TWI388219B - Image processing device and method thereof - Google Patents

Description

Image processing device and method thereof

The present invention relates to an image processing technique, and more particularly to an image processing technique that can dynamically adjust an image.

With the development of multimedia technology, people's requirements for high-quality images have increased. However, in the fast dynamic picture of the liquid crystal display, the reaction speed of the liquid crystal (Liquid Crystal, LC) is not fast enough (generally 1ms~16ms), and the backlight system adopts the holding mode of the holding type. In this way, the problem of residual image will be caused by the persistence of the human eye, and this is the so-called Motion Blur phenomenon.

In view of this, conventional techniques often use sharpening processing techniques to improve the problem of picture blur. The conventional sharpening processing technique sharpens the entire image, so that the image area that does not need to be sharpened is also forced to be sharpened.

Taking the image of a moving vehicle as an example, the overall image is sharpened by a conventional sharpening processing technique, and the blurring problem of the vehicle can be improved, but the noise of the overall image is also more serious. It is worth mentioning that if the background of the vehicle is already a clear picture, the sharpening process will over-sharp the background of the vehicle, causing the picture to be quite uncoordinated and producing noise that the human eye can't bear.

The invention provides an image processing method, which can improve the quality of an output image.

The invention provides an image processing device, which can dynamically adjust an image to improve the quality of an output image.

The invention provides an image processing method, which comprises calculating a first adjustment value according to a first pixel of a current image and a plurality of first reference pixels in the vicinity thereof. In addition, the first weight coefficient is calculated according to the difference between the gray level value of the first pixel of the current image and the first pixel of the previous image. In addition, the first weight adjustment value is calculated according to the first adjustment value and the first weight coefficient. Furthermore, the first pixel of the current image is adjusted according to the first weight adjustment value, thereby generating a first pixel of the output image.

In an embodiment of the invention, the image processing method further includes calculating a second adjustment value according to the second pixel of the current image and the plurality of second reference pixels in the vicinity thereof. In addition, a second weighting coefficient is calculated according to a grayscale value difference between the second pixel of the current image and the second pixel of the previous image. In addition, the second weight adjustment value is calculated according to the second adjustment value and the second weight coefficient. Furthermore, the second pixel of the current image is adjusted according to the second weighting adjustment value, thereby generating a second pixel of the output image.

In an embodiment of the invention, the first pixel of the current image and the first reference pixel in the vicinity thereof constitute an M×N pixel array, wherein M and N are positive integers greater than or equal to 2. In another embodiment, the first pixel of the current image may be located in the lower right corner of the pixel array. In still another embodiment, the first adjustment value is a grayscale value of the first pixel of the current image minus the current image. The average grayscale value of the first pixel and its first reference pixel.

In an embodiment of the invention, the step of calculating the first weight coefficient comprises dividing the difference between the first pixel of the current image and the gray level value of the first pixel of the previous image by the reference value to obtain the first weight coefficient. In another embodiment, the first weighted adjustment value is equal to the first adjustment value multiplied by the first weighting factor. In still another embodiment, the gray level value of the first pixel of the output image is equal to the gray level value of the first pixel of the current image plus the first weight adjustment value.

From another point of view, the present invention provides an image processing apparatus including a storage unit and a computing unit. The storage unit can store previous images. The computing unit is coupled to the storage unit, and the adjustment value is calculated according to the pixel of the current image and a plurality of reference pixels in the vicinity thereof, and the weight coefficient is calculated according to the difference between the gray level value of the pixel of the current image and the pixel of the previous image. In addition, the calculation unit may further calculate a weight adjustment value according to the adjustment value and the weight coefficient, and adjust a pixel of the current image according to the weight adjustment value, thereby generating a pixel of the output image, wherein the pixel of the current image, the pixel of the previous image, and The pixels of the output image correspond to each other.

The present invention calculates an adjustment value based on the pixels of the current image and a plurality of reference pixels in the vicinity thereof. In addition, the weight coefficient is calculated according to the difference between the gray level value of the pixel of the current image and the pixel of the previous image. In addition, the current image is adjusted according to the adjustment value and the weight coefficient. In this way, the current image can be adaptively adjusted to improve the quality of the output image.

The above described features and advantages of the invention will be apparent from the following description.

The sharpening processing technique of the prior art sharpens the entire image, and the image block that does not need to be sharpened is also forced to be sharpened, so that it is easy to generate noise that cannot be tolerated by the human eye.

In view of this, embodiments of the present invention provide an image processing technique. It can be judged that each block of the current image is a moving image or a still image. Then, the sharpening effect is enhanced for the moving image portion of the image, thereby improving the problem of dynamic blurring. In addition, for the static part of the image, the sharpening effect is reduced (even without sharpening), so noise can be avoided. In this way, the current image can be dynamically and adaptively adjusted, and the quality of the output image can be improved accordingly. The embodiments of the present invention are explained in detail below with reference to the accompanying drawings, in which FIG.

1 is a block diagram of an image processing apparatus in accordance with a first embodiment of the present invention. Referring to FIG. 1 , in the embodiment, the image processing apparatus 10 may include a storage unit 20 and a computing unit 30 . The storage unit 20 is coupled to the computing unit 30. The storage unit 20 and the computing unit 30 can receive video data. It should be noted that the storage unit 20, for example, may be a frame buffer, which can be used to store images and provide previous images to the computing unit 30. The computing unit 30 can receive the current image and adjust the current image according to the previous image provided by the storage unit 20 to generate an output image.

In this embodiment, the calculation unit 30 may include, for example, an average value calculation unit, an addition circuit, a subtraction circuit, a multiplication circuit, and a division circuit, but the invention is not limited thereto. In other embodiments, those skilled in the art are familiar with The implementation of computing unit 30 can also be changed as needed. For example, the calculation unit 30 may also include a look-up table circuit, so that the amount of calculation can be effectively reduced.

From another perspective, the storage unit 20 can receive and store the Nth frame during the Nth frame, and provide the Nth frame to the computing unit 30 during the N+1 frame, where N is a natural number . During the (N+1)th frame, the calculating unit 30 can receive the N+1th frame and adjust the N+1th frame according to the Nth frame provided by the storage unit 20 to generate an output image. By analogy, the storage unit 20 can receive and store the N+1th frame during the N+1th frame and provide the N+1th frame to the calculation unit 30 during the N+2 frame. During the N+2 frame, the calculating unit 30 can receive the N+2 frame and adjust the N+2 frame according to the N+1 frame provided by the storage unit 20 to generate an output image. The following is further explained in conjunction with the flow chart.

2 is a flow chart of an image processing method in accordance with a first embodiment of the present invention. Figure 3 is a schematic illustration of an image in accordance with a first embodiment of the present invention. 4 is a schematic diagram of the image block of FIG. 3. In this embodiment, it is assumed that the resolution of the image is 1280×1024 (as shown in FIG. 3), but the present invention is not limited thereto. In other embodiments, the image may be other resolutions, for example, 1024×768...etc. . In addition, it is assumed that the storage unit 20 has stored the previous image. The following describes in detail how the computing unit 30 adjusts the current image according to the previous image provided by the storage unit 20 when the current image is received, thereby generating an output image.

In this embodiment, the pixels of the current image (1, 1), (1, 2), The grayscale values of (1,3), (2,1), (2,2), (2,3), (3,1), (3,2), and (3,3) are, for example, 150, 160 , 170, 180, 190, 170, 170, 160, and 180. The grayscale value of the pixel (3, 3) of the previous image is, for example, 148 (as shown in FIG. 4). The following is an example of obtaining the grayscale value of the pixel (3, 3) of the output image.

First, in step S201, the calculating unit 30 can calculate an adjustment value according to a pixel of the current image and a plurality of reference pixels in the vicinity thereof. It is to be noted that there are a plurality of implementations of the step S201. In the following, an alternative embodiment is provided for those skilled in the art, and the invention is not limited thereto. In other embodiments, those skilled in the art can also change the implementation of step S201 according to their needs.

In step S201 of the present embodiment, a pixel of the current image is illustrated by taking a pixel (3, 3) as an example, and a plurality of reference pixels in the vicinity are pixels (1, 1), (1, 2). ), (1, 3), (2, 1), (2, 2), (2, 3), (3, 1), and (3, 2) are described as an example. The adjustment value can be calculated by subtracting the pixels of the pixel (3, 3) from the pixels (1, 1), (1, 2), (1, 3), (2, 1), (2, 2). Average grayscale values of ), (2, 3), (3, 1), (3, 2), and (3, 3). Due to pixels (1,1), (1,2), (1,3), (2,1), (2,2), (2,3), (3,1), (3,2) The average grayscale value of (3, 3) = (150 + 160 + 170 + 180 + 190 + 170 + 170 + 160 + 180) / 9 = 170, so in this embodiment, the adjustment value = 180-170 = 10.

It is worth mentioning that the purpose of step S201 is to calculate the adjustment value according to the contrast degree of the pixel (3, 3) of the current image and the pixels nearby, so as to perform image processing in the subsequent steps. That is to say, the adjustment obtained in step S201 The larger the whole value, the closer the pixel (3, 3) and the nearby pixels of the current image are to the high contrast image; otherwise, the smaller the adjustment value obtained in step S201, represents the pixel of the current image (3, 3) The pixels near and near it are getting closer to low contrast images. In general, when performing image processing, the sharpening effect of high contrast images is enhanced, and the sharpening effect of low contrast images is maintained or reduced. Taking the image of a portrait as an example, the above method can make the hair of the portrait more clear, and also make the skin of the portrait more feminine and avoid the occurrence of the class.

After step S201, step S202 is followed, and the weight coefficient is calculated according to the difference between the pixel of the current image and the grayscale value of the pixel of the previous image. In this embodiment, the difference between the grayscale value of the pixel of the current image and the pixel of the previous image is calculated by subtracting the grayscale value of the pixel (3, 3) of the current image from the pixel of the previous image ( 3, 3) gray scale value, that is, the above gray scale value difference = 180-148 = 32. Then, the weight coefficient can be obtained by dividing the grayscale value difference by the reference value. For example, the reference value can be, for example, 64, so the weight coefficient = 32/64 = 0.5. Those skilled in the art will appreciate that the reference value of 64 is only an alternative embodiment. In other embodiments, those skilled in the art can also change the reference value according to their needs.

The weight coefficient obtained in step S202 is the degree that the pixel (3, 3) representing the current image approaches the dynamic picture. The larger the weight coefficient, the closer the current image's pixels (3, 3) to the dynamic picture; conversely, the smaller the weight coefficient means that the current image's pixels (3, 3) approach the static picture. In general, when a liquid crystal display displays a dynamic picture, it is easy to have a problem of motion blur. Although the above problems can be solved by image sharpening, they will also generate noise. The weight coefficient obtained in step S202 can facilitate the subsequent image Therefore, it is only necessary to sharpen the dynamic picture, and the static picture can be selected without sharpening or sharpening effect.

After obtaining the adjustment value and the weight coefficient from steps S201 and S202, step S203 may be performed to calculate the weight adjustment value according to the adjustment value and the weight coefficient. In this embodiment, the weighting adjustment value may be obtained by multiplying the adjustment value and the weighting coefficient with each other, that is, the weighting adjustment value=10×0.5=5.

It is worth mentioning that step S203 is to multiply the adjustment value by the weight coefficient to obtain a weight adjustment value. Therefore, if one of the adjustment value and the weight coefficient is 0, the weight adjustment value is 0. That is to say, if the current pixel (3, 3) is a uniform smooth picture or a static picture, the weight adjustment value will be 0, which can effectively avoid the abrupt noise.

Then, step S204 is performed to adjust the pixels of the current image according to the weight adjustment value, thereby generating a pixel of the output image. For example, the grayscale value of the pixel (3, 3) of the current image can be added to the weight adjustment value to obtain the grayscale value of the pixel (3, 3) of the output image, that is, the pixel of the output image. The grayscale value of (3, 3) = 180 + 5 = 185.

In summary, steps S201 to S204 of the present embodiment may determine the degree of sharpening of image processing according to the extent to which each block of the current image approaches the dynamic picture and the extent to which each block approaches a high contrast image. Therefore, not only can the image detail be improved to improve the dynamic blurring problem, but also the abrupt noise can be avoided.

Through the teachings of the above steps S201 to S204, those skilled in the art should be able to easily obtain the grayscale values of other pixels of the output image. For example, the grayscale value of the pixel (3, 3) of the above output image is determined by the current image. Pixels (1,1), (1,2), (1,3), (2,1), (2,2), (2,3), (3,1), (3,2) and (3,3) and the pixels of the previous image (3,3) are obtained. Similarly, the steps S201 to S204 may be repeatedly performed, and the grayscale values of the pixels (3, 4) of the output image may also be from the pixels (1, 2), (1, 3), (1, 4) of the current image. , (2, 2), (2, 3), (2, 4), (3, 2), (3, 3) and (3, 4) and the previous image of the pixel (3, 4). This will give you a complete output image.

It is worth mentioning that although the image processing apparatus and the method thereof have been drawn out in a possible form in the above embodiments, those having ordinary knowledge in the art should know that each manufacturer designs the image processing apparatus and its method. They are different, so the application of the invention is not limited to this possible type. In other words, the adjustment value is calculated based on the pixels of the current image and the plurality of reference pixels in the vicinity thereof. In addition, the weight coefficient is calculated according to the difference between the gray level value of the pixel of the current image and the pixel of the previous image. In addition, adjusting the current image according to the adjustment value and the weight coefficient is in line with the spirit of the present invention. In the following, several embodiments will be described to enable those skilled in the art to further understand the spirit of the invention and to practice the invention.

Referring to FIG. 2 again, steps S201 and S202 are sequentially performed in the above embodiment, but the present invention is not limited thereto. In other embodiments, step S202 may be performed first and then step S201 may be performed, so that similar effects can be achieved.

Referring to FIG. 2 and FIG. 4 again, in the above embodiment, step S201 uses the pixels (1, 1), (1, 2), (1, 3), (2, 1) of the current image. (2, 2), (2, 3), (3, 1), (3, 2) and (3, 3) and the pixels of the previous image (3, 3) to calculate the pixels on the output image (3 3) The adjustment value, but the invention is not limited thereto. For example, FIG. 5 is a schematic diagram of an image block of another embodiment. In the embodiment of FIG. 5, step S201 can also be based on the pixels (1, 1), (1, 2), (1, 3), (2, 1), (2, 2), (2) of the current image. , 3), (3, 1), (3, 2) and (3, 3) and the pixels (2, 2) of the previous image to calculate the adjustment values of the pixels (2, 2) of the output image.

For another example, FIG. 6 is a schematic diagram of an image block of still another embodiment. In the embodiment of FIG. 6, step S201 can also be based on the pixels (1, 1), (1, 2), (1, 3), (1, 4), (2, 1), (2) of the current image. , 2), (2, 3), (2, 4), (3, 1), (3, 2), (3, 3), (3, 4), (4, 1), (4, 2 ), (4, 3) and (4, 4) and the pixels (4, 4) of the previous image to calculate the adjustment values of the pixels (4, 4) of the output image. It is worth noting that in the process of generating each pixel of the output image, it is recommended to use the same implementation, otherwise unnecessary noise may occur.

In summary, the present invention calculates an adjustment value based on the pixels of the current image and a plurality of reference pixels in the vicinity thereof. In addition, the weight coefficient is calculated according to the difference between the gray level value of the pixel of the current image and the pixel of the previous image. In addition, the current image is adjusted according to the adjustment value and the weight coefficient. In this way, the current image can be adaptively adjusted, which not only can effectively enhance the image detailing to improve the dynamic blurring problem, but also avoid the abrupt noise. Moreover, the embodiments of the present invention have the following advantages: 1. The meter reading circuit can effectively reduce the calculation amount of the calculation unit, and can add Speed calculation speed.

2. The sharpening effect of the high contrast image can be enhanced through step S201, and the sharpening effect of the low contrast image can be maintained or reduced.

3. The weight coefficient obtained in step S202 can facilitate subsequent image processing, so that only the dynamic image is sharpened, and the static image can be selected without sharpening or sharpening effect.

4. If the pixel (3, 3) of the current image is a uniform smooth picture or a static picture, the weighted adjustment value obtained in step S203 will be 0, which can effectively avoid the abrupt noise generation.

The present invention has been disclosed in several embodiments, and is not intended to limit the scope of the present invention. Any one of ordinary skill in the art can make a few changes and refinements without departing from the spirit and scope of the invention. Therefore, the scope of the invention is defined by the scope of the appended claims.

10‧‧‧Image processing device

20‧‧‧ storage unit

30‧‧‧Computation unit

S201~S204‧‧‧ steps of image processing method

(1,1), (1,2), (1,3), (1,4), (2,1), (2,2), (2,3), (2,4), (3 , 1), (3, 2), (3, 3), (3, 4) ‧ ‧ pixels

1 is a block diagram of an image processing apparatus in accordance with a first embodiment of the present invention.

2 is a flow chart of an image processing method in accordance with a first embodiment of the present invention.

Figure 3 is a schematic illustration of an image in accordance with a first embodiment of the present invention.

4 is a schematic diagram of the image block of FIG. 3.

FIG. 5 is a schematic diagram of an image block of another embodiment.

Figure 6 is a schematic diagram of an image block of still another embodiment.

S201~S204‧‧‧ steps of image processing method

Claims (8)

An image processing method includes: calculating a first adjustment value according to a first pixel of a current image and a plurality of first reference pixels in the vicinity thereof; according to the first pixel of the current image and the first image of a previous image Calculating a first weight coefficient, wherein the first weight coefficient is a difference between a first pixel of the current image and a gray level value of the first pixel of the previous image by a reference value The first weighting coefficient is obtained according to the first adjustment value and the first weighting coefficient; and the first pixel of the current image is adjusted according to the first weighting adjustment value, thereby generating an output. The first pixel of the image. The image processing method of claim 1, further comprising: calculating a second adjustment value according to the second pixel of the current image and a plurality of second reference pixels in the vicinity thereof; a difference between the two pixels and the gray level value of the second pixel of the previous image, calculating a second weight coefficient; calculating a second weighting adjustment value according to the second adjustment value and the second weight coefficient; and according to the second The weighted adjustment value adjusts the second pixel of the current image to generate a second pixel of the output image. The image processing method of claim 1, wherein the first pixel of the current image and the first reference pixels in the vicinity thereof constitute an M×N pixel array, wherein M and N are greater than A positive integer equal to 2. The image processing method of claim 3, wherein the first pixel of the current image is located at a lower right corner of the pixel array. The image processing method of claim 1, wherein the first adjustment value is a grayscale value of the first pixel of the current image minus a first pixel of the current image and the vicinity of the first pixel The average grayscale value of a reference pixel. The image processing method of claim 1, wherein the first weight adjustment value is equal to the first adjustment value multiplied by the first weight coefficient. The image processing method of claim 1, wherein the gray level value of the first pixel of the output image is equal to the gray level value of the first pixel of the current image plus the first weight adjustment value. An image processing device includes: a storage unit for storing a previous image; and a computing unit coupled to the storage unit to calculate an adjustment value according to a pixel of the current image and a plurality of reference pixels in the vicinity thereof, and The pixel of the current image is different from the grayscale value of the pixel of the previous image, and a weighting coefficient is calculated, and the calculating unit calculates a weighted adjustment value according to the adjusted value and the weighting coefficient, and adjusts the weighting adjustment value according to the weighting adjustment value. The pixel of the current image is used to generate a pixel of the output image, wherein the pixel of the current image, the pixel of the previous image and the pixel of the output image correspond to each other, wherein the first weight coefficient is the current image The difference between the first pixel and the pixel grayscale value of the previous image is divided by a reference value to obtain the first weight coefficient.