JP2009033562A - Image retrieving device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image retrieving device 3 for retrieving, in response to a request of image from an IP network system 2, images picked up by imaging apparatuses A1-An and recorded in recording devices B1-Bn in a monitoring system 1. <P>SOLUTION: A request accepting means 37 accepts information (URL information) requesting an image from the IP network system, an identification information detecting means 37 detects identification information (information on a path name or information containing the path name) of the relevant image from the accepted information requesting the image, a recording position detecting means 37, 33 detects a recording position of the image corresponding to the detected identification information, an image capturing means 37 captures the image recorded in a recording device in a monitoring system based on the detected recording position of the image, and an image output means 37 outputs the captured image to the IP network system. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image search apparatus that searches for an image, and more particularly to an image search apparatus that can search an image recorded in a monitoring system from an IP (Internet Protocol) network system.

For example, in a monitoring system, a video (image) of a monitoring target region is captured by a monitoring camera, and the captured image is monitored automatically or by human eyes.
As an example, searching for an image showing a specific person or object from a recorded image of a surveillance camera is performed while a person visually confirms an image recorded in time series for each surveillance camera. In such a work, it takes time and labor to visually confirm all the recorded images targeted by a person. In order to solve such a problem, for example, a technique for inputting a feature of a person and finding a desired image from an image database has been studied.

  On the other hand, when creating a report using the found recorded image or publishing it on the Web, download the image from the image recording device of the monitoring system to a file, and then paste the file into the application. It was done using it. This is because in the monitoring system, the image recording apparatus has a configuration suitable for image recording by the monitoring camera, and it is difficult to directly use the recorded image in a general application program.

JP 2005-236773 A

As described above, the recorded image in the conventional video surveillance system is closed to the use in the system, and when the image is reused outside, the image is temporarily downloaded from the image recording apparatus and used. It took time-consuming processing such as pasting on the paper.
The present invention has been made in view of such conventional circumstances, and an object of the present invention is to provide an image search apparatus that can search an image recorded in a monitoring system from an IP network system.

In order to achieve the above object, in the image search device according to the present invention, the image is picked up by the image pickup device in the monitoring system and recorded in the recording device in response to a request for an image from the IP network system with the following configuration. Search for an image.
That is, the request receiving unit receives information requesting an image from the IP network system. The identification information detecting unit detects the identification information of the image from the information requesting the image received by the request receiving unit. The recording position detecting means detects the recording position of the image corresponding to the identification information detected by the identification information detecting means. An image acquisition unit acquires an image recorded on the recording device in the monitoring system based on the recording position of the image detected by the recording position detection unit. The image output means outputs the image acquired by the image acquisition means to the IP network system.

  As an example of the configuration, the image search device according to the present invention includes correspondence storage means for storing the correspondence between the identification information of the image recorded in the recording device and the recording position of the image in the monitoring system. The recording position detecting means detects the recording position of the image corresponding to the identification information detected by the identification information detecting means based on the stored contents of the correspondence storage means.

  Therefore, in response to a request for an image from the IP network system, the image recorded in the recording device in the monitoring system can be searched, and the searched image can be supplied to the IP network system. It is possible to retrieve the recorded image from the IP network system.

  As a specific example, when there is a monitoring system that is an independent system having an imaging device and a recording device, and there is an IP network system that is a general Internet system, a URL is obtained from a Web browser of a computer that the IP network system has. An image is requested to the image retrieval apparatus via the specified Web server, and in response to this, the image retrieval apparatus retrieves and reads the requested image from the recording device in the monitoring system, and the image retrieval apparatus reads the image. The image can be sent to the request source of the IP network system.

Here, various configurations may be used as the IP network system and the monitoring system, respectively. For example, these systems do not have a direct interface for image search. An image search apparatus according to the present invention is used as an interface for image search.
In the monitoring system, for example, various numbers may be used as the number of imaging devices and the number of recording devices. As an example, the imaging device and the recording device may be associated with each other on a one-to-one basis. In the monitoring system, for example, a certain image captured by a certain imaging device and a recording position of the recording device on which the image is recorded (for example, when there are a plurality of recording devices, information specifying the recording device and the recording device) Correspond to each other in a one-to-one correspondence.

Various types of information may be used as the image identification information. For example, URL (Uniform Resource Locator) path name information or information including the URL path name may be used. Such image identification information can be regarded as address information of the image, for example.
Various information may be used as the path name information. For example, information including the identification information (for example, camera number) of the imaging device that captured the image and the time information of the image is used. Can do. As the time information of the image, information on various times related to the image may be used. For example, the time at the time of imaging by the imaging device, the time at the time of transmission at the transmission side, the time at the time of reception at the reception side Also, information such as the time of recording by the recording device can be used.
Various information may be used as information for requesting an image. For example, information that can specify an image to be requested is used. For example, URL information may be used.

In addition, various modes may be used for detecting the identification information of the image from the information requesting the image. For example, when the information requesting the image includes the identification information of the image, A mode of extracting identification information can be used, or as another configuration example, a predetermined conversion process (predetermined calculation process) is performed on information requesting an image to acquire the identification information of the image Various embodiments may be used.
Various information may be used as the information for specifying the recording position of the image. For example, the IP address of the recording device, the identification information (for example, camera number) of the imaging device, and time information are used. As another example, it is possible to use identification information of a recording apparatus and information on an address (address of a recording area) in the recording apparatus.

  Various modes may be used as a mode for acquiring an image recorded on a recording device in the monitoring system based on the detected recording position of the image. For example, the image is present at the detected recording position of the image. For example, when an image does not exist at the recording position of the image detected by the recording apparatus, the image can be acquired near the recording position (for example, closest). A mode in which an image existing at a certain other recording position is acquired can also be used. As an example, when time-series images are sequentially recorded at consecutive recording positions, an image close in time to an image existing at the certain recording position is present at another recording position near the certain recording position. It is recorded.

Various units may be used as the unit in which the image is handled. For example, the unit may be handled in units of image frames and may be recorded or searched, or may be used for each of a plurality of predetermined image frames. The image data may be grouped into groups, and may be recorded or searched by being handled in units of each image group.
As one configuration example, in response to a request for an image from an IP network system, an image retrieval device that retrieves an image captured by an imaging device and recorded in a recording device in a monitoring system,
Request receiving means for receiving, as information for requesting an image from the IP network system, URL information including identification information of an imaging device from which an image is captured and path name information having image time information;
Identification information detecting means for detecting path name information included in the URL information or information including the path name as identification information of the image from URL information requesting the image received by the request receiving means;
A recording position detecting means for detecting a recording position of an image corresponding to the identification information detected by the identification information detecting means;
Image acquisition means for acquiring an image recorded in the recording device in the monitoring system based on the recording position of the image detected by the recording position detection means;
Image output means for outputting the image acquired by the image acquisition means to the IP network system;
It is provided with.

The present invention can also be provided as a method, a program, a recording medium, and the like.
In the method according to the present invention, each unit executes various processes in the apparatus or system.
The program according to the present invention is executed by a computer constituting the apparatus or system, and causes the computer to function as various means.
In the recording medium according to the present invention, a program to be executed by a computer constituting an apparatus or system is recorded so as to be readable by input means of the computer, and the program executes various processes (procedures) on the computer. Let

  As described above, according to the image search device of the present invention, for example, it is possible to search an image recorded in a monitoring system, which has been an independent system in the past, from a general IP network system. it can.

Embodiments according to the present invention will be described with reference to the drawings.
FIG. 1 shows a configuration example of an image processing system having an image search device 3 according to an embodiment of the present invention.
The image processing system of this example includes a video surveillance network system 1 that is an independent system, a general IP network system 2, an image search device 3 that mediates these two systems 1 and 2, and a sensor network system 4. Have.

The video surveillance network system 1 includes, for example, a plurality of n imaging devices A1 to An each composed of a camera, a video surveillance network 11, and n image recording units B1 to Bn, for example.
Here, the imaging devices A1 to An and the image recording units B1 to Bn are connected to the video surveillance network 11.
Further, in this example, the image recording units B1 to Bn are provided for the respective imaging devices A1 to An, and the images captured by the imaging devices A1 to An are recorded in the corresponding image recording units B1 to Bn. It becomes the composition which is done.

The IP network system 2 includes a plurality of m web servers C1 to Cm, a plurality of x web browsers E1 to Ex provided in a personal computer (PC), for example, and an IP to which these are connected. A network 21 is provided.
The sensor network system 4 includes a plurality of z sensor nodes D1 to Dz and a sensor network 41 to which these are connected. Each sensor node D1 to Dz is provided with sensors for detecting various events.

  Here, in this example, the video surveillance network 11 is an independent network dedicated to the video surveillance network system 1, and the IP network 21 is a general Internet network. Also, any sensor network 41 may be used. For example, a dedicated independent network may be used, or an Internet network may be used.

The image search device 3 includes a frame selection unit 31, an address assignment unit 32, an address conversion unit 33, a feature amount storage unit 34, an image display unit 35, a search condition input unit 36, and an image search unit 37. ing. These processing units 31 to 37 are connected through, for example, a data bus, and are connected to the video monitoring network 11, the IP network 21, and the sensor network 41.
The image recording units B1 to Bn are objects to be searched for by the image search device 3 and can be considered to be included in the image search device 3, but are originally provided in the video surveillance network system 1. is there.

An outline of operations performed by the respective processing units 31 to 37 of the image search apparatus 3 of the present example is shown.
Each image recording part B1-Bn records the image image | photographed with imaging device A1-An, such as a surveillance camera, in time series in an imaging device unit. In this example, a frame of an image to be recorded is recorded in a recording position corresponding to a camera number that is information for identifying each of the imaging devices A1 to An and information on the time of shooting.
In this example, each of the image recording units B1 to Bn corresponds to a network digital recorder or the like that records images taken by the imaging devices A1 to An via a network.

The frame selection unit 31 selects one image frame from time-series continuous image frames taken by the imaging devices A1 to An.
Here, the image frames supplied to the frame selection unit 31 may be supplied directly from the imaging devices A1 to An, or may be supplied from the image recording units B1 to Bn, for example.
Various methods may be used as the method for selecting an image frame. For example, a method for selecting at regular intervals, a method for selecting an image frame when there is a change in an image, an alarm by a sensor, and the like. For example, a method of selecting in synchronization with notification from the outside can be used. As a specific example, in the method of selecting an image frame when there is a change in the image, changes in hue, shape, etc. are monitored for each image frame, and continuous image frames are divided into meaningful units by the change. , Select a representative frame in that unit. As another configuration example, a method of selecting an image frame in accordance with an instruction from a user (person) may be used.

The address assigning unit 32 assigns an address to the image frame selected by the frame selecting unit 31.
In this example, a URL that makes it easy to retrieve an image by a Web application is used as the format of the address to be assigned.
Here, the URL is composed of information of {scheme name (for example, protocol name such as http) + host name (for example, host name to be accessed) + path name}. In this example, the path name is used. To identify the image frame. For example, the path name “/ c1 / 20070227/14100105” is assigned to the frame of the JPEG image captured at the time 14: 10: 01: 05 on February 27, 2007 by the imaging apparatus A1 of the camera number 1, and “/ c1 / 20070227 / 14100105.jpg ”is assigned.
Note that the processing of the address assigning unit 32 may be performed automatically, for example, or may be performed in response to an instruction from a user (person).

The address conversion unit 33 associates the address (path name in this example) given by the address giving unit 32 with the recording position information in the image recording units B1 to Bn of the image frame selected by the frame selection unit 31. And store it in the memory.
In this example, when a path name is designated, the address conversion unit 33 responds with recording position information in the corresponding image recording units B1 to Bn. The recording position information in the image recording units B1 to Bn is, for example, when the selected image frame is read from the image recording units B1 to Bn, or when the selected image frame is written to the image recording units B1 to Bn. Can get.
Generally, when an image frame is read from an image recording apparatus (in this example, the image recording units B1 to Bn) of the video surveillance system (in this example, the video surveillance network system 1), the image recording apparatus The camera number and time information are designated to read out the corresponding image frame.

FIG. 2 shows an example of a data record format in the address conversion unit 33.
In this example, the record format is such that the IP address information, camera number, and time information of the image recording units B1 to Bn are registered in association with the path name.
That is, in this example, path name information is used as address information provided by the address assigning unit 32, and IP addresses, camera numbers, and time information of the image recording units B1 to Bn are used as recording position information.

The feature amount storage unit 34 stores the feature amount and metadata of the image frame selected by the frame selection unit 31 in the memory in association with the address assigned by the address assignment unit 32.
FIG. 3 shows an example of a record format of data stored in the feature amount storage unit 34.
In this example, the record format stores the feature amount and metadata of the selected image frame in association with the assigned path name. For example, only one of the feature amount and metadata corresponding to the path name may be registered.

Here, as the feature amount, in this example, a multidimensional vector indicating the feature of the image is used. Note that the feature amount quantitatively indicates the feature of the image such as color and shape, and the feature amount may be extracted by, for example, the frame selection unit 31 or separately. You may provide the process part (feature amount extraction part) to extract in imaging device A1-An, such as the image search apparatus 3 or a surveillance camera.
As metadata, for example, time information, sensor event types, and character information such as person information are used. Note that the metadata may be automatically created, for example, by acquiring log information acquired at the time of sensor alarm in the sensor network system 4 or at the time of imaging by the imaging devices A1 to An in the video surveillance network system 1. Alternatively, the metadata may be registered and created manually.

FIG. 4 shows an example of a processing procedure when image frames are supplied from the imaging devices A1 to An to the frame selection unit 31.
The frame selection unit 31 acquires images captured by the imaging devices A1 to An (step S1), and determines whether to select a frame of the image (step S2). When the frame of the image is selected, an address is assigned to the selected image frame by the address assigning unit 32 (step S3), and the address converting unit 33 converts the address of the selected image frame and the recording position information of the selected image frame. (Step S4). Then, the captured image frame is stored in the image recording units B1 to Bn (step S5).

FIG. 5 shows an example of a processing procedure when image frames are supplied from the image recording units B <b> 1 to Bn to the frame selection unit 31.
The frame selection unit 31 reads out the images recorded in the image recording units B1 to Bn (step S11), and determines whether to select a frame of the image (step S12). When the frame of the image is selected, an address is assigned to the selected image frame by the address assigning unit 32 (step S13), and the address converting unit 33 converts the address of the selected image frame and the recording position information of the selected image frame. (Step S14).

The image display unit 35 displays information such as an image on a screen, for example. In this example, the image display unit 35 displays information including a path name corresponding to an image (for example, an image frame) supplied from the image search unit 37.
The search condition input unit 36 inputs image search conditions in accordance with, for example, an operation by a user (person). In this example, a condition for searching for an image from the image recording units B1 to Bn is input as an image search condition, and a character string or an image is used for this input. When an image is used for this input, for example, a processing unit (a feature amount conversion unit that performs processing for converting an image of a search condition into a feature amount of an image such as a hue or a shape. May be included in the apparatus configuration.

The image search unit 37 reads image frames from the image recording units B1 to Bn using the recording position information.
Specifically, the image search unit 37 first extracts a path name of a record that matches the search condition from the feature amount storage unit 34 using the search condition. Next, using the path name, the recording position information of the image frame in the image recording units B1 to Bn corresponding to the path name is obtained from the address conversion unit 33. Next, using the recording position information, an image frame corresponding to the path name is read from the image recording units B1 to Bn. In addition, the image search unit 37 supplies information including a path name corresponding to the read image frame to the image display unit 35.
In addition, as a process which reads an image frame from image recording part B1-Bn, it is possible to use the process similar to the process in the past, for example.

  In addition, as an example of a method for extracting a path name from the feature amount storage unit 34, when using metadata, when the search condition character string is included in the metadata of the record in the feature amount storage unit 34, A method of extracting the path name can be used. As another example, when using a feature amount, the similarity between the feature amount of the image of the search condition and the feature amount of the record in the feature amount storage unit 34 is calculated (for example, a predetermined threshold value) For example, a method of extracting a path name of a record having a high similarity can be used. Furthermore, it is possible to use a method of extracting a path name using both metadata and feature amount conditions (for example, AND conditions).

Here, in this example, the image search apparatus 3 is connected to the sensor network system 4 and the IP network system 2 having general Web servers C1 to Cm. In this example, the image recording units B <b> 1 to Bn are connected to the video surveillance network 11.
Thus, by connecting the image search device 3 to various systems, for example, generation of metadata or the like can be automated, such as using a sensor event from the sensor network 41 for metadata, For example, it is possible to take out images on the Web servers C1 to Cm and use the images as search condition images in the image search apparatus 3.

Next, with reference to FIG. 6, an example of the flow (1) to (8) of search processing and display processing performed in the image search device 3 of this example is shown.
(1) The search condition input unit 36 inputs image search conditions and supplies them to the image search unit 37. In this example, conditions relating to feature quantities and metadata are used as search conditions.
(2) The image search unit 37 reads out the storage information from the feature amount storage unit 34 based on the supplied search condition, and corresponds to the storage information (in this example, feature amount and metadata) that matches the search condition. Obtained address (in this example, a path name) is obtained.
(3) That is, the image search unit 37 obtains an address response corresponding to the search condition.

(4) Based on the acquired address, the image search unit 37 designates the address to the address conversion unit 33, so that the address conversion unit 33 records the recording position information of the image frame corresponding to the address (in this example, IP addresses, camera numbers, and time information) of the image recording units B1 to Bn are obtained.
(5) That is, the image search unit 37 obtains a response of the image storage location (recording position information) corresponding to the designated address.

(6) The image search unit 37, based on the acquired recording position information, the image recording units B1 to Bn corresponding to the recording position information (for example, the image recording unit corresponding to the IP address or camera number in the recording position information) ) To request an image (for example, an image frame corresponding to the camera number and time information), the requested image frame is read from the image recording units B1 to Bn.
(7) That is, the image search unit 37 obtains a response of the requested image using the recording position information.

(8) The image search unit 37 supplies the address corresponding to the read image frame to the image display unit 35.
(9) The image display unit 35 displays an address corresponding to the supplied image frame.
By operating as described above, an image matching the search condition and its address are displayed. In this example, the address of the image is displayed so that it can be used for the next search or the like. However, the address may not necessarily be displayed.

Next, a case where a recorded image in the video surveillance network system 1 is displayed on a home page will be described.
First, a general technique for displaying an image on a home page will be described.
The home page is generally created in an HTML (Hyper Text Mark-Up Language) language. When displaying an image,
<IMG SRC = “http://abc.com/xx/zzzz.jpg”>
And in the HTML language, the scheme name http to be used, the host name abc. com (for example, a host name where the image search apparatus (or Web server) exists), path name / xx / zzzz. Specify jpg.

The Web browsers E1 to Ex are included in the HTML data.
<IMG SRC = “http://abc.com/xx/zzzz.jpg”>
If there is a tag, an image is requested to the Web servers C1 to Cm indicated by the host name using the http protocol.
The Web servers C1 to Cm analyze the http protocol from the Web browsers E1 to Ex, read the requested image, and send the image to the requesting Web browsers E1 to Ex using an http response.
When the web browsers E1 to Ex receive an http response including image data from the web servers C1 to Cm, the web browsers E1 to Ex display the received image instead of the above-described tag.
In this way, an image is displayed on the home page.

Next, a case where a recorded image in the video surveillance network system 1 is displayed on a home page will be described.
FIG. 7 shows an example of operations performed by the Web browsers E1 to Ex.
The Web browsers E1 to Ex determine, for example, whether or not there has been a request for an image from a user (person) or the like (step S21), and if so, request the image from the Web servers C1 to Cm (step S21). S22).
Specifically, if the Web browsers E1 to Ex have a tag <IMG SRC = "http://abc.com/xx/zzzz.jpg"> in the HTML data, the Web server indicated by the host name Request images from C1 to Cm using the http protocol.
Then, the web browsers E1 to Ex determine whether or not there is an image response from the web servers C1 to Cm (step S23), and if there is, receives a response including image data from the web servers C1 to Cm. The received image is displayed on a screen, for example (step S24).

FIG. 8 shows an example of operations performed by the Web servers C1 to Cm.
The Web servers C1 to Cm determine whether or not there is a request for an image from the Web browsers E1 to Ex (step S31), and if there is, the image search unit 37 is activated (step S32).
Then, the Web servers C1 to Cm determine whether or not there is an image response from the image search unit 37 (step S33). If there is, the Web server C1 to Cm receives the image from the image search unit 37 and sends the received image to the request source. To the Web browsers E1 to Ex (step S34).

  Here, the Web servers C1 to Cm deliver the processing to the image search unit 37 when an image having a predetermined path name is requested from the Web browsers E1 to Ex. As an example of this method, when a predetermined path name is designated in the Web servers C1 to Cm, a program of cgi (Common Gateway Interface) is executed to start interface processing with the image search unit 37. It is realizable by comprising.

FIG. 9 shows an example of an operation performed by the image search unit 37.
When the processing is delivered from the Web servers C1 to Cm, the image search unit 37 uses the path name of the URL included in the above HTTP protocol message to record the recording position information of the image frame requested from the address conversion unit 33. Is acquired (step S41).
Then, the image search unit 37 reads out the image frames requested from the image recording units B1 to Bn using the acquired recording position information (Step S42), and delivers the read image frames to the Web servers C1 to Cm (Step S42). S43).

With reference to FIG. 10, an example of a flow (1) to (8) of processing for displaying a recorded image of the video surveillance network system 1 on a homepage performed in the image processing system of the present example is shown.
(1) When an image request is generated, the Web browsers E1 to Ex specify URLs for specifying the images to the Web servers C1 to Cm.
(2) The Web servers C1 to Cm notify the image search unit 37 of the URL specified by the Web browsers E1 to Ex.
(3) The image search unit 37 specifies the path name included in the notified URL to the address conversion unit 33.
(4) The address conversion unit 33 responds to the image search unit 37 with the recording position information of the image frame corresponding to the designated path name.

(5) The image search unit 37 requests an image frame from the image recording units B1 to Bn using the acquired recording position information of the image frame.
(6) The image recording units B1 to Bn return the requested image frame to the image search unit 37.
(7) The image search unit 37 returns the acquired image frame to the requesting Web servers C1 to Cm.
(8) The Web servers C1 to Cm return the received image frames to the requesting Web browsers E1 to Ex.
By operating as described above, the recorded image of the video surveillance network system 1 is displayed on the homepage that can be browsed by the Web browsers E1 to Ex.

Here, in this example, since the image of the video surveillance network system 1 can be handled in the same way as the image on the Web, for example, an application that targets an existing image on the Web is used as the image of the video surveillance network system 1. Can be easily applied.
In addition, for example, the camera may be configured such that the image is called by specifying a URL including the camera number and time information from the application by including the camera number and time information in the path name by the address assigning unit 32. . As an example, the host names abc. com and a path name “/c1/20070227/14100105.jpg” including the camera number and time information is specified, and a URL such as “http://abc.com/c1/20070227/14100105.jpg” is specified. Then you can call up the image.

  Further, in the image search unit 37, for example, when there is no image frame at the time corresponding to the record of the address conversion unit 33, the closest record in time in the address conversion unit 33 is selected, and the image frame is selected as an image. The recording unit B1 to Bn may be taken out and responded. Further, in the image recording units B1 to Bn, for example, when the image frame requested by the image search unit 37 does not exist, the image search unit 37 responds to the image search unit 37 with an image frame that is temporally closest to the requested image frame. You may be set as such a structure. With such a configuration, the image calling process on the application side is simplified, and the processing efficiency can be improved.

The address conversion unit 33 does not necessarily have a function of storing a record including the URL path name. As another configuration example, the recording position of the image frame is derived from the path name. May be. As an example, the recording position information for reading the image frame from the image recording units B1 to Bn may be generated using the camera number and time information included in the path name.
In addition, each functional unit in the image search device 3 or the like of this example may be realized by, for example, independent hardware, or a plurality of functional units may be realized by one hardware.

As described above, the image search device 3 of this example searches for images that meet the search conditions from the image recording units B1 to Bn that record images in time series in the imaging devices A1 to An. A frame selection unit 31 for selecting one image frame from consecutive image frames, an address addition unit 32 for assigning an address to the selected image frame, and an image recording unit for the assigned address and the selected image frame An address conversion unit 33 that associates the recording position information in B1 to Bn with each other, and a feature amount storage unit 34 that stores arbitrary information (in this example, feature amounts and metadata) related to the image frame selected in association with the address. For example, a search condition input unit 36 for inputting a character string, an image, or the like as a search condition, and an image frame using recording position information based on the input search condition. Comprises an image retrieval unit 37 for reading over beam from the image recording unit Bl to Bn, the image display unit 35 for displaying, for example, it reads image frames and granted address. In this example, a URL path name (or a URL including a URL path name) is used as an address.
As an example of the configuration, the path name includes the camera number of the imaging devices A1 to An where the image frame is captured, and time information such as when the image frame is captured.

Further, in the image processing system including the image search device 3 of the present example, in the image calling method for calling an image, an image recording step for recording an image in time series in units of the imaging devices A1 to An, and a continuous image in time series A frame selection step for selecting one image frame from the frame, an address assignment step for assigning an address to the selected image frame, and an address conversion step for associating the assigned address with the recording position information of the selected image frame And a URL specifying step of specifying a URL by an application such as a browser, and an image reading step of reading an image frame from the recording position using the recording position information based on the specified URL. In this example, a URL path name (or a URL including a URL path name) is used as an address.
As an example of the configuration, the path name includes the camera number of the imaging devices A1 to An where the image frame is captured, and time information such as when the image frame is captured.

In the image processing system including the image search device 3 of the present example, in the image calling method for calling an image, an image recording step for recording images in time series in units of the imaging devices A1 to An, and recording positions of addresses and image frames An address conversion step for associating information, a URL specifying step for specifying a URL by an application such as a browser, and an image reading step for reading an image frame from a recording position using recording position information based on the specified URL; ,have. In this example, a URL path name (or a URL including a URL path name) is used as an address.
As an example of the configuration, the path name includes the camera number of the imaging devices A1 to An in which the image frame is captured, and time information such as when the image frame is captured.

Further, in the image processing system including the image search device 3 of the present example, the above-described image reading step is performed based on the time information associated with each image frame when the designated image frame does not exist. The image frame whose time is closest to the selected image frame is read out.
In this example, the case where the images are handled in units of frames is shown. For example, the images are grouped into a plurality of consecutive frames, and identification information is attached to the images of each group. It is also possible to adopt a configuration that treats as a unit.

  Therefore, the image search apparatus 3 of this example can provide an interface suitable for a Web application to the video surveillance network system 1, and a recorded image in the video surveillance network system 1 can be directly received from, for example, a general Web application. Can be used by calling. In this case, for example, a recorded image in the video surveillance network system 1 can be called using a URL, and thereby the usability for the user (person) can be improved.

As described above, the image search apparatus 3 of the present example can provide an interface that can handle an image in the video surveillance network system 1 in the same manner as an image on the Web. Can be easily applied to images in the video surveillance network system 1. Further, for example, the images of the image recording units B1 to Bn can be used from the Web application without changing the conventional input / output interfaces of the image recording units B1 to Bn.
Here, in this example, the process which reads the image recorded on image recording part B1-Bn based on the recording position information which consists of IP address information, camera number, and time information of image recording part B1-Bn shown in FIG. For example, when an image is read, the recording location of the image is specified by the identification information of the image recording units B1 to Bn in which the image is recorded and the memory address in the image recording unit B1 to Bn. In an arbitrary processing unit (for example, a processing unit that performs processing for reading an image from a plurality of image recording units B1 to Bn, or an image search unit 37) before reading an image, the image recording unit is determined based on the recording position information. The identification information of B1 to Bn and the conversion to the memory address in the image recording units B1 to Bn are performed. Various types of information may be used as the identification information of the image recording units B1 to Bn. For example, the IP address information may be used. In addition, when one image recording unit B1 to Bn is specified and an image recorded in the image recording unit B1 to Bn is read, for example, the image recording unit B1 to Bn can read the image before reading the image. What is necessary is just to acquire the memory address in which the image was recorded by arbitrary processing parts.

  In the image search apparatus 3 of the present example, the address conversion unit 33 associates the identification information of the image (in this example, the image frame) (address information including the URL path name in this example) with the information on the recording position. The correspondence storage means is configured by the function of storing the request, and the image search unit 37 receives the request by the function of receiving information (URL information in this example) for requesting the image from the Web servers C1 to Cm of the IP network system 2. The identification information detection means is configured by a function of detecting image identification information (address information including a URL path name in this example) from information requesting an image received by the image search unit 37. The image search unit 37 has a function for detecting the recording position of the image corresponding to the image identification information detected by using the function of the address conversion unit 33. Position detection means is configured, and recording is performed in the image recording units B1 to Bn (an example of a recording device) of the video surveillance network system 1 (an example of a monitoring system) based on the recording position of the image detected by the image search unit 37. The image acquisition means is configured by the function of acquiring the acquired image (image frame in this example), and the image acquired by the image search unit 37 is sent to the Web servers C1 to Cm of the IP network system 2 that is the request source. An image output means is configured by the output function.

Here, the configuration of the system and apparatus according to the present invention is not necessarily limited to the configuration described above, and various configurations may be used. The present invention can also be provided as, for example, a method or method for executing the processing according to the present invention, a program for realizing such a method or method, or a recording medium for recording the program. It is also possible to provide various systems and devices.
The application field of the present invention is not necessarily limited to the above-described fields, and the present invention can be applied to various fields.
In addition, as various processes performed in the system and apparatus according to the present invention, for example, the processor executes a control program stored in a ROM (Read Only Memory) in hardware resources including a processor and a memory. A controlled configuration may be used, and for example, each functional unit for executing the processing may be configured as an independent hardware circuit.
The present invention can also be understood as a computer-readable recording medium such as a floppy (registered trademark) disk or a CD (Compact Disc) -ROM storing the control program, and the program (itself). The processing according to the present invention can be performed by inputting the program from the recording medium to the computer and causing the processor to execute the program.

It is a figure which shows the structural example of the image processing system which has an image search device concerning one Example of this invention. It is a figure which shows an example of the record format of the data in an address conversion part. It is a figure which shows an example of the record format of the data in a feature-value memory | storage part. It is a figure which shows an example of the procedure of the process which supplies an image frame from an imaging device to a frame selection part. It is a figure which shows an example of the procedure of the process which supplies an image frame from an image recording part to a frame selection part. It is a figure which shows an example of the flow of a search process and a display process. It is a figure which shows an example of operation | movement of a Web browser. It is a figure which shows an example of operation | movement of a Web server. It is a figure which shows an example of operation | movement of an image search part. It is a figure which shows an example of the flow of a process which displays the recorded image of a video surveillance network system on a homepage.

Explanation of symbols

1. Video surveillance network system 2. IP network system 3. Image search device 11. Video surveillance network 21 IP network 31 Frame selection unit 32 Address assignment unit 33 .. Address conversion unit, 34 .. Feature amount storage unit, 35 .. Image display unit, 36 .. Search condition input unit, 37 .. Image search unit, 41 .. Sensor network,
A1-An ... Imaging device, B1-Bn ... Image recording unit, C1-Cm ... Web server, D1-Dz ... Sensor node, E1-Ex ... Web browser,

Claims (1)

In response to a request for an image from an IP network system, the image search device searches for an image captured by an imaging device and recorded in a recording device in a monitoring system,
Request receiving means for receiving, as information for requesting an image from the IP network system, URL information including identification information of an imaging device from which an image is captured and path name information having image time information;
Identification information detecting means for detecting path name information included in the URL information or information including the path name as identification information of the image from URL information requesting the image received by the request receiving means;
A recording position detecting means for detecting a recording position of an image corresponding to the identification information detected by the identification information detecting means;
Image acquisition means for acquiring an image recorded in the recording device in the monitoring system based on the recording position of the image detected by the recording position detection means;
Image output means for outputting the image acquired by the image acquisition means to the IP network system;
An image search apparatus comprising:

Images