JP5398301B2 - Traffic management device and traffic management method - Google Patents

Description

  The present invention relates to a traffic management device and a traffic management method for recognizing a face of a person to be photographed based on a human face image acquired by a camera and managing traffic to a specific area.

  2. Description of the Related Art In recent years, many face recognition apparatuses that capture a human face image with a camera and recognize a person based on the captured face image have been put into practical use. A traffic control device (entrance / exit management device) that manages the entrance / exit of a person into a specific area using such a face recognition device has been put into practical use.

  There is also provided a ticket gate system that uses a face recognition device for the entrance side ticket gate and the entrance side ticket gate, recognizes the face of a person, associates the entrance and exit, and performs the ticket gate processing (for example, Patent Document 1).

JP 2007-241853 A

  When the person enters the specific area, the ticket gate system and the entrance / exit management apparatus capture a face image of the person with a camera, extract a facial feature from the face image, and store the face feature. When a person enters from a specific area, the entrance / exit management device captures the person's face image again with the camera, extracts the face feature from the face image, and whether there is a match with the stored face feature. Perform the association.

  However, when there are many people passing through a specific area, there are many face features stored, and there is a problem that it takes an enormous amount of time for association processing. As a result, there is a problem in that the process is not successful and an error may be output.

  An object of one embodiment of the present invention is to provide a traffic management apparatus and a traffic management method capable of managing the traffic of a person with a short processing time and with high accuracy.

A traffic management device according to an embodiment of the present invention includes: a first image input unit that inputs an image including at least a face of a person passing through a first location in a specific area; and the first image input unit. First face detection means for detecting a face image from the input image; first face feature extraction means for extracting first face feature information from the face image detected by the first face detection means; Storage means for storing the first facial feature information extracted by the first facial feature extraction means, and selection means for selectively deleting the first facial feature information stored in the storage means. To do.

  According to one aspect of the present invention, it is possible to provide a traffic management device and a traffic management method capable of managing the traffic of a person with a short processing time and with high accuracy.

FIG. 1 is a block diagram schematically illustrating a configuration example of a traffic management system to which the traffic management device according to the first embodiment of the present invention is applied. FIG. 2 is an explanatory diagram for schematically explaining an example of a specific area to which the traffic management system shown in FIG. 1 is applied. FIG. 3 is an explanatory diagram for explaining an example of image data captured by the camera. FIG. 4 is a flowchart for explaining the face area detection processing performed by the face area detection unit. FIG. 5 is an explanatory diagram for schematically explaining the face area detection processing. FIG. 6 is an explanatory diagram for explaining an example of data stored in the face feature storage unit. FIG. 7 is a flowchart for explaining the matching process performed by the matching unit. FIG. 8 is an explanatory diagram for explaining an example of a traffic confirmation screen displayed on the display unit of the traffic management device shown in FIG. 1. FIG. 9 is a block diagram for schematically explaining a configuration example of a traffic management system to which the traffic management device according to the second exemplary embodiment of the present invention is applied. FIG. 10 is a block diagram for schematically explaining a configuration example of a traffic management system to which the traffic management device according to the third exemplary embodiment of the present invention is applied. FIG. 11 is an explanatory diagram for describing an example of a traffic confirmation screen displayed on the display unit of the traffic management device illustrated in FIG. 10. FIG. 12 is a block diagram for schematically explaining a configuration example of a traffic management system to which the traffic management device according to the fourth exemplary embodiment of the present invention is applied. FIG. 13 is an explanatory diagram for describing an example of a traffic confirmation screen displayed on the display unit of the traffic management device illustrated in FIG. 12. FIG. 14 is a block diagram schematically illustrating a configuration example of a traffic management system to which the traffic management device according to the fifth exemplary embodiment of the present invention is applied. FIG. 15 is a block diagram for schematically explaining another configuration example of the traffic management system to which the traffic management device according to the fifth exemplary embodiment of the present invention is applied.

Hereinafter, the traffic management device and traffic management method according to the first embodiment of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a block diagram for schematically explaining a configuration example of a traffic management system 1 to which the traffic management device according to the first exemplary embodiment of the present invention is applied. For example, the traffic management system 1 manages the state of a person who passes through a specific area.

  As illustrated in FIG. 1, the traffic management system 1 includes imaging terminals 110, 120, and 130, a window terminal 140, and a traffic management device 150. The imaging terminals 110 and 120 are provided at a passage place (first place) in a specific area. In addition, it is assumed that the imaging terminal 130 and the window terminal 140 are provided at a passage place (a second place different from the first place) in a specific area.

  The imaging terminals 110, 120, and 130 and the window terminal 140 are connected to the traffic management device 150 by, for example, a LAN. The traffic management device 150 can transmit and receive data to and from the imaging terminals 110, 120, and 130 and the window terminal 140.

  The imaging terminals 110, 120, and 130 include cameras 111, 121, and 131, respectively. As shown in FIG. 2, the imaging terminal 110 is provided in the passage A where the location A is located. In addition, the imaging terminal 120 is provided in the passage B where the location B is located. Furthermore, the imaging terminal 130 is provided in the passage C where the place C is located.

  The window terminal 140 acquires an image used for authentication from a person passing through the passage based on an operation of an attendant. The window terminal 140 includes a camera 141. The window terminal 140 is installed at a window provided in the emergency passage of the place C. The window terminal 140 includes a display unit 142. The display unit 142 displays various information based on control from the traffic management device 150.

  The cameras 111, 121, 131, and 141 are configured by, for example, an industrial television (ITV) camera. The cameras 111, 121, 131, and 141 digitally convert captured images using an A / D converter (not shown) and transmit the images to the traffic management device 150.

  The camera 111 acquires an image of a person passing through the passage A. The camera 121 acquires an image of a person passing through the passage B. The camera 131 acquires an image of a person passing through the passage C. The camera 141 acquires an image of a person visiting the window.

  For example, as illustrated in FIG. 3, the cameras 111, 121, 131, and 141 acquire image data as grayscale information of 512 pixels in the horizontal direction and 512 pixels in the vertical direction.

  The traffic management device 150 includes face area detection units 151, 152, and 153, and face feature extraction units 154, 155, and 156. Further, the traffic management device 150 includes a face feature storage unit 157, a collation unit 158, a collation determination unit 159, a selection unit 160, a selected face feature storage unit 161, a display unit 162, a collation unit 163, and a collation determination unit 164. Yes. Furthermore, the traffic management device 150 includes interfaces 165, 166, and 167.

  The interfaces 165, 166, and 167 each function as an image input unit. The interfaces 165 and 166 are connected to imaging terminals such as the imaging terminal 110, the imaging terminal 120, and the imaging terminal 130. The interfaces 165 and 166 receive images from the imaging terminal 110, the imaging terminal 120, the imaging terminal 130, and the like. The interface 167 is connected to the window terminal 140. The interface 167 receives an image from the window terminal 140.

  The face area detection units 151, 152, and 153 function as face detection means. The face area detection units 151, 152, and 153 detect areas (face areas) in which a person's face is shown in the input images input to the interfaces 165, 166, and 167, respectively.

  For example, the face area detection units 151, 152, and 153 detect the face area by obtaining a correlation value while moving a standard face pattern (face detection template) prepared in advance in the input image. Here, the face area detection units 151, 152, and 153 detect the position where the highest correlation value is calculated as the face area.

  FIG. 4 is a flowchart for explaining the face area detection processing performed by the face area detection units 151, 152, and 153. FIG. 5 is an explanatory diagram for schematically explaining the face area detection processing.

  The face area detection units 151, 152, and 153 detect a face area by obtaining a correlation value while moving a plurality of face detection templates Gs (i, j) prepared in different sizes S. That is, there are as many face detection templates Gs as there are sizes, for example, S = 1, S = 2,... S = 10.

  The face area detection units 151, 152, and 153 first select a face detection template Gs (i, j) having a size of S = 1 (step S11).

  The face area detection units 151, 152, and 153 set the Y coordinate when the face detection template Gs (i, j) is set to 0 (step S12).

  The face area detection units 151, 152, and 153 set the X coordinate when the face detection template Gs (i, j) is set to 0 (step S13).

The face area detection units 151, 152, and 153 set the area where the face detection template is installed as a local area. The face area detection units 151, 152, and 153 calculate a correlation Ms (x, y) between the input image f (x, y) in the local area and the face detection template Gs (i, j) (step S14). . Note that the face area detection units 151, 152, and 153 calculate the correlation Ms (x, y) using the calculation formula shown in Formula 1.

  The face area detection units 151, 152, and 153 compare the correlation Ms (x, y) with a threshold value for detecting a face area that can be determined in advance (step S15). Here, when it is determined that the correlation Ms (x, y) is equal to or greater than the threshold, the face area detection units 151, 152, and 153 determine the area where the face detection template Gs (i, j) is installed as the face area. Is determined. In this case, the face area detection units 151, 152, and 153 output information (x, y, s) relating to the coordinates and size of the face area to the subsequent face feature extraction unit as a result (step S16).

  When the correlation is compared with the threshold value, the face area detection units 151, 152, and 153 add 1 to the value of the X coordinate of the face detection template Gs (i, j) (step S17).

  Here, the face area detection units 151, 152, and 153 determine whether or not the X coordinate of the face detection template Gs (i, j) exceeds the width (WIDTH) of the input image (step S18). When the X coordinate of the face detection template Gs (i, j) does not exceed the width (WIDTH) of the input image (step S18, NO), the face area detection units 151, 152, and 153 loop to step S14.

  When the X coordinate of the face detection template Gs (i, j) exceeds the width (WIDTH) of the input image (step S18, YES), the face area detection units 151, 152, and 153 determine the face detection template Gs (i , J) is added to the Y coordinate value (step S19).

  Here, the face area detection units 151, 152, and 153 determine whether or not the Y coordinate of the face detection template Gs (i, j) exceeds the height (HEIGHT) of the input image (step S20). If the Y coordinate of the face detection template Gs (i, j) does not exceed the height (HEIGHT) of the input image (step S20, NO), the face area detection units 151, 152, and 153 loop to step S13. .

When the Y coordinate of the face detection template Gs (i, j) exceeds the height (HEIGHT) of the input image (step S20, YES), the face area detection units 151, 152, and 153 detect the face detection template Gs ( Change the size of i, j). That is, the face area detection units 151, 152, and 153 add 1 to the value of S (step S21).
Here, the face area detection units 151, 152, and 153 determine whether or not a face detection template Gs (i, j) having a size corresponding to the value of S exists (step S22). When the face detection template Gs (i, j) having a size corresponding to the value of S exists (step S22, YES), the face area detection units 151, 152, and 153 loop to step S12.

  That is, as shown in FIG. 5, the face area detection units 151, 152, and 153 move the face detection template in the input image and calculate the correlation at each coordinate. When the face area cannot be detected by the selected face detection template, the face area detection units 151, 152, and 153 switch the size of the face detection template and perform face area detection processing.

  When the face detection template Gs (i, j) having a size corresponding to the value of S does not exist (step S22, NO), the face area detection units 151, 152, and 153 perform face area detection processing on the input image. finish.

  There are various methods for detecting the face area, and the method is not limited to the above-described method. The face area detection unit can also be realized by using another method for detecting a face area such as eigenspace method or subspace method.

  The face feature extraction units 154, 155, and 156 function as face feature extraction means. The face feature extraction units 154, 155, and 156 extract face feature information H (x, y) based on the face area image (face image) detected by the face area detection unit.

  The face feature extraction units 154, 155, and 156 convert the face image into, for example, an image having horizontal 32 pixels, vertical 32 pixels, the highest luminance value 255, and the lowest luminance value 0. Thereby, the face feature extraction units 154, 155, and 156 extract the face feature information.

  The face feature extraction unit 154 shown in FIG. 1 transmits the face feature information extracted by the above processing to the face feature storage unit 157.

  The face feature storage unit 157 stores the face feature information received from the face feature extraction unit 154. That is, the face feature storage unit 157 stores the face feature information extracted based on the face image acquired by the camera of the input-side imaging terminal.

  FIG. 6 is an explanatory diagram for explaining an example of data stored in the face feature storage unit 157. As illustrated in FIG. 6, the face feature storage unit 157 includes ID data, location information indicating a passage location, information indicating a passage time, and face feature information as registration data Rc (i, j) of one category, The registration data is stored. In addition, when referring to registration data from the collation part 158, the collation part 158 designates the category C with a value, and selects the category used for collation.

  The traffic management device 150 receives an image captured by the imaging terminal installed in the first place via the interface 165. The traffic management device 150 extracts face feature information (first face feature information) based on an input image input from an imaging terminal installed in the first place. The traffic management device 150 stores the extracted face feature information in the face feature storage unit 157. In this case, the traffic management device 150 automatically generates serial number information and stores it as an ID.

  Further, the traffic management device 150 identifies the traffic time (the traffic time of the first location) based on the time when the image is captured by the imaging terminal installed in the first location, and displays information indicating the traffic time as a face. Store in the feature storage unit 157. Furthermore, the traffic management device 150 stores in the face feature storage unit 157 the location where the imaging terminal that captured the image is installed, that is, location information indicating the first location (first location information).

  Further, the traffic management device 150 extracts face feature information (second face feature information) based on the input image input from the imaging terminal installed in the second place. The collation unit 158 performs collation between the facial feature information extracted based on the image captured by the imaging terminal installed in the second place and the registration data stored in the facial feature storage unit 157, Calculate similarity.

FIG. 7 is a flowchart for explaining the matching process performed by the matching unit 158.
The collation unit 158 first sets the value of C, which is information indicating a category, to 0. (Step S31).

  The collation unit 158 is installed at the second place with the registration data Rc (i, j) of the category whose information indicating the category is C among the registration data stored in the face feature storage unit 157. The facial feature information H (x, y) extracted based on the image captured by the imaging terminal is collated to calculate the similarity degree simc (step S32).

In this case, the collation unit 158 calculates the similarity simc by obtaining the correlation as shown in Equation 2.

  The collation unit 158 outputs the similarity degree simc between one piece of registered data and face feature information, for example, to a RAM memory (not shown) (step S33).

  The collation unit 158 adds 1 to the value C, which is information indicating the category (step S34).

  Here, the collation unit 158 determines whether or not the similarity has been calculated for the registration data of all categories stored in the face feature storage unit 157 (step S35).

  When it is determined that the similarity is not calculated for the registration data of all categories stored in the face feature storage unit 157 (NO in step S35), the collation unit 158 loops to step S32.

  When it is determined that the similarity has been calculated for all the category registration data stored in the face feature storage unit 157 (step S35, YES), the matching unit 158 has the highest similarity calculated. (Maximum similarity) and the value C of the category of the registered data used for calculating the similarity are output to the collation determination unit 159 (step S36).

  There are various methods for collating the facial feature information based on the input image with the registered facial feature information, and the method is not limited to the method described above.

  The collation determination unit 159 makes a final determination regarding entrance / exit management based on the result output from the collation unit 158, that is, the maximum similarity and the value of C. The collation determination unit 159 compares the maximum similarity with a predetermined threshold value.

  The collation determination unit 159 determines that “collation is successful” when the maximum similarity is equal to or greater than the threshold. In this case, the traffic management device 150 determines that the person who has passed the first place in the specific area has passed the second place. The collation determination unit 159 outputs the determination result “collation success” and the value of C received together with the similarity to the selection unit 160.

  The collation determination unit 159 determines “collation failure” when the maximum similarity is less than the threshold. That is, the collation unit 158 and the collation determination unit 159 function as a collation unit that collates the face feature information stored in the face feature storage unit 157 and the face feature information extracted by the face feature extraction unit.

  The sorting unit 160 functions as a sorting unit. The selection unit 160 manages the traffic state of the person. The selected face feature storage unit 161 stores registration data of a person staying in a specific area.

  The face feature storage unit 157 copies registration data to the selected face feature storage unit 161 every time it receives face feature information. Note that the face feature storage unit 157 may be configured to copy the registration data of the face feature information received during the predetermined time to the selected face feature storage unit 161 at fixed time intervals.

  That is, the selected face feature storage unit 161 and the face feature storage unit 157 function as a storage unit that stores the face feature information extracted by the face feature extraction unit.

  When the determination unit 160 receives the determination result “verification success” and the value C indicating the category from the verification determination unit 159, the selection unit 160 determines that the person corresponding to the registration data of this category does not exist in the specific area. The selection unit 160 deletes the registration data corresponding to the person determined not to exist from the selection face feature storage unit 161. That is, the selecting unit 160 selects and deletes the registered data of the category corresponding to the value of C received from the matching determination unit 159 together with the determination result “successful matching” from the selected face feature storage unit 161.

  Through the above-described processing, registration data that is no longer needed is deleted from the selected face feature storage unit 161 each time a person passes through the second place. That is, the selected face feature storage unit 161 is configured to store only registration data of a person staying in a specific area.

  The display unit 162 displays the stay status based on the registration data stored in the face feature storage unit 157 and the selected face feature storage unit 161.

  The display unit 162 can display a traffic confirmation screen 180 as shown in FIG. The traffic confirmation screen 180 displays a plurality of windows 181 to 185.

  The window 181 displays the face feature information of the person who has passed through the place A in a list. The window 182 displays a list of facial feature information of persons passing through the location B. The window 183 displays the face feature information of the person who passed the place C in a list.

  The window 184 displays a list of facial feature information of persons staying in a specific area among persons passing through the place A. The window 185 displays a list of face feature information of persons staying in a specific area among persons passing through the place B.

  As shown in FIG. 8, it can be confirmed that one of the two persons who passed the place A is passing the place C. In addition, it can be confirmed that two people who have passed the place B are passing the place C. From here, the person staying in the specific area can be easily confirmed. In addition, it is possible to easily confirm which place the person staying in the specific area has passed.

  Although the display unit 162 has been described as displaying face feature information, the display unit 162 may be configured to display a face image used to extract face feature information.

  The collation unit 163 performs collation between the face feature information extracted based on the image captured by the camera 141 of the window terminal 140 provided at the window and the registration data stored in the selected face feature storage unit 161. To calculate the similarity. That is, the collation unit 163 performs collation between the facial feature information extracted from the input image and the facial feature information corresponding to the person staying in the specific area.

  The collation determination unit 164 determines either “collation success” or “collation failure” by the same processing as the collation determination unit 159, and outputs the determination result to the window terminal 140. When the collation determination unit 164 determines that “collation is successful”, the registration data (ID data, information indicating the traffic location, information indicating the traffic time, and face feature information) used for the verification is determined as “successful verification”. The result is output to the window terminal 140.

  The display unit 142 of the window terminal 140 displays various information based on the information received from the traffic management device 150. That is, the display unit 142 displays information indicating a passage place indicating which place the person imaged by the camera 141 has passed. In addition, the display unit 142 displays information indicating a travel time indicating at which time the person imaged by the camera 141 has traveled.

  Further, the display unit 142 may further display the face feature information received from the matching determination unit 164.

  The face area detection units 151, 152, and 153, the face feature extraction units 154, 155, and 156, the collation unit 158, the collation determination unit 159, the selection unit 160, the collation unit 163, and the collation determination unit 164 are not illustrated. It is realized by a CPU, a RAM, a ROM, and a nonvolatile memory.

  The CPU governs overall control of the traffic management device 150. The CPU functions as various processing units as described above by operating based on the control program and control data stored in the ROM or nonvolatile memory.

  The RAM is a volatile memory that functions as a working memory for the CPU. That is, the RAM functions as a storage unit that temporarily stores data being processed by the CPU. The RAM temporarily stores data received from the input unit. The ROM is a non-volatile memory in which a control program, control data, and the like are stored in advance.

  The nonvolatile memory is configured by a storage medium that can write and rewrite data, such as an EEPROM or an HDD. In the non-volatile memory, a control program and various data are written according to the usage of the traffic management device 150.

  As described above, the traffic management device according to the present embodiment accumulates the first facial feature information acquired at the first location in the specific area in the facial feature storage unit. The face feature storage unit copies the face feature information to the selected face feature storage unit. The traffic management device collates the second facial feature information acquired at the second location with the first facial feature information stored in the facial feature storage unit, and selects matching facial feature information. Delete from the department.

  When performing the process of confirming the passage location of the person, collation is performed based on the facial feature information stored in the selected facial feature storage unit from which the facial feature information of the person who has already passed the second location has been deleted. . For this reason, the amount of data to be processed can be reduced. As a result, it is possible to provide a traffic management device and a traffic management method capable of managing the traffic of a person with a short processing time and with high accuracy.

  In this embodiment, the face feature information is accumulated in the face feature storage unit, the face feature information is copied to the selected face feature storage unit, and the face feature information of the person who has passed through the second place is selected as the face feature information. Although described as a configuration to be deleted from the storage unit, it is not limited to this. The configuration may be such that the face feature information of the person who has passed the second place is deleted from the face feature storage unit without copying the face feature information to the selected face feature storage unit. In this case, the process of confirming the place where the person has passed is performed based on the face feature information stored in the face feature storage unit from which the face feature information of the person who has already passed the second place has been deleted.

  In the present embodiment, the face area detection units 151, 152, and 153 and the face feature extraction units 154, 155, and 156 are described as being included in the traffic management device, but the present invention is not limited to this. The imaging terminals 110, 120, and 130 may each include a face area detection unit and a face feature extraction unit.

Next, a traffic management device and a traffic management method according to the second embodiment of the present invention will be described in detail with reference to the drawings.
FIG. 9 is a block diagram for schematically explaining a configuration example of the traffic management system 2 to which the traffic management device according to the second exemplary embodiment of the present invention is applied. Note that components similar to those in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted.

  As illustrated in FIG. 9, the traffic management system 2 includes imaging terminals 110, 120, and 130, a window terminal 140, and a traffic management device 150. It is assumed that the imaging terminals 110 and 120 are provided at the first location in the specific area, and the imaging terminal 130 and the window terminal 140 are provided at the second location in the specific area.

  The imaging terminals 110, 120, and 130 include ID input units 112, 122, and 132, respectively. The ID input units 112, 122, and 132 are ID input means for causing an operator to input an operation. The ID input units 112, 122, and 132 are configured by, for example, input keys or card readers. The ID input units 112, 122, and 132 read information from input information input by an input key or a card (for example, an IC card) possessed by a pedestrian by a card reader.

  The ID input units 112 and 122 transmit the read information to the face feature storage unit 157 as ID data. The traffic management device 150 stores the facial feature information extracted based on the input image input in the facial feature storage unit 157, and further stores the information received from the ID input unit as ID data shown in FIG.

  The ID data information input by the ID input units 112 and 122 is unique information obtained from, for example, a ticket and an ID card.

  The ID input unit 132 transmits the read information to the selection unit 160. When receiving the ID data from the ID input unit 132, the selection unit 160 collates the received ID data with the ID data stored in the face feature storage unit 157.

  When the matching ID data is found, the selection unit 160 determines that the person corresponding to the registration data including this ID data has already passed the second place in the specific area. The selection unit 160 deletes the registration data corresponding to the person who has been determined to have passed the second place from the selected face feature storage unit 161.

  Note that the selection unit 160 also performs collation based on the similarity of face feature information as shown in the first embodiment. In this case, the selection unit 160 determines an integrated collation result based on the logical sum or logical product of the collation result of the face image and the collation result of the ID data collation.

  As described above, the traffic management device according to the present embodiment acquires the facial feature information and the ID data at the first location in the specific area and accumulates them in the facial feature storage unit. The face feature storage unit copies the face feature information and the ID data to the selected face feature storage unit. The traffic management device collates the face feature information and ID data acquired at the second location with the face feature information and ID data stored in the face feature storage unit, and selects the face feature storage based on the collation result. Delete registration data from the copy.

  As a result, it is possible to provide a traffic management device and a traffic management method capable of managing the traffic of a person with a shorter processing time and with higher accuracy.

  In the present embodiment, the facial feature information and the ID data are acquired at the second location and are compared with the registered data. However, the present invention is not limited to this. The configuration may be such that the camera 131 of the imaging terminal 130 in the second location is omitted and only the acquired ID is used for collation with registered data.

Next, a traffic management device and a traffic management method according to the third embodiment of the present invention will be described in detail with reference to the drawings.
FIG. 10 is a block diagram for schematically explaining a configuration example of the traffic management system 3 to which the traffic management device according to the third exemplary embodiment of the present invention is applied. Note that components similar to those in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted.

  As shown in FIG. 10, the traffic management system 3 includes imaging terminals 110, 120, and 130, a window terminal 140, and a traffic management device 150. It is assumed that the imaging terminals 110 and 120 are provided at the first location in the specific area, and the imaging terminal 130 and the window terminal 140 are provided at the second location in the specific area.

  The window terminal 140 includes a time estimation unit 143. The time estimation unit 143 functions as a time estimation unit. The time estimation unit 143 estimates the travel time at the location A and the travel time at the location B based on the time t when the person appears at the window of the location C. The time estimation unit 143 estimates the travel time according to the operation of the operator who operates the window terminal 140.

  That is, the time estimation unit 143 determines the travel time (estimated travel time) at the location A based on the shortest time t1 at which the user can move from the location A to the location C and the time t when the person appears at the window of the location C. ). In addition, the time estimation unit 143 estimates the travel time at the place B based on the shortest time t2 at which the person can move from the place B to the place C and the time t when the person appears at the window of the place C.

  The time estimation unit 143 transmits information indicating the estimated traffic time at each estimated location to the traffic management device 150.

  The traffic management device 150 includes a search unit 168 and a search face feature storage unit 169. The search unit 168 functions as search means. The search unit 168 receives information indicating the estimated travel time at each location from the time estimation unit 143. The search unit 168 searches the registered data stored in the selected face feature storage unit 161 based on the information indicating the estimated travel time at each location received from the time estimation unit 143.

  That is, the search unit 168 includes transit time information indicating the time when the registered data stored in the selected face feature storage unit 161 has passed through the first location, and the estimated transit time at each location received from the time estimation unit 143. And the registered data passing before the estimated passing time at each place is searched.

  The search unit 168 stores the registered data searched as described above in the search face feature storage unit 169. That is, the search face feature storage unit 169 stores only registration data that can be reached at the location C where the window terminal 140 is installed.

  Based on the registration data stored in the face feature storage unit 157, the selected face feature storage unit 161, and the search face feature storage unit 169, the display unit 162 stores the stay situation and registration data that can reach a specific place. Display status.

  The display unit 162 can display a traffic confirmation screen 180 as shown in FIG. The traffic confirmation screen 180 displays a plurality of windows 181 to 185. Further, the traffic confirmation screen 180 displays a frame 186 and a frame 187 for specifying and displaying the registration data searched by the search unit 168.

  As described above, the window 184 and the window 185 are windows that display a list of facial feature information of persons who have not been confirmed to pass in the second place among the persons who have passed the place A or B.

  The frame 186 identifies and displays the facial feature information of the registered data corresponding to the search condition of the search unit 168 among the facial feature information displayed in the window 184. That is, here, the frame 186 identifies and displays the facial feature information of the registration data that passes through the location A and can reach the location C.

  The frame 187 identifies and displays the facial feature information of the registered data corresponding to the search condition of the search unit 168 among the facial feature information displayed in the window 185. That is, here, the frame 187 identifies and displays the facial feature information of the registration data that passes through the location B and can reach the location C.

  As shown in FIG. 11, it can be confirmed that two of the four people who have passed through the place A are staying in the specific area, and one of them may have reached the place C. In addition, it can be confirmed that three of the four people passing through the place B are staying in the specific area, and two of them may have reached the place C.

  Although the display unit 162 has been described as displaying face feature information, the display unit 162 may be configured to display a face image used to extract face feature information.

  The collation unit 163 performs collation between the face feature information extracted based on the image captured by the camera 141 of the counter terminal 140 provided at the counter and the registration data stored in the search face feature storage unit 169. To calculate the similarity. That is, the collation unit 163 performs collation between the facial feature information extracted from the input image and the facial feature information of a person who can reach the place C.

  As described above, the traffic management device according to the present embodiment searches for registration data that meets a specific condition from registration data stored in the selected face feature storage unit. The traffic management device stores the searched registration data in the search face feature storage unit. When confirming the passage location of a person based on an image captured by the camera of the window terminal, the traffic management device uses the face feature information stored in the search face feature storage unit for collation. For this reason, it is possible to narrow down registered data to be collated.

  As a result, it is possible to provide a traffic management device and a traffic management method capable of managing the traffic of a person with a shorter processing time and with high accuracy.

Next, a traffic management device and a traffic management method according to the fourth embodiment of the present invention will be described in detail with reference to the drawings.
FIG. 12 is a block diagram for schematically explaining a configuration example of the traffic management system 4 to which the traffic management device according to the fourth exemplary embodiment of the present invention is applied. In addition, the same referential mark is attached | subjected to the structure similar to other embodiment, and the detailed description is abbreviate | omitted.

  As illustrated in FIG. 12, the traffic management system 4 includes imaging terminals 110, 120, and 130, a window terminal 140, and a traffic management device 150. It is assumed that the imaging terminals 110 and 120 are provided at the first location in the specific area, and the imaging terminal 130 and the window terminal 140 are provided at the second location in the specific area.

  The window terminal 140 includes a location information input unit 144. The location information input unit 144 functions as location information input means. The place information input unit 144 is an input unit for inputting information (place information) indicating a place to be reported by a person who appears at the window of the place C. The location information input unit 144 is realized by a keyboard or a touch panel formed integrally with the display unit 142.

  The window terminal 140 transmits location information indicating the location input by the location information input unit 144 to the traffic management device 150.

  The traffic management device 150 includes a search unit 168 and a search face feature storage unit 169. The search unit 168 receives the location information input by the location information input unit 144. The search unit 168 functions as search means. The search unit 168 searches the registered data stored in the selected face feature storage unit 161 based on the location information input by the location information input unit 144.

  That is, the search unit 168 includes information (first location information) indicating the passage location of registered data stored in the selected face feature storage unit 161 and information (first location information) input by the location information input unit 144. 3) and search for matching registration data.

  The search unit 168 stores the registered data searched as described above in the search face feature storage unit 169. That is, the search face feature storage unit 169 stores only the registration data of the person who has passed the place indicated by the third place information input by the place information input unit 144.

  The display unit 162 displays the stay status based on the registered data stored in the face feature storage unit 157, the selected face feature storage unit 161, and the search face feature storage unit 169.

  The display unit 162 can display a traffic confirmation screen 180 as shown in FIG. The traffic confirmation screen 180 displays a plurality of windows 181, 183, and 184. As shown in FIG. 13, it can be confirmed that two of the four people passing through the place A are staying in the specific area.

  Although the display unit 162 has been described as displaying face feature information, the display unit 162 may be configured to display a face image used to extract face feature information.

  The collation unit 163 performs collation between the face feature information extracted based on the image captured by the camera 141 of the counter terminal 140 provided at the counter and the registration data stored in the search face feature storage unit 169. To calculate the similarity. That is, the collation unit 163 performs collation between the face feature information extracted from the input image and the face feature information extracted from the image captured at the location input by the location information input unit 144.

  As described above, the traffic management device according to the present embodiment searches for registration data that meets a specific condition from registration data stored in the selected face feature storage unit. The traffic management device stores the searched registration data in the search face feature storage unit. When confirming the passage location of a person based on an image captured by the camera of the window terminal, the traffic management device uses the face feature information stored in the search face feature storage unit for collation. For this reason, it is possible to narrow down registered data to be collated.

  As a result, it is possible to provide a traffic management device and a traffic management method capable of managing the traffic of a person with a shorter processing time and with high accuracy.

Next, a traffic management device and a traffic management method according to the fifth embodiment of the present invention will be described in detail with reference to the drawings.
FIG. 14 is a block diagram for schematically explaining a configuration example of the traffic management system 5 to which the traffic management device according to the fifth exemplary embodiment of the present invention is applied. In addition, the same referential mark is attached | subjected to the structure similar to other embodiment, and the detailed description is abbreviate | omitted.

  As illustrated in FIG. 14, the traffic management system 5 includes imaging terminals 110 and 120, a window terminal 140, and a traffic management device 150. It is assumed that the imaging terminals 110 and 120 are provided at the first location in the specific area, and the window terminal 140 is provided at the second location in the specific area.

  The traffic management device 150 according to the present embodiment includes the number of face feature storage units 157 corresponding to the number of imaging terminals installed in the first place. As shown in FIG. 14, the face feature storage unit 157 has an imaging terminal at a location corresponding to itself. That is, each of the plurality of face feature storage units 157 stores only the face feature information acquired at a predetermined location.

  In addition, the traffic management device 150 according to the present embodiment includes a plurality of matching units 158 so as to correspond to the plurality of face feature storage units 157. The collation unit 158 collates the registration data stored in the face feature storage unit 157 corresponding to the collation unit 158 with the face feature information extracted based on the image captured by the camera 141 of the window terminal 140. That is, the collation unit 158 calculates the similarity for each registered data stored in the corresponding face feature storage unit 157.

Further, the traffic management device 150 according to the present embodiment includes a transmission control unit 170 as transmission control means. The transmission control unit 170 receives the location information input by the location information input unit 144. Based on the location information input by the location information input unit 144, the transmission control unit 170 includes a collation unit 158 that transmits the facial feature information extracted based on the image captured by the camera 141 of the window terminal 140. Select and transmit The display unit 142 of the window terminal 140 displays various information based on the information received from the traffic management device 150. The display unit 142 according to the present embodiment displays the registration data used for collation by the collation unit 158 in order from the highest similarity value.

  FIG. 15 is a block diagram for schematically explaining another configuration example of the traffic management system to which the traffic management device according to the fifth exemplary embodiment of the present invention is applied. In addition, the same referential mark is attached | subjected to the structure similar to other embodiment, and the detailed description is abbreviate | omitted.

  As illustrated in FIG. 15, the traffic management system 6 further includes a plurality of sorting units 160 and a plurality of selected face feature storage units 161 in the traffic management device 150 illustrated in FIG. 14. Each imaging terminal 110, 120, and 130 includes an ID input unit 112, 122, and 132, respectively.

  The face feature storage unit 157 stores ID data input from the imaging terminal installed in the first location in association with the face feature information. The registration data stored in the face feature storage unit 157 is sequentially copied to the corresponding selected face feature storage unit 161.

  As shown in FIG. 15, a plurality of sorting units 160 are provided so as to correspond to a plurality of face feature storage units 157 and a plurality of selected face feature storage units 161. The sorting unit 160 sorts the registration data of the selected face feature storage unit 161 based on the ID data input to the ID input unit 132 of the imaging terminal 130 installed at the second location.

  That is, the sorting unit 160 receives ID data from the imaging terminal 130 installed at the second location. The selection unit 160 compares the received ID data with the ID data of the registration data stored in the selection face feature storage unit 161. When the matching ID data is found, the sorting unit 160 determines that the person corresponding to the registration data including the ID data has passed through the second place in the specific area. The selection unit 160 deletes from the selection face feature storage unit 161 the registration data that has been passed through the second place, that is, determined not to exist in the specific area.

  A plurality of matching units 158 are provided so as to correspond to the plurality of selected face feature storage units 161. The collation unit 158 collates the registration data stored in the selected face feature storage unit 161 corresponding to the collation unit 158 with the face feature information extracted based on the image captured by the camera 141 of the window terminal 140. That is, the collation unit 158 calculates the similarity for each registered data stored in the corresponding selected face feature storage unit 161.

  The display unit 142 of the window terminal 140 displays various information based on the information received from the traffic management device 150. The display unit 142 according to the present embodiment displays the registration data used for collation by the collation unit 158 in order from the highest similarity value.

  As described above, the traffic management device 150 according to the present embodiment stores the facial features in different storage units for each place. When confirming the traffic location, the traffic management device 150 can shorten the processing time by performing verification using the registration data stored in the storage unit corresponding to the requested location.

  As described above, the traffic management device according to the present embodiment stores facial features in a separate storage unit for each location. When confirming a traffic location, the traffic management device performs verification using registration data stored in a storage unit corresponding to the requested location. For this reason, it is possible to narrow down registered data to be collated.

  As a result, it is possible to provide a traffic management device and a traffic management method capable of managing the traffic of a person with a shorter processing time and with high accuracy.

In addition, this invention is not limited to the said embodiment as it is, It can implement by changing a component in the range which does not deviate from the summary in an implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine the component covering different embodiment suitably.
The invention described in the scope of the claims at the beginning of the present application is added below.
[1]
First image input means for inputting an image including at least a face of a person passing through a first location in a specific area;
First face detection means for detecting a face image from an image input from the first image input means;
First face feature extraction means for extracting first face feature information from the face image detected by the first face detection means;
Storage means for storing first facial feature information extracted by the first facial feature extraction means;
Screening means for selecting first facial feature information stored in the storage means;
A traffic management device comprising:
[2]
A second image input means for inputting an image including at least a face of a person passing through a second location different from the first location in the specific area;
Second face detection means for detecting a face image from the image input from the second image input means;
Second face feature extraction means for extracting second face feature information from the face image detected by the second face detection means;
Collation means for collating the first facial feature information stored by the storage means and the second facial feature information extracted by the second facial feature extraction means;
Further comprising
The selecting means selects the first facial feature information stored in the storage means based on the matching result of the matching means;
The traffic management device according to [1], characterized in that:
[3]
The traffic management apparatus according to [2], wherein the selection unit deletes the first face feature information that matches the second face feature information in the collation by the collation unit from the storage unit.
[4]
First ID input means for inputting first ID data by the person;
Second ID input means for inputting second ID data by a person passing through a second place different from the first place in the specific area;
Further comprising
The storage means stores the first facial feature information extracted by the first facial feature extraction means and the first ID data inputted by the first ID input means in association with each other;
The selecting means deletes the face feature information associated with the first ID data that matches the second ID data input by the second ID input means from the storage means. The traffic management device according to [1].
[5]
Location information input means for inputting third location information;
Search means for searching for first facial feature information stored in the storage means based on third place information input by the place information input means;
Further comprising
The first image input means inputs an image including at least a face of a person passing through a plurality of places to the specific area,
The storage means stores the first facial feature information extracted by the first facial feature extraction means in association with first location information indicating a location where the person has passed,
The search means searches the storage means for facial feature information associated with the first place information that matches the third place information input by the place information input means [1. ]. The traffic management device described in the above.
[6]
Time estimation means for estimating an estimated transit time when the person has passed through the first location based on current location and second location information indicating a second location different from the first location;
Search means for searching for first facial feature information stored in the storage means based on the estimated travel time estimated by the time estimation means;
Further comprising
The storage means stores the first facial feature information extracted by the first facial feature extraction means in association with passage time information indicating a time when the person has passed the first place,
The search means searches the storage means for face feature information associated with the transit time information of the first place indicating a time before the estimated travel time estimated by the time estimation means. The traffic management device according to [1].
[7]
A plurality of first image input means for inputting an image including at least a face of a person passing through a plurality of locations in a specific area;
First face detection means for detecting a face image from images input from the plurality of first image input means;
First face feature extraction means for extracting first face feature information from the face image detected by the first face detection means;
A plurality of storage means for storing the first facial feature information extracted by the first facial feature extraction means for each location;
A second image input means for inputting an image including at least a face of a person passing through another place of the specific area;
Second face detection means for detecting a face image from the image input from the second image input means;
Second face feature extraction means for extracting second face feature information from the face image detected by the second face detection means;
Corresponding to the plurality of storage means, the first face feature information stored by the corresponding storage means and the second face feature information extracted by the second face feature extraction means are collated. A plurality of matching means;
A place information input means for inputting place information indicating the place;
Based on the location information input by the location information input unit, the verification unit that transmits the second facial feature information extracted by the second facial feature extraction unit is selected from the plurality of verification units. Transmission control means for transmitting
A traffic management device comprising:
[8]
Enter an image that includes at least the face of a person passing through a specific location in a specific area,
A face image is detected from the input image;
Extracting facial feature information from the detected face image;
Storing the facial feature information;
Selecting the stored facial feature information;
A traffic management method characterized by that.

  DESCRIPTION OF SYMBOLS 1 ... Traffic management system, 2 ... Traffic management system, 3 ... Traffic management system, 4 ... Traffic management system, 5 ... Traffic management system, 6 ... Traffic management system, 110 ... Imaging terminal, 111 ... Camera, 112 ... ID input part , 120 ... imaging terminal, 121 ... camera, 122 ... ID input section, 130 ... imaging terminal, 131 ... camera, 132 ... ID input section, 140 ... window terminal, 141 ... camera, 142 ... display section, 143 ... time estimation section , 144: Location information input unit, 150: Traffic management device, 151 ... Face region detection unit, 152 ... Face region detection unit, 153 ... Face region detection unit, 154 ... Face feature extraction unit, 155 ... Face feature extraction unit, 156 ... Face feature extraction unit, 157 ... Face feature storage unit, 158 ... Verification unit, 159 ... Verification determination unit, 160 ... Selection unit, 161 ... Selected face feature storage unit, 162 ... Display unit, 163 ... Engaging portion, 164 ... verification determining unit, 165 ... interface, 166 ... interface, 167 ... interface, 168 ... search section, 169 ... search facial feature storage unit.

Claims (2)

First image input means for inputting an image including at least a face of a person passing through a first location in a specific area;
First face detection means for detecting a face image from an image input from the first image input means;
First face feature extraction means for extracting first face feature information from the face image detected by the first face detection means;
Storage means for storing first facial feature information extracted by the first facial feature extraction means;
Screening means for selecting first facial feature information stored in the storage means;
Time estimation means for estimating an estimated transit time when the person has passed through the first location based on current location and second location information indicating a second location different from the first location;
Search means for searching for first facial feature information stored in the storage means based on the estimated travel time estimated by the time estimation means;
Comprising
The storage means stores the first facial feature information extracted by the first facial feature extraction means in association with passage time information indicating a time when the person has passed the first place,
The search means searches the storage means for face feature information associated with the transit time information of the first place indicating a time before the estimated travel time estimated by the time estimation means. Traffic management device to do. A plurality of first image input means for inputting an image including at least a face of a person passing through a plurality of locations in a specific area;
First face detection means for detecting a face image from images input from the plurality of first image input means;
First face feature extraction means for extracting first face feature information from the face image detected by the first face detection means;
A plurality of storage means for storing the first facial feature information extracted by the first facial feature extraction means for each location;
A second image input means for inputting an image including at least a face of a person passing through another place of the specific area;
Second face detection means for detecting a face image from the image input from the second image input means;
Second face feature extraction means for extracting second face feature information from the face image detected by the second face detection means;
Corresponding to the plurality of storage means, the first face feature information stored by the corresponding storage means and the second face feature information extracted by the second face feature extraction means are collated. A plurality of matching means;
A place information input means for inputting place information indicating the place;
Based on the location information input by the location information input unit, the verification unit that transmits the second facial feature information extracted by the second facial feature extraction unit is selected from the plurality of verification units. Transmission control means for transmitting
A traffic management device comprising:

Images