JP4165095B2 - Information providing apparatus and information providing method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an information providing apparatus and an information providing method for providing information to an unspecified passerby.
[0002]
[Prior art]
Conventional information providing apparatuses or information providing methods can be broadly classified into those that provide information for unspecified persons and those that provide information for specific persons.
[0003]
As an example of the former, an information providing apparatus having an electronic bulletin board / display installed in a store or a street is known. This type of device is mainly used to attract customers and expand the number of customers. For example, it is designed to repeatedly display information such as store introductions, recommended product guidance, and new products and new books. It has become. As other uses, it is also used for the purpose of displaying guidance in the facility.
[0004]
As the latter example, an information providing service using the Internet or a mobile phone is known. For example, Japanese Patent Laid-Open No. 2001-203811 proposes a system in which a user of a mobile phone can enjoy an information providing service that suits his / her lifestyle and taste.
[0005]
[Problems to be solved by the invention]
By the way, the former type cannot provide information tailored to the preference of each individual because of the relationship with unspecified persons. Also, since passersby only receive information passively, they are rarely interested in the displayed information. Therefore, there has been a problem that even if advertisement information is displayed, a high appeal effect cannot be obtained.
[0006]
On the other hand, since the latter type targets a specific person, there is an advantage that it is possible to provide an information providing service in accordance with individual preference. However, in this case, since registration of personal information is essential, there is a problem that it is difficult to expand the number of users. In other words, there are many people who feel resistance to registration of personal information from the viewpoint of privacy protection and a sense of crisis such as leakage, outflow and misuse of personal information, and such people hesitate to use the information providing service itself. There are many cases.
[0007]
The present invention has been made in view of the above circumstances, and the purpose of the present invention is to provide information that can provide unspecified passers-by with information of content that suits their preference It is to implement an apparatus and an information providing method.
[0008]
[Means for Solving the Problems]
In order to achieve the above object, an information providing apparatus of the present invention is an information providing apparatus that provides information to an unspecified passer-by, an image information acquiring unit that acquires image information of a passer-by, and the acquired Feature amount extraction means for extracting a feature amount relating to an attribute of a person from image information, information determination means for determining provision information to be provided to the passer-by based on the extracted feature amount, and the determined provision An information output means for outputting information to the passerby is adopted.
[0009]
According to this configuration, it is possible to estimate an attribute unique to a person from an unspecified passer (feature amount) without acquiring personal information related to privacy. Various attributes can be adopted, such as gender, age group (age), race, height, physique, hairstyle, clothes, walking style, belongings, whether walking alone or with multiple people Is preferred. If such an attribute can be known, it is possible to estimate the preference and lifestyle trend of the person based on the attribute, and further, it is possible to determine provision information that matches the preference and lifestyle trend. . Therefore, it is possible to provide information with content that suits the preference of an unspecified passerby without acquiring personal information.
[0010]
As a result, information providers can efficiently provide information that is highly appealing, while passers-by can obtain useful information for themselves without any special effort. The profit for both is great.
[0011]
The information determining unit includes a classifying unit that classifies the passerby into a predetermined category based on the extracted feature amount, and an information selection unit that selects the provision information to be provided according to a category to which the passer belongs. It is preferable to have
[0012]
According to this configuration, it is possible to classify unspecified passers into a predetermined category without acquiring personal information related to privacy. There are various ways of classification. For example, the classification may be performed for each of the above attributes or by a combination of a plurality of attributes. By classifying passers-by in this way, preference and lifestyle trends can be easily estimated for each category (in other words, categories can be set based on preferences and lifestyle trends). . And according to a category, the provision information suitable for the taste and lifestyle tendency can be selected. Therefore, it is possible to provide information with content that suits the preference of an unspecified passerby without acquiring personal information.
[0013]
Provided information storing means for storing a plurality of types of provided information classified in advance corresponding to the category, wherein the information selecting means displays the provided information of the type corresponding to the category to which the passerby belongs to the provided information storing means. It is preferable to select from.
[0014]
As described above, since the preference and lifestyle tendency can be estimated for each category, if the provision information is classified and stored in advance in the provision information storage means corresponding to the category, the information decision processing in the information decision means Becomes simple. As a result, since it is possible to shorten the processing time required from the acquisition of the image information of the passer-by to the provision of information, it is possible to provide the optimum information in real time to the passer-by who is moving. The process of classifying the provided information may be performed automatically, or may be performed manually by an information provider (device manager).
[0015]
The provided information output by the information output means preferably includes summary information indicating the gist of the content of the information and reference information indicating a more detailed content acquisition method.
[0016]
By including the summary information, the gist of the contents of the information can be instantaneously transmitted to a passerby who is moving. Therefore, there is a high possibility that the output provided information will attract the attention of the passerby. Further, a passerby who is interested in the summary information can acquire more detailed contents based on the reference information, and therefore can provide information with contents that are substantial both quantitatively and qualitatively. Therefore, a high appeal effect can be obtained.
[0017]
As the image information acquisition means for acquiring the image information of the passerby, an image pickup means such as a camera that acquires the image information by imaging the face, half or whole body of the passerby can be suitably used. In addition, detection means such as an infrared sensor and a distance sensor that acquire shape information and unevenness information of the face, half or whole body of a passerby can be suitably used. Note that the image information acquisition means is not limited to those exemplified here, and various means can be suitably used as long as they can acquire image information that can extract a feature amount related to an attribute of a person. Can do.
[0018]
Examples of the information output means include a display means such as an electric bulletin board or a display for outputting information in characters or images, an audio output means such as a speaker for outputting information by voice, a mobile phone or a portable information terminal possessed by a passerby A transmitting means for transmitting digital information to the mobile terminal, a printing means for printing information on a paper medium or the like, and other similar information output means can be preferably used.
[0019]
A configuration that includes a moving speed detection unit that detects the moving speed of the passerby and provides information to the passerby only when the detected moving speed is a predetermined value or less is also preferable. Passers who move at a relatively fast speed are expected to be less interested in the output information, while passers who move at a relatively slow speed or those who are stationary can receive the output information It is considered that the nature Therefore, if it is determined whether or not to provide information to the passer-by based on the travel speed of the passer-by, it is possible to provide information with high efficiency and high appealing effect.
[0020]
It is preferable that the moving speed detection unit calculates the moving speed of the passer-by based on a plurality of pieces of image information acquired at predetermined time intervals. If the moving speed is detected using the image information acquired by the image information acquiring means, it is not necessary to separately provide a sensor for measuring the moving speed of a passerby, so that the apparatus configuration becomes simple and low. Cost can be reduced.
[0021]
Note that the information providing apparatus of the present invention can be realized as a single-unit apparatus, or as a system including a plurality of information processing apparatuses including a terminal and a server. For example, an information providing apparatus (system) as one embodiment thereof is an information providing apparatus that provides information to an unspecified passer, and includes a terminal and a server, and the terminal Image information acquisition means for acquiring image information, image information transmission means for transmitting the acquired image information to the server, and information output means for outputting provided information received from the server to the passers-by The server includes image information receiving means for receiving image information from the terminal, feature quantity extracting means for extracting a feature quantity relating to a human attribute from the received image information, and the extracted feature quantity. A configuration comprising information determining means for determining provided information to be provided to the passer based on the information, and providing information transmitting means for transmitting the determined provided information to the terminal is preferable.
[0022]
Further, the information providing method of the present invention is an information providing method for providing information to an unspecified passer-by, wherein the image information of the passer-by is obtained, and the attribute of the person is obtained from the acquired image information. Extracting the feature amount, determining the provision information to be provided to the passer based on the extracted feature amount, and outputting the determined provision information to the passer-by It is characterized by including.
[0023]
The step of determining the provision information to be provided includes the step of classifying the passer-by into a predetermined category based on the extracted feature amount, and the provision information to be provided is selected according to the category to which the passer-by belongs. Preferably comprising the steps of:
[0024]
In the step of selecting the provision information to be provided, the provision information corresponding to the category to which the passer-by belongs is provided from provision information storage means for storing a plurality of kinds of provision information classified in advance corresponding to the category. It is preferable to select.
[0025]
The provision information to be output preferably includes summary information indicating a gist of information content and reference information indicating a more detailed content acquisition method.
[0026]
It is also preferable to include a step of detecting the travel speed of the passer-by and a step of providing information to the passer-by only when the detected travel speed is a predetermined value or less.
[0027]
DETAILED DESCRIPTION OF THE INVENTION
Preferred embodiments of the present invention will be illustratively described below with reference to the drawings.
[0028]
First, a schematic configuration of an information providing apparatus 1 according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a conceptual diagram of an information providing apparatus 1 according to an embodiment of the present invention.
[0029]
As shown in the figure, the information providing apparatus 1 includes an image information acquisition unit 100 that acquires image information such as a face, a half body, and a whole body of a passerby 3 and a feature amount relating to a person's attribute from the acquired image information. Feature amount extracting means 101 to be extracted, information determining means 102 for determining provision information to be provided to the passerby 3 based on the extracted feature amount, and information output means for outputting the determined provision information to the passerby 3 103.
[0030]
According to this configuration, an attribute specific to the person is estimated from an unspecified passerby 3 without acquiring personal information related to privacy, and the preference or lifestyle of the person is estimated based on the attribute. Trends can be estimated. In other words, if the information determining means 102 determines provision information that matches the preference estimated from the feature amount and the lifestyle, the preference of the person can be obtained without acquiring personal information for an unspecified passerby 3. It is possible to provide information that matches the content.
[0031]
As a result, the information provider can efficiently provide information with a high appeal effect, while the passerby 3 can acquire useful information for the passerby without any special effort. , The benefits for both are great.
[0032]
Next, a detailed configuration for realizing the information providing apparatus 1 will be described.
[0033]
Here, an example will be described in which the present invention is applied to an information providing apparatus used for providing advertisement information related to a newly published magazine in a bookstore or a convenience store. The scope of application of the present invention is not limited to this. For example, in order to provide advertising information such as store introductions, recommended product guidance, and new product information installed in stores or streets. Appropriately applied to devices used, devices installed in theme parks, shopping malls, etc., used to provide guidance information in facilities, and other devices that provide information providing services to unspecified passers-by Can do.
[0034]
(First embodiment)
First, the hardware configuration of the information providing apparatus according to the first embodiment of the present invention will be described with reference to FIG. 2 and FIG.
[0035]
FIG. 2 is a block diagram schematically illustrating an internal configuration of the information providing apparatus 1 according to the present embodiment, and FIG. 3 is an explanatory diagram schematically illustrating an installation example of the information providing apparatus 1.
[0036]
The information providing apparatus 1 generally includes a camera 11, a display 12, a communication unit 13, a memory unit 14, a CPU 15, and a storage unit 16. These components are connected to each other via a bus 20. Specifically, it can be configured by connecting the camera 11 and the display 12 to an information processing apparatus such as a personal computer.
[0037]
The information providing apparatus 1 having such a configuration is installed in a manner as shown in FIG. FIG. 3 shows a state in which the inside I and the passage O of the bookstore are viewed from above. In this example, two information providing devices 1a and 1b are installed in a store I in a bookstore. The information providing apparatus 1a has a configuration in which a camera 11a and a display 12a are connected to an information processing apparatus 10a including the communication unit 13, a memory unit 14, a CPU (Central Processing Unit) 15, and a storage unit 16. On the other hand, the information providing apparatus 1b has a configuration in which a camera 11b and a display 12b are connected to an information processing apparatus 10b having a similar configuration.
[0038]
In the following description, the configuration or operation common to both of the two information providing apparatuses 1a and 1b will be described without using the symbols “a” and “b”, and is unique to either apparatus. The configuration or operation will be described with reference to “a” or “b”.
[0039]
The camera 11 is image information acquisition means for acquiring image information of a passerby 3 (3a, 3b) passing through the passage O outside the store. The camera 11 is installed in the show window 2 facing the passage O, and acquires image information by imaging the face, half body or whole body of the passerby 3 existing in the angle of view (the dashed line in the figure). Image information obtained by imaging is input to the information processing apparatus 10. As the camera 11, various imaging means capable of imaging the passerby 3 and taking it into the apparatus as digital data, such as a CCD (Charge Coupled Device) camera, a video camera, and a digital camera can be used.
[0040]
The display 12 is an information output unit that is installed adjacent to the show window 2 and outputs provided information to the passerby 3. As the display 12, various display means capable of outputting information in characters or images, such as a liquid crystal display, a plasma display, a CRT (Cathode Ray Tube), an FED (Field Emission Display), and a projector can be suitably used. Further, display means such as an electric bulletin board may be used instead of the display.
[0041]
In the present embodiment, the display means (display 12) is used as the information output means. However, instead of the display means, voice output means such as a speaker for outputting information by voice, a passerby possesses the information output means. A transmitting means for transmitting digital information to a mobile terminal such as a mobile phone or a portable information terminal, a printing means for printing information on a paper medium or the like, or other similar information output means may be used.
[0042]
The communication unit 13 is a communication unit for controlling data communication using a computer network such as the Internet or an intranet, and corresponds to, for example, a network adapter, a terminal adapter, or a modem. The information providing apparatus 1 can exchange data with a server or another information providing apparatus via the communication unit 13.
[0043]
The memory unit 14 is a main storage unit including a readable / writable memory such as a RAM (Random Access Memory) and a read-only memory such as a ROM (Read Only Memory). The ROM stores a basic input / output system (BIOS) of the information processing apparatus 10. On the other hand, the RAM temporarily stores and holds a program and data to be executed by the CPU 15 and is a part that becomes an execution area of a program that executes processing to be described later.
[0044]
The CPU 15 receives data stored in the storage unit 16, data input from the camera 11 or data received via the communication unit 13, or a program or communication unit 13 stored in the ROM or the storage unit 16. It is an arithmetic means for reading the received program into the RAM and performing arithmetic processing based on the read contents. The CPU 15 is also a control unit that controls devices such as the camera 11 and the display 12 based on the result of the arithmetic processing.
[0045]
The storage unit 16 is auxiliary storage means for supplementing the storage capacity of the memory unit 14. For example, a magnetic disk device (hard disk, etc.), an optical disk device (CD-ROM, CD-R, CD-R / W, DVD-ROM, DVD-RAM, etc.), a magneto-optical disk device (MO, etc.), a semiconductor disk device (memory) Card). The storage unit 16 stores a program necessary for achieving the purpose of use of the apparatus, that is, a program necessary for executing a series of processes described later. Basic software such as an operation system and a device driver is also stored in the storage unit 16. In the storage unit 16, databases such as an attribute category database 17, a customer database 18, and a provision information database 19 are constructed. A detailed description of these databases will be described later.
[0046]
Next, a functional configuration of the information providing apparatus 1 will be described with reference to FIG. FIG. 4 is a functional block diagram of the information providing apparatus 1 according to the present embodiment.
[0047]
As shown in the figure, the information providing apparatus 1 includes functional blocks such as an outline, a feature amount extraction unit 30 and an information determination unit 31.
[0048]
These functional blocks are realized by the program and data stored in the storage unit 16 being read into the memory unit 14 and processed by the CPU 15. The program for realizing these functional blocks may be configured by a single program or may be realized by combining a plurality of module programs.
[0049]
The feature amount extraction unit 30 is composed of functional blocks of an image input unit 32, a face detection unit 33, and a face feature extraction unit 34, and extracts feature amounts relating to human attributes from image information acquired by the camera 11. It functions as a feature amount extraction means.
[0050]
There are various ways to classify people, that is, how to select attributes. Which one should be selected may be determined as appropriate according to the content of information to be provided, the target layer to which information is provided, and the like. For example, it is conceivable to classify people based on attributes such as gender, age group, race, height, physique, hairstyle, and clothing. In the present embodiment, two attributes of sex and age group are used.
[0051]
The image input unit 32 functions as an image input unit that captures image information captured by the camera 11 into the apparatus in a digitally processable format. For example, when an analog imaging means is used as the camera 11, the image input unit 32 converts it into a digital image (hereinafter, digitized image information is referred to as “image data”). Alternatively, when moving image capturing means such as a video camera is used as the camera 11, the image input unit 32 captures image data for one frame.
[0052]
The face detection unit 33 functions as a face area detection unit that detects an area corresponding to a passer's face from image data and generates information (face area information) such as coordinate points representing the area. In addition, the face detection unit 33 calculates the center line of the face from the position of the parts constituting the face such as the eyebrows, eyes, nose, and mouth in the face area, and estimates the face direction, and the eye part May have a function of estimating the direction of the line of sight based on the ratio of the black-eye area and the white-eye area. Such a function is realized by applying the technology disclosed in, for example, “Bad Protection Support System, OMRON TECHNICS, Vol.39, No.2, pp.108-113, (1999)”. Is possible.
[0053]
The face feature extracting unit 34 functions as a face feature extracting unit that extracts a feature amount related to the attribute from face area image data determined by face area information. The feature amount is, for example, a parameterization of the size, position, relative arrangement of parts constituting the face such as eyebrows, eyes, nose, mouth, and the face outline. Such functions include, for example, `` Classifying Facial Attributes using a 2-D Gabor Wavelet Representation and Discriminant Analysis, Proceedings of the 4th International Conference on Automatic Face and Gesture Recognition, pp. 202-207, (2000) '' and `` Modeling the This can be achieved by applying the technology disclosed in “Process of Aging in Face Images, Proceedings of the International Conference on Computer Vision, pp. 131-136, (1999)”.
[0054]
The information determination unit 31 includes functional blocks of the category classification unit 35 and the information selection unit 36, and determines information to be provided to passersby based on the feature amount extracted by the feature amount extraction unit 30. Functions as a means.
[0055]
The category classification unit 35 functions as a classification unit that classifies passers-by into predetermined categories based on the feature amounts extracted by the feature amount extraction unit 30. Specifically, with reference to attribute category information stored in the attribute category database 17 in advance, it is estimated to which category the passerby related to the extracted feature amount belongs.
[0056]
As shown in FIG. 5, the attribute category information stored in the attribute category database 17 has a data structure in which attributes (gender, age group), categories, and parameters are associated with each other.
[0057]
The category classification unit 35 compares the extracted feature quantity with the parameter relating to each category, and calculates which category parameter the feature quantity most closely matches.
[0058]
In the figure, a to o and A to O represent parameters for convenience. How to have parameters, such as a method for having parameters for each attribute, such as gender estimation parameters, age group estimation parameters, race estimation parameters, or a method for having one parameter as a combined category estimation parameter Various methods can be employed. Each parameter is composed of one or more numerical values or numerical ranges.
[0059]
For example, as parameters for estimating the age group, 0.0 to 0.2 is less than 15 years old, 0.2 to 0.4 is 15 to 25 years old, 0.4 to 0.6 is 25 to 40 years old, The parameter normalized in the numerical range of 0.0-1.0 is employ | adopted so that 0.6-0.8 may be 40-60 years old, and 0.8-1.0 may be 60 years old or more. At this time, the category classification unit 35 checks in which numerical range the value of the feature value related to the age group among the extracted feature values is included. For example, in the case of “0.12”, it is determined that it corresponds to “under 15 years old”. Then, the determination process is similarly performed for the other parameters, and finally, it is determined to which category the passerby related to the extracted feature amount belongs.
[0060]
The information selection unit 36 functions as information selection means for selecting provided information to be provided according to the category to which the passerby belongs. Specifically, the information selection unit 36 selects the type of provision information corresponding to the category from the provision information database 19 serving as provision information storage means, using the category to which the passerby belongs as a key.
[0061]
The provision information database 19 stores a plurality of types of provision information classified in advance corresponding to each category. FIG. 6 shows an example of the data structure in a tabular format. In the figure, each column represents a classification based on gender, each row represents a classification based on an age group, and provided information corresponding to each category is represented in each cell where the row and the column intersect. For example, the provided information corresponding to the category “girls student group” includes three magazines “Fashion A (3/10)”, “Hair H (March / October)”, and “Interior M (Biweekly)”. It is information about. The classification of magazines is set according to the readership assumed by the publisher.
[0062]
The category classification unit 35 and the information selection unit 36 store the category to which the passerby belongs and the provision information selected corresponding to the category in the customer database 18 as customer data. At this time, the date and time when the image data was acquired and the name of the store where the apparatus is installed are stored together. The customer data accumulated in this way can be used as useful marketing information on the store side.
[0063]
Next, a specific example of the information providing method according to the information providing apparatus 1 having the above configuration will be described with reference to FIGS. 3 and 7 to 9.
[0064]
7 to 9 are flowcharts showing the processing flow of the information providing apparatus 1. FIG. 7 shows a feature amount extraction process, FIG. 8 shows a category classification process, and FIG. 9 shows an information selection process. FIG. 3 shows a state in which two passers-by 3 a and 3 b are passing in front of the show window 2. The passers-by 3a and 3b are moving in the direction of the arrows in the figure, and the passer-by 3a has come within the angle of view of the camera 11a, and the passerby 3b has just reached the angle of view of the camera 11b. Here, it is assumed that the passerby 3a is a woman in her 40s and the passerby 3b is a man in her 20s.
[0065]
When the information providing apparatus 1a (1b) is activated, the CPU 15 reads the program and data stored in the storage unit 16 into the memory unit 14 and starts the feature amount extraction process. The category classification process is executed after the feature quantity extraction process is normally completed, and the information selection process is executed when the category classification process is normally completed.
[0066]
When the feature amount extraction process starts, first, the camera 11a (11b) captures an image and acquires image information of the passerby 3a (3b) (step S100). This image information is taken into the information processing apparatus 10a (10b) as digital image data by the image input unit 32 (step S101).
[0067]
Subsequently, the face detection unit 33 detects an area corresponding to the face from the image data, and generates face area information (step S102). At this time, the face direction and the line-of-sight direction may be estimated. Here, when the generation of the face area information has failed (for example, when no passerby exists in the image) or when it is determined that the passerby is not facing the direction of the show window 2, It is assumed that the detection has failed. If the detection of the face portion fails, the processing of the current image data is stopped, and the process returns to step S100 to acquire new image information (step S103).
[0068]
When the detection of the face portion is successful, the face feature extraction unit 34 extracts a feature amount from the face area image data (step S104), and stores the feature amount in the storage unit 16 (step S105). The stored feature amount is used for the category classification process of the next process and is stored in a person feature information database (not shown) provided in the storage unit 16. The feature amount data accumulated in this way is used as sample data for grasping the tendency and variation of the face feature amount, learning data for improving the accuracy of face detection and face feature extraction, and the like.
[0069]
Next, the category classification unit 35 is activated to start the category classification process. The category classification unit 35 reads the feature amount extracted in step S104 from the storage unit 16 (step S110) and also reads attribute category information from the attribute category database 17 (step S111). Then, the category is estimated by comparing the feature amount and the parameter relating to each category (step S112).
[0070]
In the case of the information providing device 1a, it is determined that the feature amount of the passerby 3a most closely matches the parameter “j, k, l”, the gender is “female”, the age group is “40-60 years”, and the category is “ Classified as “middle-aged housewives”. On the other hand, in the case of the information providing apparatus 1b, it is determined that the feature amount of the passerby 3b most closely matches the parameters “D, E, F”, the gender is “male”, the age group is “15 to 25 years”, the category Are classified as “students”.
[0071]
The category classification unit 35 stores the category to which the passers-by 3a and 3b belong in the customer database 18 (step S113), and ends the process.
[0072]
Next, the information selection unit 36 is activated to start the information selection process. The information selection unit 36 reads the category estimated in step S112 from the customer database 18 (step S120), and selects the provision information to be provided from the provision information database 19 using this category as a key (step S121).
[0073]
In the case of the information providing device 1a, the category to which the passerby 3a belongs is “middle-aged housewives”, so the provided information is “Cooking B (April issue)”, “Health A (April issue)”, “Women C Information on three magazine names “Every Week” is selected. On the other hand, in the case of the information providing apparatus 1b, the category to which the passerby 3b belongs is “student group”, so the provided information includes “fashion M (April issue)”, “sports S (April issue)”, “comic B”. (Weekly Month) "information on the three magazines is selected.
[0074]
By displaying the provision information determined in this way on the display 12, the provision information is output to the passers-by 3a and 3b (step S122). FIG. 10 shows a display example on the display 12a, and FIG. 11 shows a display example on the display 12b. Three magazine names of “Cooking B (April issue)”, “Health A (April issue)”, and “Women C (Weekly month)” are displayed on the display 12a as “Today's release magazine information”. . On the other hand, three magazine names “Fashion M (April issue)”, “Sports S (April issue)”, and “Comic B (Monthly month)” are displayed on the display 12b as “Today's release magazine information”. Is done.
[0075]
Here, only the magazine name is displayed as the summary information 40 indicating the gist of the content of the information without displaying the specific content of the magazine (the summary information such as the headline of the special feature page is displayed together with the magazine name). Also good). The URL (Uniform Resource Locator) “http://www.xxx.net” (FIG. 10), the sales location in the store (FIG. 11), and the like are displayed as reference information 41 indicating a more detailed content acquisition method. To do. Passers-by 3a, 3b who are interested by looking at the magazine name can obtain detailed information such as the table of contents of the magazine and introductory articles by accessing the homepage of the above URL from a mobile phone or portable information terminal. You can learn more about the magazine by visiting the sales location or by asking the store clerk.
[0076]
Thus, the information providing process is completed.
[0077]
According to the configuration of the present embodiment described above, the feature amount is extracted from the face image information acquired by the camera 11 and the attribute of the passerby is automatically estimated based on the feature amount. Therefore, the personal information related to privacy An unspecified passerby can be classified into a predetermined category without obtaining the password. The passersby can be divided into two categories: gender and age group, female youth group, female student group, housewife / office worker group, middle-age housewife group, female elderly group, young group, student group, businessman group, Since it was classified into 10 categories of middle-aged males and male seniors, it was possible to estimate the preference and lifestyle trends for each category, and matched the preferences and lifestyle trends according to the categories Offer information can be selected. Therefore, it is possible to provide information with content that suits the preference of an unspecified passerby without acquiring personal information.
[0078]
This makes it possible for information providers to efficiently provide highly appealing information, while for passers-by, information useful to them without having to enter the store or search for information on their own. Can be obtained, so the profits for both are great.
[0079]
Further, if the direction of the passer's face and the direction of the line of sight are estimated and it is determined that the direction of the show window 2 (device installation side) is facing, and the information is provided, the output provided information is the passerby. This increases the possibility of catching your eyes and attracting interest, so it is possible to provide more efficient information.
[0080]
Further, since the provision information is classified and stored in advance in the provision information database 19 corresponding to the category, and the provision information to be provided is selected from the category as a key, the information determination process is simplified. As a result, since it is possible to shorten the processing time required from the acquisition of the image information of the passer-by to the provision of the information, it is possible to provide the optimum information in real time to the passer-by who is moving.
[0081]
Moreover, since the summary information 40 such as the magazine name is displayed on the display 12, the gist of the information content can be instantly transmitted to a passerby who is moving. Therefore, there is a high possibility that the output provided information will attract the attention of the passerby. Further, by displaying the reference information 41 such as a URL in addition to the summary information 40, a passerby who is interested in the magazine name can acquire more detailed contents based on the URL. It is possible to provide rich information. Therefore, a high appeal effect can be obtained.
[0082]
Furthermore, since the category to which the passerby belongs, the provision information selected corresponding to the category, the acquisition date and time of the image data, etc. are automatically stored as customer data, a marketing survey can be performed without incurring labor costs. In addition, if the number of categories or how to divide the category is changed based on the customer data, or the content of the provided information is updated, it is possible to provide information with content that better matches the preference of the passerby.
[0083]
(Second Embodiment)
FIG. 12 shows a second embodiment of the present invention. In the first embodiment, an information providing apparatus having a single configuration is shown. However, in the present embodiment, an information providing apparatus configured by a terminal having a camera and a display and a server responsible for special quantity extraction processing and information determination processing. An information providing system is shown.
[0084]
Since other configurations and operations are the same as those of the first embodiment, the same components are denoted by the same reference numerals, and the description thereof is omitted.
[0085]
FIG. 12 is a system configuration diagram of the information providing system 4 according to the present embodiment.
[0086]
As shown in the figure, the information providing system 4 is configured by connecting a terminal 5 and a server 6 via a computer network 7 such as the Internet.
[0087]
The terminal 5 has a configuration in which a camera 11 and a display 12 are connected to an information processing apparatus 10 such as a personal computer. In addition to the function of the image input unit 32 shown in FIG. 4 of the first embodiment, the information processing apparatus 10 has a function of transmitting acquired image data to the server 6, a function of receiving provided information from the server 6, and reception The provided information is output to the display 12.
[0088]
On the other hand, the server 6 includes an information processing apparatus such as one or a plurality of personal computers, workstations, or host computers, and includes an attribute category database 17, a customer database 18, and a provision information database 19. The server 6 has functions of receiving image data from the terminal 5 in addition to the functions of the face detection unit 33, the face feature extraction unit 34, the category classification unit 35, and the information selection unit 36 shown in FIG. 4 of the first embodiment. And having the function of transmitting the determined provided information to the terminal 5.
[0089]
In the information providing system 4 configured as described above, image information captured by the camera 11 of the terminal 5 installed at a storefront or the like is transmitted to the server 6 through the computer network 7. The server 6 performs passerby extra-volume extraction processing and provision information determination processing, and the provision information is transmitted to the terminal 5 through the computer network 7. The terminal 5 displays the received provision information on the display 12 and outputs the provision information to a passerby.
[0090]
Even with the configuration of the present embodiment described above, the same effects as those of the first embodiment can be obtained.
[0091]
Furthermore, in this embodiment, since the function is divided into the terminal 5 and the server 6, the structure which installs a some terminal with respect to one server is attained. According to this configuration, the provision information is managed in an integrated manner, so that the provision information can be easily updated. In addition, since customer data from a plurality of terminals is accumulated in the server, the value as marketing information increases. Furthermore, since the equipment cost of the terminal can be kept low, introduction of the device is promoted.
[0092]
Conversely, a configuration in which one terminal transmits image data to a plurality of servers and obtains provided information from the plurality of servers is also possible. If the contents of the provision information database 19 are changed for each server, the terminal user can acquire provision information of various contents by selecting a server to be used according to the purpose and application.
[0093]
(Third embodiment)
FIG. 13 shows a third embodiment of the present invention. FIG. 13 is a flowchart illustrating the flow of the moving speed detection process and the feature amount detection process in the information providing process of the information providing apparatus according to the present embodiment.
[0094]
The information providing apparatus according to the present embodiment includes a movement speed detection unit that detects a movement speed of a passerby. Since other configurations and operations are the same as those of the first embodiment, the same components are denoted by the same reference numerals, and the description thereof is omitted.
[0095]
In the present embodiment, a movement speed detection unit that functions as movement speed detection means is further provided in the feature quantity extraction unit 30 in FIG. 4 (not shown). The moving speed detection unit calculates the moving speed of the passerby based on a plurality of pieces of image information (time-series images) obtained by imaging with the camera 11 with a predetermined time interval (comma several seconds to several seconds). The specific method will be described with reference to the flowchart of FIG.
[0096]
When the feature amount extraction process starts, first, the moving speed detection unit resets the number counter to 0 (step S130). And the camera 11 images and acquires the image information of the passerby 3 (step S131). This image information is taken into the information processing apparatus 10 as digital image data by the image input unit 32 (step S132).
[0097]
Subsequently, the face detection unit 33 detects an area corresponding to the face from the image data, and generates face area information (step S133). At this time, when the number counter is 1 or more, the moving speed detection unit determines whether the detected face belongs to the same passerby as the face included in the previously processed image data. This process does not need to be performed so strictly, and it is sufficient to compare the face area image data and determine whether the similarity exceeds a predetermined threshold value. Further, since the direction of movement of the passerby can be estimated from the face direction or the like, it may be determined whether or not the same passerby is based on the transition of the position of the detected face area.
[0098]
Here, even if the generation of face area information has failed (such as when no passerby is present in the image) or the generation of face area information has succeeded, it was not that of the same passerby In this case, it is considered that the detection of the face portion has failed. If the detection of the face portion fails, the current process is stopped, and the process returns to step S130 to start a new process (step S134).
[0099]
On the other hand, if the face portion is successfully detected in step S134, the moving speed detection unit adds 1 to the number counter (step S135), and determines whether the count value of the number counter has reached a predetermined number. (Step S136). If the predetermined number has not been reached, the processes in steps S131 to S135 are repeated. If the predetermined number has been reached, the process proceeds to a process for extracting the feature amount of the passer's face (step S104). Subsequent processing is the same as that of the first embodiment.
[0100]
The processing by the moving speed detection unit described above has the following meaning. That is, checking whether or not the same passerby is included in the predetermined number of time-series images means that the passerby was present within the angle of view of the camera 11 while the images were captured. For example, this corresponds to detecting whether the speed of movement of the passerby is slower than a predetermined value.
[0101]
If the moving speed is less than or equal to a predetermined value (when a predetermined number of images can be acquired), the process of step S104 is continued to provide information to the passerby. If it is too early (when the predetermined number cannot be obtained), the process is stopped and the process returns to step S130, so that information is not provided to the passerby.
[0102]
Passers who move at a relatively fast speed are expected to be less interested in the output information, while passers who move at a relatively slow speed or those who are stationary can receive the output information It is considered that the nature Therefore, if it is determined whether or not to provide information to the passer-by based on the travel speed of the passer-by as in the present embodiment, information can be provided efficiently and with a high appealing effect.
[0103]
Needless to say, the configuration of the present embodiment can achieve the same effects as those of the first embodiment.
[0104]
Furthermore, in the present embodiment, since the movement speed of the passer-by is calculated based on a plurality of pieces of image information, it is not necessary to separately provide a sensor for measuring the movement speed of the passer-by. As a result, the apparatus configuration is simplified and the cost can be reduced.
[0105]
In addition, although the apparatus configuration is complicated, a configuration in which a sensor or the like for measuring the moving speed of a passerby is separately provided may be employed. Also in this case, it is possible to control whether or not to provide information based on the moving speed of the passerby, and it is possible to achieve an effect that information can be provided efficiently and with a high appeal effect.
[0106]
In the present embodiment, the configuration is such that the moving speed of the passerby is detected based on whether or not a predetermined number of pieces of image data relating to the same passer have been acquired. For example, the same passerby included in a plurality of image data It is also preferable to adopt a method of calculating the moving speed based on the moving distance of the face part and the image capturing interval.
[0107]
The first to third embodiments have been illustrated as specific examples of the present invention. However, the scope of the present invention is not limited to these embodiments, and various modifications can be made based on the technical idea described above. Is possible.
[0108]
For example, in each of the above embodiments, feature amounts are extracted using a passer's face image, but feature amounts are extracted using an image of a person's half or whole body in addition to or instead of the face. It is also suitable to perform. By using half-body or whole-body images, it is possible to acquire various attributes that cannot be obtained from face images, such as physique, clothes, how to walk, belongings, whether you are walking alone or with multiple people, etc. This makes it possible to subdivide categories and provide information with a higher appeal effect.
[0109]
In the above-described embodiment, one camera is used as the image information acquisition unit. However, it is also preferable to use a plurality of cameras, or another image information acquisition unit in addition to the camera or instead of the camera. is there. The image information acquisition means may be, for example, an infrared sensor that acquires shape information and unevenness information of a passer's face, half body or whole body, detection means such as a distance sensor, and other means that can acquire image information similar to this. Various things are applicable.
[0110]
In the first embodiment, the face direction and line-of-sight direction of the passerby are estimated from one piece of image data. However, the passerby determines the direction of the apparatus installation side based on a plurality of pieces of time-series image data. It is also preferable to check the time that the person is facing and to provide information to the passer-by only when it is determined that the person is facing the predetermined time or longer.
[0111]
【The invention's effect】
As described above, according to the present invention, it is possible to provide unspecified passers-by with information with content that matches the preference of the person without acquiring personal information. As a result, information providers can efficiently provide information that is highly appealing, while passers-by can obtain useful information for themselves without any special effort. The profit for both is great.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram of an information providing apparatus according to an embodiment of the present invention.
FIG. 2 is a block diagram schematically showing an internal configuration of the information providing apparatus according to the first embodiment of the present invention.
FIG. 3 is an explanatory diagram schematically illustrating an installation example of the information providing apparatus in FIG. 1;
4 is a functional block diagram of the information providing apparatus in FIG. 1. FIG.
FIG. 5 is a diagram showing an example of a data structure of an attribute category database in a tabular format.
FIG. 6 is a diagram showing an example of a data structure of a provided information database in a table format.
7 is a flowchart showing a flow of feature amount extraction processing in the information provision processing of the information provision apparatus in FIG. 1;
8 is a flowchart showing a flow of a category classification process in the information providing process of the information providing apparatus of FIG.
9 is a flowchart showing a flow of an information selection process in the information providing process of the information providing apparatus of FIG.
FIG. 10 is a display example of provided information on a display;
FIG. 11 is a display example of provided information on a display;
FIG. 12 is a system configuration diagram of an information providing system according to a second embodiment of the present invention.
FIG. 13 is a flowchart showing a flow of a movement speed detection process and a feature amount detection process in the information provision process of the information provision apparatus according to the third embodiment of the present invention.
[Explanation of symbols]
1,1a, 1b Information providing device
2 show window
3,3a, 3b Passerby
4 Information provision system
5 Terminal
6 servers
7 Computer network
10, 10a, 10b Information processing apparatus
11, 11a, 11b Camera
12, 12a, 12b display
13 Communication Department
14 Memory part
15 CPU
16 Memory unit
17 Attribute category database
18 Customer database
19 Provided information database
20 buses
30 feature extraction unit
31 Information decision part
32 Image input section
33 Face detection unit
34 Face feature extraction unit
35 Category classification part
36 Information selector
40 Summary information
41 Reference information
100 Image information acquisition means
101 feature amount extraction means
102 Information determining means
103 Information output means
I Inside the store
O passage

Claims (14)

An information providing device that provides information to unspecified passers,
Imaging means for acquiring image information of the face, half or whole body of a passer-by,
Feature quantity extraction means for parameterizing the face configuration from the acquired image information and extracting it as a feature quantity;
Information determining means for determining provided information to be provided to the passer-by based on the extracted feature amount;
Information output means for outputting the determined provided information to the passers-by;
In an information providing apparatus comprising:
The direction of the passer's face or line of sight is obtained from the image information acquired by the imaging means, and information is provided to the passerby when the passer's face or line of sight faces the device installation side. To provide information. An information providing device that provides information to unspecified passers,
Imaging means for acquiring image information of the face, half or whole body of a passer-by,
Feature quantity extraction means for parameterizing the face configuration from the acquired image information and extracting it as a feature quantity;
Information determining means for determining provided information to be provided to the passer-by based on the extracted feature amount;
Information output means for outputting the determined provided information to the passers-by;
In an information providing apparatus comprising:
An information providing apparatus characterized in that a travel speed of the passer-by is obtained from image information acquired by the imaging means, and information is provided to the passer-by when the travel speed of the passer-by is equal to or less than a predetermined value.   The information providing apparatus according to claim 2, wherein a moving speed of the passerby is obtained from a plurality of pieces of image information acquired at predetermined time intervals.   The information providing apparatus according to claim 3, wherein when a predetermined number of pieces of image information including the same passers-by are obtained, the movement speed of the passers-by is equal to or less than the predetermined value. The information determining means includes
Classification means for classifying the passerby into a predetermined category based on the extracted feature amount;
Information selection means for selecting the provision information to be provided according to a category to which the passerby belongs;
The information providing apparatus according to claim 1, comprising: Provided information storage means for storing a plurality of types of provided information classified in advance corresponding to the category,
The information providing apparatus according to claim 5, wherein the information selection unit selects the type of provision information corresponding to the category to which the passerby belongs from the provision information storage unit.   The information providing apparatus according to any one of claims 1 to 6, wherein the provision information output by the information output unit includes summary information indicating a gist of the content of information and reference information indicating a more detailed content acquisition method. An information providing method for providing information to unspecified passers,
Acquiring image information of a passer's face, half body or whole body by an imaging means;
Parameterizing the face configuration from the acquired image information and extracting it as a feature quantity;
Determining providing information to be provided to the passer-by based on the extracted feature amount;
Outputting the determined provided information to the passers-by;
In the information providing method including
The direction of the passer's face or line of sight is obtained from the image information acquired by the imaging means, and information is provided to the passerby when the passer's face or line of sight faces the device installation side. How to provide information. An information providing method for providing information to unspecified passers,
Acquiring image information of a passer's face, half body or whole body by an imaging means;
Parameterizing the face configuration from the acquired image information and extracting it as a feature quantity;
Determining providing information to be provided to the passer-by based on the extracted feature amount;
Outputting the determined provided information to the passers-by;
In the information providing method including
An information providing method comprising: obtaining a travel speed of the passer-by from the image information acquired by the imaging means; and providing information to the passer-by when the travel speed of the passer-by is a predetermined value or less.   The information providing method according to claim 9, wherein a moving speed of the passer-by is obtained from a plurality of pieces of image information acquired at predetermined time intervals.   The information providing method according to claim 10, wherein when a predetermined number of pieces of image information including the same passers-by can be obtained, a moving speed of the passers-by is equal to or less than the predetermined value. Determining the provision information to be provided;
Classifying the passerby into a predetermined category based on the extracted feature amount;
The information providing method according to claim 8, further comprising: selecting the provision information to be provided according to a category to which the passerby belongs. In the step of selecting the provision information to be provided,
13. The information providing method according to claim 12, wherein providing information of a type corresponding to the category to which the passer belongs is selected from providing information storage means for storing a plurality of types of providing information classified in advance corresponding to the category.   The information providing method according to any one of claims 8 to 13, wherein the provided information to be output includes summary information indicating a gist of information content and reference information indicating a more detailed content acquisition method.

Images