. Detecting the chemical analysis value.

In addition, detecting a statistical analysis value of the pixels included in the received frame may include detecting a ratio of pixels in which the color component characteristic of each pixel of the received frame is white. Detecting a ratio of pixels having high luminance and high saturation detecting a ratio of pixels having a color component characteristic of each pixel of the received frame being green and a ratio of pixels having a color component characteristic of each pixel of the received frame having skin color; Characterized in that it comprises the step of detecting.

In addition, detecting a statistical analysis value of a pixel whose color component characteristic of each pixel is green among the received frames may include detecting an average luminance value of a plurality of pixels of which the color component characteristic of each pixel among the received frames is green. Detecting an average saturation value of a plurality of pixels of which the color component characteristic of each pixel is green among the received frames. Detecting an average luminance change value of a plurality of pixels of which the color component characteristic of each pixel is green among the received frames, and detecting a histogram of the G channel of the plurality of pixels of which the color component characteristic of each pixel among the received frames is green; Characterized in that it comprises a step.

In addition, the determining of the content type of the received frame according to the analyzed feature may include determining a color component of each pixel of the received frame when the ratio of pixels having a green color characteristic of each pixel of the received frame is less than or equal to a reference value. If the average saturation value of the pixel whose characteristic is green is less than or equal to the reference value, the color saturation value of the pixel of which the color component of the received frame is green is less than or equal to the reference value, and the color component characteristic of each pixel of the received frame is green. When the average value of the B channel of the pixel is greater than or equal to the reference value, if the color saturation characteristic of the pixel of the received frame is less than or equal to the reference value, the average color saturation value or average luminance value of the pixel that is green is less than or equal to the reference value. The average saturation value of a pixel is a value between a first reference value and a second reference value, and the pixel-by-pixel color of the received frame. If the width of the graph is greater than or equal to the reference value in the histogram of the pixel having the minute characteristic, the average color saturation value of the pixel with the color component characteristic of the pixel of the received frame is greater than or equal to the reference value, and the color component characteristic of each pixel of the received frame When the width of the graph is less than or equal to the reference value in the histogram of the green pixel and the average color saturation value of the pixel whose color component characteristic of the pixel of the received frame is green is greater than or equal to the reference value, the color component characteristic of each pixel of the received frame is green. If the average change value of the luminance of the in-pixel is at least one of the reference value, characterized in that it comprises the step of determining that the content type of the received frame is not a field game.

In addition, the determining of the content type of the received frame according to the analyzed feature may include a ratio of pixels having a color component characteristic of each pixel of the received frame to be less than or equal to a reference value, and the color component of each received pixel of the received frame. The ratio of pixels having high luminance and high saturation is equal to or greater than a reference value, and the ratio of pixels having color component characteristic of each pixel of the received frame is white or that of pixels whose color component characteristic of pixel of the received frame is skin color is less than the reference value. And the ratio of the pixel in which the color component characteristic of each pixel of the received frame is green is greater than or equal to the reference value, and the ratio of the pixel in which the color component characteristic of each pixel of the received frame is high luminance and high saturation is less than or equal to the reference value, and the received frame The pixel component of the color component property of the pixel that is white or the pixel property of the pixel of the received frame. If the attribute is available for the at least one of not more than a ratio of the pixel color reference value, the received content type of the frame is characterized in that it comprises the step of determining to be a game field.

1 is a flowchart illustrating a method of determining a content type of video content according to an embodiment of the present invention.
2 is a flowchart illustrating a method of detecting color component characteristics of each pixel of video content according to an exemplary embodiment.
3 is a flowchart illustrating a method of detecting a statistical analysis value included in one frame of video content according to an embodiment of the present invention.
4 is a flowchart illustrating a method of determining a content type of video content according to another embodiment of the present invention. According to another embodiment of the present invention, the content type of the current frame may be determined according to the content type of the previous frame.
5 is a block diagram illustrating a structure of a content type determination device for determining a content type of video content according to an exemplary embodiment.
6 is a block diagram illustrating a structure of a content type determination device for determining a content type of video content according to another exemplary embodiment.
7 is a block diagram illustrating a structure of a color component characteristic detection unit for each pixel of the content type determination apparatus according to an exemplary embodiment.
8 is a block diagram illustrating a structure of a pixel classification unit of a color component characteristic detection unit for each pixel according to an exemplary embodiment.
9 is a block diagram illustrating a structure of a type detection unit of a content type determination device according to an embodiment of the present invention.
10 is a block diagram illustrating a structure of a scene change detection unit of a content type determination device according to another embodiment of the present invention.
11 is a block diagram illustrating a structure of a final type detection unit of a content type determination device according to another embodiment of the present invention.
12 is a block diagram illustrating a content type determination system according to an embodiment of the present invention.
FIG. 13 illustrates an area of a graph that may include field game episodes according to one embodiment of the invention.
14 is a graph illustrating a range of RGB channel data values that may be determined to be white pixels according to an embodiment of the present invention.
FIG. 15 is a graph illustrating a range of RGB channel data values in which a pixel may be determined to be a skin color pixel according to an exemplary embodiment.
16 is a graph illustrating a range of RGB channel data values that may be determined as being a yellow pixel according to an embodiment of the present invention.
FIG. 17 is a graph illustrating a range of RGB channel data values that may be determined to be a green pixel according to an embodiment of the present invention.
18 is a graph illustrating luminance values of green pixels according to an embodiment of the present invention.
19 and 20 illustrate graphs of images and images that do not correspond to field game episodes according to an embodiment of the present invention.
21 is a diagram for a video and a video not corresponding to a field game episode according to one embodiment of the present invention.
22 to 23 illustrate graphs of images corresponding to field game episodes and images according to an embodiment of the present invention.
FIG. 24 is a graph illustrating an image when an image determined to not be a field game episode is inserted between an image determined to not be a field game episode and an image corresponding to a field game episode according to an embodiment of the present invention. It is shown.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, in the following description and the accompanying drawings, detailed descriptions of well-known functions or configurations that may obscure the subject matter of the present invention will be omitted. In addition, it should be noted that like elements are denoted by the same reference numerals as much as possible throughout the drawings.

The terms or words used in the specification and claims described below should not be construed as being limited to the ordinary or dictionary meanings, and the inventors are properly defined as terms for explaining their own invention in the best way. It should be interpreted as meaning and concept corresponding to the technical idea of the present invention based on the principle that it can. Therefore, the embodiments described in the present specification and the configuration shown in the drawings are only the most preferred embodiments of the present invention, and do not represent all of the technical ideas of the present invention, and various alternatives may be substituted at the time of the present application. It should be understood that there may be equivalents and variations.

The present invention relates to a method of determining a content type of video content, and more particularly, to a method of detecting whether a field game episode of video content is included.

Content type detection according to the present invention can detect the type of content based on the fact that the ratio of the green pixels of the content including the field game episode is relatively high, compared to other content not including the field game episode. .

In general, content that includes field game episodes is relatively high when compared to content that does not contain field game episodes, the saturation of the green pixels or the percentage of green pixels included in the frame of the content.

However, even if the content includes the field game episode, there may be a case where the saturation of the green pixel is lower than the content not including the field game episode. In other words, the high saturation of the green pixels cannot be absolutely determined to include field game episodes. In this case, a human may easily determine that the content including the field game episode having a relatively low green pixel saturation includes the field game episode even though the green pixel of the content has low saturation. The blue component may play a key role in the determination. This is because blue is almost the same pixel as green. Therefore, the average value of the B channel in the green area of the green pixel can be used for content type detection to detect whether a field game episode of video content is included. The area of the green pixel may mean pixels belonging to a range in which the corresponding pixel value can be recognized as green by a person.

In addition, the average luminance value of the frame of the content may also be used when determining whether the field game episode is included, since sports events may be held mainly in a bright place.

Content including field game episodes has a relatively narrow histograms of green channel in the green region of each pixel. This is because a uniformly colored lawn can be used in a field game. Therefore, when determining whether the field game episode is included, the distribution of the G channel of each pixel of the frame may be used.

Images containing field game episodes tend to have relatively high gradient values of luminance of each pixel in the green pixel region. This is because in the field game, the boundary color of the lawn is clear because the uniform color of the player's uniform line is conspicuous. Therefore, when determining whether the field game episode is included, the degree of change in the luminance value of each pixel of the frame may be used.

FIG. 13 illustrates an area of a graph that may include field game episodes according to an embodiment of the present invention.

Referring to FIG. 13, when the saturation of the green pixels included in the frame of the content is low or the relative number of the green pixels is low, the frame of the content may not include the field game episode. The green pixel may include a pixel belonging to an area in which the value of the RGB channel data of the pixel may be recognized as green by a person.

In the present invention, one frame can be determined as a field game episode can be largely classified into three types, in the case of a far view, in the case of a close-up view, When the frame is determined to be a field game episode, it may be classified as a case where no scene change occurs.

If the number of green pixels is relatively large, the number of bright and saturated pixels is small, and the number of white or skin color pixels is greater than or less than zero, that is, the ratio of green pixels in the frame is above the reference value, When the ratio of pixels is equal to or less than the reference value and the ratio of white pixels or the ratio of skin color pixels is equal to or smaller than the reference value, the content type of the frame may be determined to be a field game episode of a distant view. In this case, the green pixel may correspond to the color of the lawn of the field game, the pixel having high luminance and saturation, and the white pixel correspond to the uniform color of the player. In addition, the skin color pixel may correspond to the skin color of the player.

If the number of green pixels is relatively small, the number of bright and saturated pixels is large, the number of white pixels is small, and the number of skin tone pixels is greater or less than zero, that is, the ratio of the green pixels is below the reference value. When the ratio of pixels having high luminance and high saturation is greater than or equal to the reference value and the ratio of white pixels or skin color pixels is less than or equal to the reference value, the content type of the frame may be determined to be a field game episode of the close-up view. In the case of the close-up view, since the player may be displayed largely around the player, it may be determined as described above.

Depending on whether or not a scene change has been made, the content type of a frame may be determined to be a field game episode. If a previous frame is classified as a field game episode and a scene change has not occurred, the current frame is determined to include a field game episode. Can be.

Meanwhile, according to the present invention, seven types of cases may be determined that the content type of the frame is not a field game episode.

When the ratio of the green pixels included in the frame is low, when the average luminance value of the green pixel is low or when the average saturation value of the green pixel is very low, it may be determined that the field game episode is not mentioned as mentioned above. The case where the ratio of the green pixels included in the frame is low or the average saturation value of the green pixels is very low may be included in an area other than the area that may be determined not to be a field game episode in the graph of FIG. 13.

If the average saturation value of the green pixel included in the frame is low and the average value of the B channel of the green pixel is high, it may be determined that the content type of the frame is not a field game episode.

When the average saturation value of the green pixels included in the frame is medium and the graph width is wide in the histogram of the G channel value of the green pixel, it may be determined that the content type of the frame is not a field game episode.

When the average saturation value of the green pixels included in the frame is very high and the graph width is very narrow in the histogram of the G channel value of the green pixel, it may be determined that the content type of the frame is not a field game episode.

When the average saturation value of the green pixels included in the frame is high and the average value of the variance of the luminance values of the green pixels is very low, it may be determined that the content type of the frame is not a field game episode.

It does not belong to the case where it may be judged that it is not the seven field game episodes, and when it is judged that it is a field game episode, ie, when it is a far view, when it is a close-up view. When it is determined that the previous frame is a field game episode, when the scene change does not occur in one of the cases, the content type of the frame may be determined as the field game episode.

If it is determined that the seven field game episodes are not the seven cases and the three field game episodes are not included, the content type of the frame may be determined not to be the field game episodes. .

In case of belonging to the case where it may be determined that the seven field game episodes are not the above, it may be determined that the content type of the frame is not the field game episode, even though the case may belong to the case where it may be determined as the field game episode.

1 is a flowchart illustrating a method of determining a content type of video content according to an embodiment of the present invention.

Referring to FIG. 1, when the content type determination apparatus receives a frame of video content from the outside (S101), the color component characteristic of each pixel included in the received frame may be detected (S103). The color component characteristics of each pixel may include saturation, luminance, value of gradient in luminance channel, white, skin color, yellow, green, high luminance and bright and saturated.

The content type determination device may determine the content type according to the detected color component characteristic of each pixel (S105). In this case, the content type may be determined for each frame and may be determined depending on whether the content includes a field game or not.

2 is a flowchart illustrating a method of detecting color component characteristics of each pixel of video content according to an exemplary embodiment.

Referring to FIG. 2, the content type determination apparatus detects luminance and saturation of each pixel included in one frame of video content (S202). The content type determination apparatus detects color component characteristics for each pixel by using the detected luminance and saturation values of each pixel and the RGB channel value of each pixel (S203), and uses each pixel by using the detected luminance value of each pixel. The degree of change in luminance can be detected (S204). The color component characteristics of each pixel may be detected according to the luminance, saturation, and RGB channel value of each pixel. In step S203, the pixel is a color component characteristic of each pixel, and is a white, skin color, yellow, green, high luminance, and high saturation pixel. (bright and saturated pixel) can be determined whether or not.

In addition, the content type determination apparatus detects statistical analysis values of the pixels included in one frame of the video content by using the color component characteristic of each pixel detected in step S203 and the degree of change in luminance of each pixel detected in step S204. It may be (S205). The statistical analysis value of step S205 means a value obtained by analyzing the characteristics of the entire frame using the characteristics of each pixel, the statistical analysis value is the number of green pixels, the number of skin color pixels, the average luminance value, And the number of pixels that are high brightness and high saturation and the number of white pixels. According to an embodiment of the present disclosure, the content type of one frame of video content may be determined according to a statistical analysis value of one frame detected in step S205.

3 is a flowchart illustrating a method of detecting a statistical analysis value included in one frame of video content according to an embodiment of the present invention.

Referring to FIG. 3, the apparatus for determining a content type detects a statistical analysis value of pixels included in one frame (S305) and statistical analysis value of pixels belonging to green among pixels included in one frame. By detecting (S306), one or more statistical analysis values included in one frame of video content may be detected. The statistical analysis on green pixels is the average of the gradient of the luminance value of the green pixel, the average saturation value of the green pixel, the average brightness value of the green pixel, the average value of the B channel value of the green pixel, and the luminance of the green pixels. It can include the relative width of the histogram.

4 is a flowchart illustrating a method of determining a content type of video content according to another embodiment of the present invention. According to another embodiment of the present invention, the content type of the current frame may be determined according to the content type of the previous frame.

Referring to FIG. 4, when the content type determination apparatus receives a frame of video content from the outside (S401), the color component characteristic of the pixels included in the received frame is detected, and the detected color component characteristic of the current frame is determined. The content type may be detected (S403). When the current frame is the same scene as the previous frame (S405), the content type determination apparatus may determine the content type of the received frame according to the content type of the previous frame (S409). In contrast, when the current frame is not the same as the previous frame (S405), that is, when the scene is changed, the content type of the current frame may be finally determined according to the content type detected in step S403 (S407).

5 is a block diagram illustrating a structure of a content type determination device for determining a content type of video content according to an exemplary embodiment.

Referring to FIG. 5, the content type determination apparatus 500 according to an embodiment of the present invention may include a frame buffer 510, a color component characteristic detection unit 520 for each pixel, and a type detection unit 530.

The frame buffer 510 may receive the video content from the outside and transmit the frame to the color component characteristic detector 520 for each pixel.

The pixel component detection unit 520 for each pixel may receive the video content from the frame buffer 510 by one frame and detect color component characteristics of each pixel included in the received frame.

The type detector 530 may determine the content type according to the color component characteristic of each pixel detected by the color component characteristic detection unit 520 of each pixel. In this case, the content type may be determined for each frame and may be determined depending on whether the content includes a field game or not. A method of determining the content type by the type detector 530 will be described later in detail with reference to FIG. 9.

6 is a block diagram illustrating a structure of a content type determination device for determining a content type of video content according to another exemplary embodiment.

According to another embodiment of the present invention, the content type of the current frame may be determined according to the content type of the previous frame. The content type determination apparatus 600 of FIG. 6 may correspond to the content type determination apparatus 500 of FIG. 5.

Referring to FIG. 6, the content type determination apparatus 600 according to another exemplary embodiment of the present disclosure may include a frame buffer 610, a color component characteristic detector 620 for each pixel, a type detector 630, and a scene change detector 640. And a final type detector 650. The frame buffer 610, the color component characteristic detector 620, and the type detector 630 of the pixel correspond to the frame buffer 510, the color component characteristic detector 520, and the type detector 530 of FIG. 5. Duplicate contents will be omitted.

The scene change detection unit 640 outputs a False value when the current frame is the same scene as the previous frame, and outputs a True value when the current frame is not the same as the previous frame. By doing so, a value of whether or not to change the scene can be output.

The final type detector 650 may detect the content type of the current frame according to the output value of the scene change detector 640. A value that may be output from the scene change detection unit 640 may be a True or False value, and when True is changed in the scene, False may be a value output when the scene is not changed. In this case, when the output value of the scene change detection unit 640 is True, the final type detector 650 outputs the content type value detected for the current frame. If the output value of the scene change detection unit 640 is False, the last type detection unit 650 detects the previous frame. The content type value can be displayed.

7 is a block diagram illustrating a structure of a color component characteristic detection unit for each pixel of the content type determination apparatus according to an exemplary embodiment.

The color component characteristic detector 720 of FIG. 7 may correspond to the color component characteristic detectors 520 and 620 of FIGS. 5 and 6.

Referring to FIG. 7, the color component characteristic detector 720 for each pixel according to an exemplary embodiment of the present disclosure may include a saturation detector 721, a luminance detector 722, a pixel classifier 723, and a luminance gradient detector 724. And a statistical analysis unit 725.

The saturation detector 721 may detect saturation of at least one or more pixels included in one frame received by the color component characteristic detector 720 for each pixel. The saturation detector 721 may detect the saturation of each pixel using the RGB channel data R, G, and B of each pixel. In this case, the detectable saturation value S may be detected according to Equation 1 below.

In Equation 1, M0 may mean the minimum channel value among the RGB channel data of the pixel, and M1 may mean the maximum channel value among the RGB channel data of the pixel.

The luminance detector 722 may detect luminance of at least one or more pixels included in one frame received by the color component characteristic detector 720 for each pixel. The luminance detector 722 may detect the luminance of each pixel by using the RGB channel data R, G, and B of each pixel. In this case, the luminance value Y of each pixel may be detected according to Equation 2 below.

The pixel classifier 723 may classify at least one or more pixels included in one frame received by the color component characteristic detector 720 for each pixel according to their characteristics. The pixel classifier 723 may determine the RGB channel data R, G, and B of each pixel, and the chroma (S) and luminance (Y) values of each pixel detected by the chroma detector 721 and the luminance detector 722. Can be used to classify pixels. For example, it may be determined whether the pixels belong to white, high brightness and high saturation pixels, skin color, yellow, green. How the pixel classifier 723 classifies pixels according to their characteristics will be described in detail later with reference to FIG. 8.

8 is a block diagram illustrating a structure of a pixel classification unit of a color component characteristic detection unit for each pixel according to an exemplary embodiment.

The pixel classifier 800 of FIG. 8 may correspond to the pixel classifier 723 of FIG. 7.

The pixel classifier 800 may include a white pixel detector 810, a high luminance and high chroma pixel detector 820, a skin color pixel detector 830, a yellow pixel detector 840, and a green pixel detector 850.

The white pixel detector 810 may determine whether the pixel value corresponds to a value that can be recognized as white by a human by using the RGB channel data value of the pixel. In this case, the white pixel detector 810 may determine whether the pixel is a white pixel by using Equation 3 below. According to the determination, the white pixel detector 810 may output a True or False value.

는 교집합을 의미하는 것으로,

의 조건과

의 조건을 모두 만족하는 픽셀이 흰색 픽셀인 것으로 결정될 수 있고, 출력되는 값 W 는 true일 수 있다. 반면, 흰색 픽셀이 아닌 것으로 결정되는 경우, W는 false 값으로 출력될 수 있다.In equation (3)

Means the intersection,

Conditions and

The pixel satisfying all of the conditions may be determined to be a white pixel, and the output value W may be true. On the other hand, if it is determined that it is not a white pixel, W may be output as a false value.

14 is a graph illustrating a range of RGB channel data values that may be determined to be white pixels according to an embodiment of the present invention.

The high luminance and high saturation pixel detector 820 may determine whether the pixel value corresponds to a value that can be recognized as a high luminance and high saturation pixel by a human using the RGB channel data value of the pixel. In this case, the high luminance and high saturation pixel detection unit 820 may determine whether the pixel corresponds to the high luminance and high saturation pixel by using Equation 4 below. According to the determination, the high luminance and high saturation pixel detector 820 may output a True or False value.

의 조건과

의 조건을 모두 만족하는 픽셀이 고휘도 및 고채도 픽셀인 것으로 결정될 수 있고, 출력되는 값

는 true 값일 수 있다. 반면, 고휘도 및 고채도 픽셀이 아닌 것으로 결정되는 경우,

는 false 값으로 출력될 수 있다.In equation (4)

Conditions and

The pixel that satisfies all of the conditions may be determined to be a high luminance and high chroma pixel, and the output value

May be a true value. On the other hand, if it is determined that it is not high brightness and high saturation pixels,

May be output as a false value.

The skin color pixel detector 830 may determine whether the pixel value corresponds to a value that can be recognized as a skin color by a human using the RGB channel data value of the pixel. In this case, the skin color pixel detector 830 may determine whether the pixel corresponds to the skin color pixel by using Equation 5 below. According to the determination, the skin color pixel detector 830 may output a True or False value.

,

,

,

,

의 조건을 모두 만족하는 픽셀이 피부색 픽셀인 것으로 결정될 수 있고, 출력되는 값

는 true 값일 수 있다. 반면, 피부색 픽셀이 아닌 것으로 결정되는 경우,

는 false 값으로 출력될 수 있다.In equation (5)

,

,

,

,

The pixel satisfying all of the conditions may be determined to be a skin color pixel, and the output value

May be a true value. On the other hand, if it is determined that it is not a skin pixel,

May be output as a false value.

FIG. 15 is a graph illustrating a range of RGB channel data values in which a pixel may be determined to be a skin color pixel according to an exemplary embodiment.

The yellow pixel detector 840 may determine whether the pixel value corresponds to a value that can be recognized as yellow by a human using the RGB channel data value of the pixel. In this case, the yellow pixel detector 840 may determine whether the pixel corresponds to the yellow pixel by using Equation 6 below. According to the determination, the yellow pixel detector 840 may output a True or False value.

는 픽셀의 RGB 채널 데이터 값의 G 채널 값 및 R 채널 값 중 더 큰 값,

은 G 채널 값 및 R 채널 값 중 더 작은 값을 의미한다.In equation (6)

Is the greater of the G and R channel values of the pixel's RGB channel data values,

Is the smaller of the G channel value and the R channel value.

,

,

,

,

의 조건을 모두 만족하는 픽셀이 노란색 픽셀인 것으로 결정될 수 있고, 출력되는 값

는 true 값일 수 있다. 반면, 노란색 픽셀이 아닌 것으로 결정되는 경우,

는 false 값으로 출력될 수 있다.In Equation 6,

,

,

,

,

The pixel that satisfies all of the conditions may be determined to be a yellow pixel, and the output value

May be a true value. On the other hand, if it is determined not to be a yellow pixel,

May be output as a false value.

16 is a graph illustrating a range of RGB channel data values that may be determined as being a yellow pixel according to an embodiment of the present invention.

The green pixel detector 850 determines whether the pixel value corresponds to a value that can be recognized as green by a human using RGB channel data values of pixels among the pixels determined to be yellow by the yellow pixel detector 840. can do. In this case, the green pixel detector 840 may determine whether the pixel corresponds to the green pixel by using Equation 7 below. According to the determination, the green pixel detector 850 may output a True or False value.

는 픽셀의 RGB 채널 데이터 값의 R 채널 값 및 B 채널 값 중 더 큰 값을 의미하며,

는 합집합을 의미한다.In equation (7)

Means the larger of the R channel value and B channel value of the RGB channel data value of the pixel.

Means union.

,

,

,

,

=1의 조건을 모두 만족하는 픽셀이 녹색 픽셀인 것으로 결정될 수 있고, 출력되는 값

는 true 일 수 있다. 반면, 녹색 픽셀이 아닌 것으로 결정되는 경우,

은 false 값으로 출력될 수 있다.In equation (7)

,

,

,

,

The pixel that satisfies all the conditions of = 1 may be determined to be a green pixel, and the output value

May be true. On the other hand, if it is determined not to be a green pixel,

May be output as a false value.

The multiplexer 860 may integrate and output the outputs of all the pixels of the white pixel detector 810, the high luminance and high chroma pixel detector 820, and the skin color pixel detector 830.

FIG. 17 is a graph illustrating a range of RGB channel data values that may be determined to be a green pixel according to an embodiment of the present invention.

)는 커널

를 이용하여 각 픽셀의 휘도값 Y를 필터링함으로써 검출될 수 있다. 7, the luminance gradient detector 724 of FIG. 7 uses the luminance Y value of each pixel detected by the luminance detector 722 to calculate the gradient of the luminance value of each pixel. Can be detected. At this time, the degree of change in the luminance value of each pixel (

) The kernel

It can be detected by filtering the luminance value Y of each pixel using.

)가 될 수 있다. In detail, the average value of each value calculated using the kernel in the pixels of the array of 1 X 7 is the degree of change in the luminance value of each pixel (

Can be

로 결정될 수 있다.For example, when the luminance value of the fourth pixel p is Y1 and the luminance value of the seventh pixel is Y2 in the pixels of the array of 1 × 7, the change in the luminance value of the pixel p detected using the kernel. Turning

Can be determined.

), 피부색 픽셀의 수(

), 평균 휘도값(

), 녹색 픽셀의 휘도 값의 변화도의 평균(

), 고휘도 및 고채도인 픽셀의 수(

), 녹색 픽셀의 평균 채도값(

), 흰색 픽셀의 수(

), 녹색 픽셀의 평균 밝기값(

), 녹색 픽셀의 B 채널 값의 평균값(

) 및 녹색 픽셀들에 대한 휘도 히스토그램의 상대적인 너비값(

)으로 분류될 수 있다. 그리고, 통계학적 분석부(725)의 출력값은

으로 나타낼 수 있다.

내지

은 아래 수학식 8을 이용하여 검출될 수 있다. 수학식8에서 w 는 프레임의 가로길이, h는 프레임의 세로길이, i, j는 픽셀 좌표값을 의미한다.The statistical analyzer 725 may analyze the characteristics of the frame according to the color component characteristics of the pixels included in one frame received by the pixel component characteristic detector 720 for each pixel. The statistical analyzer 725 uses the classification characteristics of each pixel detected by the pixel classifier 723 and the luminance change of the luminance value of each pixel detected by the luminance detector 724. Can be analyzed. In this case, the characteristics of the frame may be classified into ten types, and the ten characteristics are the number of green pixels (

), The number of skin color pixels (

), Average luminance value (

), The average of the gradients of the luminance values of the green pixels (

), The number of pixels with high brightness and high saturation (

), The average saturation of the green pixels (

), The number of white pixels (

), The average brightness of the green pixels (

), The average value of the B channel values of the green pixels (

) And the width value of the luminance histogram relative to the green pixels (

Can be classified as). And, the output value of the statistical analysis unit 725

It can be represented as

To

May be detected using Equation 8 below. In Equation 8, w denotes the horizontal length of the frame, h denotes the vertical length of the frame, and i and j denote pixel coordinate values.

에서

은 녹색 픽셀들의 휘도 값에 대한 히스토그램이며, D는 상기 히스토그램 그래프의 너비, 즉, 최소값과 최대값의 차이이다.Equation 8

in

Is the histogram for the luminance values of the green pixels, and D is the width of the histogram graph, i.e. the difference between the minimum and maximum values.

18A is a graph illustrating luminance values of green pixels according to an embodiment of the present invention.

로 정의할 수 있다.The graph of FIG. 18A shows the luminance value on the horizontal axis and the number of green pixels on the vertical axis. At this time, a value obtained by subtracting the minimum value from the maximum luminance value among the green pixels is set to D, and the average value of the luminance values of the green pixels is set to D.

Can be defined as

18B illustrates an example of luminance graphs of green pixels of an image corresponding to a field game episode according to an embodiment of the present invention, and FIG. 18C illustrates an image not corresponding to a field game episode according to an embodiment of the present invention. This is an example of a luminance graph of green pixels.

Although the D value, which is the width value of the histogram graph of FIG. 18B, is small, the D value of FIG. 18C is wide, and the difference is that the luminance value of the green pixel is wide. Therefore, the image including the field game episode is mainly used to represent the lawn in the image, it can be confirmed that the luminance value is concentrated in a narrow range.

는

로 정의될 수 있다. 즉,

값이 true이면 1,

값이 false 이면 0을 의미할 수 있다.Also,

Is

It can be defined as. In other words,

1 if the value is true,

If the value is false, it can mean 0.

9 is a block diagram illustrating a structure of a type detection unit of a content type determination device according to an embodiment of the present invention. The type detector 900 of FIG. 9 may correspond to the type detectors 530 and 630 of FIGS. 5 and 6.

The type detector 900 may include a type determiner A to type determiner M 901 to 912 and 920. The type determiner M 920 may determine the content type of the frame using the True or False values output from the type determiners A to K 901 to 912.

)를 이용하여 컨텐츠의 타입 결정에 필요한 인자인

를 검출할 수 있다. 이때 타입 결정부 A(901)는 이하 수학식 9에 따라

를 검출할 수 있다.The type determining unit A 901 determines the average saturation value F6 of the green pixels and the number of green pixels (

) To determine the type of content

Can be detected. At this time, the type determination unit A 901 according to Equation 9 below.

Can be detected.

는

,

,

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.In this case, an integer of 0 to 4 may be inserted into i and j,

Is

,

,

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

의 값을 출력할 수 있으며, 각 출력값은 True 또는 False 값일 수 있다.Type determination unit A (901)

You can output the value of, and each output value can be either True or False.

)을 이용하여 컨텐츠의 타입 결정에 필요한 인자인

을 검출할 수 있다. 이때 타입 결정부 B(902)는 이하 수학식 10을 이용할 수 있다.

은

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determining unit B 902 has an average luminance value (

) To determine the type of content

Can be detected. In this case, the type determiner B 902 may use Equation 10 below.

silver

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

은 True 또는 False 값일 수 있다.Which is an output value of the type determination unit B

Can be a True or False value.

)를 이용하여 컨텐츠의 타입 결정에 필요한 인자인

를 검출할 수 있다. 이때 타입 결정부 C(903)는 이하 수학식 11를 이용할 수 있다.

는

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determiner C 903 indicates the number of skin color pixels (

) To determine the type of content

Can be detected. In this case, the type determiner C 903 may use Equation 11 below.

Is

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

은 True 또는 False 값일 수 있다.The output value of the type determination unit C (903)

Can be a True or False value.

)을 이용하여 컨텐츠의 타입 결정에 필요한 인자인

를 검출할 수 있다. 이때 타입 결정부 D(904)는 이하 수학식 12를 이용할 수 있다.

는

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determining unit D 904 is an average of the degree of change of the luminance value of the green pixel (

) To determine the type of content

Can be detected. In this case, the type determiner D 904 may use Equation 12 below.

Is

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

은 True 또는 False 값일 수 있다.Which is an output value of the type determination unit D

Can be a True or False value.

)를 이용하여 컨텐츠의 타입 결정에 필요한 인자인

를 검출할 수 있다. 이때 타입 결정부 E(905)는 이하 수학식 13을 이용할 수 있다.

는

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determiner E 905 indicates the number of white pixels (

) To determine the type of content

Can be detected. In this case, the type determiner E 905 may use Equation 13 below.

Is

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

은 True 또는 False 값일 수 있다.That is the output value of the type determination unit E

Can be a True or False value.

) 및 고휘도 및 고채도인 픽셀의 수(

)를 이용하여 컨텐츠의 타입 결정에 필요한 인자인

를 검출할 수 있다. 이때 타입 결정부 K(911)는 이하 수학식 14를 이용할 수 있다.The type determiner K 911 determines the number of skin color pixels (

) And the number of pixels with high brightness and high saturation (

) To determine the type of content

Can be detected. In this case, the type determiner K 911 may use Equation 14 below.

는

,

,

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.In this case, an integer of 1 or 2 may be inserted into i and j,

Is

,

,

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

의 값을 출력할 수 있으며, 각 출력값은 True 또는 False 값일 수 있다.Type determination unit K (911)

You can output the value of, and each output value can be either True or False.

) 및 고휘도 및 고채도인 픽셀의 수(

)를 이용하여 컨텐츠의 타입 결정에 필요한 인자인

을 검출할 수 있다. 이때 타입 결정부 E(905)는 이하 수학식 15를 이용할 수 있다.

및

는 미리 결정된상수로서 임의로 결정될 수 있다.The type determining unit L912 has an average luminance value (

) And the number of pixels with high brightness and high saturation (

) To determine the type of content

Can be detected. In this case, the type determiner E 905 may use Equation 15 below.

And

May be arbitrarily determined as a predetermined constant.

은 True 또는 False 값일 수 있다.Which is an output value of the type determination unit L 912

Can be a True or False value.

)를 이용하여 컨텐츠의 타입 결정에 필요한 인자인

를 검출할 수 있다. 이때 타입 결정부 F(906)는 이하 수학식 16을 이용할 수 있다.

는

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determining unit F 906 is a number of pixels having high brightness and high saturation (

) To determine the type of content

Can be detected. In this case, the type determiner F 906 may use Equation 16 below.

Is

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

은 True 또는 False 값일 수 있다.The output value of the type determination unit F (906)

Can be a True or False value.

)을 이용하여 컨텐츠의 타입 결정에 필요한 인자인

을 검출할 수 있다. 이때 타입 결정부 G(907)는 이하 수학식 17을 이용할 수 있다.

는

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determiner G 907 denotes an average brightness value of the green pixel (

) To determine the type of content

Can be detected. In this case, the type determiner G 907 may use Equation 17 below.

Is

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

은 True 또는 False 값일 수 있다.Which is an output value of the type determination unit G (907)

Can be a True or False value.

)을 이용하여 컨텐츠의 타입 결정에 필요한 인자인

을 검출할 수 있다. 이때 타입 결정부 H(908)는 이하 수학식 18을 이용할 수 있다.

은

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determiner H 908 stores an average brightness value of the green pixel (

) To determine the type of content

Can be detected. In this case, the type determiner H 908 may use Equation 18 below.

silver

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

는 True 또는 False 값일 수 있다.Which is an output value of the type determination unit H 908

Can be a True or False value.

)을 이용하여 컨텐츠의 타입 결정에 필요한 인자인

을 검출할 수 있다. 이때 타입 결정부 I(909)는 이하 수학식 19를 이용할 수 있다.

은

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determiner I 909 stores the average value of the B channel values of the green pixels (

) To determine the type of content

Can be detected. In this case, the type determiner I 909 may use Equation 19 below.

silver

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

는 True 또는 False 값일 수 있다.Which is an output value of the type determination unit I 909

Can be a True or False value.

)을 이용하여 컨텐츠의 타입 결정에 필요한 인자인

을 검출할 수 있다. 이때 타입 결정부 J(910)는 이하 수학식 20을 이용할 수 있다.

은

의 조건을 만족하는 미리 결정된 상수로서 임의로 결정될 수 있다.The type determiner J 910 is a width value of the luminance histogram for the green pixels (

) To determine the type of content

Can be detected. In this case, the type determination unit J 910 may use Equation 20 below.

silver

It may be arbitrarily determined as a predetermined constant that satisfies the condition of.

는 True 또는 False 값일 수 있다.Which is an output value of the type determination unit J

Can be a True or False value.

,

,

,

,

,

,

,

,

,

,

및

을 이용하여 프레임의 컨텐츠 타입이 필드 게임인지 여부(R)를 검출할 수 있다. 이때 타입 결정부 M(920)는 이하 수학식 21을 이용할 수 있다. The type determiner M 920 is an output value of the type determiner A to type determiner J 901 to 910.

,

,

,

,

,

,

,

,

,

,

And

R may be detected whether the content type of the frame is a field game. In this case, the type determination unit M 920 may use Equation 21 below.

R, which is an output value of the type determination unit M 920, may be a True or False value. That is, when the R value is True, the frame of the content may be determined to be a field game, and when the R value is False, the frame of the content may be determined not to be a field game.

10 is a block diagram illustrating a structure of a scene change detection unit of a content type determination device according to another embodiment of the present invention. The scene change detector 1000 of FIG. 10 may correspond to the scene change detector 640 of FIG. 6.

The scene change detection unit 100 may include a clustering block 1010, delay lines 1020 and 1030, a minimum value extraction module 1040, a maximum value extraction module 1050, and an amplification module. ) 1060, a subtractor module 1070, and a determination module 1080.

일 때, 클러스터 센터들(cluster centers),

는 수학식 22와 같이 나타낼 수 있다.The clustering module 1010 clusters one or more pixels according to the RGB channel data values of each pixel and clusters them, and uses an error that may be used to determine whether to change the scene using a cluster center among the clustered pixels. cumulative error) can be detected. Number of clusters

Cluster centers,

May be expressed as in Equation 22.

,

,

는 클러스터 1의 클러스터 센터인 픽셀의 RGB 채널 데이터 값일 수 있다. 클러스터 센터는 하나의 클러스터에 속하는 픽셀 중 하나일 수 있다.In Equation 22

,

,

May be an RGB channel data value of a pixel that is the cluster center of cluster 1. FIG. The cluster center may be one of the pixels belonging to one cluster.

The clustering module 1010 may detect an error (cumulative error, E) that may be used to determine whether to change the scene using the cluster center of each pixel, as shown in Equation 23 below.

)은 수학식24와 같이 나타낼 수 있다. 클러스터화 모듈(1010)은 갱신된 클러스터 센터들을 씬 변경 여부 판단에 이용될 수 있는 에러(cumulative error, E)를 구하는 데에 이용할 수 있다. 즉, 클러스터화 모듈(101)은 수학식 23에 따라 다음 프레임의 픽셀들에 대한 에러(E)를 갱신된 클러스터 센터들을 이용하여 구할 수 있다.Meanwhile, the updated cluster centers (

) May be expressed as in Equation 24. The clustering module 1010 may use the updated cluster centers to obtain a cumulative error (E) that can be used to determine whether to change the scene. That is, the clustering module 101 can obtain the error E for the pixels of the next frame using the updated cluster centers according to Equation 23.

The scene change detection unit 1000 may use the error value of the next frame and the error value of the previous frame to determine whether the scene change is detected.

The delay module 1030 may store the error E value of the previous frame received from the clustering module 1010 and output the same to the minimum value extraction module 1040 and the maximum value extraction module 1050.

The minimum value extraction module 1040 outputs a smaller value E_min between an error value of the next frame and an error value of the previous frame, and the maximum value extraction module 1050 further outputs an error value of the error value of the next frame and the previous frame. A large value E_max can be output. Amplification module (gain block) 1060 may multiply the input value by a constant greater than 1, and the subtraction module 1070 may perform subtraction using the input value and output the result value. Can be.

The determination module 1080 may determine whether the scene change is detected by comparing the E_max-a * E_min value with a predetermined value. * Means multiply, a is a constant greater than 1. For example, when the value of E_max-a * E_min is smaller than the predetermined value, it may be determined that the scene has not been changed. The determination module 1080 may output a True or False value depending on whether the scene is changed.

11 is a block diagram illustrating a structure of a final type detection unit of a content type determination device according to another embodiment of the present invention. The final type detector 1100 of FIG. 11 may correspond to the final type detector 650 of FIG. 6.

The final type detector 1100 may include a disjunction block 1110, a switch 1120, and a delay line 1130.

The logical OR module 1110 may output a True value if one or more True values are included among the input values. That is, the logical sum module 1110 may output a 'true' value when there is a 'true' value among the type values detected by the determination module 1080 for the previous frame and the current frame.

The delay unit 1130 may store the type value detected by the determination module 1080 for the previous frame and output the stored value when detecting the type value for the current frame.

The switch 1120 may detect and output the content type 1160 of the current frame according to a value 1150 indicating whether the scene is changed. In this case, information on whether to change the scene may be included in the content data.

The value 1150 indicating whether the scene is changed may be True or False, and when True is changed, False may be a value that is output when the scene is not changed. The switch 1120 outputs the type value 1140 detected for the current frame if the value 1150 indicating whether or not the scene is changed is true, and the output value 1150 of the scene change detection unit 640 is False, the OR module. The output value of 1110 can be output.

12 is a block diagram illustrating a content type determination system according to an embodiment of the present invention.

Content type determination system according to an embodiment of the present invention is a receiver (1220), frame buffer (frame buffer) 1230, video enhancement block (video enhancement block) 1240, field game detection unit (field game episodes) It may include a detector 1250, an adaptation block 1260, and a display unit 1270.

The receiver 1220 may receive and output the video content 1210 from the outside.

The frame buffer 1230 may store the video content 1210 received from the receiver 1220 and output one frame at a time.

The video enhancement block 1240 can process the video content received from the frame buffer 1230. For example, the video enhancement block 1240 can perform noise reduction, contrast enhancement, or sharpening of the video content.

The field game detector 1250 may detect the content type of each frame by determining whether the game is a field game for each frame using the video content 1210 received from the receiver 1220. In this case, the method for detecting the content type of each frame by the field game detection unit 1250 may be applied to the method of determining the content type by the content type determination apparatus 500 or 600 described above.

The optimization block 1260 provides the video enhancement block 1240 with information necessary for processing the video content so that the video enhancement block 1240 can process the video content according to the content type detected by the field game detection unit 1250. can do.

The display 1270 may display video content processed by the video enhancement block 1240.

19 and 20 illustrate graphs of images and images that do not correspond to field game episodes according to an embodiment of the present invention.

Referring to FIGS. 19B and 20B, it can be seen that the images of FIGS. 19A and 20A are not determined to be field game episodes due to a low ratio of green pixels.

21 is a diagram for a video and a video not corresponding to a field game episode according to one embodiment of the present invention.

Referring to FIG. 21B, it can be seen that the image of FIG. 21A is determined to be not a field game episode because the ratio of green pixels is on the lower side. Although the ratio of green pixels is sometimes high in the graph of FIG. 21B, when the current frame is determined to be the same scene as the previous frame according to the content information, the previous frame is a field game episode since the type of the current frame is determined by the determined type of the previous frame. According to the present invention, it may be determined that the current frame also does not correspond to the field game episode.

22 to 23 illustrate graphs of images corresponding to field game episodes and images according to an embodiment of the present invention.

Referring to FIG. 22B, the image of FIG. 22A belongs to a case where the ratio of green pixels is high, and the ratio of high brightness and saturation pixels to white pixels or skin color pixels is relatively small, and corresponds to a field game episode of a distant view. It can be confirmed that it is determined.

Referring to FIG. 23B, the image of FIG. 23B belongs to a case where the number of green pixels is relatively small, the number of pixels having high luminance and high saturation, the number of bright or white pixels is small, and the number of skin color pixels is greater than or less than zero, It can be confirmed that it corresponds to the field game episode of the view.

FIG. 24 is a graph illustrating an image when an image determined to not be a field game episode is inserted between an image determined to not be a field game episode and an image corresponding to a field game episode according to an embodiment of the present invention. It is shown.

Referring to FIG. 24B, it can be seen that the image of FIG. 24A is determined to be not a field game episode because there are few green pixels. However, it can be seen that the scenes before or after the video are determined to be field game episodes. Although the image of FIG. 24B belongs to one scene of the field game episode, since the lawn is not displayed, since the number of green pixels is little, it may be determined that the image of FIG. 24B does not correspond to the field game episode. Accordingly, the apparatus for determining a content type according to an embodiment of the present invention determines whether the corresponding frame belongs to the same scene as the previous frame, and determines the content type of the corresponding frame according to the content type of the previous frame. have.

According to an embodiment of the present disclosure, since the content type may be determined for each frame, the content type may be determined in real time.

According to an embodiment of the present invention, the content type of the video content may be determined at the same level as human recognition.

According to an embodiment of the present invention, since the functions used to determine the content type of the video content are linear and logical, they can be driven simply and quickly. (in real time) The content type may be determined.

The present invention can be embodied as code that can be read by a computer (including all devices having an information processing function) in a computer-readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording devices include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like.

Although the foregoing description has been focused on the novel features of the invention as applied to various embodiments, those skilled in the art will appreciate that the apparatus and method described above without departing from the scope of the invention. It will be understood that various deletions, substitutions, and changes in form and detail of the invention are possible. Accordingly, the scope of the invention is defined by the appended claims rather than in the foregoing description. All modifications within the scope of equivalents of the claims are to be embraced within the scope of the present invention.

Claims (17)

In the method of determining the content type of the video content,
Receiving a frame of the video content;
Detecting color component characteristics of each pixel of the received frame; And
Determining a content type indicating whether the received frame includes content for reproducing a scene of a predetermined genre according to the detected color component characteristic of each pixel;
Determining the content type,
Whether the video content is a field game episode of a distant view, in accordance with at least one of the ratio of pixels in which the color component characteristic of each pixel is green and the ratio of pixels in which the color component characteristic of each pixel is skin color in the received frame; or Determining whether the field game episode of the near view is a content type of the video content. The method of claim 1,
If the received frame and the previous frame are the same scene, determining the content type of the received frame according to the determined content type of the previous frame. The method of claim 1, wherein the detecting of the color component characteristic of each pixel of the received frame comprises:
Detecting luminance and saturation of a plurality of pixels included in the received frame;
Detecting color component characteristics of each pixel using the detected luminance and saturation and RGB channel values of each pixel;
Detecting a degree of change in luminance of each pixel by using the detected luminance value; And
And detecting a statistical analysis value of the received frame by using the detected change in luminance of each pixel and color component characteristic of each pixel. The method of claim 3, wherein detecting the statistical analysis value of the received frame comprises:
Detecting a statistical analysis value of a pixel included in the received frame; And
And detecting a statistical analysis value of a pixel of which the color component characteristic of each pixel among the received frames is green. The method of claim 4, wherein detecting the statistical analysis value of the pixel included in the received frame comprises:
Detecting a ratio of pixels whose color component characteristic of each pixel of the received frame is white;
Detecting a ratio of pixels having high luminance and high saturation of color component characteristics of the received frame;
Detecting a ratio of pixels in which the color component characteristic of each pixel of the received frame is green; And
And detecting a ratio of pixels whose color component characteristic of each pixel of the received frame is skin color. The method of claim 4, wherein the detecting of the statistical analysis value of the pixel of which the color component characteristic of each pixel of the received frame is green is performed.
Detecting an average luminance value of a plurality of pixels of which the color component characteristic of each pixel among the received frames is green;
Detecting an average saturation value of a plurality of pixels of which the color component characteristic of each pixel among the received frames is green;
Detecting an average B channel value of a plurality of pixels of which the color component characteristic of each pixel among the received frames is green;
Detecting an average luminance change value of a plurality of pixels of which the color component characteristic of each pixel among the received frames is green; And
And detecting a histogram of a G channel of a plurality of pixels of which the color component characteristic of each pixel among the received frames is green. The method of claim 1, wherein the determining of the content type of the received frame comprises:
When the ratio of the pixel for which the color component characteristic of each pixel of the received frame is green is less than or equal to the reference value,
When the average saturation value of the pixel whose color component characteristic of each pixel of the received frame is green is less than or equal to a reference value,
When the average saturation value of the pixel whose color component characteristic of each pixel of the received frame is green is less than or equal to the reference value, and the average value of the B channel of the pixel whose color component characteristic of the pixel of the received frame is green is greater than or equal to the reference value,
When the average color saturation value or the average luminance value of the pixel whose color component characteristic of the received frame is green is less than or equal to the reference value,
In the histogram of a pixel in which the average saturation value of the pixel whose color component characteristic of the received frame is green is a value between a first reference value and a second reference value, and the color component characteristic of the pixel of the received frame is green, If the width is above the reference value,
The average saturation value of the pixel whose color component characteristic of each pixel of the received frame is green is greater than or equal to the reference value, and the width of the graph is less than or equal to the reference value in the histogram of the pixel whose color component characteristic of the pixel of the received frame is green; and
In the case where the average saturation value of the pixel whose color component characteristic of each pixel of the received frame is green is greater than or equal to the reference value, and the average change value of the luminance of the pixel whose color component characteristic of the pixel of the received frame is green is less than or equal to the reference value.
If the at least one, the content type of the received frame comprises the step of determining that it is not a field game. The method of claim 1, wherein the determining of the content type of the received frame comprises:
The ratio of the pixel whose color component characteristic of each pixel of the received frame is green is less than or equal to the reference value, and the ratio of the pixel whose color component characteristic of each pixel of the received frame is high luminance and high saturation is greater than or equal to the reference value, and the pixel of the received frame The ratio of pixels having a white color component characteristic or a pixel whose skin component color is a color component characteristic of the received frame is less than or equal to the reference value; and
The ratio of the pixel whose color component characteristic of each pixel of the received frame is green is greater than or equal to the reference value, and the ratio of the pixel whose color component characteristic of the received frame is high luminance and high saturation is less than or equal to the reference value, and the pixel of the received frame The ratio of pixels having a white color component characteristic or a pixel whose skin component color is a color component characteristic of the received frame is equal to or less than a reference value.
If the at least one, the content type of the received frame comprises the step of determining that the content type of the video content characterized in that it comprises a field game. A frame buffer for receiving a frame of video content;
A color component characteristic detection unit for each pixel detecting a color component characteristic for each pixel of the received frame; And
And a type detector configured to determine a content type indicating whether the received frame includes content for reproducing a scene of a predetermined genre according to the detected color component characteristic of each pixel.
The type detector may determine a field of a far-field view of the video content according to at least one of a ratio of a pixel whose color component characteristic of each pixel is green and a ratio of a pixel whose color component characteristic of each pixel is a skin color in the received frame. And determining whether it is a game episode or a field game episode of a near view. The apparatus of claim 9, wherein the content type determination device is
A scene change detector configured to output whether the received frame and the previous frame have been changed; And
And a final type detector configured to determine, as the content type of the received frame, the content type determined according to the determined content type of the previous frame or the detected color component characteristic of each pixel according to the output value of the scene change detector. Device for determining the content type. 10. The method of claim 9, wherein the color component characteristic detection unit for each pixel
A luminance detector detecting luminance of a plurality of pixels included in the received frame;
A saturation detector configured to detect saturation of a plurality of pixels included in the received frame;
A pixel classifier for detecting color component characteristics of each pixel by using the detected luminance and saturation and RGB channel values of the plurality of pixels;
A luminance change detector detecting a change in luminance of each pixel by using the detected luminance values of the plurality of pixels; And
And a statistical analyzer configured to detect a statistical analysis value of the received frame using the change in luminance of each detected pixel and the color component characteristic of each pixel detected by the pixel classifier. Crystal device. The method of claim 11, wherein the statistical analysis unit
And detecting a statistical analysis value of a pixel included in the received frame and detecting a statistical analysis value of a pixel of which the color component characteristic of each pixel is green among the received frames. The method of claim 12, wherein the statistical analysis unit
Detects the ratio of the pixel whose color component characteristic of each pixel of the received frame is white, Detects the ratio of pixels having high luminance and saturation of the color component characteristic of each pixel of the received frame, and Per pixel of the received frame And detecting a ratio of pixels having a color component characteristic of green and a ratio of pixels of a skin color of the received frame. The method of claim 12, wherein the statistical analysis unit
Detects an average luminance value of a plurality of pixels of which the color component characteristic of each pixel is green among the received frames, and detects an average saturation value of a plurality of pixels of which the color component characteristic of each pixel among the received frames is green, and receives the received signal. Detects an average B channel value of a plurality of pixels whose color component characteristics are green in one frame, detects an average luminance change value of a plurality of pixels whose color component characteristics are green in the received frame, and receives the received signal. And a histogram for a G channel of a plurality of pixels whose color component characteristic of each pixel is green in one frame. The method of claim 9, wherein the content type determination unit
When the ratio of the pixel for which the color component characteristic of each pixel of the received frame is green is less than or equal to the reference value,
When the average saturation value of the pixel whose color component characteristic of each pixel of the received frame is green is less than or equal to a reference value,
When the average saturation value of the pixel whose color component characteristic of each pixel of the received frame is green is less than or equal to the reference value, and the average value of the B channel of the pixel whose color component characteristic of the pixel of the received frame is green is greater than or equal to the reference value,
When the average color saturation value or the average luminance value of the pixel whose color component characteristic of the received frame is green is less than or equal to the reference value,
In the histogram of a pixel in which the average saturation value of the pixel whose color component characteristic of the received frame is green is a value between a first reference value and a second reference value, and the color component characteristic of the pixel of the received frame is green, If the width is above the reference value,
The average saturation value of the pixel whose color component characteristic of each pixel of the received frame is green is greater than or equal to the reference value, and the width of the graph is less than or equal to the reference value in the histogram of the pixel whose color component characteristic of the pixel of the received frame is green; and
In the case where the average saturation value of the pixel whose color component characteristic of each pixel of the received frame is green is greater than or equal to the reference value, and the average change value of the luminance of the pixel whose color component characteristic of the pixel of the received frame is green is less than or equal to the reference value.
In the case of at least one of the content types, the content type determination device, characterized in that it is determined that the content type of the received frame is not a field game. The method of claim 9, wherein the content type determination unit
The ratio of the pixel whose color component characteristic of each pixel of the received frame is green is less than or equal to the reference value, and the ratio of the pixel whose color component characteristic of each pixel of the received frame is high luminance and high saturation is greater than or equal to the reference value, and the pixel of the received frame The ratio of pixels having a white color component characteristic or a pixel whose skin component color is a color component characteristic of the received frame is less than or equal to the reference value; and
The ratio of the pixel whose color component characteristic of each pixel of the received frame is green is greater than or equal to the reference value, and the ratio of the pixel whose color component characteristic of the received frame is high luminance and high saturation is less than or equal to the reference value, and the pixel of the received frame The ratio of pixels having a white color component characteristic or a pixel whose skin component color is a color component characteristic of the received frame is equal to or less than a reference value.
In the case of at least one or more, the content type determination device of claim 1, wherein the content type of the received frame is determined to be a field game. A computer-readable recording medium having recorded thereon a program that performs a method of determining a content type of video content, the method of determining the content type of the video content,
Receiving a frame of the video content;
Detecting color component characteristics of each pixel of the received frame;
Determining a content type indicating whether the received frame includes content playing a scene of a predetermined genre according to the detected color component characteristic of each pixel;
Determining the content type,
Whether the video content is a field game episode of a distant view, in accordance with at least one of the ratio of pixels in which the color component characteristic of each pixel is green and the ratio of pixels in which the color component characteristic of each pixel is skin color in the received frame; or And determining whether it is a field game episode of the near view.

Images