VIDEO PROCESSING METHOD AND APPARATUS BASED ON SMOOTH DETECTION

Abstract

The present invention discloses a video processing apparatus. The video processing apparatus includes a smooth detecting module and a video processing module. The smooth detecting module categorizes a target position in a target image of a video data as a smooth position or a non-smooth position to generate a categorizing result. The video processing module performs at least one of a plurality of video processing operations upon the target position according to the categorizing result.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagram of a video processing apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram of a video processing apparatus according to another embodiment of the present invention.

DETAILED DESCRIPTION

In practice, an image area having an artifact (e.g. color spreading or bleeding) caused by video processing operation may result from the image area being adjacent to a color edge, or the image area being a grid image. Otherwise, a smooth image area does not often contain an artifact when utilizing the video processing operations in the spatial domain. Therefore, the present invention introduces a concept of smooth detection. The concept of smooth detection is applied in video processing as a basis for selecting video processing operations in the spatial domain or in the temporal domain. Utilizing smooth detection to avoid selecting an inappropriate video processing operation will therefore reduce the artifact.

Please refer to FIG. 1. FIG. 1 is a diagram of a video processing apparatus 100 according to an embodiment of the present invention. The video processing apparatus 100 comprises a smooth detecting module 120 and a video processing module 160. The smooth detecting module 120 is utilized for performing a smooth detection to categorize a target position in a target image of input video data as a smooth position or a non-smooth position. The video processing module 160 is utilized for selectively performing at least one of a plurality of video processing operations on the target position according to a result generated by categorizing the target position utilizing the smooth detection. The above-mentioned video processing operations include video processing operations in the spatial domain and video processing operations in the temporal domain.

Various existing schemes are capable of realizing the operation of the smooth detecting module 120. For example, one of the schemes utilizes an edge detection to detect edge information included in the target image to categorize the target position as a smooth position or a non-smooth position. However, the smooth detecting module 120 can also detect differences between pixel values of a plurality of pixels in a pixel window corresponding to the target position for categorizing the target position as a smooth position or a non-smooth position. By using the above-mentioned schemes, the smooth detecting module 120 can correctly categorize the target position as a smooth position or a non-smooth position by only detecting information (pixel values or edge information) included in a single image (i.e. the target image). It is not necessary for the smooth detecting module 120 to detect information in other images. The above-mentioned schemes are examples illustrating the operation of the smooth detection. The scope of the present invention is not limited to these implementations. In this embodiment, after the result generated by the smooth detecting module 120 has been received by the video processing module 160, the video processing module 160 performs a video processing operation in the spatial domain upon the target position if the target position is categorized as a smooth position. Otherwise, the video processing module 160 performs a video processing operation in the temporal domain upon the target position if the target position is categorized as a non-smooth position.

Please refer to FIG. 2. FIG. 2 is a diagram of a video processing apparatus 200 according to another embodiment of the present invention. A smooth detecting module 220 operates in conjunction with a motion detecting module 240. As shown in FIG. 2, the video processing apparatus 200 includes the smooth detecting module 220, the motion detecting module 240, and a video processing module 260. The motion detecting module 240 is utilized for performing a motion detection to categorize a target position in a target image of input video data as a motion position or a still position. The operation of the smooth detecting module 220 is similar to that of the smooth detecting module 120, and the illustration of the smooth detecting module 220 is therefore not detailed for brevity. The video processing module 260 is utilized for selectively performing at least one of a plurality of video processing operations upon the target position according to a result generated by categorizing the target position with the smooth detecting module 220 and a result generated by categorizing the target position with the motion detecting module 240. Similarly, the above-mentioned video processing operations include video processing operations in the spatial domain and video processing operations in the temporal domain.

After receiving the results generated by the smooth detecting module 220 and the motion detecting module 240, the video processing module 260 performs a video processing operation in the spatial domain upon the target position if the target position is categorized as a smooth position, no matter the target position is categorized as a motion position or a still position by the motion detecting module 240. Otherwise, the video processing module 260 performs a video processing operation in the spatial domain upon the target position if the target position is categorized as a non-smooth position and a motion position. Additionally, if the target position is categorized as a non-smooth position and as a still position, the video processing module 260 performs a video processing operation in the temporal domain upon the target position. As mentioned above, the video processing module 260 performs a video processing operation in the temporal domain upon a target position only when the target position is categorized as a non-smooth position and a still position simultaneously. Therefore, a video artifact in output video data by erroneously utilizing the video processing operation in the temporal domain upon a target image can be avoided by the video processing apparatus 200.

Those skilled in the art will readily observe that numerous modifications and alterations of the device and method may be made while retaining the teachings of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.

Claims

1. A method for processing video data, comprising: categorizing a target position in a target image of the video data as one of a smooth position and a non-smooth position to generate a categorizing result; andperforming at least one of a plurality of video processing operations upon the target position according to the categorizing result;wherein the categorizing step utilizes a smooth detection.

2. The method of claim 1, wherein the categorizing step comprises: detecting the target image for an edge information to categorize the target position.

3. The method of claim 1, wherein the categorizing step comprises: detecting differences between pixel values of a plurality of pixels in a pixel window to categorize the target position;wherein the pixel window corresponds to the target position.

4. The method of claim 1, wherein the step of performing at least one of the plurality of video processing operations comprises:: performing a video processing operation in a spatial domain upon the target position if the target position is categorized as the smooth position.

5. The method of claim 1, wherein the step of performing at least one of the plurality of video processing operations comprises: performing a video processing operation in a temporal domain upon the target position if the target position is categorized as the non-smooth position.

6. A method for processing video data, comprising: performing a smooth detection to categorize a target position in a target image of the video data as one of a smooth position and a non-smooth position to generate a smooth-detection result;performing a motion detection to categorize the target position as one of a motion position and a still position to generate a motion-detection result; andperforming at least one of a plurality of video processing operations upon the target position according to at least one of the smooth-detection result and the motion-detection result.

7. The method of claim 6, wherein the step of performing the smooth detection to categorize the target position comprises: detecting the target image for an edge information to categorize the target position as one of the smooth position and the non-smooth position.

8. The method of claim 6, wherein the step of performing the smooth detection to categorize the target position comprises: detecting differences between pixel values of a plurality of pixels in a pixel window to categorize the target position;wherein the pixel window corresponds to the target position.

9. The method of claim 6, wherein the step of performing at least one of the plurality of video processing operations comprises: performing a video processing operation in a spatial domain upon the target position if the target position is categorized as the smooth position, wherein the target position is one of the motion position and the still position.

10. The method of claim 6, wherein the step of performing at least one of the plurality of video processing operations comprises: performing a video processing operation in a spatial domain upon the target position if the target position is categorized as the non-smooth position and the motion position simultaneously.

11. The method of claim 6, wherein the step of performing at least one of the plurality of video processing operations comprises: performing a video processing operation in a temporal domain upon the target position if the target position is categorized as the non-smooth position and the still position simultaneously.

12. A video processing apparatus, comprising: a smooth detecting module, for categorizing a target position in a target image in a video data as one of a smooth position and a non-smooth position to generate a categorizing result; anda video processing module, coupled to the video processing module, for performing at least one of a plurality of video processing operations upon the target position according to the categorizing result generated by the smooth detecting module.

13. The video processing apparatus of claim 12, wherein the smooth detecting module detects the target image for an edge information to categorize the target position as one of the smooth position and the non-smooth position.

14. The video processing apparatus of claim 12, wherein the smooth detecting module detects differences between pixel values of a plurality of pixels in a pixel window to categorize the target position as one of the smooth position and the non-smooth position, wherein the pixel window corresponds to the target position.

15. The video processing apparatus of claim 12, wherein the video processing module performs a video processing operation in a spatial domain upon the target position if the target position is categorized as the smooth position.

16. The video processing apparatus of claim 12, wherein the video processing module performs a video processing operation in a spatial domain upon the target position if the target position is categorized as the non-smooth position.

17. A video processing apparatus, comprising: a smooth detecting module, for categorizing a target position in a target image of the video data as one of a smooth position and a non-smooth position;a motion detecting module, for categorizing the target position as one of a motion position and a still position; anda video processing module, coupled to the smooth detecting module and the motion detecting module, for performing at least one of a plurality of video processing operations upon the target position according to at least one of a categorizing result generated by the smooth detection module and a categorizing result generated by the motion detection module.

18. The video processing apparatus of claim 17, wherein the smooth detecting module detects an edge information included in the target image to categorize the target position as one of the smooth position and the non-smooth position.

19. The video processing apparatus of claim 17, wherein the smooth detecting module detects differences between pixel values of a plurality of pixels in a pixel window to categorize the target position as one of the smooth position and the non-smooth position, wherein the pixel window corresponds to the target position.

20. The video processing apparatus of claim 17, wherein the video processing module performs a video processing operation in a spatial domain upon the target position if the target position is categorized as the smooth position, wherein the target position is categorized as one of the motion position and the still position.

21. The video processing apparatus of claim 17, wherein the video processing module performs a video processing operation in a spatial domain upon the target position if the target position is categorized as the non-smooth position and the motion position simultaneously.

22. The video processing apparatus of claim 17, wherein the video processing module performs a video processing operation in a temporal domain upon the target position if the target position is categorized as the non-smooth position and the still position simultaneously.