TWI420415B - Image processing apparatus and method - Google Patents

Description

Image processing device and method

The present invention relates to an image processing apparatus and method, and more particularly to an image processing apparatus and method capable of changing image resolution. While changing the resolution of an image, the image processing apparatus and method can not only eliminate edge resolution of image details. The jagged effect of the degree of change also preserves the sharpness of the edges of the detail.

In recent years, due to the rapid development of display panel technology, the size and resolution of displays have also grown rapidly. In response to different types of applications, displays of various resolutions have been developed today. However, for a fixed-size picture, if it is input to a display with different resolutions, it is necessary to perform image scaling adjustment according to the resolution of the display.

Figure 1 is a schematic diagram showing image scaling (Scaling). Image 10 represents a picture with a resolution of 720 x 480. When the image 10 is input to a display having a large resolution (for example, 800 x 600), the image 10 needs to be enlarged to an image 20 having a resolution of 800 x 600. Similarly, when the image 10 is input to a display having another large resolution (e.g., 1366 x 768), the image 10 needs to be further enlarged to an image 30 having a resolution of 1366 x 768. However, the scaling process usually causes distortion of the image. For example, in the Nearest algorithm, the pixel value closest to the interpolation point is used as the pixel value for this interpolation point. However, the result of the enlargement will be as mosaic processing on the image, so the effect of the recent algorithm is not ideal. If a more advanced algorithm, such as a bilinear interpolation algorithm, is used, the average pixel value of two pixels adjacent to the interpolation point is taken as the pixel value of the interpolation point. This method can improve the recent algorithm. The disadvantages, but the details of the magnified image are blurred and there is a severe jagged at the edge of the detail of the image.

Therefore, in order to freely adjust the image size under any display and maintain the image quality of the original image, a new image processing device and method are needed, which can not only eliminate the edge of the image detail while changing the resolution of the image. The jagged effect due to the change in resolution also preserves the sharpness of the edges of the detail.

According to an embodiment of the invention, an image processing apparatus includes a high pass filter, a low pass filter, a first scaler, a second scaler, an image enhancement unit, and an adder. The high pass filter receives an input image signal, filters the input image signal, and outputs a high pass filtered image signal. The low pass filter receives the input image signal, filters the input image signal, and outputs a low pass filtered image signal. The first scaler receives the high-pass filtered image signal, adjusts a resolution of the high-pass filtered image signal according to a resolution adjustment signal, and outputs a first output image signal. The second scaler receives the low-pass filtered image signal, adjusts a resolution of the low-pass filtered image signal according to the resolution adjustment signal, and outputs a second output image signal. The image enhancement unit receives the first output image signal, enhances the first output image signal, and outputs an enhanced image signal. The adder adds the enhanced image signal and the second output image signal to generate an output image signal.

According to another embodiment of the present invention, an image processing method includes: receiving an input image signal; processing the input image signal in different image processing paths to respectively generate a high-pass filtered image signal and a low-pass filtering Image signal; adjusting a resolution of the high-pass filtered image signal and the low-pass filtered image signal according to a resolution adjustment signal to output a first output image signal and a second output image signal; The first output image signal outputs an enhanced image signal; and the enhanced image signal and the second output image signal are added to generate an output image signal.

In order to make the manufacturing, operating methods, objects and advantages of the present invention more apparent, the following detailed description of the preferred embodiments and the accompanying drawings

Example:

The invention provides a novel image processing method and device. When changing the resolution of the image, the input image signal is divided into frequency bands for scaling, to reduce the jagged effect of the edge of the image detail due to the change of resolution. And enhance the sharpness of the edge of the detail to maintain the quality of the original image.

2 is a flow chart showing an image processing method according to an embodiment of the present invention. First, an input video signal is received (step S202). Then, the input image signal is processed to generate a high pass filtered processed image signal and a low pass filtered processed image signal (step S204). Then, adjusting the resolution of the high-pass filtered image signal and the low-pass filtered image signal according to a resolution adjustment signal to output a first output image signal and a second output image signal (step S206). Then, the first output image signal is enhanced, and an enhanced image signal is output (step S208). Finally, the enhanced image signal and the second output image signal are added to generate an output image signal (step S210). In conjunction with the image processing apparatus shown in Fig. 3, the image processing apparatus and method of the present invention will be described in more detail in the following paragraphs.

Figure 3 is a diagram showing an image processing apparatus according to an embodiment of the present invention. The image processing device 300 includes filters 301 and 302, scalers 303 and 304, an image enhancement unit 305, and an adder 306. According to an embodiment of the invention, the image processing device 300 receives the input image signal S _{IM_IN} and processes the input image signal in two different image processing paths. As shown in FIG. 3, the two different image processing paths are respectively the first image processing path passing through the high pass filter 301, the scaler 303, the image enhancing unit 305, and the adder 306, and the low pass filter 302, zooming The second image processing path of the processor 304 and the adder 306.

As shown, the high pass filter 301 receives and filters the input image signal S _{IM_IN} to output a high pass filtered image signal S _{IM_F1} . Similarly, the low pass filter 302 receives and filters the input image signal S _{IM_IN} to output a low pass filtered image signal S _{IM — F2} .

Then, the scalers 303 and 304 respectively adjust the resolution of the high-pass filtered image signal S _{IM_F1} and the low-pass filtered image signal S _{IM_F2} according to a resolution adjustment signal S _ADJ on different signal processing paths to respectively output The first output image signal S _{IM_O1} and the second output image signal S _{IM_O2} . It is worth noting that in order to enhance the fineness of the image, to enhance the sharpness of the detail edge of the image, the pixel values of the high-frequency first output image signal S _{IM_O1} are further enhanced by the image intensifying unit 305 to generate a reinforcement. The image signal S _{IM_EN is passed} . Finally, the adder 306 adds the enhanced image signal S _{IM_EN} to the second output image signal S _{IM_O2} to generate a final output image signal S _{IM_OUT} .

The scalers 303 and 304 can adjust the image resolution according to various interpolation algorithms, including a Nearest algorithm, a Bilinear Interpolation algorithm, and a Bicubic Interpolation algorithm. , a B-Spline algorithm, or other algorithms that achieve the same degree of image resolution.

As described above, in order to enhance the sharpness of the edge of the detail, the image intensifying unit 305 is further added to the high frequency signal processing path. According to an embodiment of the invention, the image enhancement unit 305 can amplify each pixel value of the first output image signal according to a magnification to generate the enhanced image signal S _{IM_EN} .

Figure 4 shows an example of a one-dimensional (1D) image. To simplify and clearly illustrate the image processing concept of the present invention, the one-dimensional image 40 includes only four pixel points, each having pixel values of 28, 50, 60, and 66. It should be noted that one of the dimensional images shown in FIG. 4 is not intended to limit the scope of the present invention. Anyone skilled in the art will appreciate that the image processing method and apparatus of the present invention can also be applied to images having one, two, or three dimensions of various resolutions, and thus anyone skilled in the art is not The scope of protection of the present invention is defined by the scope of the appended claims.

According to an embodiment of the invention, the high pass filter 301 can be a high pass filter [-1, 2, -1] and the low pass filter 302 can be a low pass filter [1, 2, 1], however It should be noted that the filters 301 and 302 may also be other one-dimensional, two-dimensional or three-dimensional filters in accordance with the dimensions of the input image signal, and thus the above embodiments are not intended to limit the scope of the present invention. Assuming that a new pixel is to be interpolated between pixels 50 and 60 of the one-dimensional image 40, the pixels 50 and 60 can be filtered by the high pass filter 301 and the low pass filter 302, respectively, to obtain the filtered pixel. point. Taking high-pass filtering as an example, the filtering result of pixel 50 is (50*2+60(*-1)+28*(-1)=12), and the filtering result of pixel 60 is (60*2+50). (*-1)+66*(-1)=4). After filtering the filtered results [12, 4] and dividing them by the absolute value of the high-pass filter, the normalized operation is performed, and the filtered pixel points [3, 1] can be obtained. Similarly, the filtered pixel points [47, 59] are obtained after the pixels 50 and 60 pass through the low pass filter.

Then, the interpolated algorithm generates the pixel value of the interpolation point according to the filtered result. For example, using bilinear interpolation, the pixel value of the interpolation point can be obtained on the high-frequency signal processing path (3*0.5+ 1*0.5=2), and the pixel value of the interpolation point can be obtained on the low-frequency signal processing path (47*0.5+59*0.5=53). According to an embodiment of the present invention, in order to enhance the sharpness of the edge of the image detail, the high-frequency interpolation point can be first amplified according to a magnification (for example, 1.5 times), and then added to the low-frequency interpolation point, so that this can be obtained. The pixel value of the last output of the interpolation point is (2*1.5+53=56). According to the image processing steps described above, according to the indication of the resolution adjustment signal, the above steps are applied to each pixel of the input image signal that needs to generate an interpolation value, and then an output having the required output resolution is generated. Image signal. In this way, not only the jagged effect of the output image signal due to the change of resolution at the edge of the image detail can be avoided, but also the fineness of the image can be preserved, and the image quality of the original image can be maintained.

The present invention has been described above with reference to the preferred embodiments thereof, and is not intended to limit the scope of the present invention, and the invention may be modified and modified without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

10, 20, 30, 40. . . image

300. . . Image processing device

301. . . High pass filter

302. . . Low pass filter

303, 304. . . Scaler

305. . . Image enhancement unit

306. . . Adder

S _ADJ , S _{IM_EN} , S _{IM_F1} , S _{IM_F2} , S _{IM_IN} , S _{IM_O1} , S _{IM_O2} , S _{IM_OUT} . . . signal

Figure 1 is a schematic diagram showing image scaling (Scaling).

2 is a flow chart showing an image processing method according to an embodiment of the present invention.

Figure 3 is a diagram showing an image processing apparatus according to an embodiment of the present invention.

Figure 4 shows an example of a one-dimensional (1D) image.

300. . . Image processing device

301. . . High pass filter

302. . . Low pass filter

303, 304. . . Scaler

305. . . Image enhancement unit

306. . . Adder

S _ADJ , S _{IM_EN} , S _{IM_F1} , S _{IM_F2} , S _{IM_IN} , S _{IM_O1} , S _{IM_O2} , S _{IM_OUT} . . . signal

Claims (6)

An image processing apparatus includes: a high-pass filter that receives an input image signal, filters the input image signal, and outputs a high-pass filtered image signal; a low-pass filter that receives the input image signal and filters the input image And outputting a low-pass filtered image signal; a first scaler receiving the high-pass filtered image signal, adjusting a resolution of the high-pass filtered image signal according to a resolution adjustment signal, and outputting a a first output image signal; a second scaler receiving the low pass filtered image signal, adjusting a resolution of the low pass filtered image signal according to the resolution adjustment signal, and outputting a second output image signal An image enhancement unit receives the first output image signal, enhances the first output image signal, and outputs an enhanced image signal; and an adder that adds the enhanced image signal to the second output image The signal is used to generate an output image signal. The image processing device of claim 1, wherein the first scaler and the second scaler adjust the resolution according to an interpolation algorithm, and the interpolation algorithm is a nearest (Nearest) calculation. Method, a bilinear interpolation (Bilinear Interpolation) algorithm, a bicubic interpolation algorithm, or a B-Spline algorithm. The image processing device of claim 1, wherein the image enhancement unit amplifies each pixel value of the first output image signal according to a magnification to generate the enhanced image signal. An image processing method includes: receiving an input image signal; processing the input image signal to respectively generate a high-pass filtered image signal and a low-pass filtered image signal; respectively adjusting the high-pass filtering according to a resolution adjustment signal And translating the image signal and the low-pass filtered image signal to output a first output image signal and a second output image signal; enhancing the first output image signal, and outputting a enhanced image signal And adding the enhanced image signal and the second output image signal to generate an output image signal. The image processing method of claim 4, wherein the step of adjusting the resolution of the image signal is adjusted according to an interpolation algorithm, and the interpolation algorithm is a nearest algorithm, A bilinear interpolation algorithm, a bicubic interpolation algorithm, or a B-Spline algorithm. The image processing method of claim 4, wherein the step of enhancing the first output image signal further comprises: amplifying each pixel value of the first output image signal according to a magnification to generate the enhanced image signal; .