JPH06268998A - Still image color videophone - Google Patents

Abstract

(57) [Summary] [Structure] The reference registration pattern was stored in the secondary storage medium of the videophone, and the content was called as appropriate to provide a function for comparison with the current image. [Effect] Error recognition and coincidence matching can be performed in the same circuit according to the comparison timing, improving usability.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color videophone apparatus capable of automatically recognizing and reporting a large change situation or coincidence matching between a video object and an audio object quickly.

[0002]

2. Description of the Related Art Conventionally, in the case of a still image videophone, a means for detecting a signal change in a monitoring application of a remote monitoring videophone apparatus includes a plurality of current camera video signals and audio signals in a predetermined time interval sequentially in units of frames. Whether to capture the image and voice memory and compare the contents of the previous frame image and voice memory with the contents of the current memory to detect whether there is a difference, or in the case of a video-phone call, set a preset time Whether to capture only the change in the moving portion of the video signal or to detect only the change in the volume at each interval. According to this method, movement or voice change is detected in real time, so there is a feature that even minute changes can be detected, but applications that only need to detect large changes include many redundant circuits. However, there is a drawback that the circuit configuration is complicated and costly. Also, if only the movement change of the object needs to be detected, it is not necessary to detect the color change of the color videophone, and although the color information of the image is not necessary, the change must be detected together with it. Of course, this also made the circuit configuration complicated.

Further, the above-mentioned conventional system does not include the function of the comparison / matching circuit such as person verification with the already registered standard pattern or the like. According to the present invention, it is possible to carry out a comparison error and a comparison error between the contents registered in the secondary storage medium and the currently input video and audio by mode switching.

Incidentally, the ones related to this type of device are:
JP-A-3-178288 and JP-A-3-249860 are mentioned.

[0005]

According to the prior art, since the movement or volume change input in real time is detected, it is effective for detecting a very small change. Since it is always necessary to consider error recognition without error by considering changes in shadows, etc., many redundant circuits are included in applications that need to detect only large changes, so the circuit configuration is complicated and costly. was there. Further, in the conventional method, the comparison error between the current images is always detected even if there is a time difference, and therefore the already registered standard pattern cannot be compared with the current image. Therefore, there is an inconvenience that it cannot be applied to the comparison and collation having the opposite meaning to the comparison error detection such as the person collation check.

An object of the present invention is to provide a color still image videophone apparatus capable of both comparison error recognition and comparison comparison between an already registered image and voice pattern and a video and voice constantly input from a camera. To do.

[0007]

In order to solve the above problems, timings at which comparison error recognition, reference image patterns for comparison and matching and reference voice patterns are stored in plural removable secondary storage media and comparison is made ， According to the comparison mode, it is called to the built-in memory of the main body, and error recognition and comparison match detection are adopted. When selecting one of the contents stored in the secondary storage medium and calling it to the internal memory of the main body, it is possible to quickly realize precise error recognition, coarse error recognition that is not comparison match, and comparison match detection. For this reason, the reference image and the reference sound are binarized and stored before being stored in the built-in memory as the threshold value set in advance. At the same time, the video and audio to be compared are binarized and stored in the internal memory of the main unit. Next, a function for comparing the contents of the internal memory of the main body in which the contents of the reference image and the reference sound are stored is provided, and the result of error recognition or coincidence recognition is preliminarily stored in the secondary storage medium according to the comparison mode. A function is provided that allows the stored image data and voice message to be transmitted to a specified telephone destination. In addition, the phone will be equipped with a loudspeaker to provide voice messages directly from the phone.

[0008]

As a result, by simply changing the mode, for example, the error recognition mode is used when an abnormality such as a fire is detected, and the comparison match mode is used at the time of person verification and fingerprint verification. It can be used properly. Further, the implementation circuit may have almost the same circuit configuration in both modes, and the conflicting comparison depending on whether only the result of the comparison circuit is used in the positive logic or the negative logic can be realized in the almost common circuit configuration. Further, since the contents of the reference image and the reference voice are stored in a removable secondary (non-volatile) storage medium, a plurality of them can be selected and called out according to the purpose, which is convenient. . Further, since the comparison error recognition or the comparison coincidence recognition is performed not by comparing the color image itself or the sound itself but by the binarized image and sound, this is also a factor to simplify the circuit configuration. .

The above is largely due to the fact that the reference image and the reference voice are stored in the secondary storage medium, and the contents are retrieved and compared each time.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram of a still picture color video telephone apparatus embodying the present invention, FIG. 2 is an operation flowchart of the circuit configuration diagram of FIG. 1, and FIG. 3 uses the error recognition mode of the present invention. An application example of anomaly detection is shown, (a) of the figure shows a reference image and a registered voice (silence) already registered in the secondary storage device of the apparatus as a normal state, and (b) of the figure is captured by a camera. FIG. 4 shows the current abnormal state, and FIG. 4 shows the abnormality detection method and the detection result of the application example of FIG.

FIG. 5 is an application example of person match detection using the comparison match mode of the present invention. FIG. 5A shows a reference person image registered in the secondary storage memory, and FIG. Shows an example in which the same person is in front of the camera of the videophone with a different facial expression from the reference person image. FIG. 6 shows a person coincidence detection method and a detection result of the application example of FIG.

The still image color videophone device 1 shown in FIG. 1 includes a video camera 2, an IC memory card 6 as a secondary storage device 6, a videophone switch 12, a videophone function circuit 15, a speaker 17, a microphone 18, and a reference. It is composed of an error recognition and collation / match comparison circuit 23 for comparing the image and voice with the current input image and voice.

First, the operation when applied to fire abnormality detection will be described with reference to FIGS. 1, 2, 3, and 4.

The reference image 31 and the reference voice (silent) are registered as follows. The telephone switch button 12 is used to perform a predetermined operation to register the normal kitchen image and voice input from the camera 2 and the microphone 3 in the IC memory card 6 which is the secondary storage memory.
The registered reference image 31 and reference voice (silence) are appropriately I
It is stored in the reference image and voice registration memory 8 via the C memory card interface circuit 7, the input selection circuit 5, and the binarization circuit 22 (step 41).

Error recognition is performed as follows. Via the camera 2, the camera signal input circuit 3, and the binarization circuit 22, the images 31, 32, and audio 34 of the current state of FIG.
Are simultaneously taken into the current image and audio recording memory 9 (step 42), and the comparison error recognition between the reference data and the current input data and the comparison / matching circuit 10 compare the contents of the reference memory and the current recording memory (step 43). Is done. At this time, as shown in FIGS. 4 (a), (b), and (c),
A threshold level is provided as an error index for determining the disagreement, and when the area of the binarized image for the image and the sound pressure level for the voice exceed the threshold level, the reference image or the voice is different from the current image and the voice. It is determined (steps 44, 46, 47).

In this case, as shown in FIG. 3 (b), since the gas stove has a fire, depending on the setting mode,
A voice message that has already been registered in the secondary storage memory, for example, "fire occurred", or an image of "evacuation route" as an image message is automatically transmitted to a telephone number that is also registered in the secondary storage memory, or the speaker 17 is used. From the user, interrupt and display on the TV screen 13 (step 4)
8, 49, 50).

Next, in the case of application to person verification,
The operation will be described with reference to FIGS. 1, 2, 5, and 6.

As described in the abnormality detection, the reference image 51 and the reference voice 52 are registered in the IC memory card 6 which is the secondary storage memory in the same procedure. The registered reference image 51 and reference voice 52 are appropriately used as the IC memory card interface circuit 7, the input selection circuit 5, and the binarization circuit 22.
It is stored in the reference image and voice registration memory 8 via (step 41).

The comparison and collation is performed as follows. Via the camera 2, the camera signal input circuit 3 and the binarization circuit 22, the image 53 and the audio 54 of the current state in FIG. 5B are simultaneously taken in the current image and audio recording memory 9 (step 42), refer to The comparison error recognition / comparison circuit 10 between the data and the current input data performs a content comparison process between the reference memory and the current recording memory (step 43). At this time, as shown in FIGS. 6 (a), (b), and (c), a threshold level is provided as a matching index for determining the matching, and the area of the binarized image for an image When the sound pressure level exceeds the threshold level, it is determined that the reference image or voice matches the current image or voice (steps 44, 45, 47).

As described above, conventionally, the input image from the camera and the input sound from the microphone are sequentially stored in real time in a plurality of frame memories built in the main body of the videophone, and the contents of the previous frame memory and the current frame memory are stored. Compared with the case where the contents are compared and the error is recognized, the method of storing the image or the sound in the secondary storage memory and comparing the contents with the current image makes the error degree clear and the standard pattern. Since it is possible to use reference images such as, it is possible to perform matching and matching with a registered pattern in addition to error recognition, the number of comparison modes is increased, and it is possible to properly use them according to the timing and situation to be compared, thus improving usability. Further, since it is a comparison with image or audio binary information, it is suitable for confirming a large change or a rough degree of coincidence.

[0021]

The present invention has the following effects. (1) The reference image and the reference voice data are stored in the secondary storage medium, and the contents are extracted to the built-in memory and compared with the current camera input image. It is possible to compare the opposite modes, which is convenient. (2) Since it is not a method of comparing in real time, but a method of comparing reference data converted into binary data and current data,
The comparison circuit configuration is simple.

(3) Since the secondary storage medium is used for storing the reference image and the reference voice data, it is possible to register a plurality of reference images and reference voice data, and register according to the purpose and situation. Data can be retrieved and usability is improved.

[Brief description of drawings]

FIG. 1 is a block diagram of a circuit of the present invention.

FIG. 2 is an operation flowchart of the circuit configuration diagram of FIG.

FIG. 3 is an explanatory diagram showing an application example of abnormality detection using the error recognition mode of the present invention.

FIG. 4 is an explanatory diagram showing an abnormality detection method and a detection result of the application example of FIG.

FIG. 5 is an explanatory diagram showing an application example of person matching detection using the comparison matching mode of the present invention.

FIG. 6 is an explanatory diagram showing a person matching detection method and a detection result of the application example of FIG. 5;

[Explanation of symbols]

1 ... Still image color TV telephone device, 2 ... Color camera for video input, 3 ... Camera signal input circuit, 4 ... Display screen display selection circuit, 5 ... Secondary storage memory input selection circuit, 6 ... IC memory card, 7 ... IC memory card interface circuit, 8 ... Reference image and voice recording memory, 9 ... Current image and voice recording memory, 10 ... Comparison error recognition / comparison circuit between reference data and current input data, 11 ... Registration control circuit, 12 ... Video phone push button switch, 13 ... Display screen, 14 ... Automatic transmission control circuit, 15 ... TV phone function control circuit, 16 ... Speaker control circuit, 17 ... Speaker, 18
... microphone, 19 ... transmission / reception circuit, 20 ... telephone outlet, 21 ... TV telephone transmission / reception antenna, 22 ... binarization circuit, 23 ... comparison / matching circuit section, 24 ... speech synthesis circuit.

Claims (2)

[Claims] 1. A still image color videophone having a camera for capturing an image and a microphone for capturing an audio, and having a function of transmitting a video signal and an audio signal obtained from the camera and the microphone to a partner device. In the apparatus, a first function for binarizing a camera video signal in black and white by using a predetermined set value as a threshold level, and binarizing the audio signal without the audio signal by using a predetermined set value as a threshold level The second function to perform, and the current video and audio input from the camera and the microphone are binarized by using the first and second functions, and the image and the audio are simultaneously stored in a storage medium built in the main body. A third function that can be recorded, a fourth function that can be recorded in a secondary storage medium that can be separated from the storage medium by using the third function, and A fifth function capable of selecting one of a plurality of cases stored in the secondary storage medium by using the fourth function and transferring the content to another storage medium built in the main body; and the third function A sixth function capable of simultaneously comparing and recognizing a comparison error between the content of the storage medium contained in the main body stored by using and the content of the storage medium contained in the main body stored by using the fifth function; A seventh aspect capable of simultaneously comparing and recognizing the comparison match between the contents of the storage medium built in the main body stored by using the third function and the contents of the storage medium built in the main body stored by using the fifth function. A function and a function of transmitting an image and voice already registered in the secondary storage medium separately from the comparison contents to the telephone number of the partner device when an error is recognized by using the sixth function. Recognize a match using the 7th function above Still image color TV telephone apparatus characterized by the images and sound above the comparison contents are separately already registered in the secondary storage medium equipped the ability to transmit to the telephone number of the other party apparatus when. 2. When the error recognition is caused by the third function according to claim 1, a voice message registered in advance in the secondary storage medium is extracted and the voice message is uttered to the external loudspeaker. Image color videophone device equipped with means.