Backlight brightness adjusting method based on edge detection

Abstract

The invention belongs to the technical field of LED driving, and particularly relates to a backlight brightness adjusting method based on edge detection. The method mainly comprises three parts of edge detection and region division, and local refinement and weighted calculation adjustment of a virtual matrix. The invention aims to optimize the use of system resources and improve the brightness smooth transition effect. According to the method, smooth transition with low calculation cost is realized in a high-contrast area, and the method is suitable for being applied to equipment such as high-resolution televisions and displays which have high requirements on brightness dimming precision.

Description

Technical Field

[0001] The invention belongs to the technical field of LED driving, and in particular relates to a backlight brightness adjustment method based on edge detection. Background Art

[0002] Local dimming technology divides the backlight of the display into multiple areas, and adjusts the backlight brightness of these areas according to the brightness of different areas in the picture. In this way, the local backlight can be turned off or weakened when displaying black or dark areas, thereby improving the contrast of the picture and showing deeper black and brighter white.

[0003] A common problem with Local Dimming is the halo effect. This is because the backlight area control cannot be accurate to each pixel. When there is a dark black area next to the highlight area, the surrounding black area may be affected by the light leakage of the highlight part, resulting in a halo effect around the bright spot, which is especially obvious in high-contrast scenes.

[0004] If the display has a small number of Local Dimming Zones, each zone may contain many pixels. Since the backlight is based on zone control, it is difficult for a larger zone to adapt to the different needs of bright and dark areas at the same time, so the light of the highlight part will overflow into the adjacent dark area, causing a more obvious halo. Displays with high zones perform better in this regard, but the halo effect is still difficult to completely eliminate. Summary of the invention

[0005] In view of the above problems, the present invention proposes a backlight brightness adjustment method based on edge detection.

[0006] The technical solution of the present invention is:

[0007] The backlight brightness adjustment method based on edge detection comprises the following steps:

[0008] S1. Use edge detection algorithm to detect the output image of the display device, obtain the area that needs dimming, and define it as the target area;

[0009] S2. According to the obtained target area, local adjustment is performed in the following manner:

[0010] Introduce a virtual matrix to expand the backlight bead matrix of the target area to obtain more pixels;

[0011] The brightness of each virtual matrix pixel is set by combining contrast, gradient and distance weighting factors, so that the brightness between pixels can achieve smooth transition and obtain the adjustment parameters of the target area;

[0012] S3. Based on the obtained adjustment parameters of the target area, obtain the adjusted backlight matrix parameters of the display device.

[0013] Furthermore, the specific method of obtaining the area requiring dimming in S1 is:

[0014] The edge detection algorithm is used to obtain the bright and dark edges in the picture, and the areas with significant light and dark contrast are marked as target areas as the difference that needs to be dimmed.

[0015] Furthermore, the method of introducing the virtual matrix in S2 is:

[0016] The row and column matrix of the original backlight lamp beads is used as the initial backlight matrix, virtual pixels are added between two adjacent lamp beads in each row and column, and virtual pixels are also added between virtual pixels between adjacent rows, so as to obtain a new backlight matrix for the target area.

[0017] Furthermore, the specific method of setting the brightness of each luminous pixel by combining contrast, gradient and distance weighting factors is:

[0018]

[0019] Among them, L P is the final brightness of the new backlight matrix midpoint P, ω i =1+α·C P +β·G P is the contrast and gradient weighting factor, C P is the contrast factor, G P is the gradient amplitude, α and β are adjustment coefficients, L neighbore(i) is the brightness value of each neighboring point, is the distance weighting factor, d x and d y It is the row and column distance between point P and the adjacent lamp beads.

[0020] The beneficial effects of the present invention are as follows: the present invention realizes a smooth transition with low computational cost in a high contrast area, and is suitable for application in devices such as high-resolution televisions and displays that require high brightness dimming accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Schematic diagram of virtual matrix refinement in the target area. DETAILED DESCRIPTION

[0022] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings.

[0023] The method of the present invention mainly includes edge detection and region division, local refinement of the virtual matrix and weighted calculation:

[0024] 1) Edge detection and region division

[0025] 1. Input signal processing: First, obtain the input high-resolution signal image and use the edge detection algorithm (such as Sobel or Canny algorithm) to process the image to identify the bright and dark edges in the picture.

[0026] 2. Output edge area: Based on the edge detection results, output the area with significant light-dark contrast, marking it as the boundary area M that needs further dimming BUFF .

[0027] 3. Region division: Divide the image into two types of regions:

[0028] ① Edge area M BUFF : Areas with obvious light-dark contrast need further refinement.

[0029] ②Non-edge area N BUFF : For areas with small color differences, the original backlight matrix is ​​retained without refinement to reduce system resource usage.

[0030] 2) Local refinement and weighted calculation of virtual matrix

[0031] 1. Matrix refinement: In the edge area M BUFF In the display, the backlight matrix is ​​refined from the original 3x3 to 5x5, and the refined virtual matrix points will improve the accuracy of dimming.

[0032] 2. Calculation of weighting factors:

[0033] ① Contrast factor C P : At point P, extract the local contrast value C P The higher the contrast, the more obvious the difference between light and dark.

[0034] ② Gradient amplitude G P : Calculate the gradient amplitude of point P (i.e. the rate of brightness change). The larger the gradient amplitude, the more significant the brightness difference.

[0035] ③Weighting factor ω i : Indicates the smoothness of edge transition: ω i =1+α·C P +β·G P , where α and β are adjustment coefficients that control the weighted influence of contrast and gradient, respectively.

[0036] 3. Distance weighting factor calculation:

[0037] To make the dimming closer to the real environment, we introduce a distance weighting factor so that the influence of neighboring points on the center point decreases as the distance increases. It is usually set to an inverse relationship: where d x and d y It is the X and Y axis distance between point P and the adjacent lamp beads.

[0038] 4. Distance weighting factor calculation:

[0039] Combining contrast, gradient and distance weighting factors, we get the final brightness L of point P. P , the formula is as follows:

[0040]

[0041] Where i is the set of all neighboring points, L neighbor(i) is the brightness value of each neighboring point, ω i =1+α·C P +β·G P is a weighting factor to ensure smooth transition of edge brightness.

[0042] According to the obtained luminous brightness value of each pixel point in the target area, the backlight matrix that needs to be adjusted is obtained, and the display device is adjusted.

Claims

1. A backlight brightness adjustment method based on edge detection, characterized in that: The following steps are involved: S1. Use edge detection algorithm to detect the output image of the display device, obtain the area that needs dimming, and define it as the target area; S2. According to the obtained target area, local adjustment is performed in the following manner: Introduce a virtual matrix to expand the backlight bead matrix of the target area to obtain more pixels; The brightness of each virtual matrix pixel is set by combining contrast, gradient and distance weighting factors, so that the brightness between pixels can achieve smooth transition and obtain the adjustment parameters of the target area; S3. Based on the obtained adjustment parameters of the target area, obtain the adjusted backlight matrix parameters of the display device.

2. The backlight brightness adjustment method based on edge detection according to claim 1, characterized in that: The specific method of obtaining the area requiring dimming described in S1 is: The edge detection algorithm is used to obtain the bright and dark edges in the picture, and the areas with significant light and dark contrast are marked as target areas as the difference that needs to be dimmed.

3. The backlight brightness adjustment method based on edge detection according to claim 1, characterized in that: The method of introducing the virtual matrix in S2 is: The row and column matrix of the original backlight lamp beads is used as the initial backlight matrix, virtual pixels are added between two adjacent lamp beads in each row and column, and virtual pixels are also added between virtual pixels between adjacent rows, so as to obtain a new backlight matrix for the target area.

4. The backlight brightness adjustment method based on edge detection according to claim 1, characterized in that: The specific method of setting the brightness of each luminous pixel by combining contrast, gradient and distance weighting factors is: Among them, L P is the final brightness of the new backlight matrix midpoint P, ω i =1+α·C P +β·G P is the contrast and gradient weighting factor, C P is the contrast factor, G P is the gradient amplitude, α and β are adjustment coefficients, L neighbor(i) is the brightness value of each neighboring point, is the distance weighting factor, d x and d y It is the row and column distance between point P and the adjacent lamp beads.

Images