JP2002135731A - Image information conversion device and image information conversion method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To to provide an image information conversion device by which excellent compression factor can be obtained in compression processing and moreover an array of frame images can be generated without omitting image information. SOLUTION: In the image information conversion device 10, an acquisition part 100 acquires a field image, one by one in order, from image information in which a pair of field images and one of the pair of field images are formed plurally as a line in time series, a decision part 400 decides whether the two field images acquired continuously are generated from the same frame image or not, a generation part 500, when all of the three continuous field images are determined that they are generated from different frame images, detects an intermediate field image as an isolated field image and generates an interpolate field image to the isolated field image, and then, a composition part 600 restores a frame image by compositing the isolated field image and the interpolated image.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video information conversion apparatus and a video information conversion method, and more particularly to a video information conversion method which converts a film image into an interlaced scan image by a 2: 3 pull-down, and sequentially scans the video information including an isolated field image. The present invention relates to a technique for restoring a frame image.

[0002]

2. Description of the Related Art In recent years, with the rapid development of digital signal processing technology, a technology for improving image quality by digitally processing an existing analog television signal has become widespread. As one of the apparatuses using the technology, there is a video information conversion apparatus that generates a sequentially scanned image by combining interlaced scanned images.

[0003] Current analog television signals include a sequence of interlaced field images. The video information conversion device, by combining the scanning line information included in the field image by digital signal processing,
A frame image having twice the definition of the field image information is created. The video information conversion device is incorporated in, for example, a television receiver including a display device capable of sequentially scanning and displaying, and generates a frame image from a received analog television signal. The television receiver reduces flickering of a display image by displaying the generated frame image on the display device, as compared with a general receiver that displays an analog television signal by interlaced scanning. At the same time, the definition of the displayed image is increased, and the image quality is improved.

[0004] A similar video information conversion device is also incorporated in a digital recording / reproducing device that records video information on a hard disk device. The recording / playback device generates a frame image from a current analog television signal using the video information conversion device. Further, the generated frame image is compressed into a compressed frame image and recorded on the hard disk device. Further, the recorded compressed frame image is expanded to restore the original frame image, and the frame image is output.

[0005] The conversion process performed by the conventional video information conversion apparatus will be described by way of an example of converting a movie video represented using a conventional analog television signal into a frame image. FIG. 4 is a schematic diagram showing the relationship between video information used for the description, 50 is a movie video recorded on a movie film, 60 is a field image sequence representing the movie video, and 70 is a field image sequence by the video information conversion device. This is a frame image sequence converted from the image sequence 60.

The relationship between the movie image 50 and the field image sequence 60 will be described. Each frame in the movie video 50 is interlacedly converted by a device (not shown) into a pair of field images including a field image composed of odd-numbered scanning line information and a field image composed of even-numbered scanning line information. In FIG. 4, thin arrows indicate that frame A is converted into field images a1 and a2, and frame B is converted into field images b1 and b2.

The field image sequence 60 is configured such that, for each frame of the movie video 50, two or three field images obtained by performing interlaced scan conversion from the frame are alternately connected.
The two or three field images include at least the pair of field images. This regularity is to make the playback time of the field image sequence 60 coincide with the playback time of the movie video 50 and to prevent scanning line information from being lost.

[0008] The field image sequence 60 alternately includes a field image composed of the odd-numbered scanning line information and a field image composed of the even-numbered scanning line information. For ease of understanding, the former is described in the upper part and the latter is described in the lower part. The conventional video information conversion apparatus sequentially compares the degree of difference between adjacent field images with a predetermined threshold value, and determines that field images having a degree of difference equal to or less than the threshold value are generated from the same frame. Is determined to have been generated from a different frame, and a frame image is generated by combining the odd-numbered and even-numbered scanning line information included in the field image determined to have been generated from the same frame. By doing so, a frame image sequence 70 is generated from the field image sequence 60.

In FIG. 4, the conventional video information converter recognizes that the field images a1 and a2 are generated from the same frame, and combines the scanning line information included in both to display the image for two fields. One frame image a is generated. In addition, it recognizes that the following three field images b1 and b2 and b1 are generated from the same frame again, and combines the scanning line information included in the previous two images to form one frame image to be displayed for three field times. Generate b. Thereafter, by repeating the same processing, a frame image sequence 70 is generated from the field image sequence 60.

A frame image obtained by combining field images generated from different frames cannot be used in the above-described digital recording / reproducing apparatus because a good compression ratio cannot be obtained when compressed. The conventional video information conversion device generates a frame image sequence suitable for use in the digital recording / playback device by the above-described processing. FIG. 5 shows an example of a field image sequence that has been edited and has no regularity. Field image sequence 8
At 0, c1 including the odd-numbered scanning line information is deleted from the field image generated from the frame C, and only c2 including the even-numbered scanning line information remains. In this manner, among the field images generated from the same frame, an image in which only one of the field image including the odd-numbered scanning line information and the field image including the even-numbered scanning line information remains is referred to as an isolated field image. Called an image.

[0011] The video information conversion apparatus determines that a field image whose difference between adjacent and preceding field images exceeds the predetermined threshold value is an isolated field image. If a frame image is generated by combining an isolated field image with another field image, the compression ratio is impaired in the compression process as described above.Therefore, the video information conversion apparatus discards the isolated field image and replaces the immediately preceding frame image with the frame image. Extend the display time by one field time.

As described above, the conventional video information conversion apparatus generates a frame image sequence from which a good compression ratio can be obtained in the compression processing, from a field image sequence obtained by interlaced scanning.

[0013]

However, the conventional video information conversion apparatus has a problem in that since the isolated field image is discarded, the video information is lost and the video quality is reduced. Also, a field image generated from the same frame may be erroneously determined to be a field image generated from a different frame, and a field image that should be originally combined to generate a frame image is regarded as an isolated field image. Over-discarding the image to promote deterioration of video quality.

[0014] In view of the above problems, an object of the present invention is to provide a video information conversion apparatus which can obtain a good compression ratio in a compression process and generates a frame image sequence without any loss of video information.

[0015]

According to another aspect of the present invention, there is provided a video information conversion apparatus for restoring one frame image for each pair of field images. Acquiring means for sequentially acquiring field images one by one from video information in which one of the pair of field images is arranged in time series, and an isolated field image in which only one of the pair of field images is included in the video information Detecting means for detecting a detected isolated field image, generating means for generating an interpolated field image for the detected isolated field image, and restoring the frame image as a pair of the detected isolated field image and the generated interpolated field image Combining means.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A video information converter according to an embodiment of the present invention acquires a field image sequence by interlaced scanning, converts the field image sequence into a frame image sequence, and outputs the frame image sequence. Hereinafter, a video information conversion device according to an embodiment of the present invention will be described with reference to the drawings. (Video Information Converter 10) FIG.
0 shows an example of the entire configuration. Video information converter 1
Reference numeral 0 denotes an image processing unit 800 and a control unit 900.

The video information conversion device 10 is, specifically, a ROM (Read) storing a processor and a program.
Only Memory), RAM for work (Rand)
omAccess Memory) and the like. The function of each component is realized by the processor executing a program stored in the ROM. Transfer of data between components is performed via hardware such as a RAM. (Image Processing Unit 800) The image processing unit 800
0, field image storage unit 110, first evaluation unit 200,
First evaluation result storage section 210, second evaluation section 300, second evaluation result storage section 310, determination section 400, determination result storage section 4
10, a generating unit 500 and a synthesizing unit 600.

The image processing section 800 executes a main part of a process for converting the acquired field image sequence into a frame image sequence. (Acquisition Unit 100) The acquisition unit 100 acquires a field image sequence by interlaced scanning from a device (not shown) and alternately includes a field image including odd-numbered scanning line information and an even-numbered field image sequence. The field image including the scanning line information is sequentially output to the field image storage unit 110. (Field Image Storage Unit 110) The field image storage unit 110 includes six field memories 111 to 116, and stores the six field images recently input from the acquisition unit. The suffix τ attached to each field memory indicates the acquisition order of the field images stored in the field memory. The newest field image is t + 2, and the oldest field image is t-3.

This value is also a value representing the display time of each field image in units of one field display time. When the latest field image is input from the acquisition unit 100, the field image storage unit 110 first transfers the contents of each of the field memories 111 to 115 to the oldest field memories 112 to 116, respectively. Is stored in the field memory 111.

Hereinafter, the field image recorded in the field memory (τ) is referred to as fd (τ). (First Evaluation Unit 200) The first evaluation unit 200 calculates the degree of difference between fd (t + 1) and fd (t) after the contents of each field memory of the field image storage unit 110 have been updated, and calculates the first difference. The degree of difference is output to the first evaluation result storage unit 210. The dissimilarity is calculated as a value obtained by summing the dissimilarities between the (2i-1) -th scanning line information and the 2i-th scanning line information included in the two field images with respect to all the scanning line information.

The first difference is low between field images generated from the same frame included in the movie image,
It has been empirically known that there is a property of increasing between field images generated from different frames. (First Evaluation Result Storage Unit 210) First Evaluation Result Storage Unit 21
0 includes three first evaluation registers 211 to 213,
The three evaluation results recently input from the first evaluation unit 200 are stored. Using the subscript τ attached to each first evaluation register, the evaluation result stored in each first evaluation register
Notated as V1 (τ).

The first evaluation result storage unit 210 transfers the contents of the first evaluation registers 211 and 212 to the first evaluation registers 212 and 213 at the same time as the contents of each field memory of the field image storage unit 110 are updated. Then, the latest evaluation result input from the first evaluation unit 200 is stored in the first evaluation register 211. After the latest evaluation result is stored in the first evaluation register 211, the EV
1 (t + 1) is the difference between fd (t + 1) and fd (t), EV1 (t) is the difference between fd (t) and fd (t-1), and EV1 (t-1) is fd (t-1). ) And fd (t
-2). (Second Evaluation Unit 300) The second evaluation unit 300 calculates the degree of difference between fd (t + 2) and fd (t) after the contents of each field memory of the field image storage unit 110 have been updated, and The degree of difference is output to the second evaluation result storage unit 310. The dissimilarity is calculated as a value obtained by summing the dissimilarities of the scanning line information of the same number included in the two field images with respect to all the scanning line information.

The second degree of difference is low between field images generated from the same frame included in the movie image,
It becomes higher between field images generated from different frames, and further, for the first, second, third and fourth field images acquired continuously, the first and second field images and the third and fourth When the field images are respectively generated from the same frame image, the first and third
It has been empirically known that the difference between the field images and the difference between the second and fourth field images are substantially equal. (Second Evaluation Result Storage Unit 310) Second Evaluation Result Storage Unit 31
0 has four second evaluation registers 311 to 314,
The four evaluation results recently input from the second evaluation unit 300 are stored. Using the subscript τ attached to each second evaluation register, the evaluation result stored in each second evaluation register
Notated as V2 (τ).

The second evaluation result storage unit 310 transfers the contents of the second evaluation registers 311 to 313 to the second evaluation registers 312 to 314 at the same time as the contents of each field memory of the field image storage unit 110 are updated. And
After that, the latest evaluation result input from the second evaluation unit 300 is stored in the second evaluation register 311. After the latest evaluation result is stored in the second evaluation register 311, EV2
(T + 2) is the difference between fd (t + 2) and fd (t),
EV2 (t + 1) is fd (t + 1) and fd (t-1)
, EV2 (t) is fd (t) and fd (t−
The difference 2), EV2 (t-1), is the difference between fd (t-1) and fd (t-3). (Determination Unit 400) The determination unit 400 determines fd (t) and fd (t) based on the properties of the first degree of difference and the second degree of difference.
It is determined whether or not -1) is a field image generated from the same frame as 0 or 1, respectively, and the determination result is output to the determination result storage unit 410.

More specifically, after the latest evaluation results are stored in the first evaluation register 211 and the second evaluation register 311, respectively, the first evaluation registers 211 to 213 and the second evaluation register 211 are stored.
Using the evaluation results stored in the evaluation registers 311 to 314, and the previous and previous determination results stored in the determination registers 412 and 413, respectively, when the following logical expression is true, fd ( t) and fd (t−
It is determined that 1) is generated from the same frame.

Logical expression: ((CORR (t−2) = 1) AND (CORR (t−1) = 1) AND (EV1 (t) ≪EV1 (t−1)) AND (NOT (EV1 (t)) ≫EV1 (t + 1))) AND (NOT (EV2 (t + 2) ≫EV2 (t + 1))) OR ((CORR (t-2) = 0) AND (CORR (t-1) = 1) AND (EV1 ( t) ≪EV1 (t-1)) AND (NOT (EV1 (t) ≫EV1 (t + 1))) AND (EV2 (t) ≒ EV2 (t-1))) OR ((CORR (t-2) = 1) AND (CORR (t-1) = 0) AND (EV1 (t) ≒ EV1 (t-1)) AND (NOT (EV1 (t) ≫EV1 (t + 1))) AND (EV2 (t) ≪EV2) (T-1)) AND (NOT (EV (T) ≫EV2 (t + 1)))) OR ((CORR (t-2) = 0) AND (CORR (t-1) = 0) AND (EV1 (t) ≒ EV1 (t-1)) AND ( NOT (EV1 (t) ≫EV1 (t + 1))) AND (NOT (EV2 (t) ≫EV2 (t-1))) AND (NOT (EV2 (t) ≫EV2 (t + 1))) where AND Is a logical product, OR is a logical sum, NOT is a logical negation, x≪y is αx <y (α is a positive number greater than 1), x≫y is x> βy (β is greater than 1 (Positive number), x ≒ y represents y−δ <x <y + δ (δ is a positive number). CORR (t-2) and CORR (t-1) are the results of the previous determination two times before, respectively, as described below. (Determination Result Storage Unit 410) The determination result storage unit 410
Judgment registers 411 to 413.
The four evaluation results that have recently been input from are stored. The determination result stored in each determination register is denoted as CORR (τ) using the subscript τ attached to each determination register.

The determination result storage unit 410 transfers the contents of the determination registers 411 and 412 to the determination registers 412 and 413, respectively, at the same time as the contents of each field memory of the field image storage unit 110 are updated.
The latest determination result input from the determination unit 400 is stored in the determination register 411. After the latest determination result is stored in the determination register 411, CORR (t) determines whether or not fd (t) and fd (t-1) are generated from the same frame.
CORR (t-1) is fd (t-1) and fd (t-
2) is generated from the same frame, CORR (t
-2) is a judgment result indicating whether or not fd (t-2) and fd (t-3) are generated from the same frame by 0 or 1, respectively. (Generating Unit 500) The generating unit 500 determines that both CORR (t) and CORR (t−1) are 1 after the latest determination result is stored in the determination register 411, ie, fd
If it is determined that (t), fd (t-1) and fd (t-2) are all field images generated from different frames, a frame image is generated in combination with fd (t-1). And generates the interpolated field image of
Output to 00. (Synthesizing Unit 600) After the latest judgment result is stored in the judgment register 411, the synthesizing unit 600 converts the frame image corresponding to fd (t-1) into CORR (t) and CORR.
Output as follows according to (t-1).

(1) CORR (t-1) = 1 and COR
When R (t) = 0, that is, fd (t−2) and fd (t−
1) are generated from different frames, and fd (t-1) and fd (t-1)
If it is determined that d (t) is generated from the same frame, fd (t-1) and fd (t) are obtained from the field memories 114 and 113, respectively, and the scanning line information contained therein is obtained. Are combined to generate a frame image, which is stored in the frame memory 601 and output to a device (not shown).

(2) CORR (t-1) = 1 and COR
When R (t) = 1, as described above, the generation unit 500
To fd (t-1) are input. A frame image is generated by combining scanning line information included in the interpolation field and fd (t-1) obtained from the field memory 114, and stored in the frame memory 601 and output to a device (not shown).

(3) If CORR (t-1) = 0, that is, if it is determined that fd (t-2) and fd (t-1) are generated from the same frame, fd (t-1) -2), and outputs the frame image stored in the frame memory 601 again. (Control Unit 900) The control unit 900 controls the video information conversion device 1
0 controls the overall operation. The control unit 900 supplies the field image storage unit 110, the first evaluation result storage unit 210, the second evaluation result storage unit 310, and the determination unit by supplying a clock signal to each component of the image processing unit 800. The update timing of the result storage unit 410 is synchronized. (Details of Generation Unit 500) FIG. 2 shows an example of the configuration of the generation unit 500. The generation unit 500 includes the first selector 51
0, a second selector 520, a scanning line information generator 530, a different area recognizer 540, a first area extractor 550, a second area extractor 560, and an interpolation field synthesizer 570.

When fd (t-1) is an isolated field image, the generation section 500 generates an interpolated field image for generating a frame image in combination with fd (t-1). The first selector 510 obtains a field image fd (t−3) obtained immediately before and after fd (t−1) and configured by scanning line information at the same vertical position as the scanning line information included in fd (t−1). ) And fd (t + 1)
EV2 (t) representing the difference between each and fd (t-1)
A field image having smaller -1) and EV2 (t + 1) is selected and output to the different area recognition unit 540.

The second selector 520 sets the value of EV2 (t-1)
Is smaller, fd (t-2) is selected, and EV2
If (t + 1) is smaller, fd (t) is selected, and any of the selected field images is converted to the first region extracting unit 560.
Output to The scanning line information generation unit 530 generates fd (t−
For each set of adjacent scanning line information included in 1), the average value of the pixel information located at the center of the vertical position of the adjacent scanning line information and included in the same horizontal position of both is set as new pixel information. Interpolated scanning line information is generated and output to the first area extraction unit 550.

The different area recognizing section 540 calculates fd (t-1)
And an area in which an image is different between the image and the field image selected by the first selector 510. Specifically, for pixel information included in the same horizontal position of the same number of scanning line information included in both field images, if there is a difference equal to or more than a predetermined threshold value, the display position of the pixel information on the screen is changed. The adjacent area is recognized as having a difference, the other areas are recognized as having no difference, and the recognition result is output to the first area extracting section 550 and the second area extracting section 560.

The first area extracting section 550 extracts, from the interpolated scanning line information generated by the scanning line information generating section 530, the scanning line information displayed in the area recognized as different by the different area recognizing section 540. , Interpolation field synthesis section 57
Output to 0. The second area extracting unit 560 extracts, from the scanning line information included in the field image selected by the second selector, the scanning line information displayed in the area recognized as having no difference by the different area recognizing unit 540, and performs interpolation. Output to the field combining unit 570.

The interpolated field synthesizing unit 570 combines the scanning line information extracted by the first area extracting unit 550 and the second area extracting unit 560 to generate an interpolated field image. FIG. 3 is an image example for specifically describing the processing in each unit of the generation unit 500. 501 and 502 are
This represents a frame included in a movie image in which a black car runs away from a black and white striped building. 503 and 504 are field images generated from the frame 501 by interlaced scanning and composed of odd-numbered and even-numbered scanning line information, respectively.
Reference numeral 505 denotes a field image generated from the frame 502 by interlaced scanning and including odd-numbered scanning line information. In the figure, a scanning line number is attached to the left side of each scanning line information for reference.

Here, it is assumed that the field image 505 is an isolated field image, and the field images 503 and 504 are outputs from the first selector 510 and the second selector 520, respectively. The scanning line information generation unit 530 generates the second scanning line information using the average value of the pixel information included in the same horizontal position of the first and third scanning line information of the field image 505 as new pixel information, and sequentially similarly. Generate fourth and sixth scanning line information. The eighth scanning line information is generated from new pixel information obtained by adding a difference between pixel information included in the same horizontal position of the sixth and seventh scanning line information to the pixel information of the seventh scanning line information, for example. I just need. Reference numeral 507 denotes the generated scanning line information, and black stripes of the building included in the original frame 502 and to be represented by the scanning line information are missing.

The different area recognizing section 540 compares the pixel information included in the field images 505 and 503, and determines that there is a difference in the pixels displayed in the lower right area of the screen. Recognition result 5 with no difference in area
06 is output. The first area extraction unit 550 extracts the scanning line information 509 included in the area 507 and displayed in the area indicated by the difference 506, and the second area extraction unit 560
4, the scanning line information 508 displayed in the area indicated by 506 in which there is no difference is extracted.
70 generates an interpolated field image 511 by merging the extracted scanning line information 508 and 509.

As described above, the generation unit 500 uses the scanning line information included in the adjacent field image for the area where there is no motion in the image, and secures the black color of the building, while using the isolated field image itself for the area where there is motion. , An appropriate interpolated field image is generated by excluding the black color of the car that is not included in the original frame 502 by interpolating and using the scanning line information included in.

The combining section 600 generates a frame image 512 that is faithful to the original frame by combining the scanning line information included in the generated interpolated field image 511 and the isolated field image 505. (Summary) As described above, the video information conversion apparatus 10 adds to the degree of difference between adjacently acquired field images,
In order to detect an isolated field image using also a field image obtained by placing another, that is, a difference degree between the latest field images formed of scanning line information at the same vertical position, a conventional video information conversion apparatus is used. In comparison, the detection accuracy of the isolated field image is improved.

As a typical example, when a field image in which odd-numbered scan line information and even-numbered scan line information are significantly different from each other is repeated, such as a field image generated from a frame representing a fine stripe pattern, a conventional image is used. The video information conversion device erroneously determines that adjacent field images have been generated from different frames. However, the video information conversion device 10
Is generated from the same frame when the difference between the most recent field images consisting of odd-numbered scanning line information and the difference between the most recent field images consisting of even-numbered scanning line information are substantially equal. It can be determined that the field image has been processed.

For an isolated field image, depending on the presence or absence of image movement, for each region, either one of the scanning line information included in the isolated field image itself or the scanning line information included in the adjacent field image is used. To generate an interpolated field image and combine the interpolated field image and the isolated field image to generate a frame image faithful to the original frame. (Postscript) (1) In the above embodiment, when the determination unit 400 performs the determination process on the two field information fd (t) and fd (t−1) obtained first, the past evaluation result EV
1 (t-1), EV2 (t) and EV2 (t-1), and the past determination result CORR (t-1) CORR (t-
2) may be processed by giving a predetermined initial value.

For example, EV1 (t-1), EV2 (t)
And EV2 (t-1) are both larger than the maximum evaluation value output from the first evaluation unit 200 and the second evaluation unit 300, and CORR (t-1) and CORR (t
By performing the determination process by setting each of -2 to 1, a favorable determination result can be obtained. (2) In the above embodiment, the generation unit 500 uses the scanning line information included in the adjacent field image for the area where the image does not move, and converts the scanning line information included in the isolated field image itself for the area where the image moves. Although the interpolation field image is generated using the scanning line information obtained by interpolation, the interpolation field image is obtained by using the scanning line information obtained by interpolating the scanning line information included in the isolated field image itself over the entire image. An image may be generated.

According to this configuration, especially in a natural image in which odd-numbered scan line information and even-numbered scan line information rarely differ significantly, an interpolated field image that achieves a fixed image quality can be more easily formed. Can be generated. (3) In the above embodiment, the synthesizing unit 600 outputs the frame image recorded in the frame memory 601 again for the field image determined to be continuously generated from the same frame. Synthesis unit 600
Can be used to count the number of field images determined to be continuously generated from the same frame, and output one frame image and the counted number as the field time in which the frame image should be displayed. Good.

The frame image sequence output by this configuration is suitable for processing in the video information compression device.

[0045]

(1) The video converter of the present invention is a video information converter for restoring one frame image for each pair of field images, and one of the pair of field images and the pair of field images. Acquiring means for sequentially acquiring field images one by one from a plurality of time-series video information; detecting means for detecting an isolated field image in which only one of the pair of field images is included in the video information; Generating means for generating an interpolated field image with respect to the isolated field image, and synthesizing means for restoring the frame image as a pair of the detected isolated field image and the generated interpolated field image.

According to this configuration, the video information conversion device
By generating an interpolated field image for an isolated field image and combining them to generate a frame image, a frame image sequence in which video information included in the isolated field image is not lost can be generated. The frame image sequence generated by this configuration can obtain a better compression ratio when subjected to a compression process than the frame image sequence generated by combining an isolated field image and an adjacent field image. (2) In the video information conversion device of (1),
The generating means may generate the interpolated field image using the detected isolated field image.

According to this configuration, the same effect as (1) can be obtained. (3) In the video information conversion device of (2),
Each field image is composed of a plurality of pieces of scanning line information,
The generation unit may generate interpolation scanning line information by taking an average value of adjacent scanning line information included in the isolated field image, and generate the interpolation field image from the interpolation scanning line information.

According to this configuration, the same effect as (1) can be obtained. (4) In the video information conversion device of (1),
The generation unit detects, among field images acquired before and after the isolated field image, a field image having a lower degree of difference from the isolated field image as a similar field image, and detects the isolated field image and the detected isolated field image. Comparing the similar field image with a similar field image to determine a display area in which the image has changed; the isolated field image displayed in the display area determined to have changed; and the similar field displayed in another display area Using the image and
The interpolation field image may be generated.

According to this configuration, particularly, in a video in which odd-numbered scanning line information and even-numbered scanning line information are significantly different, a similar field image is used for an area where there is no motion in an image, and an isolated field is used for an area where there is motion. Since the frame image is generated using the field image itself, a frame image that is faithful to the frame of the original movie video is generated, and the video quality is further improved. (5) In the video information conversion device of (4),
Each field image is composed of a plurality of pieces of scanning line information,
The generating means generates interpolation scanning line information by taking an average value of adjacent scanning line information included in the isolated field image, and generates interpolation scanning line information to be displayed in a display area determined to have changed. The interpolated field image may be generated from scanning line information included in the similar field image and displayed in another display area.

According to this configuration, the same effect as (4) can be obtained. (6) In the video information conversion device of (5),
The detecting means sequentially determines whether or not the obtained field image and the field image obtained immediately before the field image are generated from the same frame image. If it is determined that both of the former two and the latter are not, the intermediate field image may be detected as an isolated field image.

According to this configuration, the same effect as (5) can be obtained. (7) In the video information conversion device of (6),
The detecting means may further include a difference between the first and third field images and a difference between the second and fourth field images for the first, second, third, and fourth field images that are continuously acquired. When are substantially equal, it may be determined that the first and second field images and the third and fourth field images have been generated from the same frame image, respectively, and the isolated field image may be detected.

According to this configuration, in a field image sequence alternately including a field image composed of odd-numbered scanning line information and a field image composed of even-numbered scanning line information, the latest image data composed of odd-numbered scanning line information is used. When the difference between the field images and the difference between the most recent field images including even-numbered scan line information are substantially equal, it is determined that the adjacent field images are generated from the same frame image. For an image in which odd-numbered scan line information and even-numbered scan line information are significantly different, the error of excessively detecting an isolated field image is reduced. Since the frame image for the isolated field image is not excessively output, the compression ratio in the compression processing is not impaired. (8) In the video information conversion device of (6),
The detection means may determine that the t-th and t-1st acquired field images have been generated from the same frame image when the following logical expression is true.

Logical expression: ((CORR (t−2) = 1) AND (CORR (t−1) = 1) AND (EV1 (t) ≪EV1 (t−1)) AND (NOT (EV1 (t)) ≫EV1 (t + 1))) AND (NOT (EV2 (t + 2) ≫EV2 (t + 1))) OR ((CORR (t-2) = 0) AND (CORR (t-1) = 1) AND (EV1 ( t) ≪EV1 (t-1)) AND (NOT (EV1 (t) ≫EV1 (t + 1))) AND (EV2 (t) ≒ EV2 (t-1))) OR ((CORR (t-2) = 1) AND (CORR (t-1) = 0) AND (EV1 (t) ≒ EV1 (t-1)) AND (NOT (EV1 (t) ≫EV1 (t + 1))) AND (EV2 (t) ≪EV2) (T-1)) AND (NOT (EV (T) ≫EV2 (t + 1)))) OR ((CORR (t-2) = 0) AND (CORR (t-1) = 0) AND (EV1 (t) ≒ EV1 (t-1)) AND ( NOT (EV1 (t) ≫EV1 (t + 1))) AND (NOT (EV2 (t) ≫EV2 (t-1))) AND (NOT (EV2 (t) ≫EV2 (t + 1))) where AND Is a logical product, OR is a logical sum, NOT is a logical negation, x≪y is αx <y (α is a positive number greater than 1), x≫y is x> βy (β is greater than 1 (Positive number), x ＜ y represents y−δ <x <y + δ (δ is a positive number), and EV1 (τ) is an index indicating the degree of difference between the τth and τ−1th acquired field images. And EV2
(Τ) is an index indicating the difference between the τth and τ−2th acquired field images, and CORR (τ) is
This is a determination result indicating whether or not the τth and τ−1th acquired field images are generated from the same frame image by 0 or 1, respectively.

According to this configuration, in addition to the same effect as (7), the determination is made in consideration of the degree of difference between adjacent field images, so that the difference between the source frame images is more accurately determined. . (9) In order to solve the above problem, a video information conversion method of the present invention is a video information conversion method for restoring one frame image for each pair of field images, wherein the pair of field images and the pair of field images are A field image is sequentially converted from video information in which one of a plurality of images is arranged in time series.
An obtaining step of obtaining an isolated field image in which only one of the pair of field images is included in the video information; and a generating step of generating an interpolation field image for the detected isolated field image. And a synthesizing step of restoring the frame image as a pair of the detected isolated field image and the generated interpolated field image.

According to this configuration, the video information conversion method has the same effect as (1). (10) In order to solve the above problem, a program storage medium of the present invention is a computer readable program storage medium that stores a program used in a video information conversion apparatus that restores one frame image for each pair of field images. The recorded program includes: an acquisition step of sequentially acquiring field images one by one from video information in which one of the pair of field images and the one of the pair of field images are arranged in chronological order; A detecting step of detecting an isolated field image in which only one of the images is included in the video information; a generating step of generating an interpolated field image for the detected isolated field image; The frame image as a pair with the interpolated field image And a synthesis step that.

According to this configuration, the video information conversion device using the program stored in the program storage medium has the same effect as (1).

[Brief description of the drawings]

FIG. 1 shows an example of the overall configuration of a video information conversion device 10.

FIG. 2 shows an example of a configuration of a generation section 500.

FIG. 3 is an image example for specifically explaining processing in a generation unit 500;

FIG. 4 is a schematic diagram showing a relationship between video information.

FIG. 5 shows an example of a field image sequence having no regularity.

[Explanation of symbols]

 Reference Signs List 10 video information conversion device 50 movie video 60 field image sequence 70 frame image sequence 80 field image sequence 90 frame image sequence 100 acquisition unit 110 field image storage units 111 to 116 field memory 200 first evaluation unit 210 first evaluation result storage unit 211 To 213 first evaluation register 300 second evaluation unit 310 second evaluation result storage unit 311 to 314 second evaluation register 400 determination unit 410 determination result storage unit 411 to 413 determination register 500 generation unit 501 to 502 frames 503 to 505 field image 506 Recognition result 508-509 Scan line information 510 First selector 511 Interpolated field image 512 Frame image 520 Second selector 530 Scan line information generation unit 540 Different region recognition unit 550 First region extraction unit 560 Second region extraction Output unit 570 Interpolated field synthesis unit 600 Synthesis unit 601 Frame memory 800 Image processing unit 900 Control unit

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5C059 KK01 LA07 LB13 LB18 NN01 NN29 PP04 PP11 SS11 SS20 TA08 TB05 TC02 TC42 TD05 TD06 TD12 UA02 5C063 AA20 AB03 AC01 BA04 BA10 BA12 CA11

Claims (10)

[Claims] 1. A video information conversion device for restoring one frame image for each pair of field images, comprising: a video information in which one of the pair of field images and one of the pair of field images are arranged in time series. Acquiring means for sequentially acquiring field images one by one; detecting means for detecting an isolated field image in which only one of the pair of field images is included in the video information; and an interpolation field for the detected isolated field image. A video information conversion apparatus comprising: a generation unit that generates an image; and a synthesis unit that restores the frame image by pairing a detected isolated field image and a generated interpolation field image. 2. The video information conversion apparatus according to claim 1, wherein said generating means generates said interpolated field image using a detected isolated field image. 3. Each field image is composed of a plurality of pieces of scanning line information, and the generating means generates interpolation scanning line information by taking an average value of adjacent scanning line information included in the isolated field image, The video information conversion device according to claim 2, wherein the interpolation field image is generated from the interpolation scanning line information. 4. The isolated field image, wherein the generation unit detects, as a similar field image, a field image having a lower degree of difference from the isolated field image among field images acquired before and after the isolated field image. And the detected similar field image is compared to determine the display area in which the image has changed, and the isolated field image displayed in the display area determined to have changed and displayed in another display area The video information conversion apparatus according to claim 1, wherein the interpolated field image is generated using the similar field image to be generated. 5. Each field image is composed of a plurality of pieces of scanning line information, and said generating means generates interpolation scanning line information by taking an average value of adjacent scanning line information included in said isolated field image, Generating the interpolated field image from the interpolated scanning line information displayed in the display area determined to have changed and the scanning line information included in the similar field image and displayed in another display area. The video information conversion device according to claim 4, wherein 6. The detecting means sequentially determines whether or not the acquired field image and the field image acquired immediately before the field image are generated from the same frame image. 6. The video information conversion apparatus according to claim 5, wherein an intermediate field image is detected as an isolated field image when it is determined that none of the former two fields and the latter two fields of the three field images are present. 7. The apparatus according to claim 1, wherein said detecting means further comprises: first, second, third, and fourth continuously acquired data.
When the difference between the first and third field images is substantially equal to the difference between the second and fourth field images, the first and second field images and the third and fourth field images are 7. The video information conversion apparatus according to claim 6, wherein it is determined that the isolated field images are generated from the same frame image, and the isolated field image is detected. 8. The detecting means determines that the t-th and t-1st acquired field images are generated from the same frame image when the following logical expression is true, and determines the isolated field image. The video information converter according to claim 6, wherein the video information is detected. Logical expression: ((CORR (t−2) = 1) AND (CORR (t−1) = 1) AND (EV1 (t) ≪EV1 (t−1)) AND (NOT (EV1 (t) ≫EV1 ( t + 1))) AND (NOT (EV2 (t + 2) ≫EV2 (t + 1)))) OR ((CORR (t-2) = 0) AND (CORR (t-1) = 1) AND (EV1 (t)} EV1 (t-1)) AND (NOT (EV1 (t) ≫EV1 (t + 1))) AND (EV2 (t) ≒ EV2 (t-1))) OR ((CORR (t-2) = 1) AND (CORR (t-1) = 0) AND (EV1 (t) ≒ EV1 (t-1)) AND (NOT (EV1 (t) ≫EV1 (t + 1))) AND (EV2 (t) ≪EV2 (t- 1)) AND (NOT (EV2 (t) ≫EV 2 (t + 1)))) OR ((CORR (t-2) = 0) AND (CORR (t-1) = 0) AND (EV1 (t) ≒ EV1 (t-1)) AND (NOT (EV1 ( t) ≫EV1 (t + 1))) AND (NOT (EV2 (t) ≫EV2 (t−1))) AND (NOT (EV2 (t) ≫EV2 (t + 1))) where AND is a logical product, OR is a logical sum, NOT is a logical negation, x≪y is αx <y (α is a positive number greater than 1), x≫y is x> βy (β is a positive number greater than 1), x ≒ y represents y−δ <x <y + δ (δ is a positive number), EV1 (τ) is an index indicating the difference between the τth and τ−1th acquired field images, and EV2
(Τ) is an index indicating the difference between the τth and τ−2th acquired field images, and CORR (τ) is
This is a determination result indicating whether or not the τth and τ−1th acquired field images are generated from the same frame image by 0 or 1, respectively. 9. A video information conversion method for restoring one frame image for each pair of field images, comprising: a video information in which one of the pair of field images and one of the pair of field images are arranged in time series. An acquisition step of sequentially acquiring field images one by one; a detection step of detecting an isolated field image in which only one of the pair of field images is included in the video information; and an interpolation field for the detected isolated field image. A video information conversion method, comprising: a generating step of generating an image; and a synthesizing step of restoring the frame image as a pair of a detected isolated field image and a generated interpolated field image. 10. A computer-readable program recording medium on which a program used in a video information conversion apparatus for restoring one frame image for each pair of field images is recorded, wherein the recorded program comprises: An image and an acquisition step of sequentially acquiring field images one by one from video information in which one of the pair of field images is arranged in time series; and only one of the pair of field images is included in the video information. A detecting step of detecting an image, a generating step of generating an interpolated field image for the detected isolated field image, and restoring the frame image as a pair of the detected isolated field image and the generated interpolated field image A synthesizing step. Lamb recording medium.