JPH10336593A - Synchronous video processing system and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain avoidance of deviations from an actual video by storing a frame data column and plural number of time codes, calculating their feature amount and generating a record which takes the time codes representing an event as a group, when retrieval conditions and the feature amount for detecting frame data in which a certain event appears satisfy the conditions for retrieval. SOLUTION: An analog video signal 12 is supplied to a computer 2 via a video signal input port P, video data 12A are converted into digital data 13 by an A/D converter 7 and held in a memory 4. Non-video data 12B are converted into a time code 14 which has plural number of frame address information by an address information decoder and are held in the memory 4. A CPU 3 executes a detection processing, in accordance with an event detection program 15, and records frame data as a group when the frame data are detected and correctly makes a frame address correspond to the frame.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a video signal processing system and method for video indexing that supports editing and searching of video data such as video data and movies recorded on an electronic medium.

[0002]

2. Description of the Related Art In recent years, with the background of high-speed and large-capacity computers, database systems and presentation tools that can handle video information such as movies and video data recorded on electronic media, which could not be handled conventionally, have been developed. Attention has been paid. It is a time-consuming task for a user to search for and process only a necessary part from video information having a huge amount of information. Therefore, as a method of reducing the work load by a computer, for example, IPSJ Transactions Vol.33, No.4, No.543-
549, “Automatic Indexing Method and Object Searching Method for Color Video Images” and Japanese Patent Application Laid-Open No.
No. 111181, "Method for detecting a change point of a moving image", Japanese Patent Laid-Open No. 9-6 published March 7, 1997 previously proposed by the inventors.
No. 5,287, "Method and apparatus for detecting characteristic scenes of moving images".

[0003] This is because the video is automatically divided into cut units, divided into special effect points such as dissolves, and various events such as an appearance section of a subject and a subtitle display section are automatically detected. This is a method of assisting in search and editing of a moving image by displaying a list of images of the detected portion as icons. According to this, the contents of the video information are displayed in a list on the screen, and the entire configuration content can be grasped at a glance, so that the desired scene can be easily found, and the editing can be easily performed because it can be handled in a well-divided unit called a cut. Will be able to

[0004]

However, an image is composed of a sequence of 30 frames per second. Continuously processing an image sequence of 30 frames per second in real time using a generally available processing device is limited to about one type of video signal processing due to the processing speed. At present, when performing a plurality of video signal processes, three methods are conceivable.

In the first method, a single apparatus repeatedly inputs and outputs a video of the type of video signal processing, performs individual video signal processing, and converts a time code, which is the address of a frame at an event occurrence point, into a standard time code. This is a method of reading and recording from an input source device such as a video deck via a serial port. In this method, the time relationship of the processing result can be surely grasped, but the video must be re-input several times, and the processing time becomes longer as a whole. In other words, a plurality of types of video signal processing are performed. There is a problem that it cannot be performed in real time.

A second method is a method in which a single device prepares an expansion board dedicated to processing for each event and operates in parallel for each function by a conventional method. However, such a device requires a new development of a dedicated board according to the type of event, and has a problem that the development cost is high.

The third method is a method in which processing devices are individually prepared in accordance with the type of video signal processing, and the same video is distributed to each processing device by a video distributor and processed simultaneously. In this method, the processing time as a whole is the same as the time length of the video, but in order to arrange the processing results accurately in time, the time of all devices must be precisely adjusted separately, and It is necessary to read the actual time and the time code of the video being reproduced from the video deck via the standard serial port in the processing device and take necessary measures accordingly. However, when the individual processing devices are realized by a computer or the like, the times of the plurality of processing devices are set in frame units (30
It is difficult to adjust the time in units of 1 / second) or in units of fields (1/60 second).

Therefore, it is conceivable to send a time code from one VCR to all the processing devices via a standard serial port. However, since a serial port normally follows a predetermined communication procedure, a plurality of time code read requests are simultaneously sent. When the request is issued, the processing unit is queued, and the processing device in the queue receives an address shifted from the true time code at the time when the request is generated. Therefore, even with this method, perfect video synchronization is difficult.

An object of the present invention is to achieve a plurality of video signal processings in real time, and to accurately recognize a frame address of an event occurrence point detected by each video signal processing. It is an object of the present invention to provide a video signal processing method and system in which video signals are easily and correctly arranged.

[0010]

According to one aspect of the present invention, unique address information is added to each of a plurality of frames included in an analog video signal. First, an analog video signal to be processed is input to a processing device via a video signal input port in a time series in a frame unit, and in the processing device, video data in the video signal is recorded in a memory as a digital frame data sequence, In addition, the address information included in the non-video data in the video signal is digitized, converted into a plurality of time codes specifying one corresponding digital frame data, and recorded in the memory. The digital frame data is calculated in terms of their features in chronological order, and it is determined whether or not the features satisfy search conditions for detecting frame data in which a certain event appears. Alternatively, a record in which an event code and a time code representing an event are paired is created.

Thus, an accurate address of a processed frame can be recorded irrespective of the length of the processing time or the time when the arrival of the video signal is delayed from the moving picture reproducing apparatus to the video signal processing apparatus.

Further, in another aspect of the present invention, when performing a plurality of video processings in real time, a plurality of processing devices are provided, and a target video signal is distributed to the plurality of processing devices.
Each processing device executes processing independently, arranges processing results in a correct time relationship using time codes, and integrates them.

Thus, even if a plurality of video processes are processed by a plurality of processing devices, the same result as a result of processing by one virtual ultra-high-speed processing device can be obtained. Further, it is not necessary to provide a special signal line from outside in addition to the video signal input port in order to correctly grasp the time relation of the processing result. Each of these processing units can be configured using a general-purpose and inexpensive computer.

[0014]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1A is an example of a block diagram of a video processing system according to an embodiment of the present invention. 1 is CRT
And the like, and displays an output screen of the processing device 2 configured by, for example, a computer. Instructions to the computer 2 can be given using an input device 5 such as a pointing device. The moving image reproducing device 10 is a tuner device for receiving a broadcast program such as a terrestrial broadcast, a satellite broadcast, and a cable television, or a device for reproducing a moving image recorded on an optical disk, a video tape, or the like.

An analog video signal 12 output from the moving picture reproducing apparatus 10 is supplied to the computer 2 via a video signal input port P and includes video data 12A and non-video data 12B. For example, closed caption information or frame address information of a video is a specific scanning line (for example, a scanning line during a blanking period).
Are superimposed and recorded on the signal portion corresponding to. Closed captioning was developed in the United States for the hearing impaired and is now included in most television broadcasts and commercial video software. Note that Internet address information is also superimposed on non-video data of a video signal for broadcasting called recent Internet text.

In this embodiment, the video data 12A is A / D
The data is converted into digital data (frame data) 13 indicating a frame by the converter 7 and stored in the memory 4 via the interface. The frame data temporarily stored in the memory 4 is, for example, data for at most 15 frames. For the non-video data 12B of the video signal, the address information decoder 8 performs frame address information included in the non-video data 12B (hereinafter, simply referred to as “non-video portion” when clear from the context).
Are converted into a plurality of time codes 14 which are digital data, and are stored in the memory 4. In addition, if necessary,
A scanning line for storing closed caption information can also be read, converted into digital data, and sent to the memory 4. What is necessary is just to add the setting of the scanning line to read.

The digital frame data 13 recorded in the memory 4 is subjected to event detection processing by the CPU 3 in accordance with an event detection program 15 stored in the memory 4. The event detection program comprises: means for setting a search condition for detecting frame data in which an event appears; means for comparing each frame data with the search condition to determine whether or not the event matches; And at least event code data to be executed. When the frame data is detected as a result of the process, an event code representing the event and a time code 14 as the detected frame data address information are combined into a record, and the record is stored in the memory 4 or the external information as necessary. The information is recorded in the storage device 6. In addition, the computer 2 can send a stop and playback instruction to the moving image playback device 10 via the control line 11. The control line 11 is connected to a serial port interface of the computer 2.

In the computer 2 shown in FIG. 1 (a), the address information decoder 8 is realized by an A / D converter, and (1) may be formed into a chip and formed in a computer system at the time of manufacture. (2) It may be mounted on a board substrate and installed in an expansion slot of a general-purpose computer such as a personal computer.
This is effective when implementing the present invention on an existing personal computer). In order to configure the address information decoder, for example, a well-known decoder technology for reading closed captions can be used. That is, the data to be read by the decoder that decodes the closed caption is changed so that the data is read not from the scanning line portion containing the caption data of the video signal but from the scanning line portion containing the frame address information. I just need. Also, in FIG. 1A, a moving image (video signal) is directly input to the address information decoder 8. However, as shown in FIG. May be input to the decoder 8 from the bus line via the interface.

FIG. 2 specifically shows a configuration of a video signal when processing a video on the system shown in FIGS. 1 (a) and 1 (b). The video signal 12 includes a plurality of frames, and is visualized by sending 30 frames per second in the US NTSC standard. Each frame includes visualized video data 12A and non-video data 12B for synchronization control and data transmission. Video data 12
A is converted into digital frame data 13 by the A / D converter 7. In addition, non-video data 12B
Is converted into a digital time code 14 by the address information decoder 8.

In a conventional apparatus for inputting video information from a moving image reproducing apparatus to a computer, a frame and its associated time code are separated from each other and supplied through independent signal lines. Therefore, when video processing is completed due to processing delays on the computer side or the like, the time code to be acquired for the video may be shifted. At the time of video editing of a broadcasting station, this shift is fatal.

On the other hand, according to the present invention, a video signal containing video data and non-video data is received via a video signal input port of the processing device, and the video signal and the non-video data of the video signal thus received are received. , And a digital frame data and a time code attached to the digital frame data, respectively. Therefore, even if a delay occurs in video processing, the frame address acquired for a frame correctly corresponds to the frame.

FIG. 3 is a block diagram for explaining the function of a video signal processing system for detecting frame data in which a plurality of events appearing in a video signal and each of which is characterized by a search condition is intended to support video editing. This is an example. As will be described later, events handled in the present embodiment include, for example, a video signal supplied from a moving image reproducing apparatus, for example, the appearance of a transition between cuts divided for each series of scenes or a frame input continuously. There is the appearance of subtitles. The event detection unit 304 in FIG.
A function to detect a plurality of events (cut point detection unit 306, identical composition detection unit 308, color tone detection unit 310, subtitle detection unit 312, dissolve detection unit 314, replay detection unit 316, slow playback detection unit 318) is provided. However, the present invention is not limited to the type and number of events (search conditions).

The input video signal 12 is transmitted to the video editing unit 3
The video signal is supplied from the moving image reproducing apparatus or the tuner based on the instruction from the video signal generator 30, and is input to the event detection unit 304 and the video editing unit (non-linear editing unit) 330 as the video signals 121 to 128 having the same property by the video signal distribution unit 302. You. The video signal distribution unit 302 may have a known circuit configuration. In each detection processing unit of the event detection unit 304, the video signal 12
First, digital frame data 13 and time code 14 are generated from 1 to 128, and video signal processing (event detection) relating to various events is performed. When a frame is detected, each detecting unit adds an event code representing the event to be detected to the detected frame.
A detection frame (data) or an event code representing an event appearing in the detection frame is collected in the video editing unit 330 together with the time code, and then used for editing work in the editing unit 332. Since a time code in which the time relationship is correctly arranged is recorded in each of the results individually processed by the event detection unit 304, the video of the result of performing each processing can be uniformly managed. The event code may be, for example, a binary code.

Next, the event detection unit 30 in FIG.
The specific detection processing in each block included in No. 4 will be described in detail.

The cut point detecting unit 306 detects a change in the cut, and generates a record in which a representative image (frame data) at that time or an event code and a time code representing the change (event) of the cut to be searched are paired. Then, recording is performed in a storage area provided inside and outside the cut point detection unit 306. If the system is accessible,
The recording location does not matter. For the detection method of the cut point,
For example, a method described in the above-mentioned Japanese Patent Application Laid-Open No. 4-111181, "Method of detecting change point of moving image" may be used.

The same composition detecting unit 308 detects whether a picture of the same composition or a similar composition appears in a frame by going back in the past within a predetermined time. For this, an image comparison method represented by template matching may be used. Specifically, for each of the pixels at the same coordinate position of the two frames to be compared, a luminance difference or a color difference is obtained, and the sum of all the screens is calculated, and this is set as the difference between the images. . If the degree of difference is smaller than a predetermined threshold value, it can be determined that they are the same or similar. As a result of this processing, a set of a time code composed of a start time code and an end time code, and an event code representing a representative image or a search target composition (event) at that time are set as section information of the same composition. Record is recorded. The recording location may be any location that can be accessed by the system similarly to the processing result of the cut point detection unit 306.

The color tone detecting section 310 detects whether a picture of the same color tone or a similar color tone appears in a frame retroactively within a predetermined time. For this purpose, for example, a chromaticity distribution for the entire frame can be used. This is a feature quantity that indicates which color is used and how much, regardless of the composition. Specifically, for each of the two frames to be compared, the colors of the pixels representing the image are classified into about 64 colors, and the number of those colors present in each frame is counted. Then, the sum of the absolute values of the differences between the respective frequencies of the obtained frequency distribution is defined as the color difference. If the degree of difference is smaller than a predetermined threshold value, it can be determined that they are the same or similar. This processing result is recorded as a record in which section information of the same color tone is set as a set of a time code composed of a start time code and an end time code, and an event code representing a detection target color tone (event). The recording location may be any location accessible by the system.

The subtitle detection unit 312 detects whether subtitles appear in a frame in the video. A specific detection method is described in, for example, Japanese Patent Application Laid-Open No. 7-1995 published on July 28, 1995.
What is necessary is just to use the technique shown in 192003 "Moving image search apparatus and method". This processing result is recorded as a record in which the same subtitle generation section is composed of a set of a time code composed of a start time code and an end time code and an event code representing a subtitle (event) to be detected. The recording location may be any location accessible by the system.

The dissolve detector 314 detects a special effect such as a dissolve between successive frames in the video signal.
As a specific detection method, the technology of Japanese Unexamined Patent Publication No. 9-65287, “Method and Apparatus for Detecting Characteristic Scenes of Moving Image” may be used. This processing result is recorded as a record in which a set of a time code composed of a start time code and an end time code in a dissolve generation section and an event code representing a detection target dissolve (event) are set. The recording location may be any location accessible by the system.

The replay detecting section 316 detects whether the same frame appears retroactively in the past within a predetermined time. This can be performed by comparing the frames by template matching or the like, as in the same composition detection unit 308. However, if template matching was performed for each frame between moving images to be compared, it would take too much processing time. Therefore, each frame was converted to a code of about several characters, and the matching of the code sequence was performed to match the moving image. And Although the amount of information of a single code corresponding to one frame is extremely small, since a moving image is composed of many frames, the number of codes included in one moving image is large, and a series of codes in the moving image Has a sufficient amount of information to identify a piece of moving image. A moving image collation method based on this concept is as follows.
This is disclosed in Japanese Unexamined Patent Publication No. Hei 7-114567 published on May 2, 995, “Video Search Method and Apparatus”. This processing result is a record in which a section in which the same video exists is a set of a time code set including a start time code and an end time code, a label number, and an event code representing a detection target frame (event). Is recorded. The label number is a serial number assigned to a plurality of types of replay scenes included in the video signal in the order of detection. The recording location may be any location accessible by the system.

The slow reproduction detecting section 818 detects a slow reproduction frame. Slow playback is realized by displaying frames continuously at longer intervals (twice in 1/2 slow, 4 times in 1/4 slow) than in standard playback. The video data digitized by the / D converter 7 has a feature that a plurality of exactly the same frames continue (two pieces in ス ロ ー slow and four pieces in ４ slow).
Sheet). Therefore, to determine whether or not the playback is slow, two consecutive frames are checked, and the image difference is checked by template matching. Then, the transition of the degree of difference for a certain time is checked, and if the degree of difference repeats a large value and a small value in a specific cycle, it is determined that slow reproduction is performed. For example, in the case of a ス ロ ー slow motion, the same frame continues every two frames, so that the degree of difference alternately repeats a small value and a large value. In the case of 1/4, the smaller value is 3
After that, the larger value is repeated once. However,
In the case of a moving image, two consecutive frames are similar even if they are not in slow playback, so that it is necessary to determine the magnitude of the difference with a lower threshold value. This processing result is recorded as a record in which the section information of the slow reproduction is set as a set of a time code composed of a start time code and an end time code and an event code representing a slow reproduction frame (event) to be detected. . The recording location may be any location accessible by the system.

In the above, each detection process in the event detection unit 304 has been described in detail. In the specific description of each process, it has been described that the process result recording location may be any location accessible by the system. However, when editing processing is performed using the processing results in the present embodiment, these processing results, that is, the image is automatically divided into cut units, divided into special effect points such as dissolve, and the appearance of the subject The results of automatically detecting various events such as a section and a subtitle display section are collected in the editing unit 332, and when the user actually performs an editing process, the detected frames are listed as icons on the display screen. Displaying or displaying a section of the video signal where the event is present in a bar chart helps the user to edit the video. The composition of the video to be edited is displayed in a list on the display screen,
Since the user can grasp the structure of the video at a glance, the desired scene can be easily found, and the editing can be easily performed because the user can handle the scene in a well-divided unit called a cut.

FIG. 4 shows an example in which the event detection results collected in the editing unit 332 are displayed as a list on a display screen. 1) Cut point, 2) Composition, 3) Color tone, 4)
The event type of subtitle, 5) dissolve, 6) replay, and 7) slow playback is shown on the vertical axis, and time is displayed on the horizontal axis, so that the entire configuration of the video can be understood at a glance. As described above, while detecting a plurality of events in real time, the time code specifying the detected frame is stored as a set with the event code representing the event appearing in the frame or the frame. The respective detection results can be displayed together with the time code, and the frames can be managed and displayed with a correct time relationship. In FIG. 4, examples of the time point represented by the start time code are t _s1 and t _s1 .
_{In s2} , an example of the time point represented by the end time code is t _e1 ,
Each is indicated by t _e2 .

Using this, when the user designates the icon of the cut point with the pointing device, the editing unit 332 instantaneously and immediately displays the image of the video recording unit 3 corresponding to the time code designated by the icon. , It will be possible to search, find, and play back accurately.

In order to perform the above-described video signal processing for event detection accurately and in real time with respect to a video signal generated 30 frames per second like a general television signal,
An ultra-high-speed video processing device is required. In FIG. 5, which realizes this by using a plurality of processing devices such as inexpensive personal computers, a moving image reproducing system 10 is a tuner device for receiving a broadcast program such as a television, and a moving image reproducing device as in FIG. And so on. The video distributor 20 has the same function as the video distributor 302 in FIG. 3, and converts the input signal into a plurality of signals 12 having the same property.
-1, 12-2,..., 12-n. The processing devices 2-1, 2-2,..., 2-n correspond to the computer 2 in FIG.

In FIG. 5, the processing device 2-1 functions as a master processing device. This master processing device 2-1
Controls the entire system shown in FIG. 5 and also performs video editing as needed. Therefore, the function of the video editing unit 330 in FIG. 3 is also fulfilled. Processing unit 2
-2 to the processing devices 2-n perform the event detection processing assigned respectively. These processing devices 2-2 to 2-n are slave processing devices that operate based on instructions from the master processing device. Which event detection is assigned to which slave processing device can be set in advance. When an instruction to reproduce a moving image is issued from the master processing device 2-1 to the moving image reproducing device 10 via the control line 11, the moving image reproducing device 10 reproduces a video recorded on a medium such as a video tape. And outputs the received broadcast video to the video distributor 20. The reproduced video signal 12 is divided into n identical video signals 12-1 to 12-n by the video distributor 20, and supplied to the processing devices 2-1 to 2-n, respectively.
Each of the processing devices 2-1 to 2-n has the same configuration as the computer 2 shown in FIG. 1, and thus stores and holds digitized frame data and time code in a memory. The data is read from the memory, and the predetermined video signal processing (event detection and editing processing according to a user instruction) described in the description of FIG. 3 is executed.

In each of the processing units 2-1 to 2-n, FIG.
The individual event detection process described in the above is performed by, for example, the event detection program 1 stored in the memory 4 in FIG.
The event detection program 15 is read out and executed by the CPU 3. In the description of the present embodiment, each processing device 2-2
Event detection processing (30) shown in FIG.
6,308,310,312,314,316,31
8), but if the computer 2 has a high performance capable of performing image processing of 30 frames or more per second, a plurality of event detection processes can be performed as long as the condition of the moving image processing in real time is satisfied. Can be assigned to one computer. Until a termination instruction is issued by the instruction of the master processing device 2-1, each of the processing devices 2-2 to 2-n performs an event detection process in charge.
In response to a termination instruction from the master processing device 2-1, each of the processing devices 2-2 to 2-n terminates the processing, specifies the processing result (detected frame or an event code representing an event that appears in the detected frame), and identifies the detected frame. The time code is read out from the memory 4 (FIG. 1) together with the time code to be transferred to the master processor 2-1 via the network 50.
The master processor edits the video using these results. Although not shown, the system of FIG. 5 is provided with at least the same display device and input instruction device as those shown in FIG. 1 for various processes.

FIG. 6 is a control flowchart of the whole system of FIG. First, in the box 60, each of the slave processors 2-2, 2-3,...
An instruction to start an event detection processing program is issued to 2-n. Next, in box 62, the master processing device 2
From -1, an instruction to reproduce a video is issued to the moving image reproducing apparatus 10.

In this state, various event detections are executed on each of the processing devices 2-1 to 2-n. When the video for a predetermined time is processed, an instruction to stop video reproduction is issued from the master processing device 2-1 to the moving image reproducing device 10 in box 64. Subsequently, in box 66, the master processing device 2-1 issues an instruction to stop processing of event detection to the processing devices 2-2 to 2-n of the slaves. Then, in box 68, the master processor 2-1 is connected to each slave processor 2-2.
The time code attached to the detected frame is read out from the memory 4 and the processing result of .about.2-n is transferred together therewith. Finally, the user edits the video while displaying the processing result on the master processing device 2-1 in box 70.

FIG. 7 shows another embodiment of a system realized by using a plurality of processing devices 2-1 to 2-n such as a personal computer shown in FIG.

The video signal from the moving picture reproducing apparatus 10 includes
In some cases, the time code is not included, and even when the time code is included, the frame addresses are not continuous. In such a case, as shown in FIG. 7, a video signal from the moving image
Add the continuous frame address information generated in
The video signal 12 is input to the video distributor 20.
Other configurations and various functions are the same as those of the device shown in FIG. In this way, a continuous frame address is added to the video signal, and even a video without a frame number or a video with a discontinuous frame number can be processed by a plurality of processing devices. Processing results can be integrated.

FIG. 8 shows a configuration example of a time code generator which can be used in the embodiment of FIG. The video signal 12-a and the vertical synchronizing signal 12-b obtained from the moving image reproducing apparatus 10 are input, and the video signal 12 to which frame address information is added is output. The time code generator has a counter 71 that counts a vertical synchronization signal or a pulse generated based on the signal. The value of this counter increases by one. The contents of the counter are used as frame address information added for each frame. The content of this counter is converted into an analog signal by a D / A converter 72 and added to a scanning line portion of a non-video data portion of the video signal 2-a by a mixing section 73. Note that the counter 0
Reset is performed by a reset signal 11. A reset signal is supplied from the master processor only once when the system is started up via the moving image reproducing apparatus. Alternatively, a reset signal may be issued directly from the master processor.

[0044]

According to the above embodiment, the time code is directly generated from the non-video data included in the video signal supplied from one input port. Thus, there is no problem that the read time code is shifted from the actual video. further,
Since a plurality of detection processes can be independently processed by a plurality of processing devices, the processing capability of each processing device may be low, and thus a low-cost and scalable system can be realized.

[Brief description of the drawings]

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

FIG. 2 is a configuration diagram of a video signal when a video is processed on the system.

FIG. 3 is a block diagram for explaining a function of detecting a frame in which a plurality of events characterized by a search condition appearing in a video signal occurs.

FIG. 4 is an example of a list display of events detected in a video signal.

FIG. 5 is a block diagram of a video signal processing system according to an embodiment of the present invention, which is realized by using a plurality of processing devices.

FIG. 6 is a control flowchart of the system of FIG. 5;

FIG. 7 is a block diagram of a video signal processing system according to an embodiment of the present invention, which is realized by using a plurality of processing devices.

FIG. 8 is a block diagram illustrating a configuration example of a time code generator.

[Explanation of symbols]

1 display, 2 computer, 3 CPU, 4
... Memory, 5 pointing device, 6 external information recording device, 7 A / D converter, 8 address information decoder, 11 control line, 12 video signal, 13 digital image data, 14 time code, 20 ... Video distributor, 7
0: Time code generator.

Claims (18)

[Claims] 1. A video signal including a plurality of frames each including video data and non-video data, and a video signal including frame address information specifying a corresponding frame is input to the non-video data in a time-series manner in frame units. An A / D converter for converting video data in the video signal into a digital frame data string; digitizing the address information included in the non-video data in the video signal to generate a plurality of digital data; A second A / D converter for converting the time code into a plurality of time codes, a memory for storing the digital frame data sequence and a plurality of time codes, and calculating a feature value of each frame of the frame data sequence. Means for determining whether a feature satisfies a search condition for detecting frame data in which an event appears, and a method for determining whether the feature satisfies the search condition. Means for generating a record in which a set of frame data or an event code representing an event appearing in the frame data and a time code specifying the frame data are generated. 2. A video signal comprising a plurality of processing units and a plurality of frames each including video data and non-video data, wherein the non-video data includes frame address information for specifying a corresponding frame. A distribution device that distributes the distributed video signal to the plurality of processing devices, wherein each of the processing devices includes: an input unit configured to input the distributed video signal in a time series in frame units; A first A / D converter for converting video data in a video signal into a digital frame data sequence, and digitizing the address information included in non-video data in the video signal to convert the address information into a plurality of time codes A second A / D converter, a memory for storing the digital frame data sequence and a plurality of time codes, and a feature amount of each frame of the frame data sequence. Means for calculating and determining whether or not the feature amount satisfies a search condition for detecting frame data in which an event appears; and, when the feature amount satisfies the search condition, the frame data or the frame data. Means for generating a record as a set of an event code representing an event that has appeared and a time code for specifying frame data, wherein each of the processing devices calculates mutually different feature amounts, Determining whether the feature quantity satisfies a search condition, and in one of the processing devices, records generated by the remaining processing devices are arranged in a time series based on the time code; Processing system. 3. A time code generator for adding address information specifying each frame to a non-video data of a video signal having a plurality of frames each including video data and non-video data, Video signal comprising: a plurality of processing devices; and a distribution device connected to the time code generator so as to receive the video signal added with the address information from the time code generator, and distributing the video signal to the plurality of processing devices. A processing system, wherein each of the processing devices includes: an input unit configured to input a video signal to which the address information is added in a time series on a frame basis; and converts video data in the video signal into a digital frame data sequence. A first A / D converter, and digitizing the address information included in the non-video data in the video signal A second A / D converter for converting the data into a plurality of time codes, a memory for storing the digital frame data sequence and the plurality of time codes, and calculating a feature amount of each frame of the frame data sequence. Means for determining whether or not the feature value satisfies a search condition for detecting frame data in which an event appears; and, when the feature value satisfies the search condition, the frame data or the frame data Means for generating a record in which an event code representative of the generated event and a time code specifying the frame data are set, and each of the processing devices calculates a feature amount different from each other, and the feature amount is Determine whether the search criteria are satisfied,
A video signal processing system, wherein one of the processing devices is arranged in a time series based on the time code of records generated by the remaining processing devices. 4. A means for generating a digital frame image sequence from an analog moving image signal including a plurality of frames; a means for generating a time code added to each frame image of the frame image sequence from the analog moving image signal; A processing program that includes the digital frame image sequence and event code data representing the event, and detects a frame in which the event appears in the frame image sequence; a memory that stores the time code; And performs the detection processing,
Control means for creating a record in which a set of a frame image detected as a result of the processing or an event code representing an event appearing in the frame image and a time code specifying the frame image are provided. Video signal processing system. 5. A means for generating a digital frame image sequence from an analog moving image signal including a plurality of frames; a means for generating a time code added to each frame image of the frame image sequence from the analog moving image signal; A processing program that includes the digital frame image sequence and event code data representing an event, detects a frame image in which an event appears in the frame image sequence, and a memory that stores the time code; Read out the program and execute the detection process;
Control means for creating a record in which an event code representative of a frame image detected as a result of the detection processing or an event appearing in the frame image and a time code specifying the frame image, and a display device; A video signal processing system, wherein the display device displays an appearance time range of the detected frame image corresponding to a frame image sequence displayed based on the detected time code. 6. A connection portion inserted into an expansion slot of a personal computer and connected to a bus line of the personal computer, and frame address information of the video signal is read from a video signal input through the connection portion. An extension board having an address information reading function provided with an address information decoder for outputting an address information. 7. A plurality of frames each including video data and non-video data, wherein the non-video data includes frame address information for specifying a corresponding frame.
Input means for inputting a video signal in a time series on a frame basis; a first memory area for converting video data in the video signal into a digital frame data string and storing the digital video data; and a non-video data included in the video signal A second memory area for digitizing the obtained address information and converting it into a plurality of time codes and storing the digitized address information, and a procedure for obtaining a feature amount of each frame of the frame data sequence including event code data representing an event. A third area for storing a processing program including:
A storage unit having a fourth area for recording a record in which the frame data or an event code representing an event appearing in the frame data and a time code specifying the frame data are recorded; Control means for reading out and executing the processing indicated therein. 8. A video signal comprising a plurality of processing units and a plurality of frames each including video data and non-video data, wherein the non-video data includes frame address information for specifying a corresponding frame. A distribution device that distributes the distributed video signal to the plurality of processing devices, wherein each of the processing devices includes: an input unit configured to input the distributed video signal in a time series in frame units; A first memory area for converting the video data in the video signal into a digital frame data string and storing the digital data; and converting the address information included in the non-video data in the video signal into a plurality of time codes. Including a second memory area for storing and storing event code data representing an event and obtaining a feature amount of each frame of the frame data sequence. A third area for storing a program,
A storage unit having a fourth area for recording a record in which the frame data or an event code representing an event appearing in the frame data and a time code specifying the frame data are recorded; Control means for reading and executing the processing indicated therein, wherein each of the processing devices independently executes a different processing program, and the one of the video processing devices executes the different processing program. A video signal processing system, wherein the processing results obtained are integrated based on the time code. 9. A method of processing a video signal having a plurality of frames each including video data and non-video data, wherein the non-video data includes frame address information for specifying a corresponding frame. A signal is input to a processing device in a time series in a frame unit, and in the processing device,
The video data of the video signal is converted into a digital frame data sequence, the digital frame data sequence is recorded in a memory, and the address information included in the non-video data of the video signal is digitized and converted as a time code. Recorded in a memory, calculates the feature amount of each frame of the frame data sequence, determines whether the feature amount satisfies a search condition for detecting frame data in which a certain event appears, and determines whether the feature amount satisfies the search condition. A video signal processing method comprising the steps of: generating a record in which the frame data or a time code representative of the frame data is set. 10. The method according to claim 9, wherein the step of calculating the characteristic amount and determining whether the characteristic amount satisfies the search condition in the processing device detects a change point in the frame data sequence of the input video signal. A video signal processing method. 11. A video signal input port through which an analog video signal including an analog video data sequence and analog non-video data is supplied, and a video signal input port supplied through said video signal input port. A first converter for converting an analog video data sequence into a digital frame data sequence; and a non-video data analog video signal supplied via the video signal input port, each of which corresponds to one of the digital frame data sequences. To convert a plurality of time codes representing one digital frame data into a plurality of time codes
A memory for storing the digital frame data string and the time code corresponding thereto, an event represented by an event code, and means for setting a search condition for detecting such an event; and the memory Detect digital frame data that satisfies the above search condition from the digital frame data sequence in
Means for instructing a time code specifying the detected frame data and an event code representing an event appearing in the detected digital frame. 12. A video signal processing apparatus according to claim 11, further comprising a time code generator for adding, as a part of analog non-video data of said analog video signal, to each video data of said analog video data sequence. system. 13. A video signal processing system according to claim 11, further comprising a moving picture reproducing apparatus, wherein an output terminal thereof is connected to said video signal input port for supplying said analog video signal. 14. The method according to claim 11, further comprising a tuner,
A video signal processing system, wherein the output terminal is connected to the video signal input port for supplying the analog video signal. 15. A plurality of processing devices each having a video signal input port, and an analog video signal including an analog video data sequence and analog non-video data is supplied to each video signal input port of each of the plurality of processing devices. A video signal distribution unit for distributing the video signal, wherein each of the plurality of processing devices converts an analog video data stream of the video signal supplied via the video signal input port into a digital frame data stream. And a plurality of time codes each of which specifies analog one non-video data of the video signal supplied through the video signal input port and a corresponding one of the digital frame data strings. Convert to the second
A memory for storing the digital frame sequence and the time code corresponding thereto, an event represented by an event code, and a means for setting a search condition for detecting such an event; Detect digital frame data that satisfies the above search condition from the digital frame data sequence of
Means for indicating a time code specifying the detected frame data and an event code representing an event appearing in the detected digital frame data, wherein one of the processing devices manages the remaining processing devices. A video signal processing system for collecting detection results obtained by the remaining processing devices. 16. A method for processing an analog video signal including an analog video data sequence and analog non-video data, comprising: converting an analog video data sequence of the video signal into a digital frame data sequence; Converting the analog non-video data into a plurality of time codes that respectively specify the corresponding digital frame data; storing the digital non-video data in the digital frame data sequence and a plurality of time code memories; Digital frame data that satisfies a search condition for detecting a certain event is detected from a frame data sequence, and a time code specifying the detected digital frame data and an event appearing in the detected digital frame data are detected. Indicate a representative event code And a method for processing an analog video signal. 17. An analog video signal including the analog video data sequence and the analog non-video data so that each time code correctly specifies a corresponding digital frame data in the digital frame data sequence. A method for processing an analog video signal supplied via a video signal input port. 18. The method for processing an analog video signal according to claim 16, further comprising: searching for a frame using the event code and the time code; and displaying the searched frame.