JPH0670313A - Video telephone system - Google Patents

Abstract

PURPOSE:To eliminate a background part whose display is not required in a picture displayed on a monitor of an opposite party. CONSTITUTION:Processing circuits 11, 12, 13 extract a hair portion, a face portion and a cloth portion of a portrait, and a processing circuit 14 extracts a background portion from a video signal obtained through the image pickup by a CCD camera 2 respectively and the processing circuits 11-13 adjusts a brightness level, a hue and a saturation of the extracted portions. A signal with prescribed brightness level, hue and saturation is outputted from the processing circuit 14. On the other hand, a key signal of the hair portion, a key signal of the face portion and a key signal of the cloth portion are respectively generated by a binary circuit 16 and a binary picture processing circuit 19, a binary circuit 17 and a binary picture processing circuit 20, and a binary circuit 18 and a binary picture processing circuit 21, a synthesis circuit 15 selects and synthesizes output signals from the processing circuits 11-13 and an output signal from the processing circuit 14 for the background portion respectively and the synthesized signal is sent to an opposite party via an input output terminal 10.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a videophone device, and more particularly to a videophone device having an image processing function such as background erasing.

[0002]

2. Description of the Related Art In a videophone device, an image taken by a camera is normally transmitted to the other party as it is. However, when the room is messy, there is a case where the background part of the caller is desired to be erased. An example of a videophone device having such a function is described in, for example, Japanese Patent Application Laid-Open No. 57-129076, in which a background image captured by a camera is stored in a memory and an image captured by the camera is stored. By subtracting the background image recorded in this memory from the background image, the background image is erased.

[0003]

However, in the above-described conventional videophone apparatus, in order to always erase the background image from the image displayed on the screen of the other party during the call, the background image is removed every time the call is made. There is a problem that it is necessary to input the image to the memory, which makes the usability poor.

Contrary to erasing the background image, there are cases where it is desired to display the caller in a silhouette, to change the color of clothes, or to improve the complexion, but in the conventional videophone device. , It was not possible to realize such a function.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above problem and to provide a video telephone device which can realize an image processing function capable of easily erasing a background image and displaying a desired color.

[0006]

In order to achieve the above object, the present invention comprises an extracting means for extracting a portion having luminance and chromaticity within a predetermined range from an image input by a camera,
And a processing means for applying independent signal processing to each of the extracted portions.

[0007]

The extracting means extracts, for example, a predetermined portion of a person image from the video signal input from the camera. The extraction condition at this time is to set the luminance and chromaticity ranges of the portion corresponding to this person, and to extract the luminance and chromaticity portions within the set range from the input video signal.
The processing unit performs normal signal processing on the portion extracted by the extracting unit. A portion other than the portion extracted by the extracting means is a background image, and fade processing is performed. In this way, only the caller can be photographed and the background can be erased.

[0008]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a videophone device according to the present invention, in which 1 is a lens, 2 is a CCD camera, 3 is an amplifier, and 4 is an A / D (analog / digital).
Converter, 5 YC (luminance, color signal) generation circuit, 6 C modulation circuit, 7 D / A (digital / analog) converter, 8 monitor, 9 codec, 10 input / output terminals, 11-14 Processing circuit, 15 is a synthesis circuit, 16 to 1
8 is a binarization circuit, 19 to 21 are binary image processing circuits, 22
Is a latch circuit, and 23 is an output terminal.

The operation of this embodiment will be described below, but here, up to three independent parts of the video can be extracted.

In FIG. 1, light from a subject incident through a lens 1 is received by a CCD camera 2 and photoelectrically converted, and an electric signal is output. This electric signal is amplified by the amplifier 3 and then converted into a digital signal by the A / D converter 4. This digital signal is the YC generation circuit 5
And a digital luminance signal Y and a color signal C (hereinafter collectively referred to as YC signal) are generated, and are supplied to the processing circuits 11 to 14, the binarization circuits 16 to 18, and the latch circuit 22. In the processing circuits 11 to 14, this YC
Under the control of the microcomputer 24, the signal is converted into a signal having a designated arbitrary brightness level, hue, and saturation. On the other hand, in the binarization circuits 16 to 18, the YC signal is transmitted from the microcomputer 2
It is binarized according to the binarization condition set in No. 4, and is supplied to the binary image processing circuits 19 to 21. The latch 22 can sample and hold the YC signal. The luminance and chromaticity (or hue and saturation) distribution of the video signal can be known from the sample-held luminance signal Y and color signal C, and the binarization condition can be set by this. it can.

The binary image processing circuits 19 to 21 are circuits for performing a filtering process, and each of the binarization circuits 16 to 1
From among the parts including the specified color of the video signal by 8,
Extract only the part you actually want to process. A sequential growth method or the like can be used as the extraction method, and any portion that satisfies the color condition is detected. The successive growth method is described in, for example, Japanese Patent Application No. 2-325643.

The signal thus extracted becomes a key signal. This key signal is a signal indicating what kind of processing is to be performed for which part. The synthesizing circuit 15 synthesizes the signals that have been processed differently by the processing circuits 11 to 14. The combined signal is compressed by the codec 9, output from the input / output terminal (transmission I / F) 10 and sent to the other party. The codec 9 is a circuit that performs digital compression or decompression according to the H261 recommendation, for example.

The compressed signal sent and received from the other party is input from the input / output terminal 10, restored by the codec 9, C-modulated by the C modulation circuit 6, and then the D / A converter 7 Is converted into an analog signal and supplied to the monitor 8. As a result, the video of the other party is displayed on the monitor.

Next, in such an embodiment, a method of erasing a background image when a person image is present in an image captured by a CCD camera will be described with reference to FIG. 2 and accompanying operation of each section of FIG. Will be described more specifically.
However, Fig. 2 shows the input image (image taken by CCD camera)
FIG. 3 is a diagram showing an output image (video sent to the other party) obtained by performing a fade process on the background video.

When a portion other than the human image portion in the input image is faded, the processing circuits 11 to 13 of the YC signal human image portion (FIG. 2A) are three portions of hair, face and clothes. It is divided into and extracted, and each is processed independently. The luminance signal Y and the color signal C of the respective extracted portions are processed by the processing circuits 11 to 1.
Only color information such as hair color, face color, clothing color, and other colors is sent to 4. For example, in the processing circuit 11, a range representing the color of a person's hair of the color signal C of the YC signal is set, and a portion within the set range is extracted from the YC signal from the YC generation circuit 5. The brightness level, hue, and saturation of that portion are processed to have predetermined arbitrary values. Similarly, in the processing circuit 12, the face portion is extracted and in the processing circuit 13, the clothing portion is extracted and processed. In the processing circuit 14, the background portion has a constant brightness,
Since the color fades, a YC signal with a constant brightness level, hue, and saturation is output.

On the other hand, the binarization circuits 16 to 18 are supplied with the luminance signal Y and the color signal C based on the above-mentioned binarization conditions.
From the above, the part of the person image that satisfies the color conditions of hair, face, and clothes is detected. Here, it is assumed that the binarization circuit 16 extracts the hair portion of the human image, the binarization circuit 17 extracts the face portion of the human image, and the binarization circuit 18 extracts the clothing portion of the human image. To do. At this time, since the detection is controlled by the microcomputer 24 and the OR circuit, a plurality of conditions can be designated. With this, for example, when detecting a face part,
Non-skin-colored parts such as eyebrows, eyes and mouth can also be extracted at once.

The binary image processing circuits 19 to 21 extract from the signals of the portions extracted by the binarization circuits 16 to 18 the key signals k1, k2, which indicate the temporal range of these extracted portions.
Generate k3. These key signals k1, k2, k3 are supplied to the synthesizing circuit 15.

The synthesis circuit 15 receives the key signals k1, k
2, k3, one of the output signals of the processing circuits 11 to 14 is selected, and the selected portions are combined and output. That is, assuming that the key signal k1 is supplied from the binary image processing circuit 19, the synthesizing circuit 15 selects the output signal of the processing circuit 11. Therefore, the hair portion of the human image is extracted from the YC signal. If the key signal k2 is supplied from the binary image processing circuit 19, the synthesizing circuit 15 outputs the output signal of the processing circuit 12, and if the key signal k3 is supplied from the binary image processing circuit 19, the synthesizing circuit 15 The synthesizing circuit 15 selects the output signal of the processing circuit 13, respectively. As a result, the face and clothes of the person image are extracted. When neither of the key signals k1, k2, k3 is supplied, the synthesizing circuit 15 selects the output signal of the processing circuit 14.

The synthesizing circuit 15 synthesizes these selected signals and sends them to the other party through the codec 9 and the input / output terminal 10. In the image shown in FIG. The brightness level of each of the hair portion, the face portion, and the clothing portion of the image processed by the processing circuits 12 and 13,
Displayed in hue and saturation. Further, the background portion is displayed with a constant brightness and color set by the processing circuit 14, and the background is faded.

As described above, in this embodiment, it is not necessary to store the background image in the memory in order to erase the background, and therefore the background image can be easily erased.

Next, another embodiment of the videophone device according to the present invention will be described with reference to FIGS. In the embodiment described above, the background image is faded with a constant brightness and color, but in this embodiment, conversely,
The background image is displayed as it is, and the silhouette of the person is displayed by fading.

In FIG. 1, the processing circuits 11 to 13 respectively extract the hair, face and clothes of the person image as described above. The YC signal of the same constant brightness level, hue, and saturation is output in the part of. Therefore, although the display color of the YC signals output from them is the same, this color can be black.

As a result, the input image shown in FIG. 3A is displayed as an output image, and the person image is displayed as a silhouette on the background image as shown in FIG. 3B. Therefore, it is convenient when the user does not want to display his / her face on the screen when making a videophone call.

In this case, the processing circuits 11 to 14 are controlled by a user's instruction operation, and it is possible to select whether to display the person image as it is or to display it as a silhouette as described above.

FIG. 5 is a block diagram showing still another embodiment of the video telephone apparatus according to the present invention. Reference numerals 24 and 25 are switches, and the portions corresponding to those in FIG. In this embodiment, the color of the hair, face and clothes of the human figure can be freely changed.

In FIG. 4, the processing circuits 11, 12, 13 are shown.
As shown in FIG. 5B, each of the YC signals representing the input image shown in FIG. 5A represents a hair portion A, a face portion B, and a clothing portion C, as shown in FIG. 5B. The brightness levels, hues, and saturations are converted so that each of them is displayed in a color arbitrarily determined by the user (FIG. 5C). Therefore, as shown in FIG. 5D, an output image of the person's hair, face, and clothes in any color different from the input image shown in FIG. 5A is displayed on the monitor of the other party. It will be.

The switches 24 and 25 are switched by a user's instruction operation.
When 4, 25 are closed on the Q side, the same image signal as in the previous embodiment is sent from the input / output terminal 10 to the other side, and the image signal from the other side is input from the input / output terminal 10 to monitor 8. However, when the switches 24 and 25 are closed to the P side, the output signal of the synthesizing circuit 15 is supplied to the monitor 8 via the switches 24 and 25 and the D / A converter 7 to display its own image. Is displayed. Therefore, before the call, the user closes the switches 24 and 25 to the P side so that the processing circuits 11 to 11 can be seen while watching the monitor 8.
13 can be adjusted to adjust the hair, face, and clothes colors of the self image to your liking.

As described above, in this embodiment, for example, the hair of the person image can be displayed in black and the skin color can be displayed in white on the monitor on the other side, and the color of the clothes can be changed according to preference. can do. Other parts are similar to those of the first embodiment.

In this embodiment, the switches 24 and 25 are switched by the user's operation, but they may be switched by another method. For example, until the user removes the handset and inputs the telephone number of the other party, the switches 24 and 25 are closed to the P side. When the telephone number of the other party is inputted, the switches 24 and 25 are automatically switched to the Q side. You may make it switch to.

[0030]

As described above, according to the present invention,
You can easily remove from the screen those areas that you do not want to see, such as the background of the caller, that you do not want to see.

Further, according to the present invention, even if the face is not desired to be displayed on the screen, it can be displayed as a silhouette, and personal privacy is protected.

Further, according to the present invention, it is possible to display the hair, face, clothes and the like of the human figure in arbitrary colors, and display the colors of the face and clothes as one's favorite color. You can Therefore, even if oneself is transferred to the other party, he / she can call without any anxiety.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a videophone device according to the present invention.

FIG. 2 is a diagram showing an input image and an image displayed on the other side in the embodiment shown in FIG.

FIG. 3 is a diagram showing an input image and an image displayed on the other side in another embodiment of the videophone device according to the present invention.

FIG. 4 is a block diagram showing still another embodiment of the videophone device according to the present invention.

5 is a diagram showing an operation of the embodiment shown in FIG.

[Explanation of symbols]

 1 lens 2 CCD camera 3 A / D converter 4 amplifier 5 YC generation circuit 6 C modulation circuit 7 D / A converter 8 monitor 9 compression / decompression circuit 10 input / output terminals 11-14 processing circuit 15 synthesis circuit 16-18 binarization Circuit 19 to 21 Binary image processing circuit 22 Latch circuit 23 Sample and hold 24 Microcomputer

Claims (6)

[Claims] 1. A videophone comprising image pickup means for photographing a caller, means for encoding and transmitting an image pickup signal, demodulating the received signal, and a monitor for displaying the demodulated video signal. A videophone device, comprising: an extracting means for extracting a specific portion of an image input by the image pickup means; and a processing means for performing different signal processing for each portion extracted by the extracting means. 2. The videophone device according to claim 1, wherein the extraction unit is configured to extract a portion having a luminance, a hue, and a color saturation within a certain range. 3. The method according to claim 1 or 2, wherein the extraction means extracts a portion corresponding to a person, and the processing means performs a fade process on a background portion of the person extracted by the extraction means. A videophone device characterized by being configured. 4. The method according to claim 1, wherein the extracting means extracts a person corresponding to the person, and the processing means performs a fade process on the person extracted by the extracting means to output a silhouette image of the person. A videophone device characterized in that 5. The videophone according to claim 1, wherein the processing means is configured to replace the brightness or color of the specific portion extracted by the extracting means with any brightness or color. apparatus. 6. A videophone apparatus comprising image pickup means for photographing a caller, means for encoding and transmitting an image pickup signal and demodulating the received signal, and a monitor for displaying the demodulated video signal. In the above, an extraction means for extracting a portion corresponding to a person from an image input by the image pickup means, and a processing means for performing a fade process on the extracted portion and outputting the extracted portion are provided. Videophone device.