JPH09289637A - Image recorder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device for supplying the discrimination function of a text screen to an image recorder by providing an object detection sensor, an object position calculation part, an object position storage part, an object presence judgement part, an object image input part and image recording part. SOLUTION: In the case that a visitor presses the switch 103 of an interphone device or the like, a control part 104 which receives signals S11 from the switch 103 drives a CCD camera 101 and the image recording part 102 and starts the photographing of the visitor. Further, from the information S21 of a distance to the visitor by a distance sensor 113 and the information S22 of warmth characteristics near an entrance by an warmth sensor 112, a visitor judgement part 113 judges whether or not the visitor to be photographed is present. Based on the judged result S23 of the visitor judgement part 113, the control part 104 drives the CCD camera 101 and the image recording part 102 and starts the photographing of the visitor. By a request from an operation part 108, an image reproduction part 106 extracts images S17 from an image preservation part 105, changes a signal form and outputs them to an image monitor 107.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image recording device, and more particularly to an unmanned surveillance device and a video recording device.

[0002]

2. Description of the Related Art Conventionally, in a surveillance device, the presence of a person in the surveillance area or the movement of a person into or out of the surveillance area is constantly monitored and an image thereof is recorded, or only when the presence of a person is clear. I recorded it. A conventional monitoring device will be described below by taking a home entrance monitoring device as an example.

19 and 20 show a schematic structure of a conventional home monitoring apparatus. In FIG. 19, 101 is a CCD camera for monitoring the vicinity of the entrance, 102 is an image recording unit for recording the image captured by the CCD camera 101, 10
Reference numeral 3 denotes a switch for notifying a visitor when the visitor pushes it, and 104 denotes a CCD camera 101 and an image recording unit 1.
A control unit for controlling 02, an image storage unit 105 for storing the recorded image, an image reproduction unit 106 for reproducing the image stored in the image storage unit 105, and 107 for the CCD camera 1
An image monitor for displaying the image 01 and the reproduced image by the image reproducing unit 106, and an operation unit 108 for operating the image reproducing unit 106. When the configuration of the control circuit of this conventional home entrance monitoring device is expressed, a block diagram as shown in FIG. 20 is obtained. In the conventional home entrance monitoring device, when the visitor presses the switch 103 such as the intercom device, the control unit 10
4 drives the CCD camera 101 to start photographing the visitor. The image S14 of the visitor can be monitored from inside the house by the image monitor 107. At the same time, when the switch 103 is pressed, the control unit 104 also drives the image recording unit 102, and the image recording unit 102 converts the image S15 from the CCD camera 101 into a storable format S16 and saves it in the image saving unit 105. It When the operation unit 108 issues a reproduction request S19 for the stored image, the image reproduction unit 106 extracts the image S17 from the image storage unit 105, and the image monitor 10
The signal format is converted to a signal format S18 that can be displayed in step S7 and then output to the image monitor 107.

On the other hand, in the video recording apparatus, conventionally, when the user's own operation or the image recording start time designated by the user comes, the driving of the image recording unit is started to record a video image.

21 and 22 show a schematic structure of a conventional video recording apparatus. In FIG. 21, 201 is a channel selection unit that selects a broadcast wave to be received, and 202 is a channel selection unit 20.
A demodulation unit for demodulating the broadcast wave selected in 1 into a video signal,
Reference numeral 203 is an image recording unit for recording the video signal demodulated by the demodulation unit 202, 204 is a time management unit for managing image recording by time, and 205 is a channel selection unit 201 and an image recording unit according to an instruction from the time management unit 204. A control unit for controlling 203, an image storage unit 206 for storing the recorded image, 2
Reference numeral 07 is an image reproduction unit for reproducing the image stored in the image storage unit 206, 208 is an image monitor for displaying the image reproduced by the image reproduction unit 207, and 209 is a user's command to the image reproduction unit 207 for reproduction. It is an operation unit.

The structure of the control circuit of the conventional video recording apparatus is expressed as a block diagram as shown in FIG. In the conventional video recording apparatus, the television reception wave S51 selected by the channel selection unit 201 is converted into the video signal S by the demodulation unit 202.
Converted to 52. The image recording unit 203 converts the video signal S52 into a signal S53 that can be stored in the image storage unit 206 and is stored in the image storage unit 206. Time management unit 204
Is set beforehand by the user with respect to the recording start and end times of the image and the broadcasting station for recording the image. At the recording start time, the control unit 205 receives the recording instruction S56.
Gives a broadcasting station S57 for recording to the channel selection unit 201 and an instruction S58 for starting recording to the image recording unit 203. Also, when the recording end time is reached, the time management unit 204 gives a similar command. When the user gives an instruction to reproduce an image to the operation unit 209, the image reproduction unit 207 converts the signal S54 stored in the image storage unit 206 into a video signal S55 and displays it on the image monitor 208.

[0007]

However, in the above-mentioned conventional home entrance monitoring device, when the visitor has the intention to visit and presses the switch such as the intercom, the user who is the user can confirm the existence. However, since there is no means for judging the presence or absence of a visitor or the behavior of the visitor other than the switch being pressed,
There was a problem that it was impossible to record the existence and behavior of a suspicious visitor or intruder without always monitoring it.

The present invention solves the above-mentioned conventional problems, and provides a function of detecting a visitor who does not press a switch and saving the image by introducing a sensor or a camera and evaluating the behavior thereof. With the goal.

In the above conventional video recording device,
Although it is possible to record the broadcast at the user's own operation or at the time set by the user, since there is no means for determining the scene required by the user during the broadcast, the unnecessary scene is also recorded. Therefore, there is a problem that the user must select a necessary scene at the time of reproduction.

The present invention solves the above-mentioned conventional problems, and an object thereof is to provide an image recording device such as a video recording device with a function of discriminating a text screen.

[0011]

In order to achieve this object, the present invention is to record an object detected by the sensor or an image being input from the video information input by the sensor information or the image input device. An evaluation unit for evaluating whether or not the image is a quality item is provided, and the image recording is controlled according to the evaluation result.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION

(First Embodiment) A first embodiment of the present invention will be described below with reference to the drawings.

1 to 3 show a schematic structure of a home entrance monitoring apparatus according to this embodiment, FIG. 1 is a schematic view of a home entrance monitoring apparatus, and FIG. 2 is a detailed view of an image monitor. FIG. 3 is a block diagram showing the control flow of the entire home entrance monitoring system.

In FIG. 1, 101 is a CCD camera and 1
Reference numeral 02 is an image recording unit, 103 is a switch, 104 is a control unit, 105 is an image storage unit, 106 is an image reproduction unit, 107
Is an image monitor, and 108 is an operation unit. The above configuration is the same as that of the conventional example in FIG.

The difference from the conventional example is that a distance sensor 111, a temperature sensor 112, and a visitor determination unit 113 are provided.

FIG. 2 shows a state in which the image stored in the image storage unit 105 is displayed on the image monitor 107. In this embodiment, in addition to recording an image near the entrance when a visitor presses a switch such as an intercom as in the conventional example, even if a suspicious visitor is detected even if the switch is not pressed, the image is displayed. Record. The recorded image is marked with the recorded time, and when the switch is pressed, the recorded image is marked as "visitor", and the recorded image is marked as "suspicious man". Has been done.

FIG. 4 is a block diagram of the visitor determination unit 11 described with reference to FIG.
3 shows a detailed configuration of No. 3. The visitor determination unit is a person position calculation unit 121 that calculates the presence position of the visitor.
A person position storage unit 122 that stores the position of the person, a time measurement unit 123 that measures the time that the person was present,
The determination unit 124 determines whether or not to photograph the visitor. In the person position calculation unit 121, the plurality of distance sensors 111 obtain distance information S21 to the suspicious object in the measurement direction of each distance sensor. This distance information S21 is obtained by taking the difference from the distance information when there is no suspicious object held in advance,
Since it is possible to know at what distance in which sensor direction the suspicious object exists, the outline position of the suspicious object in the surveillance area can be specified. Further, information S on the thermal characteristics detected by the plurality of thermal sensors 112.
Since the direction in which an object having a human-presumed temperature exists is known from 22 as well, only the position information of the object determined to be a human from the suspicious object detected together with the previous distance information S21 is output as the calculation result S31. To do. The present position S31 of this visitor, which changes with time, is stored in the person position storage unit 122.
Remembers. Based on the variation history S32 of the visitor's existing position stored in the person storage unit 122 and the time information S33 by the time measuring unit 123, the determining unit 124 determines that the visitor has been present near the entrance for a certain period of time or longer. It is determined that the visitor should be photographed, and the control unit 104 is instructed to take the photograph S
23 is output.

The overall control flow of this embodiment will be described with reference to FIG. The flow of control when the visitor presses the switch 103 such as the intercom apparatus is the same as in the conventional example. The control unit 104 receives the signal S11 from the switch 103.
Drives the CCD camera 101 and the image recording unit 102 to start photographing the visitor. Further, in the present embodiment, the visitor determination unit determines whether or not there is a visitor to be photographed from the distance information S21 to the visitor by the distance sensor 111 and the information S22 of the thermal characteristics near the entrance by the thermal sensor 112. Then, based on the determination result S23, the control unit 104 also drives the CCD camera 101 and the image recording unit 102 to start photographing the visitor. Regarding reproduction of the photographed image, when a reproduction request is issued from the operation unit 108, the image reproduction unit 106 extracts the image S17 from the image storage unit 105 and converts it into a signal format S18 that can be displayed on the image monitor 107. It is output to the image monitor 107 in the format already shown in FIG.

As a result, as compared with the conventional home entrance monitoring device, even a suspicious person who has been in the vicinity of the entrance for a long time without the switch is automatically judged and photographed.
It is possible to realize more secure security management.

In the present embodiment, the case where the CCD camera 101 is used as the image input section has been described, but the same effect can be obtained by using an infrared camera or other image input device.

Further, a distance sensor 111 for detecting a visitor.
And at least one of the thermal sensors 112 is 1
There should be one or more. When the number of the distance sensor 111 or the thermal sensor 112 is one, as shown in FIG. 5, the mounting angle of the sensor is changed, and the detection direction is changed so as to scan near the monitoring area. The effect is obtained.

(Second Embodiment) A second embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows a schematic structure of a home entrance monitoring apparatus which is the first embodiment already described, but this embodiment also has the same structure as that of FIG. That is, in FIG. 1, 101 is a CCD camera, 102 is an image recording unit, 103 is a switch, 104 is a control unit, 10
5 is an image storage unit, 106 is an image reproduction unit, 107 is an image monitor, 108 is an operation unit, 111 is a distance sensor, 112 is a thermal sensor, 113 is a visitor determination unit, and a display method of a recorded image on the image monitor 107 Also has the same form as in FIG.

The present embodiment is different from the first embodiment in the structure of the visitor judging section 113. FIG. 6 shows a detailed configuration of the visitor determination unit 113 described with reference to FIG. This visitor determination unit is a person position calculation unit 121 that calculates the presence position of the visitor, a person position storage unit 122 that stores the position of the person, and a fractal dimension calculation that calculates the fractal dimension of the variation of the existence position of the person. It is composed of a unit 125 and a determination unit 124 that determines whether or not to photograph a visitor.

The fractal dimension is a value indicating the complexity of behavior of a figure or system or its self-similarity. It is known that when this value is small, the behavior of figures and systems is simple and regular, when it is large, it is complicated, and when it is an intermediate value, it has strong self-similarity (Reference 1. Kazuyuki Aihara). : "Chaos-the basis of chaos theory and its application", Science Co., pp93-95, 1990). Several methods for obtaining fractal dimension from time series information have been proposed by academic societies (Reference 2. T.HIGUCHI: “APPROACHTO AN I
RREGULAR TIME SERIES ON THE BASIS OF THE FRACTAL T
HEORY ”, PhysicaD, vol.31, pp.277-283, 1988).

Here, the method of finding the fractal dimension proposed in this Document 2 will be described.

[0027]

[Equation 1]

Equation 1 is a time series signal from time 0 to T,
t is the time, and x is the value of the time series signal at the time t. For the time-series signal expressed by this equation 1, from time t = m
Consider a sequence of points every k times.

[0029]

[Equation 2]

Equation 2 is an equation for obtaining the length of the time series signal represented by Equation 1. In Expression 2, L (k) is the length of the time-series signal, m is the sampling start time, and k is the sampling time. Here, the operation shown in [a] is the maximum integer value that does not exceed a. Here, a large sampling time k means a macroscopic view of the system, and a small k means a microscopic view of the system. For systems with complex behavior this sampling time k
As the size is made smaller, the macroscopic rough structure gradually becomes smaller and the microscopic local fine structure appears.
L (k) increases. Generally, the relationship shown in Formula 3 is approximately established between the length L (k) of the time series signal and the sampling time k.

[0031]

(Equation 3)

The value represented by D in the equation 3 is the fractal dimension of the time series signal. Therefore, by calculating the value of L (k) for various values of k and using the method of least squares, it is possible to obtain D having a value between 1 and 2. In this embodiment, the fractal dimension is calculated by the above calculation method.

The fractal dimension calculated in this way is known to be a quantitative expression index of the complexity and self-similarity of the signal from previous studies in academic societies. That is, the fractal dimension can be between 1 and 2, but when the value is small, the fluctuation of the signal is moderate or periodic, and when the value is large, the fluctuation of the signal is complicated and fluctuates irregularly. Is known to do. Therefore, in the present embodiment, the behavior such as whether the visitor is stationary or moving around the entrance is estimated from the fractal dimension of the time-series variation of the visitor's location, and the image of the visitor is captured by this. Control.

The operation procedure of this embodiment will be described below. In FIG. 6, from the distance information S21 to the visitor detected by the distance sensor 111 and the information S22 of the thermal characteristics of the monitoring area detected by the thermal sensor 112, the person position calculation unit 121 detects the visitor within the monitoring area. It is calculated whether or not the visitor exists, and if the visitor exists, the location where the visitor exists. The person position storage unit 122 stores the present position S31 of the visitor, which changes with time. The fractal dimension calculation unit 125 is calculated from the history of changes S32 of the visitor position stored in the person storage unit 122.
Calculates the fractal dimension S34 of visitor behavior. This fractal dimension S34 shows a high value when the visitor shows a behavior that changes the existing position irregularly, and shows a low value when the visitor stays at one place. By using this property, it is determined that the visitor should be photographed when the visitor stays without pressing the switch 103 or the behavior is suspicious, and the control unit 104 outputs a photography instruction S23.

In this embodiment, the fractal dimension calculation unit is adopted as the signal characteristic analysis unit in claim 2,
Change history S of the visitor's location in the signal characteristic analysis unit
It can also be realized by analyzing 32 frequency characteristics or amplitude characteristics.

Although the method of Reference 2 is used as the method of calculating the fractal dimension in the present embodiment, the same effect can be obtained by other calculation methods.

(Embodiment 3) A third embodiment of the present invention will be described below with reference to the drawings.

7 and 8 show a schematic structure of a home entrance monitoring apparatus according to this embodiment. FIG. 7 shows a home entrance monitoring apparatus in outline, and FIG. 8 shows a home entrance monitoring apparatus. It is a block diagram of the flow of the overall control.

In FIG. 7, 101 is a CCD camera and 1
Reference numeral 02 is an image recording unit, 103 is a switch, 104 is a control unit, 105 is an image storage unit, 106 is an image reproduction unit, 107
Is an image monitor, 108 is an operation unit, 111 is a distance sensor,
Reference numeral 112 is a heat sensor, and 113 is a visitor determination unit, and the above configuration is similar to that of the first and second embodiments of FIG.

The difference from the first and second embodiments is that a CCD camera 114, a person feature extraction unit 115 and an illumination device 116 are provided.

FIG. 9 shows the person feature extraction unit 1 described with reference to FIG.
15 shows a detailed configuration of 15. The person feature extraction unit 115 includes an image binarization unit 131 that binarizes an input image, and a centroid calculation unit 132 that calculates the centroid of the image binarized by the image binarization unit 131. It is configured. The illuminator 116 illuminates the visitor with visible light,
CCD camera 114 images the visitor. The captured image S41 is converted into a binary image S43 by a threshold value that is the image binarization unit 131. When a visitor exists by the binarization operation, it is possible to cut out the approximate shape of the visitor portion in the image S41. The centroid calculation unit 132 calculates the centroid S42 of the image of the visitor from the rough image S43 of the visitor obtained by binarization, and sends the centroid S42 of the image of the visitor to the visitor determination unit 113.

FIG. 10 is a block diagram of the visitor determination unit 1 described with reference to FIG.
13 shows a detailed configuration of 13. This visitor determination unit 113 has a configuration similar to that of the visitor determination unit 113 according to the first embodiment already described with reference to FIG. 4, and includes a person position calculation unit 121 that calculates the presence position of the visitor, It comprises a person position storage unit 122 that stores the position of a person, a time measuring unit 123 that measures the time that a person was present, and a determination unit 124 that determines whether or not to photograph a visitor. The difference from the first embodiment is that, in addition to the distance information S21 from the distance sensor 111 and the thermal characteristic information S22 from the thermal sensor 112, the person position calculation unit 121 adds a visitor image from the human feature extraction unit 115. This is the point at which the center-of-gravity position information S42 is input. This distance information S21
And information S22 of thermal characteristics and center of gravity position information S of visitor image
From 43, the person position calculation unit 121 calculates whether or not a visitor exists in the monitoring area, and if so, what position the visitor is. The person position storage unit 122 stores the present position S31 of the visitor, which changes with time. Based on the variation history S32 of the visitor's existing position stored in the person storage unit 122 and the time information S33 by the time measuring unit 123, the determining unit 124 determines that the visitor has been present near the entrance for a certain period of time or longer. It is determined that the visitor should be photographed, and the control unit 104 is instructed to photograph the visitor in S2.
3 is output.

FIG. 8 shows an overall control flow of this embodiment. This is similar to FIG. 3 which shows the overall control flow for the first embodiment, except that CC
Regarding the image S41 of the visitor captured by the D camera 114, the difference is that the center-of-gravity information S42 of the person image obtained by the person feature extraction unit 115 is also input to the visitor determination unit 113.

As a result, compared with the conventional home entrance monitoring device, even a suspicious person who has been near the entrance for a long time without pressing the switch can automatically judge with higher accuracy than in the first embodiment. Since shooting is performed, it is possible to realize more secure security management.

In the present embodiment, the CCD camera 101 is used as the image input unit of the image to be recorded, and the CCD camera 114 is used as the image input unit for detecting the visitor. 101 and C
Even if the CD camera 114 is the same device, the same effect can be obtained.

Further, for the CCD camera 101 and the CCD camera 114, the same effect can be obtained by using an infrared camera or other image input device. When an infrared camera is used instead of the CCD camera 114, the same effect can be obtained even if the illumination device 116 uses an infrared light.

The configuration of the person feature extraction unit 115 is shown in FIG.
It does not need to be in the form shown in, but the presence or absence of visitors,
The same effect can be obtained as long as the feature amount can be used as a clue to know the location of the visitor and the distance to the visitor. FIG. 11 shows another example of the detailed configuration of the person feature extraction unit 115 described with reference to FIG. The person feature extraction unit 115 includes an image binarization unit 131 that binarizes an input image, and an area calculation unit 133 that calculates the area of the image binarized by the image binarization unit 131. It is configured. The illuminator 116 illuminates the visitor with visible light, and the CCD camera 114 images the visitor. The captured image S41 is processed by the image binarization unit 13
It is converted into a binary image S43 with a threshold value of 1. When a visitor exists by the binarization operation, it is possible to cut out the approximate shape of the visitor portion shown in the image S41. Due to the binarization, the area calculation unit 133 changes the image S43 of the visitor to the area S4 of the image of the visitor.
4 is calculated and sent to the person position calculator.

(Embodiment 4) Hereinafter, a fourth embodiment of the present invention will be described with reference to the drawings.

FIG. 7 shows a schematic structure of the home entrance monitoring apparatus according to the third embodiment which has already been described. This embodiment also has the same structure as that of FIG. That is, in FIG. 7, 101 is a CCD camera, 102 is an image recording unit, 103 is a switch, 104 is a control unit, and 10 is a unit.
5 is an image storage unit, 106 is an image reproduction unit, 107 is an image monitor, 108 is an operation unit, 111 is a distance sensor, 112 is a thermal sensor, 113 is a visitor determination unit, 114 is a CCD camera, 115 is a human feature extraction unit. , 116 are lighting devices.

The difference of this embodiment from the third embodiment is the structure of the visitor determination unit 113. FIG. 12 shows FIG.
3 illustrates a detailed configuration of the visitor determination unit 113 described in 1. The visitor determination unit 113 includes a person position calculation unit 121 that calculates the presence position of the visitor, a person position storage unit 122 that stores the position of the person, and a fractal dimension that calculates a fractal dimension of variation of the presence position of the person. Calculator 12
5, and the determination unit 12 that determines whether to photograph the visitor
4. Distance information S21 from the distance sensor 111 and information S on the thermal characteristics from the thermal sensor 112
22 and from the center-of-gravity position information S43 of the visitor image from the person feature extraction unit 115, the person position calculation unit 121 determines whether or not the visitor exists in the monitoring area, and if so, at which point the existence position is. Calculate if there is. The person position storage unit 122 stores the present position S31 of the visitor, which changes with time. The fractal dimension calculation unit 125 uses the fractal dimension calculation unit 125 to calculate the fractal dimension S3 of the visitor behavior based on the history of changes S32 of the visitor's existing position stored in the person storage unit 122.
4 is calculated, and the determination unit 12 is calculated from this fractal dimension S34.
In the case where there is a visitor 4 who moves around irregularly in the vicinity of the entrance and takes a suspicious action or a visitor who remains without pressing the switch 103, it is determined that the visitor should be photographed, and the control unit 104 takes an image. Is output to S23.

FIG. 8 shows the overall control flow for the third embodiment already described, but the control flow for this embodiment is exactly the same.

As a result, a suspicious person is automatically judged with higher accuracy than in the third embodiment and the photographing is performed, so that it is possible to realize security control with higher safety.

In the present embodiment, the CCD camera 101 is used as the image input unit of the image to be recorded and the CCD camera 114 is used as the image input unit for detecting the visitor. 101 and C
Even if the CD camera 114 is the same device, the same effect can be obtained.

Further, although the fractal dimension calculating unit is adopted as the signal characteristic analyzing unit in claim 4, the same applies by analyzing the frequency characteristic or the amplitude characteristic of the variation history S32 of the visitor's existing position in the signal characteristic analyzing unit. The effect of is obtained.

(Fifth Embodiment) Hereinafter, a fifth embodiment of the present invention will be described with reference to the drawings.

13 to 15 show a schematic structure of the video recording apparatus according to this embodiment. In FIG. 13, 201 is a channel selection unit, 202 is a demodulation unit, 203 is an image recording unit, 204 is a time management unit, 206 is an image storage unit, 20
Reference numeral 7 is an image reproducing unit, 208 is an image monitor, and 209 is an operating unit. The above configuration is the same as that of the conventional example of FIG.

The difference from the conventional example is that a channel selection control unit 211, an image feature determination unit 212 and a recording control unit 213 are provided.

FIG. 16 shows a detailed structure of the image feature determining section 212 described with reference to FIG. The image feature determination unit 212 determines a still image determination unit 221 that determines a still scene in a moving image, a fractal dimension calculation unit 222 that calculates a fractal dimension of an image, and a determination whether or not the image is an image to be recorded. It is composed of a section 223.

The fractal dimension of an image is a concept obtained by expanding the fractal dimension described in the second embodiment, and is a value representing the complexity or self-similarity of an image. Several methods for obtaining fractal dimension from image information have been proposed by academic societies (Reference 3. Hiroshi Kaneko: "Fractal Features and Texture Analysis", IEICE Transactions D Vol.J7).
0-D, No.5, pp.964-972, 1987, Reference 4. Naokikazu Yokoya: "Analysis of three-dimensional complex shapes by fractal and its application", IEEJ PRU86-23, 1986).

Here, a method for obtaining the fractal dimension proposed in Document 3, which is one of them, will be described.

[0061]

(Equation 4)

Equation 4 is an expression for obtaining the total luminance area of the image I. In Expression 4, A (r) is the total luminance area of the image I, r is the length of one side of the unit divided area, and a (r) is the luminance area of one unit divided area. FIG. 12 illustrates the luminance plane. As shown in FIG. 17A, the image I is divided into grid-like unit divided areas on one side r, and the unit divided areas are arranged on the XY plane as shown in FIG. 17B. XY when the brightness of each vertex is taken in the Z-axis direction.
The surface formed in the Z space is a luminance surface, and its area is a (r). The total brightness area A (r) is the sum of the brightness areas a (r) per unit divided area shown in FIG. 17 for the image I. Generally, the relationship expressed by the equation 5 is approximately established between the total luminance area A (r) and the length r of one side of the unit division area.

[0063]

(Equation 5)

The value represented by D in this equation 5 is the fractal dimension of the image I. Therefore A for various values of r
The value of D can be obtained by calculating the value of (r) and using the least squares method. In this embodiment, the fractal dimension is calculated by the above calculation method.

The fractal dimension of the image calculated in this way is different from the properties such as the size, brightness, and color of the image according to the research conducted so far in academic societies, and the complexity and self-similarity of the image are quantified. It is known to be a general expression index.

FIG. 18 shows the relationship between the image and the fractal dimension. Table 1 shows the values of the fractal dimension in FIGS. 18 (a) and 18 (b).

[0067]

[Table 1]

From this, comparing the two images of FIGS. 18A and 18B, the self-similarity and complexity of FIG. 18B is more complicated than the regular and simple pattern of FIG. 18A. The pattern has a higher fractal dimension. In the present embodiment, the feature of the image is evaluated using this fractal dimension.

FIG. 14 shows a display mode of a recorded image on the image monitor 208 in this embodiment. In the present embodiment, as in the conventional example, not all the video from the recording start time to the recording end time designated by the user is recorded, but as shown in FIG. 14, during the recording time designated by the user, Only images in which characters are regularly arranged in parallel lines or grids are selectively recorded, and a list of recorded images of the day is shown together with the recording time and channel information.

The operation procedure of this embodiment will be described below with reference to FIG. When the time management unit 204 reaches the time designated by the user in advance, the tuning control unit 211 is instructed to select a channel S5.
6 is output. In response to this, the tuning control unit 211 controls the tuning unit 201, and the tuning unit 201 receives the designated broadcast wave. The television reception wave S51 selected by the channel selection unit 201 is converted into video signals S52 and S61 by the demodulation unit 202. From the video signal S61, the image feature determination unit 2
Reference numeral 12 determines whether or not it is a scene of text information mainly composed of characters from the video being received, and outputs the determination result S62. Based on this determination result S62, the recording control unit 213 outputs an image recording command S58 to the image recording unit 203.
Upon receiving S58, the image recording unit 203 receives the received video signal S52 and stores the received video signal S52 in the image storage unit 206.
It is converted into 53 and stored in the image storage unit 206. When the user gives an instruction to reproduce an image to the operation unit 209, the image reproduction unit 207 causes the signal S5 stored by the image storage unit 206 to be stored.
4 is converted into a video signal S55 and displayed on the image monitor 208.

Next, the operation procedure of the image feature determining section 212 will be described with reference to FIG. The still image determination unit 221 detects the motion component by taking the difference between each frame of the video signal being received and the frames before and after the video signal S61 input from the demodulation unit 202. Based on this motion component, it is determined whether or not the scene is a still scene that does not move for a certain period of time, and an image S71 of the still scene is output to the fractal dimension calculation unit 222. The fractal dimension calculation unit calculates the fractal dimension of the image of the still image S71, and the calculated result S
72 is output to the determination unit 223. If the still image S71 is an image mainly composed of characters such as a material list of a cooking program, the characters are arranged relatively regularly over the entire image, and thus the obtained fractal dimension S72 is smaller than other images. Become. From the property of this image, the determination unit 223 determines whether the image is mainly composed of characters, and the determination result S6
2 is output to the recording control unit 213.

As a result, it becomes possible to selectively record only the scenes mainly composed of text in a certain program without recording all of the program from the start to the end.

Although the present embodiment is provided with the channel selection section 201 and the demodulation section 202, the image recording section 203 outputs the video signal generated by an external device without the channel selection section 201 and the demodulation section 202.
Also, the same effect can be obtained by inputting to the image feature determination unit 212.

Further, in the present embodiment, the method of Reference 3 is used as the method of calculating the fractal dimension, but the same effect can be obtained by other calculating methods.

[0075]

As described above, according to the present invention, a distance sensor, a heat sensor, a CCD camera, and a visitor determination unit are provided in a home entrance monitoring device, and the visitor is photographed based on the presence or absence of the visitor and its behavior. By recording, it is possible to perform highly secure security management regardless of whether the family member is at home or not. Similarly, in the video recording device, by controlling the recording based on the fractal dimension of the image of the broadcast wave being received, only the text information can be selectively recorded.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a home entrance monitoring device according to an embodiment of the first invention.

FIG. 2 is a diagram showing a display of a home entrance monitoring device in the embodiment of the first invention.

FIG. 3 is a diagram showing an operation of the home entrance monitoring device in the embodiment of the first invention.

FIG. 4 is a diagram showing a configuration of a visitor determination unit according to the embodiment of the first invention.

FIG. 5 is a diagram showing an operation in which one sensor scans a monitoring area.

FIG. 6 is a diagram showing a configuration of a visitor determination unit according to an embodiment of the second invention.

FIG. 7 is a diagram showing a configuration of a home entrance monitoring device according to an embodiment of the third invention.

FIG. 8 is a diagram showing an operation of the home entrance monitoring device in the embodiment of the third invention.

FIG. 9 is a diagram showing a configuration of a person feature extraction unit according to an embodiment of the third invention.

FIG. 10 is a diagram showing a configuration of a visitor determination unit according to an embodiment of the third invention.

FIG. 11 is a diagram showing a configuration of a person feature extraction unit according to an embodiment of the fourth invention.

FIG. 12 is a diagram showing a configuration of a visitor determination unit in the embodiment of the fourth invention.

FIG. 13 is a diagram showing a configuration of a video recording device according to an embodiment of the fifth invention.

FIG. 14 is a diagram showing a display of the video recording device in the embodiment of the fifth invention.

FIG. 15 is a diagram showing an operation of the video recording device in the embodiment of the fifth invention.

FIG. 16 is a diagram showing a configuration of an image feature determination unit according to an embodiment of the fifth invention.

FIG. 17 is a diagram showing a luminance plane of an image.

FIG. 18 is a diagram showing the nature of the fractal dimension of an image.

FIG. 19 is a diagram showing a configuration of a conventional home entrance monitoring device.

FIG. 20 is a diagram showing an operation of a conventional home entrance monitoring device.

FIG. 21 is a diagram showing a configuration of a conventional video recording device.

FIG. 22 is a diagram showing an operation of a conventional video recording device.

[Explanation of symbols]

 101 CCD camera 102 Image recording unit 103 Switch 104 Control unit 105 Image storage unit 106 Image reproduction unit 107 Image monitor 108 Operation unit 111 Distance sensor 112 Temperature sensor 113 Visitor determination unit 114 CCD camera 115 Personal feature extraction unit 116 Illumination device 121 People Position calculation unit 122 Person position storage unit 123 Time measurement unit 124 Judgment unit 125 Fractal dimension calculation unit 131 Image binarization unit 132 Center of gravity calculation unit 133 Area calculation unit 201 Channel selection unit 202 Demodulation unit 203 Image recording unit 204 Time management unit 205 Control Unit 206 image storage unit 207 image reproduction unit 208 image monitor 209 operation unit 211 channel selection control unit 212 image feature determination unit 213 recording control unit 221 still image determination unit 222 fractal dimension calculation unit 223 determination unit S11 switch Output of S12 Output of control unit S13 Output of control unit S14 Output of CCD camera S15 Output of CCD camera S16 Output of image recording unit S17 Output of image storage unit S18 Output of image reproducing unit S19 Output of operation unit S21 Output of distance sensor S22 Output of thermal sensor S23 Output of visitor determination unit S31 Output of human position calculation unit S32 Output of human position storage unit S33 Output of time measurement unit S34 Output of fractal dimension calculation unit S41 Output of CCD camera S42 Output of human feature extraction unit Output S43 Output of image binarization unit S44 Output of human feature extraction unit S51 Output of channel selection unit S52 Output of demodulation unit S53 Output of image recording unit S54 Output of image storage unit S55 Output of image reproduction unit S56 Output of time management unit S57 Output of control unit S58 Output of control unit S59 Output of operation unit 61 output of the output S72 fractal dimension calculation unit output S71 still image determining unit of the output S62 image feature judging unit of the demodulator

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location H04N 5/76 G06F 15/70 410

Claims (5)

[Claims] 1. A subject detection sensor for detecting at least one of a distance to a subject and a thermal characteristic of the subject,
A subject position calculation unit that calculates the position of the subject from the distance to the subject detected by the subject detection sensor or information on the thermal characteristics of the subject, and a history of changes in the position of the subject calculated by the subject position calculation unit are stored. A subject position storage unit, a determination unit that determines the presence or absence of a subject and its state based on the time-series variation history of the subject position stored in the subject position storage unit, and a subject based on the determination result of the determination unit. An image recording apparatus comprising: an image input unit for capturing an image; and an image recording unit for recording a still image or a moving image input from the image input unit. 2. A subject detection sensor for detecting at least one of a distance to a subject and a thermal characteristic of the subject,
A subject position calculation unit that calculates the position of the subject from the distance to the subject detected by the subject detection sensor or information on the thermal characteristics of the subject, and a history of changes in the position of the subject calculated by the subject position calculation unit are stored. A subject position storage unit, a signal characteristic analysis unit that calculates at least one of the magnitude or frequency characteristic of the fluctuation amplitude and the fractal dimension from the time series change history of the subject position stored in the subject position storage unit, A determination unit that determines the presence or absence of a subject and its state based on the analysis result of the signal obtained by the signal characteristic analysis unit, an image input unit that images the subject based on the determination result of the determination unit, and an image input unit An image recording apparatus comprising an image recording unit for recording a still image or a moving image input from 3. A subject detection sensor for detecting at least one of a distance to a subject and a thermal characteristic of the subject,
A subject image input unit that captures an image of the subject, a subject region extraction unit that extracts a rough region of the subject from the image input from the subject image input unit, and a distance to the subject detected by the subject detection sensor or Information on the thermal characteristics of the subject and a subject position calculation unit that calculates the position of the subject from the subject region in the image extracted by the subject region extraction unit, and a history of changes in the subject position calculated by the subject position calculation unit are displayed. A subject position storage unit to store, a determination unit that determines the presence or absence of a subject and its state based on a time-series variation history of the subject position stored in the subject position storage unit, and based on the determination result of the determination unit A recording image input unit for capturing an image of a subject, and an image recording unit for recording a still image or a moving image input from the recording image input unit. Image recording device. 4. A subject detection sensor for detecting at least one of a distance to a subject or a thermal characteristic of the subject,
A subject image input unit that captures an image of the subject, a subject region extraction unit that extracts a rough region of the subject from the image input from the subject image input unit, and a distance to the subject detected by the subject detection sensor or Information on the thermal characteristics of the subject and a subject position calculation unit that calculates the position of the subject from the subject region in the image extracted by the subject region extraction unit, and a history of changes in the subject position calculated by the subject position calculation unit are displayed. A subject position storage unit that stores the signal, and a signal characteristic analysis unit that calculates at least one of the magnitude or frequency characteristic of the variation amplitude and the fractal dimension from the time series variation history of the subject position stored in the subject position storage unit. A determination unit that determines the presence or absence of a subject and the state thereof based on the analysis result of the signal obtained by the signal characteristic analysis unit; Determination result and recording the image input unit for imaging a subject based on the image recording apparatus characterized by comprising an image recording unit for recording a still image or a moving image is input from the recorded image input unit. 5. An image input section such as a television receiver, and an image difference detection section for detecting a difference between each frame of a moving image input from the image input section and at least one frame before or after the frame. A still image detection unit that detects a still image from the image difference information detected by the image difference detection unit, a fractal dimension calculation unit that calculates the fractal dimension of the still image detected by the still image detection unit, and the fractal A determination unit that determines an image having a fractal dimension in a certain range from the input image based on the fractal dimension of the image calculated by the dimension calculation unit, and an image input from the image input unit based on the determination result of the determination unit. An image recording apparatus comprising an image recording unit for recording.