* NOTICE*
`
`JPO and INPIT are not responsible for any damages caused bythe useofthis translation.
`
`1. This document has been translated by computer. So the translation may notreflect the original precisely.
`
`2.“ shows a word which cannotbetranslated.
`
`3.
`
`In the drawings, any words are nottranslated.
`
`Patent Number
`
`JP2002260166A
`
`Bibliography
`
`(19) [Publication country] JP
`
`(12) [Kind of official gazette] A
`
`(11) [Publication number] 2002260166
`
`(43) [Date of publication of application] 20020913
`
`(54)
`
`[Title
`
`of
`
`the
`
`invention)
`
`INFORMATION DISTRIBUTION DEVICE AND
`
`INFORMATION DISTRIBUTION SYSTEM
`
`(51) [International Patent Classification 7th Edition]
`
`GO8G 1/04 HO4N=5/225GO6T 1/00 GO8G) 1/13 H04B 7/26
`
`
`
`
`
`
`
`HO4N 7/173
`
`/G08G 1/00
`
`GO08G
`
`1/09
`
`[Fl]
`
`GO08G
`
`1/04
`
`D
`
`GO6T 1/00
`
`330A
`
`GO08G
`
`1/13
`
`
`
`HO4N=5/225 Cc
`
`7H173=6102
`
`GO08G
`
`1/00
`
`1/09
`
`A
`
`H04B 7/26
`
`A
`
`M
`
`(21) [Application number] 2001057082
`
`(22) [Filing date] 20010301
`
`(71) [Applicant]
`
`[Name] OMRON CORP
`
`(72) [Inventor]
`
`

`

`[Full name] KOBAYASHI HIDEYUKI
`
`[Full name] ANDO TANICHI
`
`[Full name] MUKOGAWA SHINICHI
`
`[Full name] SHIMIZU ATSUSHI
`
`[Full name] MITSUDA MASA
`
`Abstract
`
`(57) [Overview]
`
`PROBLEM TO BE SOLVED: To provide an information distribution device capable of
`
`easily performing determination even by a user with less knowledge and experiences.
`
`SOLUTION: Image processing is performed on image data D from an image pick-up
`
`means to extract meaningful
`
`information | associated with the image data. The
`
`meaningful information is distributed to a terminal 5 of a user 5' together with the picked-
`
`up image D or the appropriately processed image D'. The user can easily perform
`
`determination by referring to the meaningful information.
`
`Claim
`
`[Patent Claims]
`
`[Claim 1] An information distribution device for distributing information to a user via a
`
`network, and an information extraction means for performing image processing on the
`
`image data from the
`
`imaging means
`
`and extracting meaningful
`
`information
`
`accompanying the image data;
`
`Means for generating distribution image data from the image data;
`
`Distribution information generating means for synthesizing the distribution image data
`
`and the meaningful information to generate distribution information ;
`
`An information delivery device comprising : communication means for outputting the
`
`delivery information to the network.
`
`[Claim 2] A sensor for acquiring information other than an image is provided.
`
`An information distribution apparatus according to claim 1, wherein said distribution
`
`information generating means has a function of synthesizing and outputting the
`
`meaningful information obtained from said sensor.
`
`[Claim 3] An information distribution apparatus according to claim 1 or 2, wherein said
`
`distribution information generation means determines a type of meaningful information
`
`to be combined in response to a request received via said network.
`
`[Claim 4] The imaging means is for imaging a road.
`
`An information distribution apparatus according to any one of claims 1 to 3, wherein said
`
`

`

`meaningful information is traffic relationship information of said road.
`
`[Claim 5] An information extracting means for performing image processing on the
`
`image data from the
`
`imaging means
`
`and extracting meaningful
`
`information
`
`accompanying the image data;
`
`Means for generating distribution image data from the image data;
`
`A plurality of information distribution devices including distribution information generation
`
`means for synthesizing the distribution image data and the meaningful information to
`
`generate distribution information ;
`
`An information delivery system comprising an integration processing device which
`
`receives information output from a plurality of information delivery devices, integrates
`
`the information to generate integrated meaningful
`
`information, and outputs the
`
`integrated meaningful information to a network.
`
`Description
`
`[Detailed description of the invention]
`
`[0001]
`
`[Technical field of invention] The present invention relates to an information distribution
`
`apparatus and an information distribution system, and more particularly, to a system
`
`which usesbidirectional wireless communication such as a mobile phone and distributes
`
`traffic congestion information and other traffic related information.
`
`[0002]
`
`[Prior art] Typical traffic information includes highway radios on highway and traffic
`
`information that is passed during general radio broadcasts. This information is used
`
`when the user wants to listen, because the place where the broadcast
`
`is being
`
`broadcastedis limited (in the former case), and the time to be broadcasted is determined
`
`(in the latter case).
`
`It is difficult to obtain necessary information. Furthermore,
`obtain current information in real time.
`
`it has not been possible to
`
`[0003]In order to solve such a problem, a video camera captures a situation such as a
`
`traffic jam on a road and stores the captured image data in a server. Then, the image
`
`data stored in the server is distributed to a predetermined terminal via the Internet or the
`
`like. In other words, the user operates a terminal capable of displaying image data and
`
`accessesthe server. Then, it selects image information obtained by picking up an image
`
`of a desired point. With this selection, the image data stored in the server is output to the
`
`display screen of the terminal.
`
`

`

`[0004]Then, since the image data currently being picked up by the video camerais stored
`
`in the server, it is possible to distribute the image data which is being captured by the
`
`terminal. Therefore, when desired, the user can acquire the state of the necessary
`
`location (a predetermined location where the video camera is installed) as the image
`
`information. Accordingly, the user can know the degree of congestion, the causeof the
`
`congestion, and the like by looking at the acquired image information.
`
`[0005]
`
`[Problem to be solved by the invention] However, in the conventional system, although
`
`many information can be obtained based on the distributed image data, in order to obtain
`
`suchinformation, it is necessary for the user to view the image data and understand the
`
`contents thereof. In other words, to obtain useful information, knowledge and experience
`
`are required to some extent on the user side. Further,it is still difficult for the user to
`
`effectively utilize the acquired information, and it is difficult for all users to obtain and
`utilize the correct information.
`
`[O006]To provide an information distribution device and an information distribution
`
`system capable of easily making use of image information to be distributed and easily
`
`determining even a user with little knowledge / experience.
`
`[0007]
`
`[Means for solving the problem] An information distribution apparatus according to the
`
`present invention is an apparatus for distributing information to a user through a network.
`
`An information extracting means for performing an image processing on image data from
`
`an imaging means and extracting meaningful information accompanying the image data;
`
`It has a means to generate the image data for distribution from the aforementioned image
`
`data, a delivery information generating means which synthesizes the aforementioned
`
`image datafor distribution, and the above-mentioned significant clever information, and
`
`generates delivery information, and a means of communication which outputs the
`
`delivery information to a network.
`
`[0008]"Distributing information to a user" includes a case in which an information
`
`distribution apparatus directly transmits information to a user terminal, and a case in
`
`which the information is once transmitted to another apparatus and indirectly distrib uted
`
`to a user terminal via the apparatus. The meaningful information is information suitable
`
`for performing determination accurately and quickly when judging from the distributed
`
`image.
`
`Information that is difficult to determine from an image is determined, and it takes time
`
`to determine, and a meaningful information that requires an experience, Knowledge, or
`
`

`

`the like is extracted as meaningful information in advance. In the embodiment, character
`
`information is displayed as character information, but various kinds of symbols may be
`used.
`
`[O009]The meaningful information extracting means maydirectly receive the image data
`
`sent from the imaging means and perform processing, or may temporarily store the
`
`image data in the image storage means and read the image data and perform processing.
`
`In the embodiment, the means for generating the distribution image data is realized by
`
`the preprocessing unit 17 a. Note that "generating image data for distribution” is referred
`
`to as "generating image data for distribution", but it is not always necessary to perform
`
`processing for an image as in the embodiment. Whenthe image data sent from the image
`
`pickup means can be distributed as it
`
`is, processing such as selection of necessary
`
`image data without processing is performed.
`
`[0010]According to the present
`
`invention, since the meaningful
`
`information is also
`
`distributed when the image data is distributed,
`
`the received user can quickly and
`
`accurately perform the determination regardless of the degree of the knowledge and
`
`experience by viewing the meaningful information together with the image. Furthermore,
`
`since an image is normally sent, various kinds of information which cannot be obtained
`
`by meaningful information depending on the ability of an individual can be obtained.
`
`[0011]In a preferred embodiment of the present invention, it is provided with a sensor for
`
`acquiring information other than an image, and the distribution information generating
`
`means hasa function of synthesizing and outputting the meaningful information obtained
`
`from the sensor. Of course, such a sensoris not necessarily provided. When the sensor
`
`is provided as described above,
`
`information which is difficult to be understood in an
`
`image can be distributed together, so that more accurate determination can be quickly
`
`performed.
`
`[0012]Further, various kinds of meaningful information to be distributed are prepared.
`
`Accordingly, all of them may be distributed, or the distribution information generating
`
`means may determine the type of meaningful information to be combined in responseto
`
`a request received via the network. In this case, since only information required by the
`
`user is transmitted, there is no useless information for the user, and it is possible to
`
`immediately view and determine necessary information. In particular, when the terminal
`
`has a small display screen, if there is more meaningful information to be delivered, it is
`
`difficult to see an image, and there is a possibility that the image is obstructed when it is
`
`determined that the image is viewed, but
`
`the problem is solved by sending only
`
`necessary one.
`
`[0013]It is to be noted that the image pickup means is intended to image a road, and the
`
`

`

`meaningful
`
`information may be used asthe traffic relation information of the road.
`
`According to such an application, a traffic jam situation or the like can be acquired in real
`
`time and is useful in determining a route to be traveled. Note that the traffic relationship
`
`information includes various kinds oftraffic such as the traveling state of a vehicle such
`
`as a traffic jam state and the like, and the state such as a road surface or other road.
`
`[0014]In an information distribution system according to the present
`
`invention, an
`
`information extracting unit for performing an image processing on image data from an
`
`imaging unit and extracting meaningful information associated with the image data is
`
`provided. A plurality of information distribution devices including a means for generating
`
`image data for distribution from the image data and a distribution information generation
`
`means for synthesizing the image data for distribution and the meaningful information to
`
`generate distribution information are provided. FIG. 3 is a block diagram showing a
`
`configuration of an integrated semantic information management system according to an
`
`embodiment of the present invention ;.
`
`[0015]It is not always necessary for each information distribution apparatus to be of the
`
`sametype, and it is also possible to have different functions. In addition, it is of course
`
`possible to add other functions such as an external sensor in addition to the above-
`
`described functions. By thus making an integrated judgment and generating the
`
`integrated meaningful
`advanced information.
`
`information,
`
`it
`
`is possible to distribute more complicated and
`
`[0016]The above described components of the invention can be combined as much as
`
`possible. Each means constituting an information distribution apparatus according to the
`
`present invention can be realized by a dedicated hardware circuit, or can be realized by
`
`a programmed computer.
`
`[0017]
`
`[Embodimentof invention] FIG. 1 shows an example of an entire information distribution
`
`system using a network to which the present invention is applied. As shown in FIG. 1, a
`
`plurality of information collecting apparatuses 1 and a server 2 are connected to a
`
`network 3 such as the Internet, and images and other information collected by the
`
`information collecting apparatus 1 are sequentially transmitted to the server 2 via the
`
`information is processed and arranged on the server 2 and
`network 3. The sent
`converted into a form suitable for distribution.
`
`[0018]Further, by connecting the terminal 5 to the network 3, the user downloads
`
`information collected by the desired information collecting apparatus 1 to the terminal 5
`
`and acquires the information. In other words, in response to a request issued from a user
`
`

`

`(terminal), the server 2 distributes the stored predetermined information to the user
`
`terminal 5 via the network3.
`
`[0019]Thus, the user can obtain information collected by the information collection device
`
`1 whenever the environment and situation can be connected to the network 3. Thus, by
`
`operating and storing information acquired by the information collection device 1
`
`in real
`
`time on the server 2 and storing the information, the user can obtain and determine the
`
`necessary information at any time.
`
`[0020]The information collecting device 1
`
`is provided with an imaging device whichis
`
`installed around the road and capturesthe situation of the installed road. Specifically,
`
`FIG. 2 and Fig. 3 are as shownin Fig. 1. As shownin FIG. 4, the imaging device 10 is
`
`composed of a video camera using a CCD (charge-coupled device).
`
`To output digital data (digital moving picture information) such as a nJPEG format. This
`
`output digital moving picture information is supplied to the data processing apparatus 11.
`
`[0021]An imaging device 10 composed of a video camera captures a state of a road in
`
`a monitoring area and obtains moving image data. Then, this photographing process
`
`operates so as to execute the flowchart shown in FIG. 5. In other words, the imaging
`
`device 10 has an internal timer, and takes an image by the CCD camera every time a
`
`prescribed time (photographing interval) elapses (ST 11 to ST 13). Here, a video image
`
`is assumed to be a moving image, and a resolution of 720 x 480 pixels is relatively high.].
`
`For example, the imaging interval is 30 frames / second. Thus, the image data obtained
`
`at predeterminedintervals is converted into digital data (ST 14), and the obtained image
`
`data is temporarily stored in the temporary image storage device (ST 15) and transferred
`
`to the data processing device 11.
`
`In the photographing process, the above-described
`
`processing is repeatedly executed.
`
`[0022]The data processing device 11 has a processing capability of a normal PC level,
`
`and changesthe digital moving image information captured by the imaging device 10 to
`
`extract information which is easy to be used for the processing, and also changesthe
`
`data amountto be easily distributed to a terminal connected to a network having a small
`
`line capacity by, for example, compressing, compressing or deleting unnecessary data.
`
`[0023]Further, the data processing apparatus 11 is connected to a storage 12 configured
`
`of a hard disk or other storage medium, and stores an image processed by the data
`
`processing apparatus 11 and data for processing. As data for processing, there may be
`
`used an image which is picked up by an image pickup device 10 and which is to be
`
`compared with an image to be processed, an image whoseresolution is changed, and
`extracted data.
`
`

`

`[0024]Then,
`
`the image and other information processed by the data processing
`
`apparatus 11 are given to the communication apparatus 13. The communication device
`
`13 transmits the acquired image and other information to the network 3. In other words,
`
`by generating a transmission frame or the like to which an address of the server 2 of the
`
`transmission destination is added and transmitting it on the network 3, the processed
`
`image or other information is collected and accumulated in the server 2.
`
`[0025]FIG. 3 showsan internal configuration of the data processing apparatus 11. That
`
`is,
`
`it has an input unit 15 that receives output data from the imaging device 10 and
`
`acquires information received via the communication device 13.
`
`Information (data)
`
`acquired by the input unit 15 is supplied to the CPU 17.
`
`[0026]The CPU 17 reads out predetermined data stored in an external storage device
`
`(storage) 12 and executes predetermined processing on the image data received from
`
`the input unit 15 by appropriately using a memory 14 as a work memory. Then, the
`
`generated processing data is passed to the output unit 16. The output unit 16 transmits
`
`the received processing data to the communication device 13, and enables transmission.
`*
`
`The processing unit is connected to the bus and transfers data via the bus. Further, the
`
`data processing device 11 and the communication device 13 can be realized by, for
`
`example, a personal computer.
`
`[0027]Here,
`
`in the present invention, as shown in FIG. 4, the processing in the data
`
`processing apparatus 11 performs the feature extraction based on the acquired image
`data D or the pre-processed image data D ’
`obtained from the image data D, and
`
`generates the meaningful information |. Actually, in consideration of transmission on the
`Internet, an image to be distributed is pre-processed and the image data D ’
`is
`
`the feature extraction is
`obtained in which the data capacity is reduced. Further,
`performed on the basis of the image data D (which may be acorrected D ’
`for clarity).
`Then, the image data (D ’
`) and the meaningful information | are paired and distributed.
`
`In other words, the meaningful information | extracted from the image is distributed
`simultaneously with the image.
`In practice, the related image data D ’
`and the
`meaningful information | are stored in the server 2, and the user 5 ’
`accesses the
`server 2 to download the image dataD ’
`and the meaningful information | related to
`
`the terminal 5 and display them.
`[0028]Then, the user 5 ’
`performs final determination while viewing the displayed
`image data D ’
`and the meaningful information |. At this time, since the meaningful
`
`information |
`
`is displayed together with the image, it is possible to easily and quickly
`
`

`

`perform an accurate determination based on the meaningful information | even if the
`
`experience and knowledge necessary for
`
`the determination are small. As
`
`the
`
`determination result, there is, for example, "to change a route,” "to change a route,” "to
`
`change a traffic jam, and so on, so that a traffic jam is cleared,” and so on.
`
`[0029]FIG. 6 showsa software configuration of the CPU 17 for extracting the meaningful
`
`information |. That is, since the image data is transferred every predetermined interval
`
`from the image pickup device 10 every 1 frames, the CPU 17 acquires the image data
`
`(shooting data) via the input unit 15. Specifically, it is supplied to the pre-processing unit
`
`17 a. Further, the image data is stored in an external storage device (storage) 12 asit is.
`
`[0030]In order to correctly extract the meaningful
`
`information, the imaging device 10
`
`captures image data with high accuracy. As a result, the capacity required for 1 frames
`
`becomeslarge, and ifit is left asitis, it becomesdifficult to distribute video on the premise
`
`of the Internet. Therefore, the preprocessing unit 17 a reduces the number of colors to
`
`be expressed so asto be able to distribute moving images on the premiseofthe Internet,
`
`and performs processing of reducing the resolution and the frame rate. Further, a
`
`processing for making the blurred portion of the image easy to see and improving the
`
`dynamic range of the contrast and the improved brightness and darknessof the focus is
`
`performed. This preprocessed image data is once stored in the storage 12 and used for
`
`other operations and image distribution.
`
`[0031]In this example, all of the image data that has been captured and transferred is
`
`executed, but it is also possible to perform, for example, an image that is necessary for
`
`receiving a distribution request and actually distributing information.
`
`In this case, for
`
`example, it can be implemented by a procedure shownin a flowchart shown in FIG. 7.
`
`In other words, upon receiving the input of the request image data, a necessary image
`
`is acquired from the storage (image temporary storage device) 12 (ST 21, ST 22). If there
`
`is no suitable image, the process jumps to Step 1 and waits for an input associated with
`
`the transfer of a new image (ST 23).
`
`[0032]Whena suitable image is obtained, an image processing such as a change in the
`
`numberof colors and a change in resolution is performed (ST 24, ST 25). Whether or
`
`not the request received in step 21 is a moving image is determined (ST 26), and in the
`
`case of a moving image, a necessary image is collected (ST 27). For this necessary
`
`image, a change in the number of colors, a change in resolution, and the like are
`
`performed. Then, the format is changed (ST 28). Note that image data (distribution image
`
`data) generated by changing this format may be temporarily stored in, for example, a
`
`distribution image data storage unit in the storage 12, or may be passedto the distribution
`
`

`

`information generation unit 17 c.
`
`[0033]Further,
`
`the information extracting unit 17 b for extracting the meaningful
`
`information extracts the photographed image data stored in the storage 12 and extracts
`
`the meaningful
`
`information therefrom. Specifically, a flowchart shown in FIG. 8 is
`
`implemented.
`
`[0034]In other words, as shown in FIG. 4, the data processing device 11 reads the highly
`
`accurate image data transferred from the imaging device 10 from the storage 12 which
`
`is an image temporary storage device (ST 31). Then, a necessary portion in the image
`
`is cut out (ST 32). For example, on the basis of the carriageway region data stored in the
`
`storage 12 or the like, the cutting-out portion cuts out the carriageway portion in the
`
`image.
`
`[0035]Then, a moving object is detected for the cut-out image (ST 33). The detection of
`
`the moving body can be performed by using a "background difference method", a "time
`
`difference method”, or the like provided as a general vehicle extraction algorithm. In other
`
`words, in the background difference method, a moving object is detected by preparing
`
`an image (background image) without a vehicle or a falling object in advance and taking
`
`a difference between the background image and the image to be processed. Also, the
`
`time difference method extracts an object moving between 2 images by performing a
`
`comparison operation on different images between time intervals 6.
`
`[0036]One of the 2 vehicle extraction algorithms may be used, but since each of the two
`
`vehicle extraction algorithms has a single long and short one, in this embodiment, both
`
`of the two vehicle extraction algorithms are employed and extracted by appropriately
`
`switching them according to the situation. That
`
`is, while the background difference
`
`method is suitable for detecting moving objects and stationary objects (e.g., parked
`
`vehicles), the accuracy of detection is due to the accuracyof the background image, and
`
`the reliability of the background image decreases, for example,
`
`if there is a sudden
`
`change in sunlight or shadow. Thatis,
`
`To cope with environmental variation. On the other hand, in the time difference method,
`
`there is a problem that, although correspondence to environmental variation and
`
`extraction of moving objects can be performed with high accuracy, a stationary object
`
`cannot be extracted. Therefore, the presence or absence of environmental variation
`
`(degree) is determined, and when the variation is small, extraction by the background
`
`subtraction method is performed, and when the variation is large, extraction by the time
`
`difference method is performed.
`
`[0037]Next, the vehicle is identified from the extracted moving object, and the numberof
`
`

`

`passes is counted (ST 34). Then,
`
`the number of passes is output as character
`
`information. In practice, the data is stored together with the date and time data counted
`
`in the storage unit 12.
`
`[0038]The count of the number of vehicles can be handled, for example, by the following
`
`algorithm. A rectangular passing vehicle detection area is set at a predetermined position
`
`in the cut-out region. The size of this rectangle allows the front of the vehicle to enter and
`
`is installed for each lane. Then, the vehicle uses the presence of a large number of
`
`horizontal edges to apply a differential filter in the passing vehicle detection area. As a
`
`result, when the extracted moving object is a vehicle and the front of the vehicle passes
`
`through the area, a peak is generated at an output of the differential filter, so that the
`
`presence or absence of the peak is determined. Then, the number of detected vehicles
`
`per unit time (the number of peaks) is the required numberof passes.
`
`[0039]Further, by tracking the vehicle, the speed of the vehicle is measured (ST 35).
`
`Then, the vehicle speed is output as character information. In practice, it is stored in the
`
`vehicle speed storage in the storage 12 along with the measured date and time data.
`
`[0040]This speed measurement of the vehicle can be handled, for example, by the
`
`following algorithm. In other words, a speed measurement areais set in each lane. This
`
`speed measurement area is set to correspond to a constant distance (a distance
`
`sufficiently longer than the length of 1 vehicles) along the lane. Then, the vehicle moving
`
`in this area is tracked. Then, a moving distance is calculated from the position at which
`
`tracking of the vehicle is started and the end position, and the speed is calculated from
`
`the tracking time.
`
`[0041]In actual tracking, a vehicle to be tracked is detected by a vehicle detection
`
`algorithm similar to a number counting algorithm, and image data of the vehicle is
`
`registered as a template. Then, template matching is performed on the incoming and
`
`outgoing image data, and when the vehicle matching the template is detected in the area,
`
`the vehicle is tracked. Tracking of the vehicle can be performed by assuming a moving
`
`direction of the vehicle and performing template matching on the assumedregion.
`
`[0042]Further, a congestion distance is calculated from the number of passages and the
`
`speed determined as described above (ST 36). And,
`
`The calculated congestion distance is output as character information. In practice, it is
`
`stored together with date and time data in the congestion information storage unit in the
`
`storage 12.
`
`[0043]This calculation of the congestion distance can be handled, for example, by the
`
`following algorithm. In other words, for example, when the traveling speed is high, it can
`
`

`

`be estimated that no congestion has occurred. In addition, even when the traveling speed
`
`is extremely slow, if the number of passes within a certain period of time is extremely
`
`slow, it can be said that the vehicle has traveled slowly and no traffic jam has occurred.
`
`On the other hand, even if the speedis slow, it can be estimated that a traffic jam has
`
`occurred if a certain number of passes have passed. Further, as a characteristic of the
`
`traveling state (speed) of the vehicle during the traffic jam, since the stop and the start
`
`are repeated, even when the speed variation frequently occurs within a low speed, it can
`
`be estimated that the vehicle is busy.
`
`In addition,
`
`in case of congestion,
`
`it can be
`
`estimated that the longer the stopped time and the slower the speed, the longerthe traffic
`
`jam distance. Then, the table which associatedthetraffic jam distance over the traveling
`
`condition (input condition) of vehicles, such as the number of passage and speed, can
`
`be created preliminarily, hold stores can be carried out to the storage 12 etc., and it can
`
`ask by referring to the above-mentioned table in calculation processing of Step 36.
`
`[0044]Of course, it is also possible to set a plurality of cameras along a road, for example,
`
`to determine a head ofa traffic jam and a tail, and calculate a traffic jam distance from a
`distance therebetween.
`
`[0045]Further,
`
`it
`
`is also possible to determine whether the current
`
`traffic jam is
`
`progressing or expanding from the time series data of the traffic jam distance, and to
`
`store and store the same. This may beindicated, for example, as an average of the past
`
`numberof congestion distances as 0%, and in the case of an increase, as X%, and ina
`
`case of decreasing, as -%. Although the congestion distance of the comparison
`
`reference is an average in the above example,
`
`the congestion distance may be
`
`compared with the 1 congestion distances of the previous or predetermined numberof
`
`times. Further,
`
`it is also possible to predict a congestion distance after a lapse of a
`
`predetermined time from a locus of a change in the congestion distance.
`
`[0046]Further, as the meaningful information to be extracted, other than the above, for
`
`example, a ratio of a vehicle type based on a vehicle direction, an interval, a size of a
`
`density vehicle, and the like, and a brightness of a photographing range can be
`
`determined by taking into consideration information from a camera to be photographed,
`
`and a lighting rate of a vehicle can be measured from search and tracking of a bright
`
`portion.
`
`[0047]The distribution information generation unit 17 c integrates and distributes the
`
`meaningful information obtained by the information extraction unit 17 b and the image
`
`data processed by the preprocessing unit 17 a, and specifically implements a flowchart
`
`shown in FIG. 9. In other words, the distribution request is waited (ST 41, ST 42), and if
`
`there is a request, the corresponding distribution image data is read from the distribution
`
`

`

`image data storage unit in the storage 12 (ST 43).
`
`[0048]The text which judges the existence of the text to synthesize (being usually), and
`
`corresponds in existing -- that is, -- Y
`
`Semantic information is read (ST 44, ST 45).
`
`In this case, the type of meaningful
`
`information to be read may be fixed in advance or may be read out according to a
`
`condition sent together with a distribution request from the user.
`
`[0049]Then, the read character information and the pre-processed distribution image
`
`data are combined, and the obtained distribution image (with the character information)
`
`is output (ST 46, ST 47). For example, the output destination of the distribution image
`
`may be a communication device 13 via an output unit, and may be transmitted to a
`
`networkasit is, or temporarily stored in the storage unit 12 and then transmitted. It is to
`
`be noted that the video to be distributed basically uses a live video with little delay, and
`
`provides information that does not lose real-time performance.
`
`[0050]Thus, for example, it is possible to integrate the information and the image itself,
`
`and to distribute the information according to the request of the user received through
`
`the network. For example, as shown in FIG. 10, an image obtained by taking an image
`
`of a situation of a road ina monitoring area is provided, whereas according to the present
`
`embodiment, character information is displayed on a display screen of the terminal 5 as
`
`shown in Fig. 11. By transmitting the data extracted from the video and video datain this
`
`way, it is possible to easily receive information whichis difficult to be judged only by the
`
`video.
`
`[0051]For example, the detection of the moving object (vehicle) and the accompanying
`
`processing associated therewith, for example, "Developmentof the traffic flow monitoring
`
`image processin