JP3392914B2 - TV intercom equipment - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a TV intercom apparatus, and more particularly to a TV intercom apparatus for efficiently storing and holding an image picked up by a TV camera attached to an intercom slave in a master or a slave.

[0002]

2. Description of the Related Art In addition to having a television camera on the handset side,
There is a TV intercom device having a function of recording a captured image on the base unit side. An example of such a recording device for a television intercom device is disclosed in Japanese Patent Application Laid-Open No. 62-271581. In the technology disclosed in this publication, a TV camera captures an image of a visitor in conjunction with a calling key of a slave unit, stores the image in a semiconductor memory, and enlarges the image of the visitor to display on the display. . Since the semiconductor memory is used as described above, a large-sized device unlike the recording by the video tape recorder is not required, and the operation such as rewinding of the tape is not necessary and the operation time can be shortened.

[0003]

However, in the TV intercom apparatus disclosed in the above publication, the image is recorded in the semiconductor memory even when the visitor is not visible, so that the required capacity of the semiconductor memory becomes enormous. It will increase the cost. Although a large-capacity memory has been developed due to the improvement of semiconductor technology, there is a limit, and there is a possibility that a situation cannot be recorded when there are many visitors. In particular, in the case of a visitor who walks away by simply pressing the call key, an image that the visitor does not see is recorded in the semiconductor memory, which has the drawback of wasting memory capacity.

Further, in the above-mentioned TV intercom apparatus, when a visitor recorded in a semiconductor memory is monitored on the display, it is necessary to call and confirm the screen one by one captured by the TV camera. However, there is a drawback in that it is necessary to confirm each person, even if a voice message is recorded, as well as it takes time to perform the confirmation work.

The present invention eliminates the above-mentioned drawbacks of the conventional device,
Since the intercom system is intended to communicate with visitors, it makes use of the characteristics of the visitor's photography.
An object of the present invention is to provide a television intercom device that efficiently utilizes a semiconductor memory by extracting an image of a characteristic portion from an image signal.

Further, the present invention is characterized in that when the image and the sound are reproduced,
It is an object of the present invention to provide a television intercom device in which a plurality of captured images are displayed on the screen on the same screen so that the time and effort for reproduction are simplified.

It is another object of the present invention to provide a television intercom apparatus which preferentially reproduces a sound corresponding to an instructed image for a plurality of images on a reproduction screen.

[0008]

In order to achieve the above object, the invention of claim 1 is a slave unit having an image pickup means for picking up an image of a subject, and a means for transmitting and receiving a video signal and an audio signal. In a television intercom apparatus including a master unit having a control unit for transmitting and receiving a video signal and an audio signal from a slave unit, an image display unit for displaying the video signal, and a voice processing unit for inputting and outputting the voice signal, Or a child device, an image processing means for extracting a partial image of a subject imaged by the child device, a scaling means for correcting a signal of the extracted partial image to a size within a predetermined range, and an image corrected to a predetermined size It includes a semiconductor memory for storing the signals and audio signals, from the extracted partial image, the mouth of the subject person
The image signal in this area is detected assuming the area
When it detects that there is a corresponding audio signal
It is configured to be stored in body memory .

In the television interphone apparatus according to the second aspect, the image processing means extracts a partial image, for example, the face of a person, based on the difference signal formed from different frame image signals of the same subject.

According to another aspect of the television intercom apparatus of the present invention, the scaling means corrects the partial images of a plurality of different subjects so that the partial images are simultaneously displayed on the same screen.

According to another aspect of the television intercom apparatus of the present invention, the semiconductor memory stores the image signal and the audio signal in association with each other.
Multiple playback images displayed simultaneously on the screen.
The audio corresponding to the specified image.
It is supposed to be played.

The television intercom apparatus according to claim 5 is an image display device.
The visit date and time of the visitor will also be displayed when the image is replayed.
ing.

[0013]

According to the above construction, it is not necessary to record all the images photographed by the television camera in the semiconductor memory,
The characteristic portion is extracted and stored. For example, when the subject is a person, the characteristic face portion is extracted, so that waste of memory capacity can be prevented.

Further, the partial images stored in the semiconductor memory can be confirmed efficiently because the images of a plurality of subjects are simultaneously displayed on the same screen.

Further, since the reproduction of the audio signal is performed by pointing the image displayed on the screen, it is possible to reproduce only the necessary messages or in the order of priority.

[0016]

1 is a block diagram of a television intercom apparatus according to an embodiment of the present invention. The TV intercom apparatus includes a slave unit 20 installed outside the house and operated by a visitor, and a master unit 21 connected to the slave unit 20 via a cable and installed in the house. Here, for ease of understanding, the power supplied from the base unit 21 to the handset 20 and the base unit /
The description of the operation keys of the child device is omitted.

The handset 20 is an image pickup means for taking a picture of a visitor, a television camera 1, a circuit for multiplexing a video signal from the television camera 1 and an audio signal from the microphone 19, and an audio from the parent device 21. It is configured to include a multiplexing processing circuit 9 including a separation circuit that outputs a signal to the speaker 18, a speaker 18 that outputs a voice signal, and a microphone 19 that inputs a voice.

On the other hand, the master unit 21 separates the video signals multiplexed and transmitted from the slave unit 20 and the audio signal input from the later-described microphone 17 on the master unit 21 side via the analog audio processing circuit 15. A control circuit 2 for multiplexing and transmitting to the slave unit 20 is included. Video and audio signals are input to the control circuit 2 from the slave unit 20 as analog signals.
Of these analog signals, the video signal is supplied to the digital image processing means 22 and the audio signal is supplied to the digital audio processing means 23.

The digital image processing means 22 is an image A / D conversion section 3 for converting an analog video signal into a digital signal,
The converted digital video output signal is subjected to processing such as image extraction and sizing according to a pre-programmed algorithm which will be described later, and a digital image processing section 5 for storing and accumulating digital image data connected to the digital image processing section 5 And an image D / A conversion unit 6 for converting the digital video signal processed by the digital image processing unit 5 into an analog video signal.

The analog video signal output from the digital image processing means 22 is converted into a displayable signal via the image display circuit 7 and given to the display 8.

The digital voice processing means 23 is a voice A / D converter 1 for converting an analog voice signal into a digital signal.
1. Digital audio processing unit 12 for compressing and recording the converted digital audio output signal and expanding the recorded digital audio compressed signal, audio for recording digital audio data connected to the digital audio processing unit 12 Memory 1
4. Audio D / A conversion unit 1 for converting the digital audio signal processed by the digital audio processing unit 12 into an analog audio signal
Including 3. The audio memory 14 and the image memory 4 are constituted by an integrated or separate semiconductor memory, and corresponding audio data and image data are stored in association with each other.

The analog voice signal output from the digital voice processing means 23 is output to the speaker 16 via the voice processing circuit 15, and the voice of the visitor is output. The microphone 17 on the base unit side is also connected to the voice processing circuit 15, the voice of the corresponding person is taken in, and voice signal processing for correspondence with the handset 20 is performed.

The television intercom system includes a CPU (central control unit) 10 including a microcomputer, a control circuit 2,
Digital image processing means 22 and digital audio processing means 2
3, controlling the image display circuit 7 and the audio processing circuit 15,
The operation control for executing the intercom function including the image processing described below is commanded.

FIG. 2 is a block diagram showing an embodiment of the digital image processing unit 5, FIG. 3 is an explanatory view of a method of extracting a face portion including a head from a captured image of a visitor according to the present invention, and FIG. It is explanatory drawing of the image signal processing step which converts a video input signal into the signal stored in an image memory.

The video transmitted from the slave unit 20 is converted into digital data by the image A / D conversion unit 3 and input to the digital image processing unit 5 to extract a partial image to be stored in the image memory 4. Perform the work of.

The digital image data is first stored in the first frame memory 5h via the frame difference detection section 5a provided in the digital image processing section 5. The frame difference detection unit 5a is a means for extracting a partial video of the visitor from the entire image captured by the camera 1. The background image is a still image, while the visitor is a moving image with some movement. Taken in 1. Therefore, the frame difference detection unit 5
“a” detects the difference from the comparison between the image data stored in the first frame memory 5h at the previous time and the image data input at the time when one frame has elapsed, and detects the edge of the moving portion. . The differential data of the moving portion formed by the detection unit 5a is binarized by the binarization processing unit 5b, and the binarized data is labeled by the binarization process for adjustment, and noise is removed. The position data of the visitor's face (head) is calculated from the difference detection by signal processing, a face area is set, and the image signal contained in this area is stored in the second frame memory 5g. The data of the second frame memory 5g becomes a partial image extracted from the captured image.

That is, the face coordinate calculation unit 5c connected to the second frame memory 5g calculates the coordinates by extracting the contour of the face based on the binarized difference data. FIG. 4 shows the step of calculating the characteristic coordinates of the face portion from the input video signal and setting the image area to be stored in the image memory.

In the face portion extracting operation performed by the face coordinate calculating section 5c, a difference between image signals of two different frames is obtained, and the difference is binarized by the binarizing processing section 5b.
As shown in (a), scanning is performed with predetermined timing signals in the X-axis direction and the Y-axis direction. First, the Y coordinate of the pixel in which the difference appears is recognized as the crown Yt. As the scanning progresses, pixels corresponding to the image of the moving face with respect to the static background image appear after this coordinate, but when scanning from the left in FIG. The pixel appearing at the coordinate position is recognized as the face left coordinate Xl. Similarly, the pixel appearing at the rightmost X coordinate position in the process of continuing scanning is recognized as the face right coordinate Xr.

As for the coordinates of the lower part of the face, since the visitor's torso image appears continuously, the coordinates Xl and Xr of the X axis are displayed.
The lower face Y-coordinate Yu is calculated from the distance data and the statistically obtained face length coefficient Fl. That is, Yu = (Xr-X
1) × Fl.

Next, the center point Pc of the face is calculated from the above coordinate data.
The coordinates of (Xc, Yc) are calculated according to the following formula. That is, Xc = (Xr-Xl) / 2 + Xr, and Yc = (Y
u-Yt) / 2 + Yt.

By the above arithmetic processing, the coordinates of each four points of the most protruding part of the face can be determined. The coordinates of the four points are the top Pt (Xc, Yt) and the left face Pl (Xl, Y
l), right face Pr (Xr, Yr), lower face Pu (Xc, Y)
u). The area surrounded by these four points is the face area,
The image signal in the area surrounded by the face area is transferred to the second frame memory 5g to hold the partial image.

The size of the image captured by the camera 1 varies depending on the position where the visitor faces the camera 1. In order to efficiently store the image data in the semiconductor memory, size correction is performed so that the size of the face part extracted from the captured image becomes approximately the same size. The size correction is performed by the face area scaling unit 5d, and first, before the image size is enlarged or reduced, the aspect ratio is 4: 3 including the face area as shown in FIG. 3B. Set the area that surrounds the image. The coordinates of four points indicating this area are obtained from the calculated center coordinates of the face and the face length.

Since the face length L is L = Yu-Yt, the four coordinates P1 (X
1, Y1), P2 (X2, Y2), P3 (X3, Y
3) and P4 (X4, Y4) can be calculated from FIG. 5 as follows. X1 = Xc-2 × L / 3, Y1 = Yc-L / 2,
X2 = Xc + 2 × L / 3, Y2 = Y1, X3 = X1, Y
3 = Yc + L / 2, X4 = X2, Y4 = Y3.

The face area scaling unit 5d enlarges the cut-out face area so that it fits within a predetermined range of size.
Perform the reduction operation. For example, the size correction when four screens are displayed on the screen configured by 320 × 240 pixels will be described. The description is an example for facilitating understanding, and there is no limitation on the number of screen pixels and the number of screens to be displayed.

When four screens are displayed in a size of 320 × 240 pixels, the screen size assigned to one screen is 16
The number of pixels is 0 × 120. Therefore, when the area surrounding the face is deviated from this 160 × 120 pixels, 160 × 1
The size is adjusted so that it can be displayed in 20 pixels. The fitting method is the length X in the X-axis and Y-axis directions of the area surrounding the face.
Based on L = X2-X1 and YL = Y3-Y1, this is performed by enlarging and reducing operations. FIG. 6 is a flowchart showing the steps of the fitting method, where the face area is already 4: 3.
Since it is cut out with, the length XL in the X-axis direction is 160
Set it so that the ratio of 160 to XL compared with pixels is an even number, and if XL is small, expand by interpolating according to the insufficient pixels, and if XL is large, reduce by thinning pixels to reduce the range. Adjust to the inside size. By this processing, the image is corrected to have a uniform size regardless of the size of the image at the time of shooting. After performing the fitting process, when the pixel size of the face area is not equal to 160 × 120 pixels, the remaining pixels are filled with a constant black level, for example.

The partial image signal read from the second frame memory 5g is subjected to the above size correction by the scaling unit 5d, and the partial image data having an even size is transferred to the image memory 4 via the memory control unit 5e. The image of the extracted face part of the visitor is stored. Since the image memory 4 is a semiconductor memory and the voice memory 14 is integrally configured, it also stores digital voice data transferred from the digital voice processing means 23, and the corresponding image data and voice data are shown in FIG. As shown in the memory map, the visitor number and the like are stored as a set as common data.

The voice signal is guided to the message "I'm out of office now. Please give me a message", which is emitted when the visitor presses the call button (not shown) of the child device 20. It is caused by the action of a visitor telling a message. The emitted voice is input to the microphone 19,
It is input to the audio A / D conversion unit 11 through the multiplex processing circuit 9 and the television interphone control circuit 2. The audio A / D converter 11 converts an analog audio signal into a digital PCM signal. The PCM signal is processed by ADPC in the digital audio processing unit 12.
The voice is compressed by a method such as the M method and stored in the voice memory 14 together with the same visitor number as the one attached to the image.

To reproduce the data stored in the semiconductor memory, the image data stored in the image memory 4 of the semiconductor memory is called when the confirmation button (not shown) of the master device 21 is pressed, and the read image is read. The data is expanded in the graphic memory by the digital image processing means 5 so as to display a plurality of images on one display screen as shown in FIG. 9, and the image display circuit 7 is operated through the image D / A converter 6 in accordance with the access instruction.
Is output to. FIG. 10A is a diagram schematically showing a data storage state of the graphic memory, which is indicated by 1 of the display 8.
Data is stored so that it is convenient for the operation of displaying four images on the screen. FIG. 10B shows the storage status of data in the image memory.

The image display circuit 7 converts the input image data into a video signal, allocates the screen of the display 8 to four areas, and displays a four-screen face image on the screen in ascending order of visitor number. Since the image data is normalized to the size of the display 8, a plurality of screens of each visitor can be efficiently displayed at the same time with the screens of the display 8 being evenly allocated. When the image is reproduced, the visitor number and date and time are also displayed.

The digital image processing means 5 has a built-in arrangement for arranging a cursor on the graphic screen, and a signal for operating a cursor key (not shown) provided on the operation panel of the master unit 21 is input to the CPU 10 for operation. The cursor position according to the content is set by the control of the CPU 10. When the play button of the base unit is pressed at the position where the cursor is stopped, the cursor position is captured by the CPU 10 and recognized, and the visitor's message corresponding to the image appearing at the cursor position is paired with the image data. It is read from the stored voice memory 14 and reproduced. The voice data in the voice memory 14 is output to the voice processing circuit 15 via the voice D / A converter 13, and the voice is notified from the speaker 16.

Regarding the voice of the visitor, it is possible to detect the movement of the visitor's mouth without requiring guidance by operating the call button and automatically record the voice in the memory when there is movement. That is, when the face coordinate detection unit 5c detects the vertex coordinates Pt (Xc, Yt), the mouth area is given a length statistically set in advance from the vertex coordinates,
Determine based on the head coordinates. The movement is discriminated from the difference detection of the image signals in the mouth region, and if there is movement, it is determined that the visitor is transmitting the message and the voice is stored in the voice memory 14 by the above-mentioned operation. In this case, only the voice of the visitor who generated the voice is recorded, and the use of the memory area becomes more efficient.

Although the embodiment has been described with reference to the intercom device having the image processing means in the base unit, it may be provided in the handset side.

[0043]

As described above, according to the present invention, since the characteristic portion of the subject is cut out and stored in the semiconductor memory, the memory capacity can be omitted and the object can be efficiently used. For example, if the subject is a person, the most valuable face part of the visitor is extracted and stored in memory , and
Only when there is movement in the mouth area from the image signal of the mouth area
Since the voice data is stored in the memory, efficient use of the memory is possible.
Can be used even more.

Further, in reproducing the data stored in the memory, since a plurality of screens stored at different times are simultaneously displayed on the same screen of the display, a plurality of pieces of information can be confirmed at the same time, which makes the confirmation work troublesome. I can omit it.

By pointing an image to a plurality of images on the playback screen, the corresponding message message can be played back, and the voice message can be selected and listened to.

[Brief description of drawings]

FIG. 1 is a block diagram of a television intercom apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a specific configuration of a digital image processing unit.

FIG. 3 is a coordinate relationship diagram on a screen for explaining a partial image extraction method and an image area cutout.

FIG. 4 is a flowchart illustrating an image processing step of a video input signal.

FIG. 5 is a coordinate relationship diagram for size correction of a partial image.

FIG. 6 is a flowchart for correcting the size of a partial image.

FIG. 7 is a memory map of image data and audio data.

FIG. 8 is a diagram of a reproduction screen on the display screen of the parent device.

FIG. 9 is a diagram showing graphic data on a screen.

FIG. 10 is a diagram showing a data storage state of a graphic memory and an image memory.

[Explanation of symbols]

1 TV camera 4 image memory 5 Digital image processing means 5a frame difference detector 5b Binarization processing unit 5c Face coordinate calculation unit 5d scaling section 8 display 10 CPU 12 Digital audio processor 14 Voice memory 20 cordless handsets 21 Base unit

Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N 7/18 H04M 9/00

Claims (5)

(57) [Claims] 1. A handset having an image pickup means for picking up an image of a subject, a means for transmitting and receiving a video signal and an audio signal , a control means for transmitting and receiving a video signal and an audio signal from the handset, and a video signal In a television interphone device including a base unit having an image display unit for displaying and a voice processing unit for inputting and outputting a sound signal, the base unit or the slave unit is an image for extracting a partial image of a subject captured by the slave unit. and processing means, and scaling means for modifying the signal of the extracted partial image to the size of the predetermined range includes a semiconductor memory for storing image signals and audio signals that have been fixed to a predetermined size, from the extracted partial image Assuming the subject's mouth area
Image signals in the lever area are detected to detect movement.
Storing corresponding audio signal in the semiconductor memory
A TV intercom device that is characterized by: 2. The television intercom apparatus according to claim 1, wherein the image processing means extracts a partial image from a differential signal of image signals of different frames of the same subject. 3. The television intercom apparatus according to claim 1, wherein the scaling means corrects the partial images of a plurality of different subjects so that the partial images are simultaneously displayed on the same screen. 4. A semiconductor memory stores image signals and audio signals in association with each other, and stores a plurality of replays simultaneously displayed on the screen.
If you specify the raw image with the cursor,
The television intercom device according to claim 1, wherein a corresponding sound is reproduced . 5. The visit date and time of the visitor when the image is reproduced
The television intercom device according to claim 1, wherein the television intercom device is displayed .