JP6770847B2 - Channel identification systems, devices, methods and programs - Google Patents

Description

The present invention relates to a channel identification device for identifying a broadcast channel being presented by a receiver, a channel identification method, and a channel identification program.

In recent years, domestic robots that play the role of partners who live with users have been developed. By recognizing the TV receiver and grasping the position of such a robot, it is possible to perform control using the positional relationship between the TV receiver and the robot. Furthermore, by grasping the positional relationship between the TV receiver and the robot and then identifying the broadcast channel being selected by the TV receiver, the robot can perform operations and information related to the broadcast channel being selected. Can be provided to users.

In such a situation, for example, in Patent Document 1, in order for a self-propelled robot to acquire a screen position of a television receiver, a predetermined pattern, color, two-dimensional bar code, or the like is placed at a predetermined position on the screen. A method of embedding a marker image has been proposed.
Further, Patent Document 2 proposes four methods for a self-propelled robot to specify a broadcasting channel. The first method is a method of inquiring the user by voice or characters. The second method is a method of acquiring operating state information of the television receiver by communicating with the television receiver via a communication line such as the Internet or infrared rays. The third method is a method in which a broadcast reception function is mounted on a robot, and the broadcast channels being received by the robot are sequentially switched to compare image data. The fourth method is to identify the broadcast channel received by the TV receiver by acquiring the output audio signal by sequentially switching the broadcast channels and comparing it with the audio signal propagating from the TV receiver. Is.

Japanese Unexamined Patent Publication No. 2009-11362 Japanese Unexamined Patent Publication No. 2005-4697

However, when the robot tries to recognize the position of the television receiver while the user is watching a broadcast program, the technique of Patent Document 1 embeds a marker image in the display screen of the television receiver, so that the program It interferes with viewing.

Further, when the robot identifies a broadcast channel by the technique of Patent Document 2, the robot cannot automatically identify the broadcast channel by the first method of inquiring the user about the broadcast channel being selected. In the second method in which the robot identifies the broadcast channel via communication with the television receiver, the television receiver needs to have a data communication function with the robot. In the third and fourth methods, it is necessary to mount the broadcast receiving means in the robot, and further, a situation in which the broadcast wave can be received is required, so that the position of the robot body is limited.

An object of the present invention is to provide a channel identification system, a channel identification device, a channel identification method, and a channel identification program, which have few restrictions on recognizing the position of a television receiver and identifying a broadcast channel.

The channel identification system according to the present invention has a terminal and a server that are communicated with each other, and the terminal collects a predetermined or greater difference based on the difference between the image pickup unit and the image captured by the image pickup unit. A detection unit that detects the square area of the TV receiver as an area of the TV receiver, an extraction unit that extracts image information in the area detected by the detection unit, and a communication unit that transmits the image information to the server. The server has a storage unit that stores image information of broadcast video for each channel in association with the broadcast time, and stores that store image information received from the terminal in accordance with the imaging time of the image information. It is provided with a determination unit for determining a channel selected by the television receiver based on the degree of matching by comparing with the image information stored in the unit.

The extraction unit extracts the feature points in the region and the feature amount for each feature point as the image information, and the storage unit extracts the feature points of the broadcast image and the feature amount for each feature point. You may remember.

The extraction unit may extract the image information in the region after converting the image in the region into a predetermined shape corresponding to the television screen.

When the region includes the outer edge of the image captured by the imaging unit, the terminal may include a first control unit for moving the imaging direction of the imaging unit toward the outer edge.

When the area is less than a predetermined size, the terminal may include a second control unit that adjusts the zoom setting of the imaging unit so that the area is larger than the predetermined size.

The determination unit may extract image information in a predetermined range before and after the imaging time from the storage unit.

The channel identification device according to the present invention detects a square region in which differences of a predetermined value or more are gathered as a region of a television receiver based on the difference between the imaging unit and the image captured by the imaging unit. The detection unit, the extraction unit that extracts the image information in the area detected by the detection unit, and the image information are transmitted to the server and compared with the image information of the broadcast video for each channel stored in the server. As a result, a communication unit for determining the channel selected by the television receiver based on the degree of matching is provided.

In the channel identification program according to the present invention, in a computer provided with an imaging unit, a square region in which differences of a predetermined value or more are gathered based on the differences between images captured by the imaging unit is formed on a television receiver. By detecting it as an area, extracting the image information in the detected area, transmitting the image information to the server, and comparing it with the image information of the broadcast video for each channel stored in the server, the degree of matching can be determined. Based on this, the channel selected by the television receiver is determined.

According to the present invention, there are less restrictions on recognizing the position of the television receiver and identifying the broadcast channel.

It is a figure which shows the functional structure of the channel identification system which concerns on embodiment. It is a flowchart which shows the detection process of the screen area which concerns on embodiment. It is a figure which shows the operation example of the image pickup unit which concerns on embodiment. It is a figure which shows the operation of the feature amount extraction which concerns on embodiment. It is a figure which shows the data structure example of the feature amount which concerns on embodiment. It is a flowchart which shows the channel identification process which concerns on embodiment. It is a conceptual diagram which shows the correspondence point determination process which concerns on embodiment.

Hereinafter, an example of the embodiment of the present invention will be described.
The channel identification system 1 according to the present embodiment recognizes the position of the television receiver from the image captured by the camera mounted on the device (terminal) such as a robot, and the image feature amount of the recognized television image and the broadcast image. By comparing with the image feature amount of, the channel being selected by the TV receiver is identified.
The TV receiver is not limited to a stationary dedicated machine. The television screen also includes, for example, an application window displayed on a PC display, and indicates an area on which a broadcast image is displayed.

FIG. 1 is a diagram showing a functional configuration of the channel identification system 1 according to the present embodiment.
The channel identification system 1 includes a plurality of robots 10 (terminals) as channel identification devices and a determination server 20. The plurality of robots 10 (10-1, 2, ..., N) and the determination server 20 are each communicated and connected via the Internet or the like.
The channel identification device is not limited to the domestic robot 10, and may be another information processing device equipped with a camera, for example, a smartphone or a tablet terminal.

The robot 10 includes an imaging unit 11, a detection unit 12, a control unit 13 (first control unit, second control unit), an extraction unit 14, and a communication unit 15.
The determination server 20 includes an extraction unit 21, a storage unit 22, and a determination unit 23.

The imaging unit 11 is a camera that captures a moving image or a still image, and imaging, zooming, changing the brightness, and the like are controlled by control data from the control unit 13. Further, the imaging unit 11 changes the camera direction (imaging direction) according to the control data from the control unit 13.

Based on the difference between the images (frames) captured by the imaging unit 11, the detection unit 12 detects a quadrangular region in which pixels having a difference of a predetermined value or more are gathered as a region of the television receiver.

The control unit 13 controls the imaging direction or zoom of the imaging unit 11 according to the detection result by the detection unit 12.
Specifically, when the detected region includes the outer edge of the image captured by the imaging unit 11, that is, when only a part of the television receiver is imaged, the control unit 13 (first control unit) , The imaging direction of the imaging unit 11 is moved toward the outer edge.
When the detected region is less than a predetermined size, the control unit 13 (second control unit) adjusts the zoom setting of the imaging unit so that this region is larger than the predetermined size.

FIG. 2 is a flowchart showing a detection process of a screen area of the television receiver according to the present embodiment.
In this detection process, Kenny's edge detection algorithm (see, for example, "Digital Image Processing [Revised New Edition], pp232-233, Image Information Education Promotion Association, 2015") is used as an example.

In step S1, the detection unit 12 acquires the image data captured by the imaging unit 11. The image data to be acquired may be, for example, a continuous frame in a moving image, or a plurality of still images with continuous imaging times captured at the same position.

In step S2, the detection unit 12 calculates the difference in each pixel by using the plurality of image data acquired in step S1. The pixel value in the pixel (x, y) is, for example, the luminance G (x, y). As a result, the detection unit 12 cuts out a region in which there is movement between images that are continuous in time as a region in which the difference values of each pixel are gathered.

In step S3, in order to detect the edge component in the image, the detection unit 12 first performs a smoothing process on the difference image consisting of the difference values obtained in step S2 in order to reduce the influence of noise. For example, the detection unit 12 smoothes the difference image using a Gaussian filter.
The detection unit 12 may further binarize the smoothed difference image according to a predetermined threshold value.

In step S4, the detection unit 12 calculates the gradient vectors in the x-direction and the y-direction by differentiating the luminance values in each direction, and calculates the gradient intensity and the gradient direction.
Assuming that the gradient vector in the x direction is fx (x, y) and the gradient vector in the y direction is fy (x, y), the gradient intensity g (x, y) and the gradient direction d (x, y) are as follows. It is represented by the equations (1) and (2) of.
g (x, y) = (fx (x, y) ² + fy (x, y) ² ) ^1/2 ... (1)
d (x, y) = tan ^-1 (fy (x, y) / fx (x, y)) ... (2)

In step S5, the detection unit 12 detects the edge component in the difference image by using the gradient intensity of step S4. The edge point is defined as a point where the gradient intensity of the pixel of interest is locally maximized as compared with eight points in the vicinity of the pixel of interest.
At this time, the detection unit 12 uses a constant threshold value Th, sets it as an edge point if the gradient intensity is equal to or higher than the threshold value Th, and sets only the edge point as white (G (x, y) = 1) according to the equation (3). Generate the image.
g (x, y) ≧ Th: G (x, y) = 1
g (x, y) <Th: G (x, y) = 0 ... (3)

The detection unit 12 may provide an upper threshold value and a lower threshold value with respect to the gradient intensity, and detect a point of the gradient intensity between the two threshold values as an edge point.
Further, the method for detecting the gradient of brightness is not limited to this, and for example, a Sobel operator or a Prewitt operator may be used.

In step S6, the detection unit 12 detects a quadrangular region that can be determined to be a television receiver from the edge component that is a candidate for a straight line in the image detected in step S5.
Specifically, the detection unit 12 first extracts a line segment constituting a quadrangular region that is a candidate for a television receiver. To extract the line segment, a Hough transform (see, for example, "Digital Image Processing [Revised New Edition], pp238-243, Image Information Education Promotion Association, 2015") and the like can be used.
Next, the detection unit 12 traces the intersection where another line segment intersects from the arbitrarily selected line segment, reaches the first selected line segment, and traces when there are a total of four intersections. The combination of the line segments is determined to be a quadrangle, and the area of this quadrangle is determined to be a television receiver. On the other hand, when the number of intersections is less than 4 (for example, 2), it is determined that a part of the television receiver is detected.

In step S7, the control unit 13 determines whether or not the quadrangular region could be detected in step S6. If this determination is YES, the process ends, and if the determination is NO, the process proceeds to step S8.

In step S8, the control unit 13 changes the imaging direction of the imaging unit 11 and repeats steps S1 to S7.

FIG. 3 is a diagram showing an operation example of the imaging unit 11 under the control of the control unit 13 according to the present embodiment.
The detection unit 12 determines whether or not the television receiver A exists in the image by using the number of intersections 31 of the line segments in the captured image. As shown in FIG. 3A, if the entire television receiver A is detected, the control unit 13 does not transmit control data to the image pickup unit 11.

On the other hand, as shown in FIG. 3B, if only a part of the television receiver A is present in the captured image, the control unit 13 determines the imaging unit 11 so that the television receiver A falls within the imaging range. Control the angle of.
For example, the control unit 13 transmits control data such as changing the angle at which the image pickup unit 11 is facing to the right by 15 degrees. If the television receiver A is not detected in the captured image, for example, the control unit 13 transmits control data such as changing the angle of the image pickup unit 11 to the right by 30 degrees to the image pickup unit 11. To do.

Returning to FIG. 1, the extraction unit 14 extracts feature points as image information and feature quantities for each of these feature points from the image region recognized as a television receiver by the detection unit 12.
At this time, the extraction unit 14 converts the image in the region into a predetermined shape corresponding to the screen of the television receiver, that is, a rectangle, and then extracts the feature amount.

The type of feature amount is not limited, but in the following, as an example, a method using SIFT (Scale-Invariant Feature Transfer) (see, for example, "Digital Image Processing [Revised New Edition], pp234-237, Image Information Education Promotion Association, 2015"). Will be described.

FIG. 4 is a diagram showing an operation of feature amount extraction in the extraction unit 14 according to the present embodiment.
The extraction unit 14 calculates the direction at the feature point 40 (orientation 41) and the 128-dimensional feature amount around the feature point by using the gradient intensity and the gradient direction used in the detection unit 12.

First, the extraction unit 14 quantizes the gradient direction in 36 directions (in units of 10 degrees), weights the gradient intensity in the local region 42 around the feature point higher as it is closer to the center of the feature point, and votes for weighting. The gradient direction histogram 43 is created. The peak direction 44, which is 80% or more of the maximum value of the gradient direction histogram 43, is defined as the orientation 41 in the local region.

Next, the extraction unit 14 rotates the local region 42 for acquiring the pixel values around the feature points so as to match the orientation 41 (45). Subsequently, the extraction unit 14 divides one side of the rectangular region into 4 blocks (46) and the gradient direction into 8 directions (47) to create a gradient direction histogram, and features 4 * 4 * 8 = 128 dimensions. Calculate the amount.

The feature amount extraction method is not limited to this, and for example, SURF (Scale-InvariantFairtureTransform) may be used.

The communication unit 15 adds time information to the image information acquired from the extraction unit 14 and transmits the time information to the determination unit 23 of the determination server 20 using a communication line such as the Internet.
Here, the time information is the time at which the image was taken, and is composed of, for example, the year, month, day, hour, minute, and second, and is composed of the character string YYYYMMDDhmmss. For example, when representing 23:59:59 on March 15, 2016, the time information is "20160315235959".

Further, the communication unit 15 may transmit image information of each of a plurality of images (frames) continuously captured by the image pickup unit 11. As a result, it is possible to suppress the deviation of the frame from the broadcast image stored in the determination server 20, and the possibility of matching is improved.

The extraction unit 21 of the determination server 20 selects each channel to be detected during broadcasting, and periodically extracts feature points and feature quantities for each of these feature points as image information from the broadcast image.
The storage unit 22 stores the feature amount of each broadcast image acquired from the extraction unit 21 in association with the broadcast time information.

The determination unit 23 compares the feature amount V received from the robot 10 with the feature amount V'stored in the storage unit 22 in response to the added time information to identify the channel being selected. , The specified channel number is returned to the robot 10.

At this time, the determination unit 23 may extract image information (feature amount V') in a predetermined range before and after the imaging time from the storage unit and compare each with the received feature amount V.
Further, the determination unit 23 may calculate the imaging time by subtracting a predetermined time including the processing time and the communication time of the robot 10 from the time when the image information (feature amount V) is received from the robot 10. In this case, the time information does not have to be added to the feature amount V'transmitted from the robot 10.

FIG. 5 is a diagram showing an example of a data structure of a feature amount according to the present embodiment.
Here, the data structure of the feature amount V (FIG. 5 (a)) extracted by the extraction unit 14 of the robot 10 and the feature amount V'(FIG. 5 (b)) extracted by the extraction unit 21 of the determination server 20. An example is shown.
The data regarding the feature amount V is composed of the (x, y) coordinate position where the feature amount is extracted, the orientation angle, and the 128-dimensional feature amount, and the total number of the extracted feature amounts is i. The feature amount V'has the same configuration, but the total number of extracted feature amounts is j.

FIG. 6 is a flowchart showing a channel identification process by the determination unit 23 according to the present embodiment.
In step S11, the determination unit 23 acquires the feature amount V extracted by the extraction unit 14 from the communication unit 15 of the robot 10.

In step S12, the determination unit 23 selects a channel number to be compared with the feature amount V.
In step S13, the determination unit 23 calculates the distance between the feature amount V and the feature amount V'extracted from the broadcast image of the channel selected in step S12 for each feature point. At this time, the feature amount V and the feature amount V'are compared with the information at the same time based on the added time information.

For the distance between the feature amount V and the feature amount V', for example, the Euclidean distance is used. The Euclidean distance dust (V, V') between the feature amount V and the feature amount V'is expressed by the equation (4).
dist (V, V') = √ {Σ (v _{k −} v _k ') ² }
V = (v1, v2, ..., v128)
V'= (v1', v2', ..., v128') ... (4)

In step S14, the determination unit 23 sorts each of the feature quantities V (V1, ..., Vi) in ascending order of the value of the Euclidean distance dust calculated in step S13, and determines the point with the smallest distance. Let d1 and the next smallest point be d2.
Next, the determination unit 23 determines that the points satisfying the equation (5) are the corresponding points between the images.
d1 <d2 × K ・ ・ ・ (5)
Here, the coefficient K uses, for example, K = 0.6, but may be adjusted in consideration of the characteristics of the imaging unit 11.

FIG. 7 is a conceptual diagram showing a corresponding point determination process in the determination unit 23 according to the present embodiment.
In the captured image 51 acquired from the robot 10, for example, the feature amount coordinates extracted by the extraction unit 14 are (x1, y1).
Next, in the broadcast image 52 handled by the extraction unit 21 of the determination server 20, the feature quantity coordinates to be extracted are (x1', y1'), (x2', y2'), (x3', y3'), ...・, (XJ', yj').

In FIG. 7, (x1, y1) and (x1', y1'), (x1, y1) and (x2', y2'), (x1, y1) and (x3', y3'), (x1, y1) ) And (xj', yj') Euclidean distances are calculated. As a result, the Euclidean distances of the feature points of the broadcast image 52 with respect to (x1, y1) are dist (1) = 10, dist (2) = 77, dist (3) = 65, and dist (j) = 84. ing.
In this case, d1 = dust (1) and d1 = dust (3) are replaced, and the conditional expression (5) is satisfied (10 <65 × 0.6 = 39). ', Y1') is determined to be the corresponding point.

In step S15, the determination unit 23 determines whether or not the matching rate of the feature points between the captured image and the broadcast image is equal to or higher than a predetermined value. That is, the determination unit 23 determines whether or not the equation (7) is satisfied, assuming that the corresponding points are Numm, the feature points of V'are NuMV', and the threshold value of the matching rate is PM.
(NumM / NuMV')> PM ・ ・ ・ (7)
Here, the threshold PM is set to, for example, 0.6, but may be adjusted depending on the environment at the time of imaging or the specifications of the imaging unit 11.
If this determination is YES, the determination unit 23 determines that the channels match, and the process proceeds to step S16. On the other hand, when the determination is NO, the process returns to step S12, and the determination unit 23 changes the channel number and repeats steps S12 to S15.

In step S16, the determination unit 23 returns the channel number selected in step S12 to the robot 10.

If there is no channel determined to match, the determination unit 23 determines that the area detected in the captured image is not a television receiver or the channel 20 is not the target of the determination server 20, and outputs an error. End the process.

As described above, in the channel identification system 1, the robot 10 detects the television receiver from the image captured by the imaging unit 11, calculates the feature amount of the detected image of the television receiver, and then transmits it to the determination server 20. To do. The determination server 20 identifies the channel being selected by the television receiver by comparing the feature amount transmitted from the robot 10 with the stored data for each channel.
In the channel identification method of the present embodiment, since the robot 10 makes an inquiry to the determination server 20 via the Internet, it is possible to respond to inquiries from a plurality of robots 10-1 to N.

According to the present embodiment, in the channel identification system 1, the screen portion of the television receiver is temporally displayed without embedding a marker image in the screen of the television receiver or adding a special device to the television receiver. The television receiver can be recognized from the image captured by the camera of the robot 10 because the image changes to.
Then, the channel identification system 1 extracts image information from the television image recognized by the robot 10, transmits the extracted data to the determination server 20 via a communication line such as the Internet, and together with the image information received by the determination server 20. By comparing with the image information of the broadcast image and returning the matching channel number to the robot 10, the channel being selected is identified.
In this way, the channel identification system 1 can recognize the television receiver and further identify the channel being selected without the robot 10 and the television receiver communicating with each other.

Therefore, in the channel identification system 1, there are few restrictions when recognizing the position of the television receiver and identifying the broadcast channel. As a result, it is not necessary to equip the robot 10 with a broadcast reception function, and at the same time, the TV receiver does not need a special function for communicating with the robot 10, and the channel can be specified. Therefore, the channel can be specified by the channel identification system 1 even in an environment where the TV receiver is not connected to the Internet, or in an environment where the device such as the robot 10 or the TV receiver is outdoors instead of indoors.

In addition, the channel identification system 1 extracts feature points and feature quantities for each feature point as image information. Therefore, the communication load between the robot 10 and the determination server 20 and the amount of data stored in the determination server 20 are reduced, and the matching process of image information is made more efficient.

The channel identification system 1 has an image in the screen shape when the area of the television receiver detected from the captured image is different from the rectangular screen shape, for example, when the screen is detected as a trapezoid because the screen is not captured from the front. Image information is extracted after conversion. Therefore, the channel identification system 1 can reduce the difference between the captured image and the broadcast image of the television receiver and improve the channel identification accuracy.

Further, when only a part of the television receiver is imaged, the channel identification system 1 adjusts the imaging direction to image the entire television screen and extract image information. Therefore, the channel identification system 1 can accurately identify the broadcast channel by searching for the television receiver.

Further, the channel identification system 1 adjusts the zoom of the image pickup unit 11 to extract image information when the area of the imaged television receiver is less than a predetermined size. Therefore, the channel identification system 1 can reduce the difference between the captured image of the television receiver and the broadcast image and improve the channel identification accuracy by extracting the image information from the captured image of an appropriate size.

The channel identification system 1 transmits a plurality of continuous image information from the robot 10 to the determination server 20. Therefore, the channel identification system 1 increases the possibility that the corresponding image information can be obtained even if the time information does not match between the robot 10 and the determination server 20, for example, even if the frames are different and the images are significantly different. Channel identification accuracy is improved.

Further, the channel identification system 1 attempts to match the image information in a predetermined range before and after the imaging time on the determination server 20. Therefore, even when the robot 10 and the determination server 20 do not have accurate time information, the channel identification system 1 can search for the corresponding image information and identify the broadcast channel with high accuracy.

Further, the channel identification system 1 can calculate the imaging time by subtracting the assumed processing time, communication time, and the like from the time when the image information is received from the robot 10. Therefore, the channel identification system 1 can appropriately select the corresponding image information in the determination server 20 even when the robot 10 does not have accurate time information.

Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments. Further, the effects described in the present embodiment merely list the most preferable effects arising from the present invention, and the effects according to the present invention are not limited to those described in the present embodiment.

In the present embodiment, the control unit 13 controls the orientation and zoom of the image pickup unit 11 (camera) mounted on the robot 10, but the present invention is not limited to this. When the entire television receiver is not shown, or when the shape or size is not appropriate, the position may be changed by moving the entire robot 10 to take an image.
Further, when the quadrangular region determined to be the television receiver cannot be detected, the control unit 13 may search by pointing the image pickup unit 11 in the direction of the sound.
Further, the robot 10 may start the imaging process by the imaging unit 11 with the voice as a trigger to reduce the load in the steady state.

In the present embodiment, the feature amount is used as the image information, but the robot 10 may transmit, for example, the image data itself of the detected area of the television receiver. As a result, more accurate image matching becomes possible when there is a margin in the communication band and immediacy is not required.
On the other hand, when the feature amount is extracted, it is possible to maintain high determination accuracy by matching as compared with the case where the transmission data amount is reduced and the resolution is lowered.

In the present embodiment, the terminal that captures the image of the television receiver has been described as the robot 10, but the robot 10 is not limited to this. You may.
In this case, the user can, for example, shoot a street television with a camera of a mobile device, continuously watch a broadcast program of the identified channel on the mobile device, or acquire related information of the program.
Further, in this case, the portable device may display an adjustment instruction such as the direction or zoom of the camera on the screen or output the sound in order to capture the entire image of the television receiver.

A program for realizing the function of the channel identification system 1 may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read by the computer system and executed.

The term "computer system" as used herein includes hardware such as an OS and peripheral devices. Further, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, or a CD-ROM, or a storage device such as a hard disk built in a computer system.

Furthermore, a "computer-readable recording medium" is a communication line that transmits a program via a network such as the Internet or a communication line such as a telephone line, and dynamically holds the program for a short period of time. It may also include a program that holds a program for a certain period of time, such as a volatile memory inside a computer system that serves as a server or a client in that case. Further, the above-mentioned program may be a program for realizing a part of the above-mentioned functions, and may be a program for realizing the above-mentioned functions in combination with a program already recorded in the computer system. ..

1 channel identification system 10 robot (terminal)
11 Imaging unit 12 Detection unit 13 Control unit 14 Extraction unit 15 Communication unit 20 Judgment server (server)
21 Extraction unit 22 Storage unit 23 Judgment unit

Claims (7)

It has terminals and servers that are connected to each other by communication.
The terminal
Imaging unit and
A detection unit that detects a quadrangular region in which differences of a predetermined value or more are gathered as a region of a television receiver based on the differences between the images captured by the imaging unit.
When there are less than four intersections of line segments that are edges of the region and the region includes the outer edge of the image captured by the imaging unit, it is determined that only a part of the television receiver is imaged. A first control unit that moves the imaging direction of the imaging unit toward the outer edge so that the number of intersections is four .
An extraction unit that extracts image information in the area detected by the detection unit, and an extraction unit.
A communication unit that transmits the image information to the server is provided.
The server
A storage unit that stores image information of broadcast video for each channel in association with the broadcast time,
The image information received from the terminal is compared with the image information stored in the storage unit corresponding to the imaging time of the image information, and the channel selected by the television receiver is determined based on the degree of matching. A channel identification system including a determination unit. The extraction unit extracts the feature points in the region and the feature amount for each feature point as the image information.
The channel identification system according to claim 1, wherein the storage unit stores the feature points of the broadcast video and the feature amount for each feature point. The channel identification system according to claim 1 or 2, wherein the extraction unit extracts image information in the region after converting an image in the region into a predetermined shape corresponding to a television screen. Claims 1 to 3 include a second control unit that adjusts the zoom setting of the imaging unit so that the area becomes a predetermined size or more when the area is less than a predetermined size. The channel identification system described in any of. The channel identification system according to any one of claims 1 to 4, wherein the determination unit extracts image information in a predetermined range before and after the imaging time from the storage unit. Imaging unit and
A detection unit that detects a quadrangular region in which differences of a predetermined value or more are gathered as a region of a television receiver based on the differences between the images captured by the imaging unit.
When there are less than four intersections of line segments that are edges of the region and the region includes the outer edge of the image captured by the imaging unit, it is determined that only a part of the television receiver is imaged. , A control unit that moves the imaging direction of the imaging unit toward the outer edge so that the number of intersections is four .
An extraction unit that extracts image information in the area detected by the detection unit, and an extraction unit.
By transmitting the image information to the server and comparing it with the image information of the broadcast video for each channel stored in the server, the channel selected by the television receiver is determined based on the degree of matching. A channel identification device including a communication unit. For a computer equipped with an imaging unit
Based on the differences between the images captured by the imaging unit, a quadrangular region in which differences of a predetermined value or more are gathered is detected as a region of the television receiver.
When there are less than four intersections of line segments that are edges of the region and the region includes the outer edge of the image captured by the imaging unit, a part of the television receiver so that the number of intersections is four. It is determined that only the image is captured, and the imaging direction of the imaging unit is moved toward the outer edge.
Extract the image information in the detected area and
By transmitting the image information to the server and comparing it with the image information of the broadcast video for each channel stored in the server, the channel selected by the television receiver is determined based on the degree of matching. Channel identification program for.

Images