JP2008160466A - Video signal processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve conventional problems that noise reduction and contour enhancement are performed using coding information acquired from a decoder, and to allow electronic devices which are not provided with a decoder and cannot acquire coding information, such as television receivers, to flexibly perform noise reduction and contour enhancement. <P>SOLUTION: A block noise detection circuit 2 extracts impulse state differentiated points from differentiated signals that inputted decoded video signals are differentiated as isolated differentiated points, and then detects the number of isolated differentiated points on a screen, the absolute value of which is equal to or greater than a prespecified threshold, as a block noise detection result. Corresponding to the detection result from the block noise detection circuit 2, a noise reduction circuit 3 performs block noise reduction processing for decoded video signals if required, and then outputs them to a contour enhancement circuit 4. The contour enhancement circuit 4 performs contour enhancement processing for the decoded video signals outputted from the noise reduction circuit 3, and then outputs them to an output terminal 5 if required. Thereby, a decoder is not required for noise reduction and contour enhancement. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a video signal processing apparatus, and in particular, decodes a video signal compressed and encoded by a method such as MPEG (Moving Picture Experts Group) and performs processing for improving the image quality of the decoded video signal. The present invention relates to a video signal processing apparatus.

  Video signals that are continuous on the time axis are handled as digital signal information. At that time, for the purpose of transmitting and storing information with high efficiency, motion-compensated prediction is used using temporal redundancy, An encoding method such as MPEG that performs encoding and compression using orthogonal transform such as discrete cosine transform using redundancy is now widely used.

  In such an encoding method, the encoded video signal obtained by performing the above motion compensation prediction and orthogonal transform in block units of a rectangular area composed of a predetermined number of pixels adjacent in the horizontal direction and vertical direction of the video signal. Is transmitted or stored. The encoded video signal transmitted or accumulated is subjected to decoding processing complementary to that at the time of encoding in units of blocks, and restored to the original video signal.

  However, when the compression rate is set to a high value and the amount of information is kept small, a gradation difference is generated between a certain block and an adjacent block in the decoding process, and particularly an image with a gradual gradation change. In the portion, noise (block noise) due to a difference in gradation between adjacent blocks is easily noticeable.

  On the other hand, it is known to increase the definition of an image by performing an edge emphasis process as an image quality correction process for an input video signal. When the edge emphasis process is performed at the time of encoding and decoding, Noise (block noise) is emphasized, and the image quality cannot be sufficiently improved. Noise reduction and contour enhancement are contradictory. When noise reduction is performed using, for example, a noise removal filter, noise is reduced, but the definition of image quality is impaired.

  Thus, conventionally, as a method of performing these processes adaptively, a video signal processing apparatus that performs noise reduction and contour enhancement using encoded information obtained from a decoder has been proposed (see, for example, Patent Document 1). . In the conventional video signal processing apparatus described in Patent Document 1, noise reduction means for reducing the noise of the decoded video signal, image quality correction means for performing image quality correction on the noise-reduced video signal, and video signal Weighting means for setting a weighting coefficient corresponding to the amount of noise predicted from the coding information related to the amount of noise at the time of decoding, the step of the encoded block boundary of the decoded image, and the distance from the encoded block boundary; The image quality correction amount by the image quality correction means is controlled according to the weighting coefficient from the weighting means. That is, in this conventional video signal processing apparatus, after reducing the noise of the decoded video signal, image quality correction such as edge enhancement is performed on the video signal whose noise has been reduced.

Japanese Patent No. 3800704

  However, in the above-described conventional video signal processing device, noise reduction and contour enhancement are performed using encoded information obtained from a decoder (MPEG decoder), and thus an electronic device such as a television receiver without a decoder. However, there is a problem that encoded information cannot be obtained and noise reduction and contour enhancement cannot be performed adaptively.

  The present invention has been made in view of the above points, and an object thereof is to provide a video signal processing apparatus capable of adaptively performing noise reduction and contour enhancement even in an electronic device such as a television receiver having no decoder. To do.

  In order to achieve the above object, the video signal processing apparatus of the present invention divides each screen generated from the video signal into block units, which are image regions of a predetermined number of pixels, and performs compression encoding on the block unit basis. A video signal processing apparatus that selectively performs either noise reduction processing or contour enhancement processing on a decoded video signal composed of each screen obtained by decoding an encoded signal, An impulse-shaped differential point is extracted as an isolated differential point from the differential signal obtained by differentiating the video signal, and an isolated differential point whose absolute value of the isolated differential point is equal to or greater than a preset threshold is extracted, and the isolated differential is extracted. Block noise detection means for detecting the number of dots in one screen as a block noise detection result; and at least a block where the block noise of the decoded video signal is generated with respect to the decoded video signal A noise reduction process for correcting a difference in signal level at the boundary with the block is performed, and when the value of the block noise detection result is larger than a preset first value, the larger the value, the larger the correction amount. In addition, the noise reduction means for controlling the correction amount of the noise reduction processing, the contour enhancement processing for emphasizing the edge portion of the decoded video signal are performed on the decoded video signal, and the value of the block noise detection result is the first value. And a contour emphasizing unit for controlling the emphasis amount of the contour emphasis processing so that the emphasis amount of the contour emphasis increases as the value becomes smaller when the value is equal to or smaller than a predetermined second value smaller than To do.

  In this invention, since the block noise detection result is obtained based on the differential signal obtained by differentiating the decoded video signal, the block noise detection result can be obtained without using the encoding information obtained from the decoder, Further, noise reduction processing by the noise reduction means and contour enhancement processing by the contour enhancement means can be selectively performed according to the value of the block noise detection result.

In order to achieve the above object, the present invention provides the above block noise detecting means,
Differentiating means for differentiating the decoded video signal to generate a differential signal that is a signal sequence of discrete differential points, isolated differential point extracting means for extracting an impulse-like differential point from the differential signal as an isolated differential point, and isolated differential An absolute value conversion unit that converts the isolated differential point extracted by the point extraction unit into an absolute value, and an integration that generates an integrated signal by integrating the absolute value of the isolated differential point converted by the absolute value conversion unit in units of blocks A comparison result indicating that the signal level of the integral signal is equal to or higher than the threshold value, the comparison means for comparing the signal level of the integral signal with the preset threshold value, and the comparator means; It is characterized by comprising count means for outputting the count value as a block noise detection result. In this invention, since the block noise detection result is obtained based on the differential signal obtained by differentiating the decoded video signal, the block noise detection result can be obtained without using the coding information obtained from the decoder.

  According to the present invention, a block noise detection result can be obtained without using encoded information obtained from a decoder by obtaining a block noise detection result based on a differential signal obtained by differentiating a decoded video signal. Therefore, even in an electronic device such as a television receiver that does not have a decoder, it is possible to select noise reduction and contour enhancement according to the block noise detection result, thereby achieving good image quality according to the block noise state. Can be provided to the viewer.

  Next, an embodiment of the present invention will be described with reference to the drawings. In the present invention, as a general feature of the MPEG decoded image, it is better to perform edge enhancement for a video signal with less block noise such as a still image because the image quality is relatively good with less noise. On the other hand, for video signals that are intensely moving and have a lot of block noise, it has been made by paying attention that noise reduction is a better impression.

  FIG. 1 is a block diagram showing an embodiment of a video signal processing apparatus according to the present invention. First, the outline will be described. Decoded video signals obtained by decoding an MPEG-encoded signal that has entered the input terminal 1 are supplied to a block noise detection circuit 2 and a noise reduction circuit 3, respectively. Is done. The block noise detection circuit 2 detects block noise with a configuration described later, and supplies the detection result to the noise reduction circuit 3 and the contour enhancement circuit 4 respectively.

  The noise reduction circuit 3 performs block noise reduction processing on the decoded video signal as necessary in accordance with the detection result from the block noise detection circuit 2 and then supplies it to the contour enhancement circuit 4. The contour emphasis circuit 4 performs contour emphasis processing on the decoded video signal output from the noise reduction circuit 3 as necessary, and outputs it to the output terminal 5.

  Next, details of the present embodiment will be described. 2 is a block diagram of an embodiment of the block noise detection circuit 2 in FIG. 1, and FIG. 3 is a diagram for explaining the block noise detection operation of FIG. As a feature of block noise in a decoded video signal obtained by decoding an encoded signal encoded by the MPEG method, there is a feature that a steep step appears at a block boundary between adjacent blocks. The block noise detection circuit 2 of the present embodiment uses this feature to detect block noise.

  In FIG. 2, an MPEG decoded video signal that has entered the input terminal 20 (corresponding to the input terminal 1 in FIG. 1) is first input to a differentiator 21 and differentiated. If the input decoded video signal is a discrete signal as shown in FIG. 3A, the differentiator 21 obtains a differential point as shown in FIG. 3B and supplies it to the isolated differential point extraction circuit 22. . The isolated differential point extraction circuit 22 extracts an impulse-shaped differential point whose adjacent differential point is 0 among the input differential points as an isolated differential point (FIG. 3C) and calculates an absolute value. This is supplied to the conversion circuit 23.

  The absolute value conversion circuit 23 converts the input isolated differential point into an absolute value, and supplies the obtained absolute value signal to the integrator 24. The integrator 24 integrates the input absolute value signal in units of blocks. Thus, when the isolated differential point shown in FIG. 3C is output from the isolated differential point extraction circuit 22, the integrator 24 outputs an integrated signal as shown in FIG.

  Here, for example, if one block consists of 8 pixels, the integrator 24 adds isolated differential points converted into absolute values every 8 pixels. Then, the integral signal has a large value at the steep step portion of the block boundary.

  The value of this integration signal is compared with a predetermined threshold level indicated by Th in FIG. 3D by the comparator 25, and the comparison result exceeding the threshold level Th is counted by the counter 26. Here, if the period counted by the counter 26 is, for example, one frame, how much block noise is generated in one frame can be known from the counter value, and block noise can be detected. This counter value is output from the output terminal 27 as a block noise detection signal.

  Returning again to FIG. The noise reduction circuit 3 of FIG. 1 has a configuration as shown in the block diagram of FIG. 4 disclosed by the present inventor in Japanese Patent Laid-Open No. 2001-119695, for example. As shown in FIG. 4, the noise reduction circuit 3 includes an isolated differential point extraction circuit 31 that differentiates an input video signal to obtain a differential signal and extracts and outputs an isolated differential point in the differential signal, and block noise is generated. A filter 32 that corrects a difference in signal level at the boundary between the current block and the adjacent block, a delay circuit 33 that delays the input video signal for a predetermined time, and an addition that adds the output signal of the filter 32 and the output signal of the delay circuit 33 And a vessel 34.

  In FIG. 4, the noise reduction circuit 3 first supplies the MPEG decoded video signal that has entered the input terminal 30 (corresponding to the input terminal 1 in FIG. 1) to the isolated differential point extraction circuit 31 and the delay circuit 33, respectively. The isolated differential point extraction circuit 31 generates a differential signal of the input decoded video signal, outputs only the differential value protruding from the adjacent pixel by the logical filter, and supplies it to the filter 32. Here, when the input decoded video signal has a waveform having a gradation difference at the block boundary as shown in FIG. 5 (a), the isolated differential point extraction circuit 31 is shown in FIG. 5 (b). Such isolated differential data is output.

  The filter 32 is a digital filter, which sequentially delays the isolated differential data from the isolated differential point extraction circuit 31 for each pixel, adds predetermined weights to the isolated differential data corresponding to the adjacent five pixels, and adds them. Then, the correction data as shown in FIG. 5C is generated by correcting the difference in signal level at the boundary between the block where the block noise is generated and the adjacent block, and is output to the adder 34.

  On the other hand, the delay circuit 33 delays the input decoded video signal by two pixels and outputs the decoded video signal shown in FIG. The adder 34 adds the correction data from the filter 32 and the delayed decoded video signal from the delay circuit 33, and as shown by a solid line in FIG. The decoded video signal is output to the output terminal 35. Adjustment of the noise reduction effect of the noise reduction circuit 3 can be performed by changing the number of stages and coefficients of the filter 32.

  Returning again to FIG. The decoded video signal that has been subjected to noise reduction processing according to need and output from the noise reduction circuit 3 is supplied to the contour enhancement circuit 4, where the contour has different effects depending on the detection result from the block noise detection circuit 2. After the enhancement process, the decoded video signal subjected to the noise reduction process or the outline enhancement process is output to the output terminal 5.

  The contour emphasis circuit 4 is configured, for example, as shown in the block diagram of FIG. 6 disclosed in Japanese Patent No. 3233331 by the applicant. In FIG. 6, the decoded video signal output from the noise reduction circuit 3 is input to the input terminal 40 and supplied to the secondary differentiation circuit 41 and the delay circuits 45 and 46, respectively. The secondary differentiation circuit 41 extracts a high frequency component of the input decoded video signal (luminance signal) as a secondary differential waveform using a known aperture circuit or the like. Here, assuming that the input decoded video signal (luminance signal) has a waveform having a contour portion shown in FIG. 7A, the secondary differential circuit 41 outputs a secondary differential waveform shown in FIG. 7B.

  The secondary differential waveform output from the secondary differential circuit 41 is limited in amplitude by a limiter 42 with a predetermined threshold (limiting level), and the amplitude portion below the predetermined limiting level is output as it is. As a result, a signal having a waveform shown in FIG. 7C is extracted from the limiter 42 and supplied to the differentiating circuit 43 to be a signal having a differentiated waveform shown in FIG. 7D. At this time, a sharper differential waveform is obtained for a signal having a large amplitude than a signal having a small amplitude.

  On the other hand, the decoded video signal (luminance signal) that has entered the input terminal 40 is delayed by a delay circuit 45 for a predetermined time, and then supplied to a differentiation circuit 47 to obtain a primary differential signal as shown in FIG. Is done. The correlation logic 44 is supplied with the differential signals output from the differentiating circuits 43 and 47, and outputs the differential signal from the differentiating circuit 43 only when the two input differential signals have the same polarity and have different polarities. When the signal is not output, a signal having a waveform shown in FIG. The output signal waveform of the correlation logic 44 is sharper than the original differential signal (the output differential signal of the differentiating circuit 47) as the amplitude of the input differential signal is large and the amount of the signal limited by the limiter 42 is large.

  The signal having the waveform shown in FIG. 7 (f) output from the correlation logic 44 is differentiated by the differentiating circuit 48 to be a correction signal shown in FIG. 7 (g). As a result, the decoded video signal (luminance signal) having the waveform shown in FIG. 7A from the input terminal 40 is delayed by the delay circuit 46 for a predetermined time for time adjustment with the output correction signal of the differentiation circuit 48. The addition signal 49 is added to the correction signal shown in FIG. 6G output from the differentiation circuit 48 to be output as a video signal subjected to the contour emphasis processing indicated by the solid line in FIG. It is output to the terminal 50 (corresponding to the output terminal 5 in FIG. 1). The effect of the contour emphasis circuit 4 can be adjusted by adjusting the limiting level of the limiter 42 in FIG.

  According to the present embodiment, the detection result obtained by the block noise detection circuit 2 in FIG. 1 is a count number corresponding to the block noise, and the noise reduction circuit 3 and the contour enhancement circuit 4 are simply turned on and off. However, according to the count number, as indicated by I in FIG. 8, the smaller the count number, the smaller the block noise. Therefore, the effect of edge enhancement is made larger than the effect of noise reduction (the amount of edge enhancement is increased). On the other hand, as indicated by II in FIG. 8, since the block noise increases as the count number increases, the noise reduction effect is greater than the edge enhancement effect (the correction amount of the noise reduction process) To be large).

  In addition, as shown in FIG. 8, when the count number according to the block noise obtained by the block noise detection circuit 2 is a value within a predetermined range, neither the edge enhancement process nor the noise reduction process is performed. When the value is smaller than the range, only the edge enhancement processing is performed, and when the value is larger than the range, only the noise reduction processing is performed.

  As described above, according to the present embodiment, the effects of the noise reduction circuit 3 and the contour enhancement circuit 4 are finely adjusted according to the count number corresponding to the block noise obtained by the block noise detection circuit 2. Therefore, it is possible to always provide the optimal image quality for the viewer.

  Although the noise reduction circuit 3 has been described with respect to the block noise reduction circuit in the above-described embodiment, when the block noise is generated, the mosquito noise is almost always generated. It is also possible to set a mosquito noise reduction circuit according to the detection result.

  FIG. 9 shows a block diagram of an example of a television receiver in which the video signal processing apparatus of the present invention is incorporated. In the figure, the same components as those in FIG. In FIG. 9, the television receiver 51 does not have an MPEG decoder, and the video signal processing device 60 of the present invention is provided between the HDMI receiver 52 and the processing circuit 53 of the television.

  In this television receiver 51, a decoded signal obtained by decoding an encoded signal compressed and encoded by the MPEG method is supplied to the HDMI receiver 52, and HDMI (High-Definition Multimedia Interface) which is a known interface standard is provided. After the interface is taken, the video signal processing device 60 is supplied to the video signal processing device 60, and the decoded video signal in the decoded signal is subjected to the video signal processing of the present invention described with reference to FIGS. It is supplied to the processing circuit 53. The television processing circuit 53 performs known television signal processing to display an image on a television screen (not shown) and to generate sound from a speaker (not shown).

  As described above, in the present invention, block noise is detected without using the encoded information obtained from the MPEG decoder. Therefore, the television receiver 51 having no MPEG decoder as shown in FIG. The noise reduction circuit 3 and the contour emphasis circuit 4 are adaptively operated according to the detection result of the block noise detection circuit 2 in the video signal processing device 60, so that it is always optimal for the viewer. It is possible to provide image quality.

  FIG. 10 shows a block diagram of an example of a reproducing apparatus incorporating the video signal processing apparatus of the present invention. In the figure, the same components as those in FIG. In FIG. 10, the playback device 55 does not have an MPEG decoder, and the video signal processing device 60 of the present invention is provided between the playback device 56 and the HDMI transmitter.

  In this playback device 55, the playback device 56 plays back the encoded signal compressed and encoded by the MPEG method from the recording medium, decodes the playback signal to obtain the decoded signal, and supplies it to the video signal processing device 60 of the present invention. The decoded video signal in the decoded signal is subjected to the video signal processing of the present invention described with reference to FIGS. 1 to 8 and then supplied to the HDMI transmitter 57. The HDMI transmitter 57 packetizes the decoded video signal and decoded audio signal in the input decoded signal by a known method and outputs them to the outside.

  Although the playback device 55 shown in FIG. 10 does not have an MPEG decoder, the noise reduction circuit 3 and the contour enhancement circuit 4 operate adaptively according to the detection result of the block noise detection circuit 2 in the video signal processing device 60. By doing so, it is possible to always provide the optimum image quality to the viewer who receives the packet output from the HDMI transmitter 57.

  Note that the present invention is not limited to the above embodiment, and for example, the connection order of the noise reduction circuit 3 and the contour emphasis circuit 4 may be reversed from that in FIG.

It is a block diagram of one embodiment of a video signal processing device of the present invention. It is a block diagram of an example of the block noise detection circuit in FIG. 3 is a timing chart for explaining the operation of FIG. 2. It is a block diagram of an example of the noise reduction circuit in FIG. FIG. 5 is a signal waveform diagram for explaining the operation of FIG. 4. It is a block diagram of an example of the outline emphasis circuit in FIG. FIG. 7 is a signal waveform diagram for explaining the operation of FIG. 6. It is a figure explaining adjustment of the effect of the noise reduction circuit and outline emphasis circuit by the block noise count number of the block noise detection circuit of FIG. It is a block diagram of an example of the television receiver incorporating the video signal processing device of the present invention. It is a block diagram of an example of the reproducing | regenerating apparatus incorporating the video signal processing apparatus of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Decoded video signal input terminal 2 Block noise detection circuit 3 Noise reduction circuit 4 Outline emphasis circuit 5 Output terminal 21 Differentiator 22 Isolated differential point extractor 23 Absolute value conversion circuit 24 Integrator 25 Comparator 26 Counter 51 Television receiver 55 Reproduction machine 60 Video signal processing apparatus of the present invention

Claims (2)

Each screen generated from the video signal is divided into blocks each of which is an image area having a predetermined number of pixels, and the decoding is made up of each screen obtained by decoding the encoded signal that is compression-coded in units of the blocks. A video signal processing apparatus that selectively performs either noise reduction processing or contour enhancement processing on a video signal,
An impulse-like differential point is extracted as an isolated differential point from the differential signal obtained by differentiating the decoded video signal, and an isolated differential point whose absolute value of the isolated differential point is equal to or greater than a preset threshold is extracted and extracted. Block noise detection means for detecting the number of isolated differential points on one screen as a block noise detection result;
The decoded video signal is subjected to the noise reduction processing for correcting a difference in signal level at a boundary between at least a block where the block noise of the decoded video signal is generated and an adjacent block, and the block noise detection result Noise reduction means for controlling the correction amount of the noise reduction processing so that the amount of correction increases as the value increases when the value is larger than a preset first value;
The contour enhancement process for enhancing the edge portion of the decoded video signal is performed on the decoded video signal, and the value of the block noise detection result is equal to or smaller than a preset second value that is smaller than the first value. In this case, the video signal processing apparatus further comprises contour emphasizing means for controlling the emphasis amount of the contour emphasizing process such that the smaller the value, the larger the emphasis amount of the contour emphasis. The block noise detecting means is
Differentiating means for differentiating the decoded video signal to generate a differential signal that is a signal series of discrete differential points;
An isolated differential point extracting means for extracting an impulse-shaped differential point from the differential signal as an isolated differential point;
Absolute value conversion means for converting the isolated differential points extracted by the isolated differential point extraction means into absolute values;
Integrating means for integrating the absolute value of the isolated differential point converted by the absolute value converting means in units of the block to generate an integrated signal;
Comparison means for comparing the signal level of the integral signal with the preset threshold value;
Counting means for counting a comparison result output from the comparison means indicating that the signal level of the integral signal is equal to or higher than the threshold value on one screen and outputting the count value as the block noise detection result. The video signal processing apparatus according to claim 1.

Images