KR102337829B1 - Method for logo detection and display device using thereof - Google Patents

Abstract

The present invention detects a still logo including a translucent logo and changes the luminance without degrading the image quality, thereby extending the lifespan of the display panel and improving the lifespan of the afterimage. To this end, the present invention divides the image into blocks, calculates the difference between the front and rear blocks, divides the logo candidate block and non-logo block based on this, classifies the logo candidate group by block, expands the logo candidate group, and Logo candidates are grouped together, logos and non-logos are classified within the extended logo candidate group based on an adaptive threshold, and the detected logo is compensated.

Description

Logo detection method and display device using the same

The present invention relates to a logo detection method and a display device using the same.

As information technology develops, the market for display devices, which is a connection medium between users and information, is growing. Accordingly, organic light emitting display (OLED), quantum dot display (QDD), liquid crystal display (LCD), plasma display (PDP), etc. The use of the same display device is increasing.

The display device includes a display panel including a plurality of sub-pixels arranged in a matrix form, a driver outputting a driving signal for driving the display panel, and a controller controlling the driver.

The display device is implemented as a TV, a set-top box, a navigation system, an image player, a Blu-ray player, a personal computer (PC), a home theater, a mobile phone, and the like. Images used in TV, etc. include various types of logos such as opaque logos, translucent logos, or animation logos.

Since some of the display devices change their luminance characteristics due to device deterioration, when an image including an opaque logo or a translucent logo is displayed on a display panel for a long period of time (time), the deterioration of the logo part is noticeable. In this case, an afterimage is generated on the display panel due to the difference in the degree of deterioration between the area deteriorated by the logo and the area around the area, as well as deterioration of display quality and lifespan, so improvement is required.

The present invention for solving the problems of the above-mentioned background technology detects a still logo including a translucent logo and varies the luminance without deterioration of image quality to extend the lifespan of a display panel and improve the afterimage life, and to compensate for the artifacts (artifacts) ) to improve and improve display quality by lowering the influence of In addition, the present invention improves the memory increase (or insufficient) problem by dynamically allocating a memory of a predetermined size for compensation for each pixel after deriving a logo candidate group.

As a means for solving the above problems, the present invention provides a display device including a difference value calculating unit, a logo candidate group estimation unit, a logo candidate group expansion and grouping unit, an adaptive threshold value setting unit, and a data compensation unit. The difference calculating unit divides the image into blocks and calculates a difference value between the front and rear blocks. The logo candidate group estimator classifies the logo candidate group block from the non-logo block based on the difference value between the front and rear blocks, and classifies the logo candidate group in block units. The logo candidate group expansion and grouping unit expands the logo candidate group and groups the same logo candidate group. The adaptive threshold value setting unit obtains an adaptive thresholding value and classifies a logo and a non-logo in the extended logo candidate group based on the adaptive threshold value. The data compensator compensates the detected logo.

The adaptive threshold value setting unit may obtain an adaptive threshold value by using an average of difference values between frames before and after pixels in blocks classified into the same logo candidate group.

The adaptive threshold value may change at least every N (N is an integer greater than or equal to 1) frames.

The logo candidate group expansion and grouping unit may extend the logo candidate group up to K (K is an integer greater than or equal to 1) blocks adjacent to the logo candidate group and group the same logo candidates.

The apparatus may further include a memory allocator for dynamically allocating the logo memory so that blocks classified into the logo candidate group are allocated to the logo memory and blocks excluded from the logo candidate group are deleted from the logo memory.

The data compensator may compensate the detected luminance of the logo in units of pixels, blocks, or pixels and blocks.

In another aspect, the present invention provides a logo detection method. The logo detection method divides the image into blocks, calculates the difference value between the front and rear blocks, distinguishes the logo candidate group block from the non-logo block based on the difference value between the front and rear blocks, and classifies the logo candidate group by block; Expanding the logo candidate group and grouping the same logo candidate group, obtaining an adaptive thresholding, and discriminating a logo from a non-logo within the expanded logo candidate group based on the adaptive threshold value, and the detected logo Including a step of performing compensation for.

In the step of classifying the logo candidate group by block, the block counter may be updated using the difference value between the front and rear blocks, and the logo candidate group block and the non-logo block may be distinguished using the block counter.

In the step of discriminating a logo from a non-logo in the extended logo candidate group, an adaptive threshold may be obtained by using the average of the difference values between the frames before and after pixels within the block classified into the same logo candidate group.

In the grouping of the same logo candidate group, the logo candidate group can be extended to K (K is an integer greater than or equal to 1) blocks adjacent to the logo candidate group and grouped with the same logo candidate group.

The present invention has the effect of detecting a still logo including a translucent logo and changing the luminance without degrading the image quality to extend the lifespan of the display panel and improve the lifespan of the afterimage. In addition, the present invention has the effect of improving and improving the display quality by allocating a compensation value in units of pixels to the detected logo and reducing the effect of artifacts that occur when compensating in units of blocks. The present invention has an effect of classifying the change between frames of a moving picture with an adaptive threshold, and detecting and compensating for a translucent logo based on this. In addition, the present invention has an effect of improving the memory increase (or insufficient) problem by dynamically allocating a memory of a predetermined size for compensation for each pixel after deriving a logo candidate group.

1 is a block diagram schematically showing a display device;
FIG. 2 is a configuration diagram schematically illustrating the sub-pixel shown in FIG. 1;
3 is a diagram schematically illustrating a logo detection method according to an experimental example;
4 is a view showing an opaque logo detection result according to an experimental example;
5 is a view showing a result of detecting a translucent logo according to an experimental example;
6 to 8 are diagrams for explaining a logo detection method according to an embodiment of the present invention.
9 is a diagram for explaining a logo estimation process.
10 is a diagram for explaining a logo grouping process.
11 is a view for explaining a memory dynamic allocation and storage process;
12 is a view showing the results of detecting an opaque logo and a semi-transparent logo;
13 and 14 are diagrams for explaining a block counter update process;
15 is a diagram showing an example of pixel-wise compensation;
16 and 17 are diagrams showing a difference value map between an opaque logo and a translucent logo.
18 and 19 are exemplary diagrams of compensation in units of pixels and blocks.
20 is a block diagram schematically illustrating a part of a display device according to an embodiment of the present invention;
FIG. 21 is a block diagram of the timing controller shown in FIG. 20;

Hereinafter, specific details for carrying out the present invention will be described with reference to the accompanying drawings.

The display device using the text detection method according to the present invention is implemented as a TV, a set-top box, a navigation system, an image player, a Blu-ray player, a personal computer (PC), a home theater, a mobile phone, and the like. The display panel of the display device may include, but is not limited to, a liquid crystal display panel, an organic light emitting display panel, a quantum dot display device, an electrophoretic display panel, a plasma display panel, and the like.

FIG. 1 is a block diagram schematically illustrating a display device, and FIG. 2 is a configuration diagram schematically illustrating a sub-pixel illustrated in FIG. 1 .

1 , the display device includes an image supply unit 110 , a timing controller 120 , a gate driver 130 , a data driver 140 , and a display panel 150 .

The image supply unit 110 processes the data signal and outputs it together with a vertical synchronization signal, a horizontal synchronization signal, a data enable signal, and a clock signal. The image supply unit 110 transmits a vertical synchronization signal, a horizontal synchronization signal, a data enable signal, a clock signal and a data signal, etc. to the timing controller 120 through a low voltage differential signaling (LVDS) interface or a transition minimized differential signaling (TMDS) interface. ) is supplied to For example, the image supply unit 110 outputs R, G, and B data signals.

The timing controller 120 receives the data signal DATA from the image supplier 110 , and a gate timing control signal GDC for controlling the operation timing of the gate driver 130 and the operation timing of the data driver 140 . and output a data timing control signal DDC for controlling the . The timing controller 120 outputs the data signal DATA together with the gate timing control signal GDC and the data timing control signal DDC through the communication interface, and the operation timing of the gate driver 130 and the data driver 140 . to control

The gate driver 130 outputs a gate signal (or a scan signal) while shifting the level of the gate voltage in response to the gate timing control signal GDC supplied from the timing controller 120 . The gate driver 130 supplies a gate signal to the sub-pixels SP included in the display panel 150 through the gate lines GL1 to GLm. The gate driver 130 is formed in the form of an integrated circuit (IC) or is formed in the display panel 150 in the form of a gate in panel (GIP).

The data driver 140 samples and latches the data signal DATA in response to the data timing control signal DDC supplied from the timing controller 120 , and converts the digital signal into an analog signal in response to the gamma reference voltage and outputs it . The data driver 140 supplies the data signal DATA to the sub-pixels SP included in the display panel 150 through the data lines DL1 to DLn. The data driver 140 is formed in the form of an integrated circuit (IC).

The display panel 150 displays an image in response to the gate signal supplied from the gate driver 130 and the data signal DATA supplied from the data driver 140 . The display panel 150 includes a lower substrate, an upper substrate, and sub-pixels SP formed between the lower substrate and the upper substrate.

As shown in FIG. 2 , one sub-pixel is supplied through the switching thin film transistor SW and the switching thin film transistor SW connected to (or formed at the intersection of) the gate line GL1 and the data line DL1. A pixel circuit PC operating in response to the data signal DATA is included. The sub-pixels SP include a liquid crystal device or an organic light emitting device according to the configuration of the pixel circuit PC.

When the display panel 150 is configured as a liquid crystal display panel, it is a TN (Twisted Nematic) mode, VA (Vertical Alignment) mode, IPS (In Plane Switching) mode, FFS (Fringe Field Switching) mode, or ECB (Electrically Controlled Birefringence) mode. implemented in mode. When the display panel 150 is configured as an organic light emitting display panel, it is implemented in a top-emission method, a bottom-emission method, or a dual-emission method.

The above display device displays a specific image as the display panel 150 emits or transmits light based on the gate signal and data signal DATA output from the gate driver 130 and the data driver 140 . .

The display device is implemented as a TV, a set-top box, a navigation system, an image player, a Blu-ray player, a personal computer (PC), a home theater, a mobile phone, and the like. Images used in TV, etc. include various types of logos such as opaque logos, translucent logos, or animation logos.

Since some of the display devices change their luminance characteristics due to device deterioration, when an image including an opaque logo or a translucent logo is displayed on a display panel for a long period of time (time), the deterioration of the logo part is noticeable. In this case, an afterimage is generated on the display panel due to the difference in the degree of deterioration between the area deteriorated by the logo and the area around the area, as well as deterioration of display quality and lifespan, so improvement is required.

Hereinafter, the problems of the experimental examples will be considered and the examples prepared as a way to improve them will be described.

[Experimental example]

3 is a diagram for schematically explaining a logo detection method according to an experimental example, FIG. 4 is a diagram showing an opaque logo detection result according to an experimental example, and FIG. 5 is a diagram showing a translucent logo detection result according to an experimental example .

As shown in FIG. 3 , in the experimental example, the logo is detected and the luminance is adjusted in the order of the frame comparison step ( S10 ), the logo block detection step ( S20 ), and the luminance adjustment step ( S30 ). In the experimental example, it is determined whether the image is stopped by using a specific threshold, and the logo is determined based on this.

If it is assumed that the logo is still in the image containing the logo, the presence or absence of the logo can be detected by accumulating enough difference values between the front and rear frames (N-1 frame, N frame) and comparing them.

The logo detection method of the experimental example compares the front and rear frames (N-1 frame, N frame) in block units based on these characteristics to guess and detect the block containing the logo, and reduce the luminance of the detected block. alleviate the deterioration of In the process of comparing the front and rear frames (N-1 frame, N frame), it is possible to improve the lifespan of the afterimage of the logo to some extent by distinguishing the logo block using a specific threshold and reducing the luminance of the block determined as the logo. can

As shown in Fig. 4, since the experimental example uses a method of accumulating the difference values between the front and rear frames (N-1 frame, N frame), it is easy to distinguish the opaque logo part from the non-logo part through the histogram. could pay

As shown in Fig. 5, since the experimental example uses a method of accumulating the difference values between the front and rear frames (N-1 frame, N frame), the semi-transparent logo part and the non-logo part can be easily distinguished through the histogram. couldn't pay

The experimental example was useful for judging a logo (opaque logo) having a similar change in a specified frame unit, but was found to be inappropriate for judging a logo (translucent logo) having a different amount of change depending on the screen. The reason is that a method of determining whether an image is stopped using a specific threshold value and determining whether a logo exists based on this is used.

As can be seen from the results of the experimental example, in the experimental example, the detectable logo is limited depending on how the threshold is set, and when the threshold is set high enough to detect the translucent logo, the probability of false detection increases.

[Example]

6 to 8 are diagrams for explaining a logo detection method according to an embodiment of the present invention, FIG. 9 is a diagram for explaining a logo estimation process, and FIG. 10 is a diagram for explaining a logo grouping process, 11 is a diagram for explaining a memory dynamic allocation and storage process, FIG. 12 is a diagram showing an opaque logo and a translucent logo detection result, FIGS. 13 and 14 are diagrams for explaining a block counter updating process, and FIG. 15 is a diagram showing an example of compensation in units of pixels, FIGS. 16 and 17 are diagrams showing a difference value map between an opaque logo and a translucent logo, and FIGS. 18 and 19 are diagrams illustrating compensation in units of pixels and blocks.

As shown in Fig. 6, the logo detection method according to an embodiment of the present invention includes a logo block tracking step (S110), a block expansion and grouping step (S120), and a pixel and block unit compensation step (S150) in the order of logo is detected and the luminance is adjusted. In the logo detection method, a logo candidate group is derived, a logo pixel is classified using an adaptive threshold in the candidate group, a compensation value is derived, and luminance is varied using the compensation value.

As shown in FIGS. 6 and 7 , in the logo block estimation step ( S110 ), the image is divided into blocks of an appropriate size, the logo block and the non-logo block are divided using the difference between the front and rear blocks, and the block unit to classify the logo candidates. A block is possible if M (M is an integer greater than or equal to 2) * N (N is an integer greater than or equal to 2). A block of an appropriate size can be established through experimentation. In this case, the minimum unit of one block is defined as a pixel.

The logo block estimation step (S110) includes the step of updating the block counter using the difference between the front and rear blocks (S113) and the step of distinguishing the logo block from the non-logo block using the block counter (counter value) (S115). include

The logo part of the video containing the logo has the characteristic of being still when the video has a large instantaneous change. By using this characteristic, it is possible to distinguish blocks that contain a logo and blocks that do not, in units of blocks. By dividing the image into blocks, limiting the size of the area that can contain the logo, and waiting until the cumulative change of successive frames reaches a predetermined size, it is possible to distinguish between logo candidate blocks and non-logo blocks.

The size of the area where the logo can be included can be 1% or more of the entire screen of the display panel. However, if the size of the area where the logo can be included is too small, it is difficult to derive the logo candidate group, and there is a possibility of omission. Therefore, it is preferable that the size of the area in which the logo can be included does not exceed 1% to 20% of the entire screen of the display panel, but this may vary depending on the ratio or resolution of the display panel through experiments.

As shown in Fig. 9, in one frame (1 frame), all of the logo (Running Man training race) and the logo estimation (candidate group) blocks are detected. However, in subsequent frames (2 to 3 frames), the size of the area that can contain the logo is limited (the size of the area that can contain the logo is limited to 5% of the entire screen of the display panel) and the block for the area corresponding to the logo As a counter (block counter up) is performed, non-logo blocks are excluded.

Therefore, in this process, the threshold value that determines the difference between the front and rear frames of the area where the logo can be included is increased so that the translucent logo can be sufficiently included. In this case, there is a problem of erroneous detection in which blocks that do not contain a logo are detected, but since this is simply a process of setting up a logo candidate group, it does not affect future erroneous compensation.

In the block expansion and grouping step (S120), after classifying the logo candidate group by block, the logo candidate group is up to K (K is an integer greater than or equal to 1) blocks adjacent to the left, right, top, and bottom so as to respond to logos of special shapes and sizes. to expand and group (Grouping). In this step, after the logo candidate group has been expanded, one candidate group is created between adjacent blocks so that adaptive thresholding can be applied, and the same logo candidate group is grouped to distinguish between logos and logos (to distinguish them from other logos). do

As shown in Fig. 10, the logo candidate group (C) is classified by block (a in Fig. 10) and the logo candidate group is extended (EC) to 1 block adjacent to the logo candidate group (Fig. 10 b), and the expanded When the logo candidate group is grouped (Fig. 10c), the initially detected logo candidate group (C) is expanded and grouped like the first group (1), the second group (2), and the third group (3). .

After the block expansion and grouping step S120 is performed, the pixel and block unit compensation step S150 may be performed immediately. However, in order to improve the detection capability of the translucent logo, it is preferable to perform the memory dynamic allocation and storage step S130 and the pixel counter update step S140 by applying adaptive thresholding as follows.

6, 8 and 11, in the memory dynamic allocation and storage step (S130), a block determined as a logo is allocated and stored in the memory (Memory). If memory is dynamically allocated, logo blocks of arbitrary positions and arbitrary sizes (which can be allocated smaller than the specified size, which can reduce memory usage) can be allocated to the memory by using only a predetermined size of memory. For example, 600 blocks are allocated to the memory, and one block may be set to store 16*16 pixels, but is not limited thereto.

6, 8, and 12 to 17, in the pixel counter update step (S140) through the application of adaptive thresholding, before pixels within the block classified into the same (same) logo candidate group, An adaptive thresholding value is obtained by using the average of the difference values between subsequent frames. The adaptive thresholding is a value for estimating (or detecting) a block having a relatively small amount of change within a frame. The adaptive threshold makes it easy to estimate (or detect) different transparency of the logo. If adaptive thresholding is used, logos and non-logos can be distinguished on a pixel-by-pixel basis, and a pixel counter can be stored using dynamically allocated memory.

After classifying the logo area of the image into a logo candidate group, an average of the difference values between the front and rear frames is calculated in units of pixels within a block determined as one logo candidate group to obtain an adaptive thresholding value. And when distinguishing between logos and non-logos, adaptive thresholding is used to detect translucent logos. The logo candidates are grouped with the same logo candidates. Accordingly, the same adaptive threshold value may be applied to the same logo candidate group, but different adaptive threshold values may be applied between different logo candidate group groups.

The upper left image of Fig. 12 (a) is an image (Running Man training race) that captures the opaque logo, and the lower left image of Fig. 12 (a) is the difference value map of the before and after frames for the opaque logo (Difference) map), and the right image of FIG. 12 (a) is a histogram distribution for an opaque logo.

As can be seen from the histogram of the opaque logo, logo detection is possible even if a sufficiently small threshold is used because the difference value between the front and rear frames of the pixel converges to 0.

The upper left image of Fig. 12 (b) is a captured image (SBS HD) of the translucent logo, and the lower left image of Fig. 12 (b) is the difference value map of the before and after frames for the translucent logo. , the right image of FIG. 12 (b) is a histogram distribution for a semi-transparent logo.

As can be seen from the histogram of the translucent logo, this occurs when the difference between the front and rear frames of the pixel is not 0, and in this case, if a sufficiently small threshold is used, the logo can be determined as not a logo. have. For this reason, it is necessary to use a different threshold value for each frame.

Therefore, in order to detect a semi-transparent logo, an adaptive thresholding value that changes at least every N (N is an integer greater than or equal to 1) frames should be used.

As shown in FIG. 12 , in the case of the translucent logo (b), the pixel value varies greatly depending on the image compared to the opaque logo (a), but when the degree of difference value is continuously accumulated, it converges to the average difference value. will do And since this difference converges to a value smaller than the non-logo part, it is possible to distinguish the translucent logo.

13, when the pixel counter is stored in the memory, if the difference value is smaller than the average (average 3) (0, 1 and 2), the count is incremented by one, and when the difference value is larger (3) , 4, 5) is decremented by one, and a pixel counter is created for each logo candidate block.

And if the counter value is continuously obtained for a certain period of time or longer, a sufficiently large counter value is obtained in the case of a logo, and a sufficiently small counter value is obtained in the case of a non-logo. And with the average of this counter value, the logo can be distinguished again. Therefore, as shown in FIG. 14 , the pixel and block counter values may be used as pixel and block unit compensation values (block gain) for reducing pixel luminance. By using a pixel counter or a block counter, a block gain map for compensation in units of pixels or blocks may be prepared.

15 shows an example in which a translucent logo is compensated (Compensated value) by using a pixel counter (b) for an original translucent logo (Original value) (a).

As can be seen from the opaque logo and translucent logo frame difference map shown in FIGS. 16 and 17, when determining adaptive thresholding, the average of the difference values between the front and rear frames in the group is used. effective.

However, the determined adaptive thresholding may not accurately distinguish all pixels. Therefore, by repeating the above process for several frames and storing the cumulative sum to determine logo and non-logo, logo and non-logo pixels can be distinguished based on this. And in this process, the adaptive thresholding is not fixed to a single value, but is variable to a first adaptive threshold value, a second adaptive threshold value, etc. (th1 and th2 meaning an adaptive threshold in FIG. 14 ) see) can be

16, 18, and 19, after the pixel counter is stored or updated, the luminance of the detected logo is adjusted through the pixel and block unit compensation step S150.

As shown in FIG. 18 , in the pixel and block unit compensation step S150 , the luminance of the detected logo may be lowered in units of (1) pixels, (2) blocks, or (3) pixels and blocks. Also, as shown in FIG. 19 , in the pixel and block unit compensation step S150 , the brightness of the screen may be lowered while the brightness of the detected logo is maintained.

As can be seen from FIGS. 18 and 19 , if pixel and block unit compensation is performed on the detected logo, the luminance can be partially varied (reduced), so it is possible to minimize the quality deterioration factor, thereby extending the lifespan and improving the lifespan of the afterimage. can do.

20 is a block diagram schematically illustrating a part of a display device according to an embodiment of the present invention, and FIG. 21 is a block diagram of the timing controller shown in FIG. 20 .

20 and 21 , the timing control unit 120 receives the data signal DATA from the image supply unit 110 in conjunction with the frame memory 160 and the logo memory 165 and detects and detects the logo. After compensating the logo, the compensation data signal CDATA is output.

The data signal DATA supplied to the timing controller 120 is stored in the frame memory 160 in units of frames. The timing controller 120 stores or loads the previous and subsequent frames (N-1 frames, N frames) (or the previous frame and the current frame) in the frame memory 160 . The logo memory 165 stores blocks determined as logos (position, size, etc.) and a pixel counter. The timing controller 120 stores or retrieves a block (position, size, etc.) determined as a logo and a pixel counter in the logo memory 165 .

The timing control unit 120 includes a difference value calculator 121 , a logo candidate group estimation unit 123 , a logo candidate group expansion and grouping unit 125 , a memory allocation unit 126 , an adaptive threshold value setting unit 127 , and data. Compensation unit 129 is included.

The difference calculating unit 121 divides the image into blocks of an appropriate size and calculates (calculates) a difference between the front and rear blocks. A block is possible if M (M is an integer greater than or equal to 2) * N (N is an integer greater than or equal to 2). A block of an appropriate size can be established through experimentation. In this case, the minimum unit of one block is defined as a pixel.

The logo candidate group estimator 123 classifies the logo candidate group block and the non-logo block based on the difference value between the front and rear blocks, and classifies the logo candidate group in block units. The logo candidate group estimator 123 may update the block counter based on the difference value between the front and rear blocks, and distinguish the logo candidate group block from the non-logo block using the block counter (counter value).

The logo candidate group estimator 123 divides the image into blocks, limits the size of a region that can include a logo, and waits until the cumulative change of successive frames reaches a predetermined size. Then, it is possible to distinguish between a logo candidate group block and a non-logo block.

On the other hand, the size of the area in which the logo can be included can be larger than the size of one block designated to distinguish the logo. However, if the size of the area where the logo can be included is too small, it is difficult to derive the logo candidate group, and there is a possibility of omission. Therefore, it is preferable that the size of the area in which the logo can be included does not exceed 1% to 20% of the entire screen of the display panel, but this may vary depending on the ratio or resolution of the display panel through experiments.

The logo candidate group expansion and grouping unit 125 classifies the logo candidate group by block, and then the logo up to K (K is an integer greater than or equal to 1) blocks adjacent to the left, right, top, and bottom so as to correspond to a logo of a special shape and size. Expand and group the candidate group.

The logo candidate group expansion and grouping unit 125 creates one candidate group between adjacent blocks to apply an adaptive thresholding value after the logo candidate group expansion is performed, and separates the logo from the logo (for distinguishing from other logos). For this purpose, the same logo candidates are grouped together.

The memory allocator 126 allocates and stores the block determined as the logo in the logo memory 165 . When the logo memory 165 is dynamically allocated, logo blocks of arbitrary positions and arbitrary sizes (which can be allocated smaller than the predetermined size, thereby reducing memory usage) can be allocated to the memory by using only the memory of a predetermined size. For example, 600 blocks are allocated to the logo memory, and one block may be set to store 16*16 pixels, but is not limited thereto.

The adaptive threshold value setting unit 127 obtains an adaptive thresholding value by using an average of difference values between frames before and after pixels in blocks classified into the same (same) logo candidate group. The adaptive threshold setting unit 127 distinguishes a logo from a non-logo in units of pixels through adaptive thresholding and stores a pixel counter by using the dynamically allocated logo memory 165 . .

The adaptive threshold value setting unit 127 separates a logo from a non-logo by estimating a pixel having a small amount of change compared to the surrounding color within the extended logo candidate group based on the adaptive threshold value. The adaptive threshold value setting unit 127 obtains an adaptive thresholding value in which a value is changed for at least N (N is an integer greater than or equal to 1) frames in order to detect a translucent logo when a logo is distinguished from a non-logo. The logo candidates are grouped with the same logo candidates. Accordingly, the same adaptive threshold value may be applied to the same logo candidate group, but different adaptive threshold values may be applied between different logo candidate group groups.

The data compensator 129 prepares a compensation data signal CDATA by performing pixel and block compensation on the detected logo. The data compensator 129 lowers the luminance of the detected logo in units of (1) pixels, (2) blocks, or (3) pixels and blocks, or compensates for lowering the luminance of the screen while maintaining the luminance of the detected logo. can be done

When compensation is unnecessary for the logo, the data compensation unit 129 omits the compensation process and outputs the data signal DATA corresponding to the original. The data compensator 129 can partially change (reduce) luminance by performing pixel- and block-based compensation on the detected logo, thereby minimizing image quality deterioration factors, thereby extending the lifespan of the display panel and improving the lifespan of the afterimage. can

If the logo is compensated in pixel and block units, the problem of afterimages on the display panel due to the difference in the degree of deterioration between the area deteriorated by the logo and the area around it, and the problem of deterioration of display quality and lifespan, are improved. .

In one embodiment of the present invention, since pixel and block unit compensation is made for the logo, a pixel unit memory must be provided. Accordingly, in the process of limiting the size of the logo when implementing the device (H/W) (with only a limited-sized memory and using dynamic allocation), blocks classified as logo candidates are allocated to the memory, and blocks excluded from the candidate group are removed from the memory. The problem of increasing (or running out of) memory can be improved by freeing (deleting) it.

An embodiment of the present invention is more effective when applied to an organic light emitting display device. The reason is that the organic light emitting diode device exhibits markedly reduced luminance and lifespan characteristics due to deterioration (image quality deterioration, element deterioration, etc.) when driven for a long period of time (time) compared to a liquid crystal element. Therefore, when an opaque logo and a translucent logo are detected in pixel units using an embodiment of the present invention and the brightness is partially lowered, image quality deterioration factors are minimized to improve and improve the afterimage life and display quality of the organic light emitting display device. have.

As described above, the present invention has the effect of detecting a still logo including a translucent logo and changing the luminance without degrading the image quality to extend the lifespan of the display panel and improve the lifespan of the afterimage. In addition, the present invention has the effect of improving and improving the display quality by allocating a compensation value in units of pixels to the detected logo and reducing the effect of artifacts generated when compensating in units of blocks. According to the present invention, a change between frames of a moving picture is divided by an adaptive threshold, and a translucent logo can be detected and compensated based on this. In addition, the present invention has the effect of improving the memory increase (or insufficient) problem by dynamically allocating a memory of a predetermined size for compensation for each pixel after deriving a logo candidate group.

Although the embodiments of the present invention have been described above with reference to the accompanying drawings, the technical configuration of the present invention can be changed to other specific forms by those skilled in the art to which the present invention pertains without changing the technical spirit or essential features of the present invention. It will be appreciated that this may be practiced. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. In addition, the scope of the present invention is indicated by the following claims rather than the above detailed description. In addition, all changes or modifications derived from the meaning and scope of the claims and their equivalents should be construed as being included in the scope of the present invention.

110: image supply unit 120: timing control unit
130: gate driver 140: data driver
150: display panel 121: difference value calculator
123: logo candidate group estimation unit 125: logo candidate group expansion and grouping unit
126: memory allocation unit 127: adaptive threshold value setting unit
129: data compensation unit

Claims (10)

a difference calculation unit that divides the image into blocks and calculates a difference value between before and after blocks;
a logo candidate group estimator for classifying a logo candidate group block and a non-logo block block based on the difference value between the front and rear blocks, and classifies the logo candidate group by block;
a logo candidate group expansion and grouping unit that expands the block estimated as the logo candidate group and groups the same logo candidate group;
An adaptive thresholding value of the corresponding block is obtained for each N (N is an integer greater than or equal to 1) frame by using the average of the difference values between the frames before and after the pixels in the block classified into the same logo candidate group, and calculated for each N frame an adaptive threshold value setting unit for discriminating a logo from a non-logo in the corresponding block based on the average of the adaptive threshold value and the difference value of each pixel in the corresponding block; and
A display device including a data compensator for compensating for the detected logo.
delete According to claim 1,
The adaptive threshold value changes every N frames.
According to claim 1,
The logo candidate group expansion and grouping unit
A display device for extending the logo candidate group to a block of K (K is an integer greater than or equal to 1) adjacent to the logo candidate group and grouping the same logo candidate group. According to claim 1,
and a memory allocator for dynamically allocating the logo memory so that blocks classified into the logo candidate group are allocated to the logo memory, while blocks excluded from the logo candidate group are deleted from the logo memory. According to claim 1,
The data compensator
A display device for compensating for the detected luminance of the logo in units of pixels, blocks, or pixels and blocks. dividing the image into blocks, calculating a difference value between the front and rear blocks, classifying the logo candidate group and non-logo block based on the difference value between the front and rear blocks, and classifying the logo candidate group by block;
expanding the block estimated as the logo candidate group and grouping the same logo candidate group;
An adaptive thresholding value of the corresponding block is obtained for each N (N is an integer greater than or equal to 1) frame by using the average of the difference values between the frames before and after the pixels in the block classified into the same logo candidate group, and calculated for each N frame distinguishing a logo from a non-logo in the corresponding block based on an average of the adaptive threshold value and the difference value of each pixel in the corresponding block; and
A logo detection method comprising the step of performing compensation for the detected logo. 8. The method of claim 7,
The step of classifying the logo candidate group in the block unit is
A logo detection method of updating a block counter using the difference value between the front and rear blocks, and using the block counter to distinguish between a logo candidate group block and a non-logo block. delete 8. The method of claim 7,
The step of grouping the same logo candidates
A logo detection method of extending the logo candidate group to a block of K (K is an integer greater than or equal to 1) adjacent to the logo candidate group and grouping the same logo candidate group.

Images