KR100958708B1 - Method and system for format conversion - Google Patents

Abstract

A method for format conversion includes providing (10) a first interlaced video signal comprising a non-sequential video of a first resolution and downconverting the field of the first resolution to a field of a second resolution (16) combining and interlacing the fields 18 at a third resolution. This video signal is transmitted 20 at a third resolution. The video signal of the third resolution is deinterlaced 32 to form sequential format video. Sequential format video is converted to the first non-sequential resolution by alternating up and down the position of the sequential format of the third resolution to produce an upper field and a lower field of the first resolution, respectively, for display. .

Description

Method and system for format conversion {METHOD AND SYSTEM FOR FORMAT CONVERSION}

FIELD OF THE INVENTION The present invention generally relates to low bit rate transmission, and more particularly to the low bit rate compression of non-progressive video. A method and apparatus for providing format conversion are provided.

The video field is sometimes interlaced so that a serial video bit stream can be transmitted over the network. Interlacing involves continuously transmitting alternating fields of video over a link. The video stream may be divided into progressive or non-progressive. Sequential video includes pairs of continuous fields with no object motion between successive field pairs. Out of order video may include a field with the movement of an object between successive field pairs. Progressive frames can be easily formatted, but format conversion of out of order video does not enjoy the same capabilities. Instead, out of order video is much more difficult to format convert without significant loss of vertical resolution.

One approach to this downconversion is to use intra-field interpolation to convert 240-line incoming fields into 120-line fields. . As a result, the 240 interlaced (240i) video is compressed and transmitted and then decompressed. The 240i video is displayed after intra-field converted by vertical interpolation into a field of 240 lines of the 480i signal. The vertical resolution is particularly low compared to 240 sequential image processing. The diagonal edges of the moving object are also jagged.

Therefore, there is a need for a method and system in which non-sequential video is downconverted to a lower picture resolution to allow acceptable compression for low bit rate video. Another need exists for upconverting the transmitted low bit rate video and then providing adequate picture quality.

A method for format conversion includes providing a first interlaced video signal comprising a non-sequential video of a first resolution, down converting a field of the first resolution to a field of a second resolution; Combining and interlacing a field of a third resolution. The video signal is transmitted at a third resolution. The video signal of the third resolution is deinterlaced to form sequential format video. Sequential format video is converted to the first non-sequential resolution by alternately adjusting the position of the sequential format of the third resolution up and down to make the upper and lower fields of the first resolution respectively for display.

In another embodiment, deinterlacing the out of order video signal into a sequential format having a first resolution may be included. A jagged diagonal reduction method can be used to perform deinterlacing.

A system for format conversion includes a conversion device for deinterlacing an out of order interlaced video signal having a first resolution into a sequential format having the same resolution. The conversion device down converts the sequential format to a second resolution and combines and interlaces the fields of the second resolution to provide a frame at the third resolution. A transmission link is connected to the conversion device for transmitting the video signal at the third resolution. A reconversion device is connected to the output of the transmission link. The reconversion device deinterlaces the video stream of the third resolution as progressive format video and converts the progressive format video into alternating upper field lines and lower field lines for display at the first resolution.

The advantages, features, and several additional features of the invention will appear more fully in light of the exemplary embodiments now described in detail in connection with the accompanying drawings.

1 is a block diagram illustrating an exemplary preprocessing sequence for format conversion in accordance with the present invention.                 

2 is a block diagram illustrating block 12 in more detail in accordance with an embodiment of the present invention.

3 is a block diagram illustrating blocks 16 and 18 in more detail in accordance with an embodiment of the present invention.

4 is a block diagram illustrating an exemplary postprocessing sequence for format conversion in accordance with the present invention.

5 illustratively shows adjusting the line position in step 34 in accordance with the present invention.

6 illustratively illustrates the jagged diagonal reduction method / algorithm used in accordance with the present invention.

7 illustrates an exemplary format conversion system in accordance with the present invention.

It is to be noted that the accompanying drawings are intended to illustrate the concept of the invention and are not the only configuration that can illustrate the invention.

The present invention provides a method and system for format conversion of video to enable compression of low bit rate streams. The present invention deinterlaces incoming non-progressive video into a field to provide interlaced video with a 1/2 resolution, for example vertical resolution. (deinterlaced) and downconverted. Interlaced video at half resolution is compressed, transmitted and decompressed. The interlaced video at 1/2 resolution is then deinterlaced in sequential format, and the sequential format video is converted into alternating upper and lower fields for display on a conventional TV set or display monitor. Although the present invention is described with respect to 2: 1 resolution reduction, the vertical resolution reduction can be reduced to other ratios, for example 4: 1 reduction.

Although the present invention is described with respect to a conventional 480 line interlaced (480i) video system, the present invention is much broader and may include any video format or size that can be transmitted over a wired or wireless network. It should be understood.

In one exemplary embodiment, the incoming out of order 480i video is deinterlaced and downconverted into fields of 120 lines to produce 240i video. 240i video is compressed, transmitted and decompressed. The 240i video is then deinterlaced to 240p. 240p is converted to alternating 240 line upper field and 240 line lower field, for example, to produce 480i for display on a conventional TV set. In addition, the present invention may be applied to any system including a video transmission system via a telephone, a set top box, a computer, a satellite link, the Internet, and the like. The invention is particularly useful for video on-demand applications, which can be implemented on a digital subscriber line (DSL) network or other system with medium bit rate channels. .

It is to be understood that the elements shown in the figures may be implemented in various forms of hardware, software, or a combination thereof. Preferably, these elements are implemented in software on one or more suitably programmed general-purpose devices, which may include a processor, memory, and input / output interfaces.

Referring now to the drawings and initially FIG. 1 in which like reference numerals refer to like elements or like elements throughout the several views, a flow diagram / block diagram is provided for preprocessing video in accordance with one embodiment of the present invention. The method / system is illustrated. In block 10, the incoming signal can be cropped horizontally to reduce processing. In an illustrative example, a 720 pixel (horizontal) × 480 interlaced (480i) pixel signal may be cropped to 704 × 480i as needed. In block 12, horizontal downsampling is performed to form 352 x 480i. In this example, 2: 1 downsampling is performed. Other ratios are also considered. Downsampling can be performed using conventional interpolation filtering methods.

With continued reference to FIG. 1 and with reference to FIG. 2, block 12 is described in more detail. After a low pass anti-alias filter 13 filters the video data, a subsampling step is performed at block 15. The output of block 15 is a half downsampled field.

Referring again to FIG. 1, this signal is deinterlaced at block 14, preferably at 352 × 480p for 60 Hz. Deinterlacing may be performed by a conventional deinterlacing method. In block 16, this signal is vertically directed by a 1/4, 1/2, 1/4 filter 21 as shown in FIG. It is lowpass filtered and subsampled 4: 1 according to the first phasing to produce an upper 120 line field, which is subsequently shown as lines numbered i1, i3, and so forth. Other filters may also be used for the filter 21. Also shown in FIG. 3 is a subsequent deinterlaced 480p frame that is processed through the same filter as frame i but is 4: 1 subsampled in a phase different from the first phase to produce a subsequent lower 120 line field. It is shown by lines numbered i2, i4 and the like. In this way a 240i sequence of 60 Hz field is formed. Contiguous 240i field pairs are then formed in a 30 Hz frame at block 18.

Each frame is formed in block 18 shown in FIG. In the illustrative example, the 704x480i frame was downconverted to 352x240i frame suitable for reduced bit rate compression. In block 20, the video stream is compressed and then transmitted to a receiver, such as a television or display device, over a network or transmission link (wired or wireless). Advantageously, the present invention improves the vertical resolution displayed for non-moving images. Immobilized images exhibit a 2: 1 or greater vertical resolution improvement over conventional systems. In addition, by using the method of the present invention, including, for example, jagged diagonal reduction as described below, an improved downconversion for an out of order video stream is provided.

Referring to FIG. 4, after transmission of the video, the video is decompressed at block 29. Postprocessing involves upconverting each frame in the horizontal direction. This is illustratively shown in block 30 where interpolation methods such as the bicubic interpolation method are used to upconvert 352 samples per line to 704 samples per line. The upconverted 240i signal is deinterlaced at block 32. One of a plurality of other deinterlacing algorithms may be used. In one embodiment, a jagged diagonal reduction algorithm is used for the moving part of the picture. This is described later with reference to FIG. 5. In block 34, the deinterlaced 240p / 60Hz signal alternates upward movement to generate the upper 240 line field from the 240p frame and downward movement to produce the lower 240 line field from subsequent 240p line frames. Is converted into a 240 line field. These movements are performed at block 34 to adjust the position up or down to reconstruct the 480i field. 5 is an exemplary diagram illustrating the conversion of 240p to 480i. The number associated with each line, for example 1/4 and 3/4, is a weight. The first image i is moved up by a quarter line, and the second image i + 1 is moved down by a quarter. These two shifted pictures are interlaced to produce a 480 line interlaced picture. After adjusting the position, this signal can be output as a 60 Hz field sequence, which is restored to the 704 x 480i format.

Referring to FIG. 6, an exemplary technique for deinterlacing is shown. In an illustrative example, a jagged diagonal reduction portion of deinterlacing is performed. Other deinterlacing techniques can also be used and contemplated by the inventors. In the absence of motion, previous field lines may be inserted between present field lines to perform deinterlacing. This maintains the full level of vertical detail. When the object is moving, the previous or following field cannot be used to create the middle line of pixels. So these pixels must be formed from lines directly above and below the pixel to be formed.

The conventional practice is to use simple vertical interpolation to do this. This is done by forming pixel I by taking the average L3 of the pixels immediately above and below the I pixel.

Preferably in the jagged diagonal reduction method used with the present invention, the diagonal pixel pairs L1 to L5 are tested to see which has the lowest difference in intensity, for example. The pair with this lowest difference is averaged and used as pixel I. An exception to this occurs if this mean is not in the middle of the mean, i.e. the top and bottom pixels. If not, the top pixel or bottom pixel with the value closest to this average is used instead. By using this method jagged edges are significantly reduced, which results in a better image for the moving object when the video is displayed.                 

7, a system 130 for format conversion is schematically illustrated in accordance with the present invention. System 130 includes a conversion device 132 that deinterlaces a non-sequential interlaced video stream having a first resolution (e.g., 480i) into a sequential format (e.g., 480p) having the same resolution. to provide. The conversion device 132 down converts this sequential format to a second resolution (e.g., 120) and combines the fields of the second resolution to provide a frame of the third resolution 240i. This converting device 132 is preferably a 1/4, 1/2, 1/4 filter (a 1/4, 1/2, 1/4 filter) for down converting the sequential format 480p of the first resolution. Provide 134. The conversion device 132 samples the first image i with the first phase and the second image i + 1 using the same 1/4, 1/2, 1/4 filter 134 with the second phase. Sample the. The field is output from the 1/4, 1/2, 1/4 filter in the first phase and the second phase, and the combining device uses the field of the second phase and the field of the first phase to obtain a third resolution. To combine. Jagged diagonal reduction algorithm 142 may be used to provide deinterlacing of the video signal. The conversion device 132 also preferably compresses the video data before transmission. The conversion device 132 is described with reference to blocks 10-18 in FIG. 1.

Transmission link 136 is coupled to a conversion device for transmitting the compressed video stream at a third resolution (eg, 240).

Reconversion device 138 is coupled to transport link 136 and receives the transmitted video stream. Reconversion device 138 decompresses the video data when received. The reconversion device 138 deinterlaces the decompressed video stream of the third resolution into sequential format video and alternating the sequential format video to produce a continuous field for display. Convert to field line. Reconversion device 138 is described with reference to blocks 30-34 of FIG. 4.

System 130 may include a cropping module / device 140 that crops out a portion of the out of order interlaced video stream to reduce the amount of pixels being processed. . Reconversion device 138 may include a jagged diagonal reduction algorithm or other program / device 142 to deinterlace the video stream. This video may be displayed over the display or TV set 144, for example. Compression and decompression can be performed by known methods.

Although preferred embodiments have been described for format conversion for low bit rate compression (which are intended to be illustrative and not limiting), various modifications and changes may be made to those skilled in the art from the foregoing. It should be noted that it may be possible for a person. Therefore, it will be understood that various changes may be made in the specific embodiments of the invention disclosed which are within the scope and spirit of the invention as outlined by the appended claims. Therefore, while the present invention has been described to have the requirements of clarity and requirements of clarity required by patent law, what is claimed and intended to be protected by the patent registration is disclosed in the appended claims.

As mentioned above, the present invention is available to provide format conversion of video to enable low bit rate compression of out of order video.

Claims (23)

As a method for format conversion, Providing (10) a first interlaced video signal comprising a non-progressive video of a first resolution; Downconverting the field of the first resolution to a field of the second resolution using an interpolation filtering method (15), Combining and interlacing a field of the second resolution to a third resolution, wherein the third resolution is determined according to the field combination of the second resolution; Transmitting a video signal at the third resolution (20); Deinterlacing (32) the video signal of the third resolution to form progressive format video; The first non-sequential video is adjusted by alternately adjusting the position of the sequential format of the third resolution up and down to generate the upper field and the lower field of the first resolution, respectively, for display. Converting to resolution 34, such adjustments are made by weighting; Comprising a format conversion. 2. The method of claim 1, further comprising deinterlacing (14) the out of order video signal into a sequential format having the first resolution. 3. The method of claim 2, wherein deinterlacing the out of order video signal comprises using a jagged diagonal reduction algorithm (142). 3. The method of claim 2, wherein the downconverting comprises downsampling the field of the first resolution by a factor of four (15). 5. The method of claim 4, wherein the downsampling samples the first image (i) using filter 21 in a first phase and the same filter 21 in a second phase. Sampling the image (i + 1) using < RTI ID = 0.0 > 6. The field of claim 5, wherein the field is output from the filter in the first and second phases, and the method further comprises forming the out of order video of the third resolution. Combining (18) the fields output from the filter of phases. The method of claim 1, wherein the first resolution, second resolution, and third resolution are associated with a vertical resolution. 2. The method of claim 1, wherein deinterlacing the third resolution video signal to form a sequential format video comprises using a jagged diagonal reduction algorithm (142). As a method for format conversion, Providing (10) a first interlaced video signal comprising non-sequential video at a first resolution; Deinterlacing the out of order video signal into a sequential format having the first resolution (14); Down converting the sequential format of the first resolution to a field of the second resolution using an interpolation filtering method (15), Combining and interlacing a field of a second resolution into a third resolution, wherein the third resolution is determined according to a combination of fields of the second resolution; Transmitting a video signal at the third resolution (20); Converting the sequential format video of the third resolution to the first non-sequential resolution by alternately adjusting the position of the sequential format of the third resolution up and down to produce an upper field and a lower field respectively for display. As step 34, this adjustment is made by weighting. Comprising a format conversion. 10. The method of claim 9, wherein the downconverting comprises downsampling (15) the field of the first resolution by a factor of four. The method of claim 10, wherein the downsampling step (15) samples the first image (i) using the filter (21) with a first phase and the image using the same filter (21) with a second phase. sampling (i + 1). 12. The filter according to claim 11, wherein the field is output from the filter (21) in the first phase and the second phase, and the method comprises the filter in the field of the second phase and the first phase to produce a third resolution. Combining (18) the fields output from the method. 10. The method of claim 9, wherein the first resolution, second resolution, and third resolution are associated with vertical resolution. 10. The method of claim 9, further comprising deinterlacing (32) the video signal of the third resolution to form sequential format video. 15. The method of claim 14, wherein deinterlacing the video signal of the third resolution to form a sequential format video comprises using a jagged diagonal reduction algorithm (142). 10. The method of claim 9, wherein deinterlacing the out of order video signal into a sequential format having the first resolution comprises using a jagged diagonal reduction algorithm (142). As system 130 for format conversion, A conversion device 132 for deinterlacing a non-sequential interlaced video signal having a first resolution into a sequential format having the same resolution, the conversion device converting the sequential format to a second resolution using an interpolation filtering method. A conversion device 132, wherein the conversion device 132 is combined and interlaced to provide a frame of a third resolution, the third resolution being determined in accordance with the combination of fields of the second resolution; A transmission link 136 coupled to the conversion device for transmitting a video signal at the third resolution; A reconversion device 138 coupled to the transmission link for deinterlacing the video signal of the third resolution into sequential format video and converting the sequential format video into alternating upper and lower field lines for display. Reconversion device 138, where the conversion is by weight And a system for performing format conversion. 18. The device of claim 17, wherein the conversion device for filtering phases of pixels in a sequential format of the first resolution comprises a 1/4, 1/2, 1/4 filter (a 1/4, 1/2, 1/4). a system for format conversion, including filter). 19. The device of claim 18, wherein the conversion device samples the first image i in a first phase and uses the 1/4, 1/2, 1/4 filter 134 in a second phase to produce a second image (i. A system for format conversion that samples i + 1). 20. The system of claim 19, wherein the field is output from a 1/4, 1/2, 1/4 filter in the first and second phases and the system is adapted to obtain the field of the second phase to obtain the third resolution. Combining means (18) for combining the fields of the first phase. 18. The system of claim 17, wherein the first resolution, second resolution, and third resolution are associated with vertical resolution. 22. The system of claim 21, further comprising means (140) for decimating pixels to reduce horizontal resolution. 18. The system of claim 17, wherein the reconversion device comprises a jagged diagonal reduction algorithm (142) for deinterlacing the video signal.

Images