JP4279940B2 - TV broadcasting billboard advertisement effect investigation system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a TV broadcast billboard advertisement effect investigation system that investigates the advertising effect of a billboard advertisement displayed on a TV screen for viewers in TV broadcast, for example.
[0002]
[Prior art]
For example, competitions such as baseball, soccer, volleyball, and rugby are performed in predetermined competition spaces such as a baseball stadium, a soccer field, and an indoor stadium. In these baseball stadiums, soccer stadiums and indoor stadiums, many companies and organizations, including the sponsors of the competitions to be held, have installed many signboard advertisements in a position that can be seen by the audience for their own promotion. Yes.
[0003]
In some cases, a competition held in such a stadium can be watched on a TV receiver in each home by being broadcast on TV. In some cases, TV is recorded and only the highlight portion of the sports news program is broadcast.
[0004]
Companies and organizations that install signboard advertisements in these stadiums not only verify how much the signboard appeals to the spectator spectators, but also when the competition is broadcast on TV. In many cases, a signboard appears in the background of the competitor. In this case, verifying how much the signboard appeals to TV viewers is also an important verification item for signboard installation.
[0005]
[Problems to be solved by the invention]
In general, it is clear that it is desirable to install a signboard at a position where the TV camera often shoots. However, these are qualitative judgments only, and the advertising effect of the signboard advertisement installed at that position is quantified. Cannot be verified.
[0006]
In order to avoid such a situation, a method of recognizing a still image of 1/30 second interval constituting a TV moving image and grasping a signboard included in the still image can be considered. Considering that the size of the signboard on the still image changes due to frequent use, a large amount of processing time is required for the image recognition processing, which is not practical.
[0007]
Also, in TV broadcasts of sports such as soccer, the camera orientation of the TV camera changes frequently and at high speeds, so even if a signboard appears on the TV screen at home, There is a concern that the contents of the signboard cannot be recognized from the TV screen being viewed.
[0008]
The present invention has been made in view of such circumstances, and a signboard included in a relayed TV image can be detected using a simple software method other than the image recognition technique, and the appearance of each signboard in the TV image. It is an object of the present invention to provide a TV broadcast billboard advertisement effect investigation system that can count the number of times and quantitatively verify the advertising effect of a billboard advertisement copied on a TV screen broadcasted on TV.
[0009]
[Means for Solving the Problems]
In order to solve the above problems, in the TV broadcast signboard advertisement effect investigation system of the present invention, a camera orientation detector that is attached to a TV camera installed at a predetermined position and detects the camera orientation, and a TV camera And a zoom amount detector for detecting the zoom amount of the camera, and a camera orientation and a zoom amount detector detected by the camera orientation detector at a predetermined cycle with respect to a TV image taken by the TV camera. Data insertion means for inserting the zoom amount, a camera position sign corresponding memory for storing each sign included in the TV image when the TV camera is set to each camera orientation and each zoom amount, and the camera from the TV image at a predetermined cycle Data extracting means for extracting the azimuth and zoom amount, and each signboard corresponding to the camera azimuth and zoom amount sequentially extracted at a predetermined cycle are displayed on the camera position signboard correspondence menu. It includes a reading signboard reading means from Li, a signboard appearance number counter for counting the number of times of reading of each billboard in the billboard reading means as the number of occurrences of each billboard in the TV image.
[0010]
In the TV broadcast signboard advertisement effect investigation system configured as described above, the camera orientation and zoom amount of the TV camera are always detected. That is, in a TV camera installed at a predetermined position, if the camera orientation and the zoom amount are determined, the TV image obtained by this TV camera is a fixed image except for the competitors and the audience. Of course, if the camera orientation and the zoom amount are determined, the signboard included in the TV image obtained by this TV camera is also specified.
[0011]
Therefore, each sign included in the TV image when the TV camera is set to each camera orientation and each zoom amount is stored in the camera position sign corresponding memory. Therefore, by measuring the camera orientation and zoom amount of the TV camera at each time, the number of appearances (appearance frequency) of the signboard in the TV image can be grasped.
[0012]
In another invention, in the above-described TV broadcast billboard advertisement effect investigation system, out of the number of appearances of each billboard counted by the billboard appearance frequency counter, the billboard reading means continuously reads a predetermined number of cycles. A correction appearance number counter that counts this as the number of appearances after a new correction is provided.
[0013]
As described above, if the camera orientation of the TV camera changes at high speed, even if a signboard appears on the TV screen, if it disappears at the next moment, the TV viewer cannot grasp the contents of the signboard. Therefore, in the present invention, it is determined that the signboard has been broadcast on the TV screen only when the signboard continues to appear (broadcast) for a predetermined number of cycles (predetermined time). By adopting such a configuration, it is possible to grasp the effect of the signboard advertisement that is closer to the actual situation.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a schematic diagram showing a schematic configuration of a TV broadcast billboard advertisement effect investigation system according to an embodiment.
[0015]
For example, when playing a baseball game, a plurality of TV cameras 1 with camera numbers A are arranged at a plurality of predetermined positions in a baseball stadium. Each TV camera 1 has a zoom lens 2, a camera body 3 with a built-in CCD image sensor, and a horizontal azimuth α (angle) and vertical azimuth β (angle) of the camera body 3 according to the operation of the cameraman. It is comprised with the support mechanism 4 to support.
[0016]
Then, with respect to the TV camera 1, a zoom amount detector 5 for detecting the zoom amount Z of the zoom lens 2, and a camera orientation comprising a horizontal orientation α and a vertical orientation β of the TV camera 1 as shown in FIG. A camera orientation detector 6 for detecting α and β is attached. In addition, a camera number storage unit 8 for storing the camera number A of the TV camera 1 is provided.
[0017]
The TV image signal a output from the TV camera 1, the zoom amount Z output from the zoom amount detector 5, the camera directions α and β output from the camera direction detector 6, and the camera number storage unit 8. The camera number A is input to the data insertion processing unit 7.
[0018]
The data insertion processing unit 7 inserts the input camera number A, camera orientation α, β, and zoom amount Z into the input TV image signal a as insertion data (A, α, β, Z) at a predetermined cycle. To do. A specific method for inserting the insertion data (A, α, β, Z) into the TV image signal a will be described with reference to a schematic diagram of the TV image signal a shown in FIG.
[0019]
The TV image signal a is actually 525 scanning line signals connected in series. However, in order to simplify the description, as shown in FIG. 3, it is repeated at a predetermined period of 1/30 seconds. It is assumed that the image data 11 that is a still image and a blanking section 12 are included. Then, the insertion data (A, α, β, Z) is written in the blanking section 12 that appears at a predetermined period of 1/30 seconds. Accordingly, the image data 11 that is each still image updated at a predetermined period of 1/30 seconds includes the camera number A, the camera orientation α, β, and the zoom amount Z of the TV camera 1 at the time when the image data 11 is captured. Is stored as insertion data (A, α, β, Z) in the adjacent blanking section 12. Note that the insertion data (A, α, β, Z) inserted in the blanking section 12 is not displayed on a general TV receiver.
[0020]
The TV image signal a in which the insertion data (A, α, β, Z) output from the data insertion processing unit 7 is inserted into each blanking section 12 is recorded by the VTR 9 for replay, and a program organizing device. 10 is input. The program organizing apparatus 10 selects and broadcasts the optimum TV image signal a at this time among the TV image signals a output from the data insertion processing unit 7 of each TV camera 1 according to the operation of the program director. It is sent to the relay device 13 as a TV image for use.
[0021]
The relay device 13 transmits the input TV image signal a to the broadcast station-side relay device 17 via the transmission antenna 14, the communication satellite 15, and the reception antenna 16. The TV image signal a received by the relay device 17 is recorded on the VTR 18 and sent to the broadcasting station 19. The broadcast station 19 broadcasts the received TV image signal a on the broadcast radio wave from the antenna (TV tower) 20. In the broadcasting station 19, the insertion data (A, α, β, Z) inserted in the blanking section 12 can be removed.
[0022]
The TV image signal a recorded in the VTR 18 is read out asynchronously with respect to the broadcast, and is input to the data analysis device 21 that investigates the advertising effect in the billboard advertisement.
[0023]
For example, the data analysis device 21 composed of a computer is configured as shown in FIG. In this data analysis device 21, as shown in FIG. 4, a camera for storing each sign 29 included in the TV image data 11 when each TV camera 1 is set to each camera orientation α, β and each zoom amount Z. A position signboard correspondence database (memory) 24 is provided.
[0024]
Specifically, as shown in FIG. 6, the camera number A, the camera orientations α, β, and the insertion data (A, α, β, Z) inserted in the blanking interval 12 of the TV image signal a, and For all the combinations in the zoom amount Z, in each signboard 29 attached to this baseball stadium when the TV camera 1 is set to the camera orientation α, β and the zoom amount Z corresponding to the combination, Whether or not each signboard 29 appears (included) in the image data 11 is stored. The case where it appears (is included) is “1”, and the case where it does not appear (is not included) is “0”.
[0025]
As shown in FIG. 4, the image data 11 is set to “1”, assuming that the case where the signboard 29 is included in half or more appears (included). The camera orientations α and β are set, for example, every 1 °, and the zoom amount Z, for example, every 1 mm.
[0026]
In FIG. 5, the TV image signal a shown in FIG. 3 reproduced from the VTR 18 is inserted into the blanking section 12 by the data extraction unit 22 at intervals of 1/30 second (A, α, β, Z). ) Is extracted and sent to the signboard reading unit 23. The signboard reading unit 23 searches the camera position signboard correspondence database 24 with the input insertion data (A, α, β, Z), and indicates the appearance corresponding to the corresponding insertion data (A, α, β, Z). Each signboard 29 (signboard number) set to “1” is read and sent to the appearance counter 25 for each signboard.
[0027]
The number-of-appearance counter 25 for each signboard counts the number of appearances CN of each signboard 29 set in the camera position signboard correspondence database 24.
[0028]
The appearance count counter 25 for each signboard sends out the counted appearance count CN for each signboard 29 to the next corrected appearance count counter 26. The corrected appearance number counter 26 sends the corrected track appearance number CM to the display device 27 and the printer 28. The display device 27 and the printer 28 display and output the corrected appearance frequency CM for each inputted signboard 29 (signboard number) and display it in a report.
[0029]
The overall operation of the data analysis apparatus 21 configured as described above will be described with reference to the flowchart shown in FIG.
[0030]
When one piece of image data 11 having an interval of 1/30 seconds constituting the TV image signal a reproduced from the VTR 18 is input (S1), the data extraction unit 22 starts from a blanking section 12 adjacent to the image data 11. Insertion data (A, α, β, Z) is extracted (S2). Next, the camera position signboard correspondence database 24 is searched with the extracted insertion data (A, α, β, Z), and “1” indicating the appearance corresponding to the corresponding insertion data (A, α, β, Z) is obtained. The set signboard number is read (S3).
[0031]
Then, the number of appearances CN of the sign number set to “1” read this time in the number-of-appears appearance counter 25 is updated (S4).
[0032]
CN = CN + 1
Further, the appearance number CN of the signboard number set to “0” that has not been read out in the signboard appearance frequency counter 25 is cleared to 0 (S5).
[0033]
CN = 0
Then, it is checked whether or not there is a signboard 29 (signboard number) having an appearance count CN of 15 in the updated appearance counter 25 for each signboard (S6). When the appearance count CN has reached 15, it is determined that the corresponding signboard 29 has been displayed on the TV screen of the TV receiver for 0.5 seconds or more, and the appearance count of the corresponding signboard number in the corrected appearance count counter 26 The CM is updated (S7).
[0034]
CM = CM + 1
Next, the appearance number CN of the signboard number in the signboard appearance number counter 25 is cleared to 0 (S8).
[0035]
In the TV broadcast billboard advertisement effect investigation system configured as described above, the camera orientations α and β and the zoom amount Z of the TV camera 1 are obtained by the camera orientation detector 6 and the zoom amount detector 5, as shown in FIG. Always detected. The data insertion processing unit 7 inserts the camera number A together with the camera number A into the TV image signal a output from the TV camera 1 as insertion data (A, α, β, Z).
[0036]
In the camera position signboard correspondence memory 24, each signboard 29 included in each image data 11 of the TV image signal a when the TV camera 1 is set to each camera orientation and each zoom amount is stored.
[0037]
Accordingly, by measuring the camera number A, the camera orientations α and β, and the zoom amount Z of the TV turtle 1 camera at each time, the number of appearances (appearance frequency) CN of the signboard 29 in the TV image can be grasped.
[0038]
Furthermore, it is determined that the signboard 29 has been broadcast on the TV screen of the TV receiver only when the signboard 29 continues to appear (broadcast) for a predetermined time such as 0.5 seconds. By adopting such a configuration, the signboard 29 that is not noticed by the viewer and is instantaneously reflected is not counted as the appearance frequency CM.
[0039]
Therefore, the installer of the signboard 29 can grasp the effect of the signboard advertisement that is closer to reality.
[0040]
【The invention's effect】
As described above, in the TV broadcast signboard advertisement effect investigation system of the present invention, the signboard included in the relayed TV image is detected from the detector for detecting the camera direction and zoom amount of the TV camera, and the camera position signboard correspondence memory. It is detected using and. Then, the number of appearances of each signboard in the TV image is counted. Therefore, it is possible to quantitatively verify the advertising effect of the signboard advertisement that is copied on the TV screen that is broadcast on TV.
[0041]
Furthermore, only when it is displayed on the TV screen continuously for a certain period of time, the display count (appearance count) is adopted as a factor of the advertising effect of the corresponding signboard. Therefore, the installer of the signboard can grasp the effect of the signboard advertisement that is closer to reality.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a schematic configuration of a TV broadcast billboard advertisement effect investigation system according to an embodiment of the present invention. FIG. 2 is a schematic diagram showing a zoom direction and a camera orientation of a TV camera incorporated in the system of the embodiment. Fig. 3 is a schematic diagram showing a TV image signal output from a TV camera. Fig. 4 is a diagram showing image data constituting the TV image signal. Fig. 5 is a diagram of a data analysis apparatus incorporated in the system of the embodiment. FIG. 6 is a block diagram showing the detailed configuration. FIG. 6 is a diagram showing the contents stored in a camera position signboard correspondence database formed in the data analysis apparatus. FIG. 7 is a flowchart showing the operation of the data analysis apparatus incorporated in the system of the embodiment. [Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... TV camera 2 ... Zoom lens 3 ... Camera body 4 ... Support mechanism 5 ... Zoom amount detector 6 ... Camera direction detector 7 ... Data insertion processing part 8 ... Camera number memory | storage part 10 ... Program organization apparatus 11 ... Image data 12 ... Blanking section 21 ... Data analysis device 22 ... Data extraction unit 23 ... Signboard reading unit 24 ... Camera position signboard correspondence database 25 ... Appearance counter 26 for each signboard ... Correction appearance counter

Claims (2)

A camera orientation detector that is attached to a TV camera installed at a predetermined position and detects the camera orientation;
A zoom amount detector attached to the TV camera for detecting the zoom amount of the camera;
Data insertion means for inserting a camera orientation detected by the camera orientation detector and a zoom amount detected by the zoom amount detector in a predetermined cycle with respect to a TV image taken by the TV camera;
A camera position signboard corresponding memory for storing each signboard included in the TV image when the TV camera is set to each camera orientation and each zoom amount;
Data extraction means for extracting a camera orientation and a zoom amount at the predetermined period from the TV image;
Signboard reading means for reading out each signboard corresponding to the camera orientation and zoom amount sequentially extracted at the predetermined period from the camera position signboard correspondence memory;
A billboard advertisement effect investigation system for TV broadcasting, comprising: a billboard appearance frequency counter that counts the number of times each billboard is read by the billboard reading means as the number of times each billboard appears in the TV image. Of the number of appearances of each signboard counted by the signboard appearance number counter, the corrected number of appearances is counted as a new corrected number of appearances only when the signboard reading unit continuously reads the signboard number of times. The billboard advertisement effect investigation system for TV broadcasting according to claim 1, further comprising a counter.

Images