JP2007293454A - Material presentation system and material presentation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a material presentation system for referring to materials used in the past conference for the next conference. <P>SOLUTION: This material presentation system 100 is provided with a participant recognition part 10 for recognizing a participant who participates in a first conference and a participant who participates in a second conference; a material recognition part 11 for recognizing materials used in the first conference; a DB registration part 21 for registering material data recognized by the material recognition part 11 in a conference DB 30 by associating the material data with the information of the participant; a DB retrieval part 22 for retrieving the material data to be used in the second conference from the conference DB 30 based on the information of the participant recognized in the second conference; and an information presenting part 12 for visually presenting the material data retrieved by the DB retrieval part 22. The material recognizing part 11 recognizes the materials by capturing the image of the screen when the display screen provided by the computer connected to a projector changes. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a material presentation system and a material presentation method.

  In recent years, many methods for providing and customizing information in accordance with individual preferences and characteristics have been proposed. In the example of the Web (World Wide Web), in a site that deals with books, the relevant books are presented on the homepage based on the books that the user has checked or purchased in the past, and the user's interest is drawn. It has become. In another example, when a user uses a commuter pass to exit a ticket gate, information around the destination and information on a genre registered in advance are delivered to the user's mobile phone. Patent Document 1 proposes a method of adjusting the indoor environment—temperature and humidity—based on personal preference environmental information.

On the other hand, it is considered effective to provide some information based on such personal information even in a meeting scene. For example, Patent Document 2 proposes a method capable of continuously maintaining the electronic work space of each individual even if the physical work space changes.
JP-A-6-11170 JP 2005-44159 A

  However, by applying the technique described in Patent Document 2, when a certain user enters the conference room, the ID, which is personal information, is read, and the document is taken out from the electronic work space created by the user. It is easy to think about the content of projection. In addition, even when multiple people gather, it is possible to create a collaborative work space that collects the work spaces of each individual. The required information is not always in the working space where you are currently working. In the case of a meeting, information on the previous or previous meeting may be required. In some cases, information specified by personal information of a plurality of people is more useful than information specified by personal information of one person. For example, in the case of a continuous meeting body, there are cases where it is desired to look back on materials and minutes used in past meetings. In such a case, it is only necessary to provide the conference materials and the like used when the people who participated there participated in the previous session. In this case, it is necessary to use personal information of a plurality of people, not personal information of only one person.

  Therefore, the present invention has been made in view of the above problems, and an object of the present invention is to provide a material presentation system and a material presentation method that can refer to materials used in previous conferences in the next conference.

  In order to solve the above problems, a material presentation system according to the present invention includes first recognition means for recognizing a participant who participates in a first conference, and material recognition for recognizing a material used in the first conference. Means, registration means for registering the material data recognized by the material recognition means in the storage means in association with the information of the participant, and second recognition means for recognizing the participant who participates in the second meeting And based on the material data retrieved by the retrieval means, retrieval means for retrieving material data to be used in the second meeting from the storage means based on the information of the participants recognized in the second meeting Presentation means for visually presenting the material. Thereby, the participant can refer to the material used in the past meeting without any operation in the next meeting.

  Further, it is preferable that the presenting means presents a reduced image of the material retrieved by the retrieving means so as to be visible. Accordingly, the entire image of the material can be shown to the participant by the reduced image of the material. The first recognition means may recognize the participant by detecting identification information of a portable device carried by the participant. The document recognition means may recognize the document by capturing an image of the screen when a display screen provided by a computer that outputs a video signal to a predetermined projection means changes. Further, the material recognizing means may recognize the material by capturing an image based on a video signal input to a predetermined projecting means.

  In addition, the search means may participate in the first meeting with a participant in the second meeting when a participant who participates in the first meeting does not match the participant in the second meeting. It is preferable to search for materials to be used in the second meeting based on the number of persons overlapping. For example, the search means may be obtained by dividing the number of participants of the second meeting and the number of participants at a certain time of the first meeting by the number of participants of the second meeting, Based on the degree of match obtained from adding the number of participants of the meeting and the number of overlapping participants at a certain time of the first meeting divided by the number of participants at the certain time of the first meeting To search for materials to be used in the second meeting.

  Further, the search means searches for the material so as to restrict the presentation of the material used in the second conference due to the overlap of the participants participating in the first conference and the participants participating in the second conference. Good. Further, the presenting means may enable an operation on the reduced image of the material when a predetermined password is input. The presenting means may limit the presentation of the material according to a word included in the material searched by the searching means. The presenting means may limit the presentation of the material according to a similarity obtained from a predetermined word group and the number of appearances of words included in the material searched by the searching means.

  The material presentation method according to the present invention includes a first recognition step for recognizing a participant who participates in the first meeting, a material recognition step for recognizing the material used in the first meeting, and the material recognition step. Registered in the storage means in association with the participant information, a second recognition step for recognizing a participant who participates in the second conference, and the second conference. A retrieval step for retrieving material data to be used in the second meeting from the storage means based on the information of the participants, and a presentation for visually presenting the material based on the material data retrieved in the retrieval step Steps. Thereby, the participant can refer to the material used in the past meeting without any operation in the next meeting.

  The document recognition step may be configured to recognize the document by capturing an image of the screen when a display screen provided by a computer that outputs a video signal to a predetermined projection unit changes. The material recognition step may be configured to recognize the material by capturing an image based on a video signal input to a predetermined projection unit.

  In the search step, when a participant who participates in the first conference does not match a participant in the second conference, a participant in the second conference and a participant at a certain time of the first conference It is preferable to search for materials used in the second meeting based on the overlapping number.

  Further, the searching step searches for the material so as to restrict the presentation of the material used in the second conference due to the overlap of the participant participating in the first conference and the participant participating in the second conference. To do. In addition, the presenting step includes a step of restricting the presentation of the material according to a word included in the material searched in the searching step. In addition, the presenting step includes a step of restricting the presentation of the material according to the similarity obtained by a predetermined word group and the number of appearances of the word included in the material retrieved in the retrieval step.

  ADVANTAGE OF THE INVENTION According to this invention, the material presentation system and material presentation method which can refer the material used at the past meeting at the next meeting can be provided.

  Hereinafter, the best mode for carrying out the present invention will be described. FIG. 1 is a diagram showing an overall configuration diagram of a material presentation system 100 according to a first embodiment of the present invention. The material presentation system 100 of the present invention provides information based on personal information of a plurality of persons. In particular, the material presentation system 100 provides materials and minutes used in past meetings. This material presentation system 100 is roughly divided into two parts. The first is to recognize participants and register their personal information and the documents used in the conference database for the current conference. The other part is a part for recognizing a plurality of participants and searching past meeting materials based on the personal information. According to this material presentation system 100, participants in a conference can refer to materials and minutes of past conferences without any particular instruction. This will be specifically described below.

  As shown in FIG. 1, the material presentation system 100 includes a participant recognition unit 10, a material recognition unit 11, an information presentation unit 12, an information provision unit 20, and a conference database (DB) 30. The information providing unit 20 includes a DB registration unit 21 and a DB search unit 22. First, a process of registering personal information of participants who have participated in the conference using the DB registration unit 21 and materials used in the conference DB 30 will be described.

  For example, the participant recognition unit 10 recognizes a participant who exists in the conference room and participates in the conference by detecting identification information of a portable device carried by the participant. For example, the participant recognition unit 10 attaches a reader of an employee ID card for managing entry / exit to / from a conference room, an IC tag, a tag that emits infrared rays or ultrasonic waves, and reads it with a reader. You may recognize whether you have entered or exited. In addition, after recognizing all the participants, the participant recognition unit 10 determines whether or not the participants in the current conference (second conference) match the participants in the past conference (first conference). To detect.

  The material recognition unit 11 recognizes materials and minutes used in the meeting. The material recognizing unit 11 is, for example, software that operates on a computer that provides a video signal to the projector lamp, and software that captures an image of the screen and recognizes the material when the display screen provided by the computer changes, This is realized by hardware that distributes the video signal input to the projector lamp and captures the image and recognizes the material every time the image based on the video signal changes. The material recognition unit 11 may be any one that can capture the same image projected by a projector lamp or the like. The information presentation unit 12 is configured by, for example, a display device, and presents the material visually based on the material data retrieved by the DB retrieval unit 22. Note that the display device will be described as an example, but the case where the material is printed by a printer or the like is included in the visual presentation of the material.

  The DB registration unit 21 registers the material data recognized by the material recognition unit 11 in the conference DB 30 in association with the participant information. In addition, the DB registration unit 21 receives the participant ID from the participant recognition unit 10 and the conference data captured by the material recognition unit 11, and enters the conference DB 30 together with the participant ID and the current time when the conference data is received. save. The DB registration unit 21 has been described on the assumption that the material data is automatically registered, but the material data can also be manually registered. As an example of manually registering document data, there is a person who participated without wearing an ID card, or an image captured by another document presentation system not connected to the document presentation system 100, such as a digital camera. In some cases, the images taken in step 1 are registered in the conference DB 30. The DB search unit 22 searches material data used in the current conference from the conference DB 30 based on the information of the participants recognized in the current conference.

  The conference DB 30 stores participant IDs, material data, time information, and the like. The conference DB 30 can register information in a plurality of conference rooms by placing the participant recognition unit 10 and the material recognition unit 11 in each conference room. In this case, the ID of each conference room may be recorded in the conference DB 30 at the same time.

  Next, the operation procedure of the material presentation system 100 will be described in detail. FIG. 2 is a diagram illustrating a procedure for registering a participant ID. The participant recognition unit 10 detects whether or not there is a recognition of the participant and an increase or decrease in the number of participants (step S1). The participant recognition unit 10 recognizes all the participants and detects whether or not the participants in the current conference coincide with the participants in the past conference. If there is a change in the number of participants, the participant recognition unit 10 sends the participant ID to the DB registration unit 21 (step S2). For example, in the participant recognition unit 10, the IDs of the current conference participants are “ID001”, “ID002”, and “ID003”, and the number of participants whose ID is “ID004” has increased. In this case, all IDs “ID001”, “ID002”, “ID003”, and “ID004” of the participants in the current conference are sent to the DB registration unit 21. The DB registration unit 21 registers the received participant ID and the current date and time in the conference DB 30 (step S3).

  FIG. 3 is a diagram showing a procedure for registering material data. The material recognition unit 11 detects whether there is a change in the presented material (step S11). Here, when there is a change in the presented material, for example, the material recognition unit 11 switches the slide to the next page, captures the material, and generates image data. Here, the material recognition unit 11 assigns a unique data name to the image data. An example of a unique data name may be a date and time, a URL storing image data, or the like, and any data name and data may be associated with each other. The material recognition unit 11 sends the image data name to the DB registration unit 21 (step S12). The DB registration unit 21 registers the received image data name, current date and time in the conference DB 30 (step 13).

  4A to 4C are diagrams showing records 201 to 203 registered in the conference DB 30. FIG. Records 201 to 203 include data items of the date of the conference, the time when the participant participated in the conference, the participant ID, and the image ID. A record 201 is a record of participant IDs at different times. At 13:13:00, a participant with a card of “ID0002” was detected in the conference room, and after 4 seconds, a participant with a card of “ID0001” was detected in the conference room. At 13:14:48, “ID003” is detected and at 13:15:10, “ID004” is detected and registered.

  The record 202 is updated by adding an image ID to the record 201. When there is an additional change of image data from the material authentication unit 11, as shown in FIG. 4B, an additional image ID “image 0100” is registered in the record 202 after the participant ID. . Also, when the number of participants is increased or decreased, the record 202 simultaneously registers the currently displayed image ID in addition to the participant ID. In this case as well, the image ID at the last change may be used. As a result, as shown in FIG. 4C, information is registered as in the last line of the record 203.

  In the above description, the case where the timing of registering the participant ID, date and time, and the material data in the conference DB 30 has been described, but the timing for registering each data is, for example, every second, An image ID may be registered.

  Further, the conference DB 30 may be manually registered and edited. As an example of manual registration and editing, authentication cannot be performed, for example, when an ID card is forgotten, and therefore a participant ID may be added after the time of participation. Also, images taken with a digital camera, memos taken by individuals, and materials that are not automatically captured, such as meeting minutes issued later, can be registered at any time by directly editing the meeting DB 30. Can do.

  Next, referring again to FIG. 1, a process of searching for materials etc. from the conference DB 30 using the DB search unit 22 and presenting necessary materials etc. to the participants will be described. The participant recognizing unit 10 is also used for storing in the conference DB 30 via the DB registration unit 21, but at the same time, the participant recognizing unit 10 stores past materials from the conference DB 30 via the DB search unit 22. It is also used for searching.

  The DB search unit 22 receives the ID of the participant from the participant recognition unit 10, and searches the conference DB 30 for and extracts materials used in past conferences in which the participant has participated from the participant ID. The information presentation unit 12 receives the material extracted by the DB search unit 22 and presents it to the participants. The information presentation unit 12 includes, for example, a screen connected to a projector (not shown).

  FIG. 5 is a diagram showing a procedure for searching for materials. The participant recognition unit 10 detects whether or not there is a recognition of the participant and an increase or decrease in the number of participants (step S21). The participant recognition unit 10 recognizes all the participants and detects whether or not the participants in the current conference coincide with the participants in the past conference. When there is a change in the number of participants, the participant recognition unit 10 sends the participant ID to the DB registration unit 21 (step S22). The DB search unit 22 searches the conference DB 30 for material data that matches these participant IDs using the participant ID as a key, extracts the corresponding image data, and sends it to the information presentation unit 12 (step S23). The information presentation unit 12 presents the received image data to the participant (step S24).

  FIG. 6 is a diagram illustrating records in the conference DB 30. FIG. 7 is a diagram showing a search result screen of the DB search unit 22. As shown in FIG. 6, records 211 to 213 are respectively registered in the conference DB 30, and participant ID codes and the like in the records 211 to 213 are collectively searched by the DB search unit 22. For example, when the ID of the card of the participant participating in the current conference is “ID001”, “ID002”, “ID003”, “ID004”, the portion indicated by the underline of this record 213 is summarized. To be searched. As a result, as shown in FIG. 7, the DB search unit 22 performs “image 0020”, “image 0021”, “image 0022” on August 29, 2005, and “image 0100” on September 2, 2005, “Image 0101” is searched and extracted.

  Next, an example in which the information presentation unit 12 presents a reduced image of the material searched by the DB search unit 22 in a visually observable manner will be described. FIG. 8 is a diagram illustrating a display example by the information presentation unit 12. FIG. 9 is a diagram illustrating another display example of the information presentation unit 12. As shown in FIG. 8, when there are a plurality of images to be presented to the participant, the information presentation unit 12 generates thumbnail images, arranges them by date, and displays them to the participants. When the participant indicates a thumbnail image, the enlarged image can be displayed on another window. As another display method of the information presentation unit 12, the information may be displayed on the screen one by one.

  As described above, the participant can refer to the material used in the past conference in the current conference without any operation. In addition, it is possible to retrieve related information by manually registering the minutes created after the conference is completed in the conference DB.

  Next, an example of the search method of the DB search unit 22 will be described. As an example of the search method, there is a method of calculating the fitness for all conference data in the conference DB 30. According to this calculation method, the DB search unit 22 can perform a search even in a regular meeting or the like where the participants are not necessarily the same each time, even when the participants do not completely match. The coincidence calculation method is represented by the following formula, for example.

・ Matching rate = (Reproducibility + Fit rate) / 2 (1)
・ Recall rate = (Number of participants overlapping with current participant at time T of a past meeting / Number of current participants) (2)
・ Accuracy rate = (Number of participants overlapping with current participant at time T of a past conference / Number of participants at time T of a past conference) (3)

  The above equation (1) calculates the degree of coincidence by taking the average value from the sum of the recall rate equation (2) and the matching rate equation (3). Specifically, when the participant who participates in the past conference does not match the participant in the current conference, the DB search unit 22 overlaps the participant in the current conference and the participant at a certain time in the past conference. Divided by the number of participants in the current meeting and the number of participants in the current meeting and the number of participants in a past meeting time divided by the number of participants in a past meeting time Add materials and search for materials to be used in the current meeting based on the degree of match obtained. According to this calculation method, for example, the current participants are recognized as “ID001”, “ID002”, and “ID003”, and the participants at the time T1 of the past meeting are exactly the same “ID001”, “ID002”, “ In the case of “ID003”, the reproducibility is 3/3, the relevance ratio is 3/3, and the degree of matching is (1 of the reproducibility + 1 of the relevance ratio) / 2, so the degree of coincidence is “1”. Also, when the participants at another time T2 of the past meeting are “ID002”, “ID003”, “ID004”, “ID005”, the recall rate is 2/3, the matching rate is 2/4, and the degree of matching is The degree of coincidence is “0.58” because (2/3 of the reproduction rate + 2/4 of the matching rate) / 2.

  As described above, the value of the degree of match is higher when the participant is less excessive or deficient. Note that the method of calculating the degree of coincidence is not limited to the method of taking the average value at the end, and for example, it may be calculated by taking the harmonic average. Alternatively, the degree of match may be calculated in all past time zones, and image data for a time with a high degree of match may be searched. When the search results increase, the DB search unit 22 may set a threshold value and send only image data having a matching degree equal to or higher than the threshold value to the information presentation unit 12.

  The information presentation unit 12 presents the image data obtained from the DB search unit 22 to the participant (see FIG. 8). In this case, it may be presented not only in the date order but also in a ranking with a high degree of match. Further, the size of the thumbnail image may be changed using the degree of match as shown in FIG.

  Next, in an example in which the DB search unit 22 searches for materials so as to limit the presentation of materials used in the current conference due to the overlap of participants participating in the past conference and participants participating in the current conference. is there. FIG. 10 is a diagram for explaining the degree of coincidence between participants of a conference currently being held and participants of a conference held in the past. In FIG. 10, a circle filled with gray is a set of participants of a conference currently being held. An open circle or ellipse is a set of participants at a certain time in a past meeting.

  FIG. 6A is a diagram showing a case where a current participant and a participant in a certain time zone of a past conference are exactly the same. FIG. 5B is a diagram showing a case where the current participant is included in the participants in a certain time zone of a past conference. FIG. 6C is a diagram showing a case where a current participant and a participant in a past conference partially overlap. FIG. 4D is a diagram illustrating a case where the current participant includes a participant in a certain time zone of a past conference. FIG. 5E shows a case where the current participant and the participant in a certain time zone of a past conference are completely different.

  In the case of FIG. 5A, it is only necessary to search for materials used in a time zone in which participants completely match as in the first example. Further, even if they do not completely match, the degree of matching described above can be used to search to some extent in the cases of (B), (C), and (D) in FIG. In the case of FIG. 5E, it can be said that the conference is completely irrelevant, so there is no need to search. However, from the viewpoint of security, the cases (B), (C), and (D) of FIG.

  For example, in the case of (B) in the figure, all the current participants are participating in a certain time zone of the past conference, so the materials used at that time should be seen by all the current participants. Therefore, there seems to be no problem even if it is presented as a search result. In the case of (C) in the figure, since it is a material of a past meeting in which some of the current participants have not participated, there is a problem in presenting it as a search result if the material is a confidential document. . Similarly, in the case of FIG. 4D, since a part of the participants of the current conference is a conference material that is not yet participating, there is a problem in presenting the confidential document as a search result.

  Specifically, for example, the confidential material X used when Manager A and Manager B talked about personnel matters later was a meeting where Manager A, Manager B, General Employee C, General Employee D, and General Employee E participated. This is a case where it is presented as past conference material. In order to prevent such a situation, it is necessary to restrict the presentation of information with respect to the cases shown in FIGS. In order to determine (C) and (D) in the figure, it may be determined whether or not all the current participants are included at the time of search. If all current participants are not included, the search results are not presented.

  Next, a description will be given of a method that can be presented only to a specific person who is also a meeting data that is likely to cause a confidential problem. The information presentation unit 12 enables an operation on a reduced image of the material when an appropriate password is input. As an example of this presentation method, there is a method for enabling access by a password. For example, even in the case of FIGS. 10C and 10D described above, the search results are presented as thumbnails. At this point, it is assumed that it is impossible to read information in the thumbnail, for example, characters. If a user determines that the document can be enlarged and displayed, the user designates the thumbnail. Then, a user name and password input window is opened, and the user name and password are input. Here, it is assumed that the user name and password are registered in advance. The information presentation unit 12 performs an enlarged display only when the password is input, the user name and the password match, and the user has browsed the material in the past.

  Specifically, using the above example, the confidential material X used when Manager A and Manager B talked about personnel matters will be Manager A, Manager B, General Employee C, General Employee D, and General Employee. When manager A or manager B decides that disclosure may be made at the conference in which E participates, manager A or B can enlarge the confidential document X by inputting his user ID and password. Even if the general employees C, D, and E input their user names and passwords, the confidential material X is not used in the past together with the general employees C, D, and E, and is not enlarged.

  FIG. 11 is an overall configuration diagram of a material presentation system 200 according to the second embodiment of the present invention. As the display method of the information presenting unit 12, the conference data searched by the DB searching unit 22 is classified and displayed in order of date or in descending order of coincidence, but the display method here is currently displayed. We propose a method of searching for similar conference data by using existing conference data, ranking it and displaying it. This display method is presented to the participants in the ranking order based on the similarity between the newly searched conference data and the currently displayed conference data.

  As illustrated in FIG. 11, the material presentation system 200 further includes a word index DB 31 and a word buffer 32 in addition to the configuration of the material presentation system 100 illustrated in FIG. 1. The DB registration unit 21 not only sends the participant ID, conference data, and current time to the conference DB 30, but also extracts text as a character string from each captured image data, divides the text into word character strings, And the number of appearances of the word in the image and the image name (image ID) are sent to the word index DB 31 as a set. Further, the DB registration unit 21 sends to the word buffer 32 the text extracted from the last captured image into words and the number of appearances of the word in the image. Here, OCR is used as an example of a text extraction method. As an example of the method of dividing into words, morphological analysis is used.

  The DB search unit 22 searches the conference DB 30, the word index DB 31 and the word buffer 32 in order, and sends similar conference data from the currently displayed conference data to the information presentation unit 12. Specifically, the DB search unit 22 receives the participant ID from the participant authentication unit 10 and searches the conference DB 30 for image data used in the past conference using the participant ID as a search key. Next, the DB search unit 22 searches the word index DB 31 using the searched image ID as a search key, and extracts words included in the image and the number of appearances thereof. Next, the DB search unit 22 compares the extracted word and the number of appearances with the search key and the word contained in the word buffer 32 and the number of appearances.

  That is, the DB search unit 22 searches and presents materials used in past conferences using the recognized participants as keys, and participates by ranking based on the content of the materials projected at the current conference. Present to the person.

  The word index DB 31 registers the character string, the number of appearances, and the image name (image ID) of each word sent from the DB registration unit 21. The word buffer 32 stores the character string of each word sent from the DB registration unit 21 and the number of appearances of the word in the image. The word buffer 32 clears the buffer once before saving, and then stores the character string and the number of appearances of each word.

  FIG. 12 is a diagram showing records registered in the word index DB 31 and the word buffer 32. FIG. 12A shows a record 204 registered in the word index DB 31. The record 204 includes each word code shown in the leftmost column and extracted from the image, an appearance frequency code in the image of the word in the right row, and an image ID code of the image corresponding to the right row. FIG. 12B shows the record 205 registered in the word buffer 32. The record 205 includes each word code extracted from the image of the conference data last received by the DB registration unit 21 and the appearance code in the image of each word in the right row.

  Next, the operation of the material presentation system 200 will be described. FIG. 13 is a flowchart until the submitted material data is registered in each DB 30, 31, 32. The material recognition unit 11 detects whether there is a change in the presented material, and if there is a change, sends the image ID to the DB registration unit 21 (step S31). The DB registration unit 21 sends the received image data name, current date, and time to the conference DB 30 (step S32). Further, the DB registration unit 21 extracts a character string from each image of conference data using OCR, and divides the extracted character string into character strings such as words. The DB registration unit 21 registers the divided character string, the number of appearances of the word in the image, and the image ID in the word index DB 31 (step S33). Further, the DB registration unit 21 extracts text from the last captured image, and registers each word and the number of appearances of the word in the image in the word buffer 32 (step S33). The image data is registered together with the current date and time in the conference DB 30 (step S34).

  FIG. 14 is a flowchart for presenting conference data similar to the currently displayed conference data to participants. The participant authentication unit 10 detects whether there is an increase or decrease in the number of participants (step S41). Participant authentication unit 10 sends the IDs of all participants participating in the current conference to DB search unit 22 (step S42). The DB search unit 22 searches the conference DB 30 using the participant ID as a search key, and extracts an image ID of conference data presented in the past (step S43). The DB search unit 22 searches the word index DB 31 using the extracted image ID as a search key, and extracts words included in the image and the number of appearances thereof (step S44).

  The DB search unit 22 searches and extracts the words contained in the word buffer 32 and the number of appearances using the extracted words and the number of appearances as search keys (step S45). The DB search unit 22 weights each image based on the word in the word index DB 31 and the number of appearances thereof, the word in the word buffer 32 and the number of appearances, and the image data is transferred from the conference DB 30 according to the weighting. Extracted and sent to the information presentation unit 12 (step S46). The information presentation unit 12 ranks the conference data sent from the DB search unit 22 in descending order of similarity and displays the ranking to the participants (step S47).

  The DB search unit 22 described above calculates the similarity based on the word included in the image searched in the word index DB 31 and the number of appearances thereof, and the word included in the word buffer and the number of appearances thereof. Here, the DB search unit 22 determines that the similarity is high when the word index DB 31 and the word buffer 32 have the same word and the number of appearances coincides.

  Next, a method for calculating the similarity of the DB search unit 22 described above will be described. FIGS. 15A, 15B, and 15C show the word included in the image ID searched in the word index DB 31 and the number of appearances of the word for each image ID. FIG. 15D shows the words included in the word buffer 32 and the number of appearances of the words. The word included in each image in the word index DB 31 shown in FIGS. 15A, 15B, and 15C, the number of appearances of the word, the word in the word buffer shown in FIG. The image similarity is calculated based on the number of appearances as follows.

・ Similarity of image 001: Beef (0 points) + America (2 points) + Australia (0 points) + Domestic (2 points) + Imports (0 points) = 4 points (4)
・ Similarity of image 004: Automobile (3 points) + America (2 points) + Europe (3 points) + Domestic production (2 points) + Import (0 points) = 10 points (5)
・ Similarity of image 006: Internet (0 points) + Access (0 points) + Site (0 points) + Domestic (0 points) + Overseas (2 points) = 2 points (6)

  The above-described calculation method will be briefly described. In this image similarity calculation method, the same word is present in the word for each image ID in the word index DB 31 and the word in the word buffer 32, and the number of appearances of the word. The lesser number is used as the weight of each word to give the score of each word, and the total score of those words is calculated. The DB search unit 22 calculates the similarity of images according to the total score.

  More specifically, in the above equation (4), the similarity of the image 001 in the word index DB 31 shown in FIG. 15A is calculated, and the word index DB 31 shown in FIG. 15A is calculated. When the words in the word buffer and the words in the word buffer shown in FIG. 15D are respectively compared, the words “beef”, “international”, and “import” in the word index DB 31 exist in the words in the word buffer 32. Therefore, the words “beef”, “international”, and “import” in the word index DB 31 are each calculated as 0 points. The word “America” in the word index DB 31 matches the word “America” in the word buffer 32, and each word appears twice, so “America” in the word index DB 31. Is calculated as 2 points.

  On the other hand, the word “domestic” in the word index DB 31 matches the word “domestic” in the word buffer 32, and the number of occurrences (two times) of the word “domestic” in the word buffer 32 is Since the number of occurrences of “domestic” in the word index DB 31 is less than three, it is calculated as two points. From the above, the total score of words included in the image “001” is calculated as 4 points. By the same calculation, the total score of words included in the image “004” is 10 points as understood from the above formulas (5) and (6), and the total score of words included in the image “006” is 2 points. The DB search unit 22 calculates the similarity of images according to the total score for each image.

  When the number of appearances of each word is compared in the DB search unit 22, words that appear in common in all images, for example, particles and general words, may be deleted in advance. Also, as a method of calculating the similarity of images, not only the comparison of the number of occurrences of words, but also the comparison of word colors, word font, word font size, font style, word image position, and word Alternatively, the similarity of images may be calculated by comparing logos and images used. In this case, in addition to the number of appearances, the word color, font, size, position, and the like may be recorded in the word index DB 31 and the word buffer 32 for comparison.

  As described above, when the information presentation unit 12 presents a document to a participant by limiting the presentation of the material according to the similarity obtained by the word group and the number of appearances of the word included in the retrieved material. In addition to simply presenting materials used by the same person, it is possible to sequentially present past image data similar to the image data currently used. Therefore, it is possible to present image data that can be easily referred to at the current conference.

  The material presentation method according to the present invention is realized by the material presentation systems 100 and 200. FIG. 16 is a hardware configuration diagram of the material presentation system. As shown in FIG. 16, the material presentation systems 100 and 200 include, for example, a CPU (Central Processing Unit) 61, a ROM (Read Only Memory) 62, a RAM (Random Access Memory) 63, a hard disk device (HD) 64, and the like. It is configured using.

  The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to such specific embodiments, and various modifications can be made within the scope of the gist of the present invention described in the claims. Can be changed. Note that the password protection function described in the material presentation system 100 can be applied to the material presentation system 200.

It is a figure which shows the whole block diagram of the data presentation system which concerns on 1st Example of this invention. It is a figure which shows the procedure which registers participant ID. It is a figure which shows the procedure which registers document data. It is a figure which shows the record registered into meeting DB. It is a figure which shows the procedure which searches material. It is a figure which shows the record in meeting DB. It is a figure which shows the screen of the search result of DB search part. It is a figure which shows the example of a display by an information presentation part. It is a figure which shows the other example of a display of an information presentation part. It is a figure for demonstrating the coincidence degree of the participant of the meeting currently performed, and the participant of the meeting performed in the past. It is a whole block diagram of the material presentation system which concerns on 2nd Example of this invention. It is a figure which shows the record registered into word index DB and a word buffer. It is a flowchart until registering the submitted material data in each DB. It is a flowchart until meeting data similar to the currently displayed meeting data is presented to participants. It is a figure which shows the word contained in the image ID searched by word index DB, and the appearance frequency of the word for every image ID. It is a hardware block diagram of a data presentation system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100,200 Material presentation system 10 Participant recognition part 11 Material recognition part 12 Information presentation part 20 Information provision part 21 Database registration part 22 Database search part 30 Conference database 31 Word index DB
32 word buffers

Claims (17)

First recognition means for recognizing participants participating in the first meeting;
Material recognition means for recognizing the material used in the first meeting;
Registration means for registering the material data recognized by the material recognition means in the storage means in association with the information of the participant;
A second recognition means for recognizing a participant who participates in the second meeting;
Search means for searching for material data to be used in the second meeting from the storage means based on the information of the participants recognized in the second meeting;
A material presentation system comprising: presentation means for visually presenting the material based on the material data retrieved by the retrieval unit. The material presentation system according to claim 1, wherein the presenting unit presents a reduced image of the material retrieved by the retrieval unit so as to be visible. The material presentation system according to claim 1, wherein the first recognition unit recognizes the participant by detecting identification information of a portable device carried by the participant. 4. The document recognition unit according to claim 1, wherein the document recognition unit recognizes the document by capturing an image of the screen when a display screen provided by a computer that outputs a video signal to a predetermined projection unit changes. The material presentation system described in Crab. The material presentation system according to any one of claims 1 to 3, wherein the material recognition unit recognizes the material by capturing an image based on a video signal input to a predetermined projection unit. When the search means does not match a participant who participates in the first conference and a participant in the second conference, the search means and a participant of the second conference at a certain time of the first conference The material presentation system according to any one of claims 1 to 5, wherein the material used in the second meeting is searched based on the overlapping number. The retrieval means retrieves the material so as to restrict the presentation of the material used in the second conference due to the overlap of the participant participating in the first conference and the participant participating in the second conference. The material presentation system according to any one of claims 1 to 6. The material presentation system according to claim 2, wherein the presenting unit enables an operation on a reduced image of the material when a predetermined password is input. The material presentation system according to any one of claims 1 to 7, wherein the presentation unit restricts the presentation of the material according to a word included in the material searched by the search unit. 8. The presentation unit according to claim 1, wherein the presentation unit limits presentation of the material according to a similarity obtained by a predetermined word group and the number of appearances of words included in the material searched by the search unit. The material presentation system described in any one. A first recognition step for recognizing participants participating in the first meeting;
A material recognition step for recognizing the material used in the first meeting;
A registration step of registering the material data recognized in the material recognition step in the storage means in association with the information of the participant;
A second recognition step for recognizing a participant who participates in the second conference;
A search step of searching for material data to be used in the second meeting from the storage unit based on the information of the participants recognized in the second meeting;
A material presentation method comprising: a presentation step of visually presenting the material based on the material data retrieved in the retrieval step. 12. The material presentation method according to claim 11, wherein the material recognition step recognizes the material by capturing an image of the screen when a display screen provided by a computer that outputs a video signal to a predetermined projection unit changes. . The material presentation method according to claim 11, wherein the material recognition step recognizes the material by capturing an image based on a video signal input to a predetermined projection unit. In the search step, when a participant who participates in the first conference does not match a participant in the second conference, a participant in the second conference and a participant at a certain time of the first conference The material presentation method according to any one of claims 11 to 13, wherein a material used in the second meeting is searched based on an overlapping number. The searching step searches for the material so as to restrict the presentation of the material used in the second conference due to the overlap of the participant participating in the first conference and the participant participating in the second conference. The material presentation method according to claim 11. The material presentation method according to any one of claims 11 to 14, wherein the presenting step includes a step of restricting presentation of the material according to a word included in the material searched in the search step. 12. The presenting step includes a step of restricting the presentation of the material according to a similarity obtained by a predetermined word group and the number of appearances of words included in the material retrieved in the retrieving step. Item 15. The method for presenting material according to any one of Items 14 to 14.

Images