JP7227475B2 - Video search program, video search method, and video search device - Google Patents

Description

The present invention relates to a video search program, video search method, and video search device.

Recently, the number of vehicles equipped with a drive recorder is increasing. For example, when an event such as an accident occurs, a video is recorded in the drive recorder, and by displaying the video, it is possible to confirm or prove the occurrence of the event.

A drive recorder may record, for example, position information acquired using a GPS (Global Positioning System) together with image data of an image captured by a camera. In the drive recorder, for example, position information is used as a search key to search for images shot at a position that matches the position information.

For example, there are the following technologies. That is, based on the search condition indicating the geographical range, it was determined that the image data for which the range of the public location information of the image data is included in the geographical range should be included in the search results, and the output of the image data should be restricted. In this case, there is a server device that limits the output of the corresponding image data.

According to this technology, it is possible to prevent the shooting location of an image whose position information is hidden from being specified from the position information added to the similar image.

JP 2015-49762 A

A drive recorder records video data in a single file format in memory at short time intervals, such as every minute. By recording several hundred files, several hours of video data can be stored in memory. may be recorded.

Therefore, when searching for a desired video in the drive recorder, if the video is searched in order from the oldest video, hundreds of files are searched, which may take a long time.

In the technique for limiting the output of the image data described above, the image data is searched without considering the search time. Therefore, with the above-described technique, it may take time to search for image data.

Therefore, one disclosure is to provide a video search program, a video search method, and a video search device that speed up search processing.

One disclosure is a video search program executed by a computer of a video search device having a memory in which position information and video data acquired at a position represented by the position information are recorded, wherein the position of the search start position Based on the information and the position information of the specific position, the video data from the search start position to the primary mesh containing the specific position is skipped from the video data at the search start position, and then skipped. Based on the position information of the acquisition position where the video data was acquired and the position information of the specific position, (n-1) (n is n An integer that satisfies ≧2) skips the video data up to the next mesh from the video data at the acquisition position, repeats this up to the n-th mesh having a size smaller than the (n−1)th-th mesh, and position information at the specific position A video search program that causes a computer to execute a process of searching from the memory for video data after skipping up to the n-th order mesh to video data at the specific position and outputting the video data at the specific position based on It is in.

According to one disclosure, it is possible to speed up search processing.

FIG. 1 is a diagram showing a configuration example of a communication system. FIG. 2 is a diagram showing a configuration example of a video search device. FIG. 3A is a diagram showing the size of the mesh, and FIG. 3B is a diagram showing an example of the relationship between the meshes. FIG. 4 is a diagram showing an example of a primary mesh. FIG. 5A is a diagram showing an example of a secondary mesh, and FIG. 5B is a diagram showing an example of a tertiary mesh. FIG. 6 is a flow chart showing an operation example. FIG. 7 is a diagram showing an example of a primary mesh. FIG. 8 is a diagram showing an example of a primary mesh. FIG. 9 is a diagram showing an example of a primary mesh. FIG. 10 is a diagram showing an example of a primary mesh. FIG. 11 is a diagram showing an example of a primary mesh. FIG. 12 is a diagram showing an example of file data. FIG. 13 is a diagram showing an example of a primary mesh. FIG. 14 is a diagram showing an example of a primary mesh. FIG. 15A is a diagram showing an example of a secondary mesh, and FIG. 15B is a diagram showing an example of a tertiary mesh.

EMBODIMENT OF THE INVENTION Hereinafter, the form for implementing this invention is demonstrated. It should be noted that the following examples do not limit the technology disclosed. Further, each embodiment can be appropriately combined within a range in which the processing contents are not inconsistent.

[First embodiment]
<Configuration example of communication system>
FIG. 1 is a diagram showing a configuration example of a communication system 10. As shown in FIG.

The communication system 10 includes video search devices 100 - 1 to 100 - 4 , a base station device (hereinafter sometimes referred to as “base station”) 300 , and a center 400 .

Each video retrieval device 100-1 to 100-4 is mounted on each vehicle 200-1 to 200-4. Video retrieval devices 100-1 to 100-4 are, for example, drive recorders, which are in-vehicle devices that can be mounted on vehicles 200-1 to 200-4. Video search devices 100-1 to 100-4 receive position information transmitted from center 400 via base station 300, and record video data having position information that matches the received position information in internal memory. Search from the video data that has been recorded. Then, video search devices 100-1 to 100-4 transmit the searched video data to center 400 via base station 300 as search results. Image retrieval devices 100-1 to 100-4 record image data of surrounding images in memory while vehicles 200-1 to 200-4 are moving. Video search devices 100-1 to 100-4 are also, for example, video recording devices.

For example, when an event such as sudden braking of vehicles 200-1 to 200-4 equipped with image retrieval devices 100-1 to 100-4 occurs, image retrieval devices 100-1 to 100-4 It is also possible to transmit the video data of the video at that time to the center 400 .

Base station 300, for example, converts a wired signal transmitted from center 400 into a wireless signal, transmits the signal to image retrieval devices 100-1 to 100-4, and transmits the signal transmitted from image retrieval devices 100-1 to 100-4. It is a communication device that converts a wireless signal into a wired signal and transmits it to the center 400 . Base station 300 serves as an interface between center 400 and video search devices 100-1 to 100-4, for example, for converting between wired and wireless signals.

The center 400 is, for example, a personal computer, a server device, or the like, and is also a management device that manages position information. For example, if there is a falling object on the road, the center 400 sends position information representing the position of the falling object to the base station in order to investigate when the falling object existed and how it fell on the road. It is transmitted to each video retrieval device 100-1 to 100-4 via the station 300. FIG. The center 400 receives the video data of the video corresponding to the transmitted position information from each of the video search devices 100-1 to 100-4, and the investigator checks the video to investigate the cause. becomes.

Also, for example, when an event occurs in the vehicles 200-1 to 200-4, the center 400 receives the video data of the video at that time from each of the video retrieval devices 100-1 to 100-4, can also check the video.

Note that the example of FIG. 1 shows an example of four video search devices 100-1 to 100-4, but the number of video search devices may be one, two to three, or five or more. good too. Below, video search devices 100-1 to 100-4 may be referred to as video search device 100 unless otherwise specified. Vehicles 200 - 1 to 200 - 4 may also be referred to as vehicle 200 .

<Configuration example of video search device>
FIG. 2 is a diagram showing a configuration example of the video search device 100. As shown in FIG.

The video search device 100 includes a camera 101 , a sensor 102 , a GPS receiver 103 , a recording unit 104 , a memory 105 , a video search unit 106 and a radio unit 107 .

Camera 101 captures an image around vehicle 200 , acquires image data, and outputs the image data to recording unit 104 . Therefore, the camera 101 is provided with an image sensor inside, and converts it into an electric signal according to the intensity of light, so that it is possible to acquire, for example, RGB (Red Green Blue) video data. The camera 101 can acquire video data at a frame rate of F video frames per second (for example, F=14 fps (frame per second)).

The sensor 102 is, for example, a speed sensor, an acceleration sensor, a gyro sensor, or the like. The sensor 102 detects, for example, speed information, acceleration information, angular velocity information, etc. of the vehicle 200 on which the video retrieval device 100 is mounted, and outputs the detected information to the recording unit 104 . Sensor 102 may also include, for example, a timer. In this case, sensor 102 outputs time information detected by the timer to recording unit 104 .

The GPS receiving unit 103 receives GPS signals transmitted from artificial satellites, and acquires position information representing the position of the vehicle 200 (or the video search device 100) based on the received GPS signals. GPS receiving section 103 outputs the acquired position information to recording section 104 . In the first embodiment, position information is represented by latitude and longitude, both of which are represented by decimal numbers. An example of position information is (latitude, longitude)=(35.673033, 139.633546).

The recording unit 104 records (or stores) video data, position information, time information, speed information of the vehicle 200 and the like in the memory 105 . The recording unit 104 acquires position information from the GPS receiving unit 103 when the image is captured by the camera 101 , and acquires time information and speed information of the vehicle 200 from the sensor 102 when the image is captured by the camera 101 . Then, the recording unit 104 records the acquired video data, position information, time information, speed information, etc. in the memory 105 . For example, the recording unit 104 stores position information in the memory 105 for each video frame.

At this time, the recording unit 104 collects the video data, the position information, and the like into one file format data (hereinafter sometimes referred to as a “file”) in a certain time unit (for example, one minute), and stores the data in the memory. 105. The recording unit 104 records tens to hundreds of files in the memory 105, thereby recording video data for a predetermined time period such as one hour or two hours in the memory 105. FIG.

FIG. 12 is a diagram showing an example of files recorded in the memory 105. As shown in FIG. Details will be described in an operation example, but as shown in FIG. 12, a file for a predetermined time can be recorded in the memory 105.

Returning to FIG. 2, the memory 105 is, for example, a storage medium such as an HDD (Hard Disk Drive) or a semiconductor memory. The memory 105 stores video data, position information, time information, speed information of the vehicle 200, and the like.

Video search unit 106 searches memory 105 for video data having position information that matches the position information of the specific position transmitted from center 400 via radio unit 107 or the like. At that time, the video search unit 106 searches for video data using mesh. A mesh is, for example, a rectangular area that divides an area on a map based on latitude and longitude. The meshes are, for example, 1st-order to 3rd-order meshes, with the 1st-order mesh being the largest and the 3rd-order mesh being the smallest. Meshes are classified by the Ministry of Internal Affairs and Communications and may be used for various statistics. Details of processing in the mesh and video search unit 106 will be described later.

The video search unit 106 outputs the searched video data (or a file containing the video data) to the radio unit 107, for example.

Radio section 107 receives a radio signal transmitted from base station 300 , performs frequency conversion processing and demodulation processing on the received radio signal, and extracts position information transmitted from center 400 . Radio section 107 outputs the extracted position information to video search section 106 . Further, radio section 107 performs modulation processing, frequency conversion processing, and the like on video data output from video search section 106, converts the radio signal into a radio signal, and transmits the converted radio signal to base station 300. do.

Note that, as shown in FIG. 2, the recording unit 104 and the video search unit 106 may be, for example, a CPU (Central Processing Unit) 110 . By reading a program from the memory 105 and executing the read program, the CPU 110 can implement the functions of the recording unit 104 and the video search unit 106 . In this case, a program for executing such processing can be transmitted (or downloaded) from center 400 and stored in memory 105 by CPU 110 via base station 300 and radio section 107 . A processor or controller such as an MPU (Micro Processing Unit), a DSP (Digital Signal Processor), or an FPGA (Field-Programmable Gate Array) may be used instead of the CPU.

<About mesh>
FIG. 3(A) is a diagram showing an example of sizes from the primary mesh to the tertiary mesh. This size is also determined by, for example, the Ministry of Internal Affairs and Communications.

The primary mesh (or primary mesh, hereinafter referred to as "primary mesh") has a mesh size of "40 minutes" in the latitudinal direction and a mesh size of "1 degree" in the longitudinal direction. be. Assuming that the length of one second in the latitudinal direction is "30.8 m" and the length of one second in the longitudinal direction is "25.2 m", the size of the primary mesh is approximately 73.9 km in the latitudinal direction and approx. The direction is about 90.7 km, and the length of one side is about 80 km as shown in FIG. 3(A).

The secondary mesh (or secondary mesh, hereinafter referred to as "secondary mesh") has a size of "5 minutes" (= about 9.2 km) in the latitudinal direction and "7 minutes and 30 seconds" in the longitude direction. ” (=about 11.3 km), and the length of one side is about 10 km.

The tertiary mesh (or tertiary mesh, hereinafter referred to as "tertiary mesh") has a size of "30 seconds" (=about 0.92 km) in the latitudinal direction and "45 seconds" (=about 0.92 km) in the longitudinal direction. = about 1.13 km), and the length of one side is about 1 km.

FIG. 3B is a diagram showing an example of the relationship between meshes from the primary mesh to the tertiary mesh. As shown in FIG. 3B, the primary mesh includes the secondary mesh, and the secondary mesh includes the tertiary mesh.

Such meshes are arranged at predetermined positions on a map of Japan, for example, by the Ministry of Internal Affairs and Communications.

4 shows an example of a primary mesh, FIG. 5A shows an example of a secondary mesh, and FIG. 5B shows an example of a tertiary mesh.

A unique mesh code is assigned to each mesh from the first mesh to the third mesh so that it can be distinguished from other meshes. A mesh code for a location can be calculated using the latitude and longitude of the location. The calculation formula will be explained in an operation example.

In the first embodiment, the image retrieval unit 106 retrieves images using such meshes.

<Operation example>
FIG. 6 is a flow chart showing an operation example in the first embodiment. In the first embodiment, first, processing is performed using the primary mesh, then processing is performed using the secondary mesh, and finally processing is performed using the tertiary mesh. . The order of the mesh is changed in S21. The steps will be explained in order below.

<1. Processing using primary mesh>
When starting the process (S10), the video search unit 106 sets the search position Px as the oldest data position of the recorded video (S11).

FIG. 7 expresses where the video data recorded in the memory 105 is recorded on the map. A square frame in FIG. 7 represents one mesh in the primary mesh. As described above, positional information is also recorded in the video data stored in the memory 105, and FIG. 7 also expresses the positional information on a map. FIG. 7 shows an example in which video data from a position P1 in the vicinity of Numazu City to Shinjuku Park Tower is recorded in the memory 105 as video data of recorded video.

In this case, the oldest video data recorded in the memory 105 is the video data at the position P1. Therefore, the video search unit 106 sets the position P1 as the oldest data position as the search position Px. Below, the position P1 may be referred to as a search position (or search start position) P1. Also, the video search unit 106 sets x to "1". After that, the search position Px is changed in order from P1. In this case, for example, P1 may be the search start position, and the subsequent Px may be the search positions.

It is assumed that the video search device 100 has also received the position information of the specific position P0 from the center 400 before performing this process. FIG. 7 also shows an example of the specific position P0 represented on the map. In the first embodiment, PO=(latitude, longitude)=(35.673033, 139.6333546).

FIG. 12 is a diagram showing an example of file data (hereinafter sometimes referred to as "file") recorded in the memory 105. As shown in FIG. For example, each file records time information (t1 to t0) together with video data. Therefore, the video search unit 106 reads out from the memory 105 the position information (for example, position information #1=P1) in the file containing the oldest time information among the time information recorded in the memory 105, and reads the position information. The information may be the search position P1.

As described above, the position information contained in the file is the position information obtained by the GPS receiving unit 103 when the camera 101 shot an image, and both the latitude and longitude are expressed in decimal numbers. .

Note that the video search unit 106 sets the degree of the mesh to “1” and stores the set degree “1” in the memory 105 .

Returning to FIG. 6, next, the video search unit 106 calculates a regional mesh code (hereinafter sometimes referred to as "mesh code") M0 at the specific position P0.

The video search unit 106, for example, calculates a mesh code of a primary mesh (hereinafter sometimes referred to as a “primary mesh code”) at a specific position (latitude Y, longitude X) P0 using the following formula. .

[Number 1]
Primary mesh code=M1y×100+M1y (1)
[Number 2]
M1y=INT (Y×(60.0÷40.0)) (2)
[Number 3]
M1x=INT(X-100) (3)

INT( ), for example, represents a function that returns the largest integer that does not exceed the number in ( ).

For example, when P0 = (latitude, longitude) = (Y, X) = (35.673033, 139.6333546), using equations (2) and (3) gives M1y = 53 and M1x = 39, The primary mesh code at the specific position P0 is "5339".

For example, the video search unit 106 reads out the formulas (1) to (3) stored in the memory 105, and converts the latitude and longitude of the specific position PO received from the center 400 into the formulas (2) and (3). , and the result is substituted into equation (1) to calculate the primary mesh code at the specific position P0.

FIG. 8 is a diagram showing an example of the primary mesh code M0 (=5339) including the specific position P0.

Returning to FIG. 6, next, the video search unit 106 sets x to "1" and calculates the primary mesh code M1 at the search position P1 (S13). Assuming that P1=(Y, X)=(35.107108, 138.866221), the video search unit 106 calculates "5238" using equations (1) to (3) as in S12. The primary mesh code M1 at the search position P1 is "5238". FIG. 8 shows an example of the primary mesh code M1 at the search position P1.

Returning to FIG. 6, next, the video search unit 106 determines whether or not M0=M1 (S14). Then, when M0=M1 is not true (No in S14), the video search unit 106 calculates the distance S1 from the search position P1 to the primary mesh including the specific position P0 (S18).

In S14, the video search unit 106 determines, for example, whether the search position P1 is included in the same primary mesh as the specific position P0. In the example of FIG. 8, since MO=5339 and M1=5238, the video search unit 106 determines No in S14. Therefore, in the example of FIG. 8, the video search unit 106 calculates the distance S1 from the search position P1 to the primary mesh of the primary mesh code M0.

FIG. 9 is a diagram showing an example of the distance S1. For example, the video search unit 106 calculates the distance S1 by the following processing.

That is, the video search unit 106 first reads from the memory 105 the latitudes and longitudes of the vertex positions M01 to M04 of the primary mesh at the specific position P0. Then, the video search unit 106 calculates each distance from the search position P1 to each of the vertex positions M01 to M04 based on the longitude and latitude of each of P1 and M01 to M04. Calculate as In the example of FIG. 9, the distance from the search position P1 to the vertex position M03 is the shortest, so the video search unit 106 calculates that distance as S1.

The calculation of the distance S1 described above is an example. For example, the image search unit 106 calculates the latitude and longitude of the primary mesh code M0 from the primary mesh code M0, and calculates the distance based on the latitude and longitude of the search position P1 and the calculated latitude and longitude. S1 may be calculated.

In either case, the video search unit 106 uses latitude and longitude when calculating the distance. For example, if the decimal latitude and longitude of the search position P1 is (YP1, XP1), and the decimal latitude and longitude of the vertex position M03 is (YM03, XM03), then the video search unit 106 uses the following formula ( 4) makes it possible to calculate the distance S1.

[Number 4]
Distance S1=√{(YM03−YP1)×(30.8/3600)} ² +{(XM03−XP1)×(25.2/3600)} ² (4)

For example, the image search unit 106 reads out the formula (4) stored in the memory 105, and substitutes the latitude and longitude of the search position P1 and the latitude and longitude of the vertex position M03 into the formula (4) to obtain the distance It is possible to calculate S1. For vertex positions other than vertex position M03, the distance S1 can be obtained by substituting the latitude and longitude of the other vertex position in place of vertex position M03 in equation (4).

Returning to FIG. 6, next, the video search unit 106 calculates the required time T1 required for moving the distance S1 (S19). For example, assume that the speed of the vehicle 200 is Vmax (for example, 180 km/h), the image search unit 106 reads Vmax stored in the memory 105, and divides the distance S1 calculated in S18 by the speed Vmax. Thus, the required time T1 is calculated.

FIG. 10 shows the position of the vehicle 200 after the required time T1 has elapsed, assuming that the vehicle 200 departed from the search position P1 travels on the road shown in FIG. 10 at a speed Vmax. In FIG. 10, the vehicle 200 is positioned at the circled position at T1.

Returning to FIG. 6, next, the video search unit 106 changes the recorded video to the position of the video data skipped for the required time T1 (S20). For example, in the example of FIG. 10, the video data stored in the memory 105 is skipped from the video data at the search position P1 to the video data at the position (marked with a circle T1) after the required time T1 has elapsed.

FIG. 12 is a diagram showing an example of a video data file stored in the memory 105. As shown in FIG. In the example shown in FIG. 12, the video search unit 106 retrieves video data for the required time T1 (from video data #1 to video data #(m1- 1) is skipped. After skipping, the video search unit 106 accesses the file #m1 containing the video data #m1 in the memory 105 . The position on the map represented by the position information #m1 included in this file #m1 is the position marked with a circle T1 in FIG.

FIG. 11 is a diagram showing an example of skipping on a map. As shown in FIG. 11, the circled position T1 in FIG. 10 is changed as a new search position P2. For example, as shown in FIG. 12, the video search unit 106 extracts the position information #m1 from the file #m1 after the required time T1 has elapsed, increments x, and adds the extracted position information #m1 to the search position P2. change to

Returning to FIG. 6, the process moves from S20 to S13, and the video search unit 106 repeats the above-described process. In the example of FIG. 11, for example, the video search unit 106 performs the following processes.

That is, the video search unit 106 calculates the primary mesh code M2 of the search position P2 (S13). In this case, the video search unit 106 substitutes the position information #m1 (latitude and longitude) of the search position P2 read out from the memory 105 into the formulas (1) to (3) to obtain 1 of the search position P2. Calculate the next mesh code "5239". Since the mesh code "5239" of the primary mesh including the search position P2 does not match the primary mesh code "5339" of the specific position P0 (No in S14), the video search unit 106 searches the specific position P0 from the search position P2. A distance S2 to is calculated.

FIG. 13 is a diagram showing an example of the distance S2. In this case, the video search unit 106 calculates the distance S2 based on the latitude and longitude of the search position P2 and the latitude and longitude of the mesh vertex position M03 of the primary mesh code M0. Then, the video search unit 106 calculates the required time T2 by dividing the distance S2 by the speed Vmax of the vehicle 200 (S19), and skips the video data for the required time T2 from the video data of the search position P2 (S20). . In the example of FIG. 13, the video search unit 106 calculates the required time T2 for the distance S2, and skips the video data for that time. In the example of FIG. 12, the video data for the required time T2 is skipped from the video data #m1, and skipped to the video data #m2 (including the file data #m2) at the time tm2.

FIG. 14 is a diagram showing an example of the position P3 of the vehicle 200 after the required time T2 has elapsed from the search position P2. As shown in FIG. 12, the video search unit 106 skips from file #m1 to file #m2 after the required time T2 has elapsed, and extracts position information #m2 from file #m2, thereby obtaining the position information of search position P2. Get (latitude and longitude). The video search unit 106 then increments x to change the search position P2 to P3.

Returning to FIG. 6, next, the video search unit 106 calculates the primary mesh code M3 at the search position P3 by substituting the position information #m2 into the formulas (1) to (3) (S13). In this case, it is assumed that the video search unit 106 has calculated "5339" as the primary mesh code M3 of the search position P3.

Next, the image search unit 106 determines whether or not the primary mesh code M0 including the specific position P0 and the mesh code Mx of the primary mesh including the search position Px match (S14). (Yes in S14), it is determined whether or not the order of the regional mesh (hereinafter sometimes referred to as "mesh") is smaller than "3" (S15).

In the example of FIG. 14, since M0=M3, the video search unit 106 determines Yes in S14, and determines No in S15 because the order of the mesh is "1". For example, the video search unit 106 determines S15 based on the order of the mesh stored in the memory 105 .

Returning to FIG. 6, when the degree of the mesh is smaller than "3" (Yes in S15), the video search unit 106 increments and changes the degree of the mesh (S21). In the example of FIG. 14, the video search unit 106 changes the order of the mesh from "1" to "2". For example, the video search unit 106 changes the order of the mesh stored in the memory 105 from “1” to “2” and stores “2” in the memory 105 . After this processing, the processing using the secondary mesh is performed.

<2. Processing using secondary mesh>
Returning to FIG. 6, the process moves to S12 again, and the video search unit 106 calculates a secondary mesh code (hereinafter sometimes referred to as a "secondary mesh code") M0 including the specific position P0. (S12). The video search unit 106 calculates a secondary mesh code including the specific position P0 (latitude Y, longitude X) using, for example, the following equations (5) to (7).

[Number 5]
Secondary mesh code = primary mesh code x 100 + M2y + M2x (5)
[Number 6]
M2y=INT (((Y×60.0)−(M1y×40.0))÷5.0) (6)
[Number 7]
M2x=INT (((X-100)-M1x)×(60.0÷7.5)) (7)

For example, the video search unit 106 calculates "533945" as the secondary mesh code for the specific position P0.

Next, the video search unit 106 calculates a secondary mesh code for the search position P3 (S13). For example, the video search unit 106 uses Equations (5) to (7) to calculate the secondary mesh code at the search position P3 and obtain "533920".

Next, the video search unit 106 determines whether or not the mesh code at the specific position P0 matches the mesh code at the search position P3 (S14). For example, the video search unit 106 does not match the secondary mesh code "533945" at the specific position P0 and the secondary mesh code "533920" at the search position P3. Therefore, the video search unit 106 determines that they do not match (No in S14), and performs the processing from S18 to S19.

That is, the video search unit 106 calculates the distance S3 from the search position P3 to the tertiary mesh of the specific position P0, calculates the required time T3 required to move the distance S3, and skips video data for the required time T3. (S18-S19).

FIG. 15A shows an example where the vehicle 200 is shifted from the video data at the search position P3 to the search position P4 by skipping the video data for T3.

Also in this case, the video search unit 106 performs, for example, the following processing. That is, the image search unit 106 sets the minimum distance S3 among the distances from the search position P3 to each vertex position of the tertiary mesh of the specific position P0, reads the velocity Vmax from the memory 105, and sets the distance S3. Divide by the velocity Vmax. The video search unit 106 accesses the memory 105, skips the file (or video data) for the required time T3 from the file (or video data) #m2 at the search position P3, and skips to the file #m3. The video search unit 106 extracts the position information #m3 from the file #m3 after the required time T3 has elapsed, thereby obtaining the position information (latitude and longitude) of the changed search position P4 (S20).

The video search unit 106 then calculates a mesh code including the search position P4 (S13). For example, the video search unit 106 substitutes the latitude and longitude of the search position P4 acquired in S20 into the formulas (5) to (7) read out from the memory 105, thereby obtaining the search position P4 at 2 Calculate the next mesh code "533945".

Next, when the video search unit 106 compares the secondary mesh code at the specific position P0 and the secondary mesh code at the search position P4 (S14), they match. 2”, it is determined as Yes in S15.

Then, the video search unit 106 increments the order of the mesh to change from "2" to "3" (S21). For example, the video search unit 106 stores the changed mesh degree “3” in the memory 105 .

As shown in FIG. 15A, since the search position P4 is positioned within the second-order mesh of the specific position P0, the video search unit 106 will perform processing using the third-order mesh thereafter.

<3. Processing using tertiary mesh>
Returning to FIG. 6, the process shifts to S12 again, and the video search unit 106 calculates a tertiary mesh mesh code (hereinafter sometimes referred to as "tertiary mesh code") M0 of the specific position P0. The video search unit 106 calculates a tertiary mesh code for the specific position P0=(latitude, longitude)=(Y, X) using, for example, the following equations (8) to (10).

[Number 8]
Tertiary mesh code = secondary mesh code x 100 + M3y x 10 + M3x (8)
[Number 9]
M3y=INT ((((Y×60.0)−(M1y×40.0))−(M2y×5.0))×(60.0/30.0)) (9)
[Number 10]
M3x=INT ((((X-100)-M1x)×60.0)-(M2x×7.5))×(60.0/45.0)) (10)

The tertiary mesh code for the specific position PO=(35.673033, 139.6333546) is "53394510".

Equations (8) to (9) are also stored, for example, in the memory 105, and the video search unit 106 reads out the equations (8) to (10) from the memory 105 during this process, and finds the specific position. By substituting the latitude and longitude of P0, the tertiary mesh code at the specific position P0 is calculated.

Next, the video search unit 106 calculates the tertiary mesh code M4 at the search position P4 (S13). For example, the video search unit 106 calculates "53394541" as the tertiary mesh code M4 at the search position P4.

Then, since the tertiary mesh code at the specific position P0 and the tertiary mesh code at the search position P4 are different (No in S14), the image search unit 106 calculates the distance S4 from the search position P4 to the specific position P0 ( S18). Also in this case, for example, the video search unit 106 calculates the distance S4 by calculating the shortest distance among the four vertices of the tertiary mesh of the specific position P0 and the search position P4.

The video search unit 106 calculates the required time T4 (=S4/Vmax) required for moving the distance S4, and skips the video data at the recorded video search position P4 to the position Px after the required time T4 (S19). , S20). FIG. 15B is a diagram showing an example of the position P5 of the vehicle 200 after skipping the required time T4 from the search position P4. The video search unit 106 performs subsequent processing using this position P5 as the post-change search position Px.

FIG. 12 shows an example of the file after skipping the required time T4. The video search unit 106 skips the file (or video data) for the required time T4 and accesses the file #m4.

Returning to FIG. 6, next, the video search unit 106 calculates the tertiary mesh code at the search position P5 (S13). For example, the video search unit 106 calculates "53394510" as the tertiary mesh code for the search position P5.

Since the tertiary mesh code at the specific position P0 and the tertiary mesh code at the search position P5 match (Yes in S14), the image search unit 106 determines whether or not the order of the mesh is smaller than "3" (S15). ). In this case, since the order of the mesh is "3", the video search unit 106 determines No in S15.

When it is determined No in S15, the video search unit 106 performs linear search from the search position P5 (S16). In the example of FIG. 15B, the video search unit 106 sequentially searches files from the search position P5, and searches for files having position information of the specific position P0.

FIG. 12 shows an example in which the linear search is sequentially performed from the file #m4 having the position information of the search position P5 after skipping for the required time T4. As shown in FIG. 12, the video search unit 106 sequentially searches from file #m4 to file #0 at the specific position P0 without skipping.

For example, when the video search unit 106 searches for the file at the specific position P0 by performing linear search, the video search unit 106 outputs the file at the specific position P0 to the wireless unit 107 . Alternatively, the image search unit 106 extracts image data from the file at the specific position P0 and outputs the extracted image data to the wireless unit 107. FIG. As a result, the video data of the specific position P0 is transmitted from the video search device 100 to the center 400 as a search result, and the center 400 can acquire the video data of the designated specific position P0.

Then, the video search unit 106 ends the series of processes (S17).

In this way, the video search device 100 searches video data from the search start position P1 to the primary mesh including the specific position P0 based on the position information of the search start position P1 and the position information of the specific position P0. The video data at position P1 is skipped. Then, the video search device 100 repeats the process of skipping the video data based on the position information of the acquisition position (or search position) Px where the skipped video data was acquired and the position information of the specific position P0. conduct. That is, the video search device 100 retrieves the video data from the acquisition position Px to the (n-1) (n is an integer that satisfies n≧2)-order mesh smaller in size than the first-order mesh including the specific position P0. Skip from the video data at Px and repeat this up to the n-th mesh with a size smaller than the (n-1)-th mesh.

In the operation example described above, the process is repeated up to the 3rd order mesh with n=3, but it may be repeated up to the 4th order mesh smaller than the 3rd order mesh with n=4. In this case, the order of the mesh in S15 may be changed from "3" to "4" and determined by whether or not it is smaller than "4". Furthermore, meshes of orders higher than the fifth order, such as fifth-order meshes, sixth-order meshes, etc. (the larger the order, the smaller the mesh size) may be repeated. In this case, "3" in S14 is replaced with " n" (n is an integer of 5 or more). Alternatively, with n=2, the video search device 100 may repeat up to the second mesh and not process the third mesh.

Then, based on the position information at the specific position P0, the video search device 100 searches the video data after skipping up to the video data at the specific position P0 from the memory 105, and retrieves the video data at the specific position P0 from the memory 105. read from and output.

As a result, for example, as shown in FIG. 12, the video search device 100 can skip from file #1 to file #m4 and search from file #m4 to file #0 in order. Therefore, compared to the case of sequentially searching files from file #1 to file #0, the video search device 100 skips files (or video data), so that the search processing can be speeded up. It becomes possible.

The above is summarized as follows.

(Appendix 1)
A video search program executed by a computer of a video search device having a memory in which position information and video data obtained at a position represented by the position information are recorded,
Based on the position information of the search start position and the position information of the specific position, the video data from the search start position to the primary mesh including the specific position is skipped from the video data at the search start position,
Next, based on the position information of the acquisition position where the skipped video data was acquired and the position information of the specific position, (n -1) (n is an integer that satisfies n≧2) skip the image data up to the next mesh from the image data at the acquisition position, and repeat this up to the nth mesh smaller in size than the (n−1)th mesh. ,
searching the memory for video data after skipping up to the n-th order mesh to video data at the specific position based on the position information at the specific position, and outputting the video data at the specific position; A video search program characterized in that it is executed by

(Appendix 2)
Based on the position information of the search start position and the position information of the specific position, a first shortest distance from the search start position to the primary mesh including the specific position is calculated, and the first shortest distance is calculated skipping the video data for the corresponding time from the video data at the search start position;
Next, based on the position information of the acquisition position and the position information of the specific position, calculate the (n-1)th shortest distance from the acquisition position to the (n-1) order mesh including the specific position. and skipping the image data corresponding to the (n−1)-th shortest distance from the image data at the acquisition position, and repeating this up to the n-th mesh. Video search program.

(Appendix 3)
When the primary mesh including the first search position where the video data after skipping from the video data at the search start position is acquired does not match the primary mesh including the specific position, the first a primary mesh including a search position where the video data is skipped from the first search position based on the position information of the search position and the position information of the specific position, and the video data after skipping is acquired; Repeat skipping until the primary mesh containing the specific position matches,
When the first-order mesh including the search position and the first-order mesh including the specific position match, an image from the search position to the (n-1)th-order mesh having a size smaller than the first-order mesh including the specific position. The video search program according to appendix 1, wherein data is skipped from the video data at the search position.

(Appendix 4)
The (n−1) order mesh including the second search position where the video data after skipping from the video data at the acquisition position is acquired matches the (n−1) order mesh including the specific position. If not, the video data is skipped from the second search position based on the position information of the second search position and the position information of the specific position, and the search position where the skipped video data is obtained. Repeat skipping until the (n-1)th mesh containing the specific position matches the (n-1)th mesh containing the specific position,
when the (n−1)th order mesh including the search position and the (n−1)th order mesh including the specific position match each other, the image data from the search position to the nth order mesh including the specific position; , and skipping from the video data at the search position.

(Appendix 5)
position information of the third search position when the n-th order mesh including the third search position from which the video data after skipping from the search position is acquired does not match the n-th order mesh including the specific position; and the position information of the specific position, the video data is skipped from the third search position, and an n-th mesh including the search position where the skipped video data was acquired, and the n-th mesh including the specific position Repeat skipping until the next mesh matches,
When the n-th order mesh including the search position and the n-th order mesh including the specific position match, the memory is searched for from the video data after skipping to the n-th order mesh to the video data at the specific position. The video search program according to Supplementary Note 1, characterized by:

(Appendix 6)
A primary mesh including the first search position and the specific position based on a mesh code of the primary mesh including the first search position and a mesh code of the primary mesh including the specific position. The video search program according to appendix 3, wherein it is determined whether or not the primary mesh matches.

(Appendix 7)
determining whether or not the mesh code of the (n-1)th order mesh including the second search position matches the mesh code of the (n-1)th order mesh including the specific position. A video search program according to appendix 4.

(Appendix 8)
The image search program according to claim 1, wherein file format data including the position information and the image data is stored in the memory at predetermined time intervals.

(Appendix 9)
The video search program according to appendix 8, wherein the video data is skipped based on time information included in the file format data.

(Appendix 10)
A video search method in a video search device having a memory in which position information and video data acquired at a position represented by the position information are recorded,
Based on the position information of the search start position and the position information of the specific position, the video data from the search start position to the primary mesh including the specific position is skipped from the video data at the search start position,
Next, based on the position information of the acquisition position where the skipped video data was acquired and the position information of the specific position, (n -1) (n is an integer that satisfies n≧2) skip the image data up to the next mesh from the image data at the acquisition position, and repeat this up to the nth mesh smaller in size than the (n−1)th mesh. ,
Based on the position information at the specific position, the memory is searched for video data after skipping up to the n-th order mesh to the video data at the specific position, and the video data at the specific position is output. video search method.

(Appendix 11)
a memory in which position information and video data acquired at the position represented by the position information are recorded;
and a video search unit,
The video search unit is
Based on the position information of the search start position and the position information of the specific position, the video data from the search start position to the primary mesh including the specific position is skipped from the video data at the search start position,
Next, based on the position information of the acquisition position where the skipped video data was acquired and the position information of the specific position, (n -1) (n is an integer that satisfies n≧2) skip the image data up to the next mesh from the image data at the acquisition position, and repeat this up to the nth mesh smaller in size than the (n−1)th mesh. ,
Based on the position information at the specific position, the memory is searched for video data after skipping up to the n-th order mesh to the video data at the specific position, and the video data at the specific position is output. video search device.

10: Communication system 100 (100-1 to 100-4): Video search device 101: Camera 103: GPS receiver 104: Recording unit 105: Memory 106: Video search unit 107: Wireless unit 200 (200-1 to 200- 4): Vehicle 300: Base station device 400: Center P0: Specific position P1: Search position (search start position)

Claims (4)

A video search program executed by a computer of a video search device having a memory in which position information and video data obtained at a position represented by the position information are recorded,
Based on the position information of the search start position photographed at a first time and the position information of the specific position photographed at a second time that is later than the first time , from the search start position to the specific position advance the time from the video data at the search start position and skip the video data up to the first mesh including
Next, based on the position information of the acquisition position where the skipped video data was acquired and the position information of the specific position, (n -1) (n is an integer that satisfies n≧2). Repeat until the next mesh,
Based on the position information at the specific position, the memory is searched for the video data at the specific position while advancing the time from the video data skipped to the n-th order mesh, and the video data at the specific position is output. A video search program characterized by causing a computer to execute processing. Based on the position information of the search start position and the position information of the specific position, a first shortest distance from the search start position to the primary mesh including the specific position is calculated, and the first shortest distance is calculated skipping the video data for the corresponding time from the video data at the search start position;
Next, based on the position information of the acquisition position and the position information of the specific position, calculate the (n-1)th shortest distance from the acquisition position to the (n-1) order mesh including the specific position. and skipping the image data corresponding to the (n-1)-th shortest distance from the image data at the acquisition position, and repeating this up to the n-th mesh. video search program. A video search method in a video search device having a memory in which position information and video data acquired at a position represented by the position information are recorded,
Based on the position information of the search start position photographed at a first time and the position information of the specific position photographed at a second time that is later than the first time , from the search start position to the specific position advance the time from the video data at the search start position and skip the video data up to the first mesh including
Next, based on the position information of the acquisition position where the skipped video data was acquired and the position information of the specific position, (n -1) (n is an integer that satisfies n≧2). Repeat until the next mesh,
Based on the position information at the specific position, the memory is searched for the video data at the specific position while advancing the time from the video data skipped to the n-th order mesh, and the video data at the specific position is output. A video retrieval method characterized by: a memory in which position information and video data acquired at the position represented by the position information are recorded;
and a video search unit,
The video search unit is
Based on the position information of the search start position photographed at a first time and the position information of the specific position photographed at a second time that is later than the first time , from the search start position to the specific position advance the time from the video data at the search start position and skip the video data up to the first mesh including
Next, based on the position information of the acquisition position where the skipped video data was acquired and the position information of the specific position, (n -1) (n is an integer that satisfies n≧2). Repeat until the next mesh,
Based on the position information at the specific position, the memory is searched for the video data at the specific position while advancing the time from the video data skipped to the n-th order mesh, and the video data at the specific position is output. A video search device characterized by:

Images