JPH09182059A - Method for detecting video signal luminance difference distribution change and detector therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a video signal brightness difference distribution change detector which is capable of automatically detecting the presence or absence of the change of a monitoring object by a simple constitution. SOLUTION: In a first memory 2, an image to be a reference (reference image) is preliminarily stored, and the actual image of a place to be the monitoring object is stored via an h/D converter 1 and the actual image is made to be displayed on a monitoring device via an on-screen display control 4. By an operation circuit 9, a range to be monitored can be set on a monitor screen. In a micocontroller 7, the luminance difference for every picture element is calculated between the reference image and the actual image for every range and the distribution state is made to be determined. When the change of the distribution state which is more than a prescribed change is detected, a change output signal is made to be outputted.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for detecting a change in an image captured by an image pickup apparatus such as a television camera, and more particularly to a method and apparatus for detecting a change in a desired portion of an image. .

[0002]

2. Description of the Related Art Conventionally, as a technique for detecting the movement of a specific object or a person, for example, when the target is a person, the movement of the person is detected by detecting a temperature change between the person and the surroundings by an infrared sensor. Some have been designed to detect. In addition, there is a device that detects the movement of the target object or person by the so-called Doppler effect of the reflected wave from the target object or person using ultrasonic waves, and these can realize the device at a relatively low cost. It has an advantage and is used as a device for easily detecting the movement of an object or the like.

However, in such a simple device as described above,
Usually, it is designed to detect the movement of one target within the detection range, so it is not suitable for applications that require movement detection for multiple objects within a relatively wide range. is there. Therefore, for such an application, for example, a state of a desired monitoring range imaged by an image pickup device such as a television camera is displayed on a monitor TV, and by monitoring this, movement of an object, a person, or the like in the monitoring range or some kind of There is a way to know the state change.

[0004]

However, by directly looking at the monitor screen as described above, the target object,
The method of recognizing movement of a person or the like requires at least one person to monitor the monitor screen at all times, which is inefficient and has a problem that human resources are not effectively utilized.

For this reason, various devices have been proposed that can be automatically monitored by image processing. For example, in a conventional device of this type, a previously registered screen is compared with an actual screen. Although it detects the movement of the target object etc., the circuit scale becomes large because it is configured to compare the registration screen and the actual screen as a whole to determine the presence or absence of the movement of the target object. However, there is a problem that it becomes an expensive device.

The present invention has been made in view of the above circumstances, and a video signal luminance difference distribution change detection method and a video signal luminance difference distribution distribution which can automatically detect the presence or absence of a change in a monitoring target with a simple configuration. A change detection device is provided. Another object of the present invention is to provide a video signal luminance difference distribution change detecting method and a video signal luminance difference distribution change detecting apparatus capable of automatically monitoring a plurality of objects. Another object of the present invention is to provide a video signal luminance difference distribution change detection method and a video signal luminance difference distribution change detection apparatus that can shorten the processing time. further,

[0007]

According to another aspect of the present invention, there is provided a video signal luminance difference distribution change detecting method, wherein a predetermined range within an image is provided between a reference image and an image symmetrical to comparison. Regarding, regarding the brightness difference for each pixel,
When the distribution of the number of pixels for each brightness difference is obtained, and this distribution characteristic is left-right asymmetrical about the location of the brightness difference where the maximum number of pixels is present, it changes for a predetermined range of the image that becomes the symmetry of the comparison. Is assumed to have occurred.

In such a structure, the luminance difference for each pixel between the reference image and the image to be compared which is the actual image in a specific range in the image, that is, a portion to be actually monitored. And the distribution state of pixels is calculated for each luminance difference. If there is no change between the reference image and the comparison image, the distribution characteristics for each luminance difference are line-symmetrical with respect to the portion of the luminance difference that is the maximum number of pixels. Therefore, when the distribution characteristics are asymmetrical, it can be determined that some change (abnormality) has occurred in the specific range of the comparative image.

According to a second aspect of the present invention, there is provided a video signal luminance difference distribution change detecting method, in which a luminance difference is pixel by pixel within a predetermined range between an image serving as a reference and an image symmetrical to comparison. And the distribution of the number of pixels for each brightness difference, and if this distribution characteristic is asymmetrical about the location of the brightness difference with the maximum number of pixels, the size of the asymmetric part is calculated. The multiplication value obtained by multiplying the size of the asymmetrical portion by a predetermined coefficient is added for each asymmetrical point, and when the added value exceeds a predetermined value, the image which becomes symmetrical in the comparison. It is assumed that a change has occurred in a predetermined range of.

In such a configuration, in a specific range in the image, that is, in a portion to be actually monitored, for each pixel between the reference image and the actual image to be compared. The brightness difference is calculated, and the distribution state of pixels is calculated for each brightness difference. If there is no change other than the change in brightness between the reference image and the image to be compared, the distribution characteristic for each brightness difference (luminance difference characteristic) is the location of the brightness difference that is the maximum number of pixels. It becomes line symmetric with respect to.

Therefore, when the brightness difference distribution characteristic is asymmetrical, it can be judged that some change (abnormality) has occurred in a predetermined range of images to be compared, but this method makes a judgment. In order to ensure the certainty, numerical calculation is performed and the presence or absence of a change is determined based on the calculated value. That is, in the distribution characteristics, the size of the asymmetrical part is obtained, the size of the asymmetrical part is multiplied by a predetermined coefficient according to the brightness difference, and this multiplication value at each asymmetrical part is added. Is greater than or equal to a predetermined value, it is possible to avoid erroneous determination when the luminance difference distribution characteristic is asymmetrical due to noise or the like by setting a change in a specific range of the comparison image, and to improve reliability. It is possible to detect an image change based on a high change in the brightness difference distribution.

According to a third aspect of the present invention, there is provided a video signal luminance difference distribution change detecting device, which comprises a reference image serving as a reference for comparison.
A storage unit that stores a comparison image to be compared with the reference image, and a brightness difference between the reference image and the comparison image for each pixel in a predetermined range in the image, and a pixel for each brightness difference. A luminance difference distribution calculating unit for obtaining a number distribution, and a distribution for determining whether or not the luminance difference distribution obtained by the luminance difference distribution calculating unit is line-symmetric with respect to the location of the luminance difference of the maximum number of pixels. And a change output unit that outputs a predetermined signal when the distribution state determination unit determines that the luminance difference distribution is asymmetric.

In such a configuration, it is preferable that the reference image and the comparison image are stored in the storage means in the form of, for example, a video signal and as a digital signal. In the luminance difference distribution calculation means, between the reference image and the comparison image stored in the storage means, as in the video signal luminance difference distribution change detection method according to claim 1 or 2, the predetermined range is determined. The brightness difference distribution is determined, the distribution state determination unit determines the brightness difference distribution state, and the change output unit outputs a predetermined signal based on the determination result.

That is, in the distribution state judging means,
Similarly to the video signal brightness difference distribution change detection method according to claim 1 or 2, it is determined whether or not the brightness difference distribution characteristics are bilaterally symmetric with respect to the brightness difference portion having the maximum number of pixels. When it is determined that the image is left-right asymmetrical, the change output means outputs a predetermined signal for notifying the image change.

According to another aspect of the present invention, there is provided a video signal luminance difference distribution change detecting device, which comprises a reference image serving as a reference for comparison,
A storage unit that stores a comparison image to be compared with the reference image, and a brightness difference between the reference image and the comparison image for each pixel in a predetermined range in the image, and a pixel for each brightness difference. It is determined whether or not the luminance difference distribution calculating unit for obtaining the number distribution and the luminance difference distribution obtained by the luminance difference distribution calculating unit are line-symmetric with respect to the line of the luminance difference of the maximum number of pixels. Distribution state determination means,
When the distribution state determination means determines that the luminance difference distribution is asymmetric, the size of the asymmetric portion is obtained, and the size of the asymmetric portion is predetermined according to the luminance difference at the point where the asymmetry occurs. And a judgment value calculation means for multiplying this multiplication value for a location where asymmetry occurs, and when the value calculated by the judgment value calculation means exceeds a predetermined value, it is determined that the comparison screen has changed. Change output means for outputting a signal.

In such a structure, it is preferable that the reference image and the comparison image are stored in the storage means in the form of, for example, a video signal and as a digital signal. In the luminance difference distribution calculation means, between the reference image and the comparison image stored in the storage means, as in the video signal luminance difference distribution change detection method according to claim 1 or 2, the predetermined range is determined. The luminance difference distribution is determined, and the distribution state determination means determines the luminance difference distribution state.

That is, in the distribution state judging means,
It is determined whether or not the brightness difference distribution characteristics are bilaterally symmetric with respect to the line of the brightness difference of the maximum number of pixels. Then, when it is determined that the brightness difference distribution is asymmetric, the determination value calculating unit numerically calculates the degree of asymmetry, and the calculated size causes some change in the specific range of the comparison image. It is determined that (abnormality) is present, and a necessary signal is output from the change output means.

In particular, the calculation in the judgment value calculating means obtains the size of the asymmetrical portion for each asymmetrical portion of the luminance difference distribution, and obtains a value obtained by multiplying this by a coefficient determined according to the luminance difference. The integrated value of each multiplication value at each asymmetrical position is obtained. The change output means outputs a predetermined signal indicating that there is a change (abnormality) in a specific range of the comparison image when the value obtained by the judgment value calculation means exceeds the predetermined value. It is possible to avoid erroneous determination when the luminance difference distribution characteristic becomes asymmetric due to noise or the like, and to detect an image change based on a change in the luminance difference distribution with higher reliability.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. The members, arrangements, and the like described below do not limit the present invention and can be variously modified within the scope of the gist of the present invention.

First, a video signal luminance difference distribution change detecting device (hereinafter referred to as "this device") in the embodiment of the present invention.
The configuration will be described with reference to FIG. This device obtains a brightness difference distribution, which will be described later, between a video signal input from the outside and a pre-registered image for each predetermined range to be monitored, and determines the change state of the brightness difference distribution. It is configured to detect a state change in a predetermined range to be monitored and obtain a predetermined output signal.

That is, the present apparatus is an analog / digital converter that converts a video signal input from the outside into an analog / digital signal.
A digital converter (denoted as “A / D” in FIG. 1 and hereinafter referred to as “A / D converter”) 1 is provided, and a video signal input from the outside is sequentially converted into a digital signal, It is adapted to be stored in the first memory 2. The first memory 2 is also used as a memory for previously storing an image serving as a reference for obtaining a brightness difference distribution described later.

The video signal is input to the timing generator 3 and the on-screen display controller 4. The timing generator 3 is
The horizontal sync and vertical sync timings are extracted from the video signal and the horizontal sync signal (HSYNC) and vertical sync signal (V
SYNC) is generated. The on-screen display controller 4 performs necessary signal conversion for displaying an externally input video signal on a monitor device (not shown), and simultaneously displays characters on the monitor screen.
This is for displaying a figure or the like in an overlapping manner. Data such as characters and figures to be displayed on the monitor screen is so-called R
It is stored in the second memory 5 using the OM.

Further, the present apparatus comprises a memory controller 6, a microcontroller 7, a third memory 8 and an operation circuit 9. Memory controller 6
Is for controlling the write timing, address, etc. when the video signal converted into the digital signal by the A / D converter 1 is written and stored in the first memory 2.

The micro-controller 7 is for performing a brightness difference distribution change detection process as will be described later and for controlling the operation of the entire apparatus, and is constructed by using a so-called microprocessor. The program executed by the microcontroller 7 is stored in the third memory 8 using a ROM and can be read by the microcontroller 7 at any time.

The operation circuit 9 is a so-called pointing device for setting a box indicating a range to be monitored and a so-called pointing device for setting various variables on the monitor screen as shown in FIG. A keyboard is provided, and signals corresponding to these operations are output to the microcontroller 7.

Next, the brightness difference distribution change detection processing in this apparatus will be described with reference to the flow chart shown in FIG. First, the operation of this device will be described generally. This device uses a television camera (not shown in the figure) that uses an image pickup device such as a CCD device or an image pickup tube installed at an appropriate place in a place where some kind of monitoring is desired. Video signal from (1), input this video signal to a so-called monitor TV, and set a suitable range including the target to be monitored as a box on the monitor screen at the place where you want to monitor it. The change in the state of the monitoring target can be detected from the change in the brightness difference distribution (details will be described later) in the box.

The operation of the present apparatus will be specifically described below with reference to FIG. 2. As a premise, an image of a place to be monitored is preliminarily taken by a television camera (not shown) and the A / D converter 1 is set. It is assumed that the image is stored as a reference image in a predetermined storage area of the first memory 2 via the memory.
Further, it is assumed that the setting of the box for specifying the monitoring target has already been made by referring to the monitor screen. That is, as shown in FIG. 3, on the monitor screen showing a place to be monitored, a box, which is a rectangular frame indicating a specific range desired to be monitored by operating the operation circuit 9, is displayed.
Is set multiple times. After setting a desired number of boxes by operating the operation circuit 9, by pressing a predetermined function key (not shown) of the operation circuit 9, the information about the plurality of boxes is The data is stored in a predetermined storage area of the memory 2.

Under these assumptions, the initial values of various variables and flags are first set (step 100 in FIG. 2).
reference). In the embodiment of the present invention, since a plurality of boxes are set and the brightness difference distribution described later is obtained for each box, the flag F1 for recognizing the number of processes is
In this initial setting, it is set to "1".

Next, an image is input by a video signal from the outside (see step 102 in FIG. 2). That is, the video signal converted into a digital signal by the A / D converter 1 is stored as a current image in a predetermined area of the first memory 2 (an area different from the area in which the reference image is stored). . Subsequently, for the box, the luminance difference distribution is calculated between the relevant portion on the reference screen and the relevant portion of the image at the current time input in step 102 described above (see step 104 in FIG. 2).

That is, the brightness difference is calculated for each pixel between the reference image corresponding to the range specified by the box and the current image corresponding to the range similarly specified by the box, and each brightness is calculated. The distribution state of the number of pixels with respect to the difference is calculated, and the luminance difference distribution is obtained.

Here, assuming that there is no change in the brightness difference distribution between the reference image and the image at the present time, the number of pixels with a brightness difference of zero is the maximum, and the point is symmetrical about this point ( The shape becomes so-called line symmetry, and the distribution state is close to a so-called normal distribution. Note that, when the reference image and the image at the present time are completely coincident with each other, theoretically, the luminance difference distribution is a so-called line spectrum distribution in which a predetermined number of pixels appear at points where the luminance difference is zero. However, in actuality, due to the influence of noise or the like, the distribution characteristic has a certain extent of spreading around the zero brightness difference as described above.

Although there is no change in the imaged object in the box or the person itself, the overall brightness between the reference image and the actual image is changed by the operation of the automatic focusing function of the television camera. When a change occurs, the luminance difference that maximizes the number of pixels is not zero as shown in FIG.
The distribution shape does not change only when it becomes a specific value, and is similar to the distribution shape when there is no change between the reference image and the actual image.

After the brightness difference distribution is calculated as described above, it is judged whether the distribution states match (see step 106 in FIG. 2). That is, as described above, if the distribution shapes are similar only in that the value of the brightness difference that maximizes the number of pixels is different, the distribution state match (YE
On the other hand, when the distribution states are not similar to each other, it is determined that they do not match (NO).

Here, when the distribution states do not match, that is, when the object or person in the box moves from the state of the reference screen to the outside of the box, the luminance difference distribution is as shown in FIG.
As shown in, the distribution is asymmetrical about the point with the maximum number of pixels. When it is determined that the distribution states match in the determination in step 106 described above (YE
In the case of S), while proceeding to step 114 described later,
When it is determined that the distribution states do not match (in the case of NO), the determination value is calculated (see step 108 in FIG. 2).

If the judgment value is judged to be inconsistent in step 106, it is determined that there is a change between the reference image and the image at the present time, and a change output signal is not output immediately, but an image due to noise or the like is output. In order to avoid an erroneous determination due to a change in the above condition, the change output signal is output when the change state is equal to or more than a predetermined value, and is calculated as described below. For example, the brightness difference distribution is shown in FIG.
As shown in, the distortion occurs at two points of the luminance difference −Y1 and the luminance difference + Y2, and the distribution shape around the point Yc where the number of pixels at the center of the distribution is the maximum is the luminance difference. In the case of left-right asymmetry, the size of each asymmetric portion is first obtained.

Here, the size of the asymmetric portion is obtained as follows. For example, in the example of FIG. 5, the size of the asymmetric portion at the point of the luminance difference −Y1 is
First, a point that is point-symmetric with respect to the point of luminance difference −Y1 is obtained with the point of luminance difference Yc, which is the center of the distribution, as the center of point symmetry (in the case of this example of FIG. It is located on the positive brightness difference side), and is defined as the difference between the brightness difference at this point and the brightness difference −Y1. With respect to the brightness difference + Y2, similarly, a point that is point-symmetric with respect to the point Yc is obtained, and the difference between the brightness difference and the brightness difference + Y2 at this point is the size of the asymmetric portion. The reason for determining the size of the asymmetrical part is to consider the fact that the spread of the distribution changes due to the effect of changes in noise due to temperature changes, etc., and it is difficult to determine the normal distribution. is there.

Then, the size of each of the obtained asymmetric portions is multiplied by a so-called weighting coefficient which is predetermined according to the brightness difference. Here, the weighting coefficient is set to be successively increased as the distance from the point of the brightness difference where the number of pixels becomes maximum.
With such a setting, as the luminance difference becomes larger, the number of pixels originally distributed is smaller. Therefore, when the weighting as described above is not considered, even if the distribution change occurs at the position where the luminance difference is large, This is for avoiding that such a result cannot be obtained when the probability of exceeding the predetermined value is low and it should be judged that there is an inherent change.

Then, the sum of the two multiplication values obtained as described above is obtained to obtain the step 108.
The calculation of the judgment value in is ended.

After the judgment value is calculated as described above, it is judged whether or not the judgment value exceeds the predetermined value (see step 110 in FIG. 2), and it is judged that the predetermined value is not exceeded. If it is (NO), the process proceeds to step 114 described later, while if it is determined that the predetermined value is exceeded (YES), a change output signal is output from a predetermined input / output terminal of the microcontroller 7. Will be output (see step 112 in FIG. 2).

Here, the change output signal is for notifying the outside when the predetermined change occurs in the box to be monitored as described above, and its specific form is various. Conceivable. For example, an opening signal (or closing signal) of a mechanical contact called a so-called relay contact is output, and a lamp is lit by using this relay contact in an external circuit, or a change occurs. It is preferable to output the serial data in a predetermined format including the box number and the box number.

Further, depending on the box in which the change has occurred,
The output level may be changed stepwise. Also,
The output time of this change output signal can be various forms such as output for a predetermined time or output until a reset command is issued from the operation circuit 9.

After the change output signal is output as described above, the flag F1 for recognizing the box which is the target of the brightness difference distribution calculation processing is the predetermined value K (= the number of boxes). It is determined whether or not (step 11 in FIG. 2).
4), when it is determined that F1 = K (in the case of YES)
In this case, since the above-mentioned processing has been completed for all the boxes to be processed, the series of processing is completed, the process returns to the main routine (not shown), and after the predetermined processing in the main routine is performed, The series of processing shown in FIG. 2 is performed.

On the other hand, when it is determined that F1 = K has not yet been reached (in the case of NO), "1" is added to the current F1 (see step 116 in FIG. 2), and the process goes to the previous step 104. Returning to the next box, the series of processes described above is similarly performed.

In the embodiment of the invention described above, the storage means in claims 3 and 4 is shown in FIG. 2 by the A / D converter 1, the first memory 2, the memory control 6 and the microcontroller 7. This is realized by executing step 102. The brightness difference distribution calculating means in claims 3 and 4 is realized by the execution of step 104 shown in FIG.

Further, the distribution state judging means in claims 3 and 4 is realized by the execution of step 106 shown in FIG. 2 by the microcontroller 7. Still further, the judgment value calculating means according to claim 4 is the step 106, 1 shown in FIG.
It is realized by executing step 08, and the change output means described in the same paragraph is realized by executing steps 110 and 112 shown in FIG.

In the embodiment of the invention described above, when it is determined that the brightness difference distribution is not symmetrical as shown in the flowchart of FIG. 2, a predetermined calculation is performed (step 106 in FIG. 2). , 108), the change output signal is output according to the calculated value (see steps 110 and 112 in FIG. 2). However, the simplest configuration is that the luminance difference distribution is calculated in step 106 in FIG. The change output signal may be output when it is determined to be asymmetric (in the case of NO).

In particular, when the change of the image always appears in a clear state and the possibility of misjudgment is small, the arithmetic processing of steps 108 and 110 shown in FIG. 2 can be omitted. With the simplification of the configuration, the processing of the entire device becomes faster. And in this case, steps 106 and 11
By executing step 2, the change output means according to claim 4 is realized.

[0048]

As described above, according to the first and third aspects of the present invention, the brightness difference for each pixel between two images is obtained for a predetermined range of the images, that is, a monitored range. Further, unlike the conventional case, the pixel number distribution for each brightness difference is obtained and the change in the predetermined range of the two images is detected based on the brightness difference distribution characteristic. Since it is not necessary to determine whether or not there is a change in all the pixels of the image, the processing time can be shortened, and the presence or absence of a change is detected by the shape of the brightness difference distribution characteristic, so a simpler configuration than before Therefore, an inexpensive device can be provided.

According to the second and fourth aspects of the present invention, when the luminance difference distribution is obtained after obtaining the luminance difference distribution as in the first and third aspects of the present invention, if the distribution characteristic becomes asymmetric, the asymmetry It is necessary to judge presence / absence of change for all pixels of two images by differentiating from the conventional one by adopting a configuration in which the degree of asymmetry is numerically expressed for a part and the presence / absence of image change is detected by its size. Since there is no error, the processing time can be shortened, and erroneous judgments due to changes in the brightness difference distribution characteristics due to factors such as noise that are not the original changes in images can be avoided, further improving reliability. It can be done.

Further, in any of the first to fourth aspects of the invention, since the presence or absence of the image change is detected based on the change of the shape of the brightness difference distribution characteristic, for example,
When the change between the two images is only the overall brightness, it is possible to prevent the image change and the erroneous determination that the monitoring target has moved within a predetermined range from being erroneously determined with a simple configuration, unlike the conventional case. However, it can be used as a monitoring method and a monitoring device with practicality.

Furthermore, if a portion to be particularly monitored in the image is once set as a predetermined range, it is possible to recognize that a change has occurred by the change output signal when any change occurs. Therefore, it is not necessary to constantly monitor the screen of the monitor device, and efficient monitoring can be performed.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a configuration example of an embodiment of a video signal luminance difference distribution change detection device according to the present invention.

FIG. 2 is a flowchart for explaining a procedure of brightness difference distribution processing in the video signal brightness difference distribution change detection device shown in FIG.

FIG. 3 is a schematic diagram schematically showing an example of setting a box on a monitor screen.

FIG. 4 is a characteristic diagram showing an example of a luminance difference distribution when the change between the reference image and the actual image is only luminance.

FIG. 5 is a characteristic diagram showing an example of a brightness difference distribution when a change occurs in an actual image.

[Explanation of symbols]

 1 ... A / D converter 2 ... First memory 3 ... Timing generator 4 ... On-screen display controller 5 ... Second memory 6 ... Memory controller 7 ... Microcontroller 8 ... Third memory 9 ... Operation circuit

Claims (4)

[Claims] 1. A luminance difference is obtained for each pixel in a predetermined range within the image between a reference image and an image that is symmetrical to be compared, and a distribution of the number of pixels for each luminance difference is calculated. A video signal characterized in that, when the distribution characteristic is left-right asymmetrical about the location of the brightness difference having the maximum number of pixels, a change occurs in a predetermined range of the symmetrical image of the comparison. Luminance difference distribution change detection method. 2. A luminance difference is obtained for each pixel in a predetermined range within the image between a reference image and an image that is symmetrical to be compared, and a distribution of the number of pixels for each luminance difference is calculated. If this distribution characteristic is left-right asymmetrical about the location of the brightness difference that is the maximum number of pixels, the size of the asymmetrical part is obtained, and it is obtained by multiplying the size of this asymmetrical part by a predetermined coefficient. It is characterized in that the multiplied values are added for each asymmetrical point, and when the added value exceeds a predetermined value, a change occurs in a predetermined range of the image symmetrical to the comparison. Video signal brightness difference distribution change detection method. 3. A reference image serving as a reference for comparison, storage means for storing a comparison image to be compared with the reference image, and a predetermined range in the image between the reference image and the comparison image. A luminance difference distribution calculating unit that obtains a luminance difference for each pixel and also obtains a distribution of the number of pixels for each luminance difference, and a luminance difference distribution obtained by the luminance difference distribution calculating unit is a luminance difference portion of the maximum number of pixels A distribution state determination unit that determines whether or not the line is symmetrical about the center, and a change output unit that outputs a predetermined signal when it is determined that the brightness difference distribution is asymmetric by the distribution state determination unit, A video signal luminance difference distribution change detection device comprising: 4. A reference image as a reference for comparison, a storage unit for storing a comparison image to be compared with the reference image, and a predetermined range in the image between the reference image and the comparison image. A luminance difference distribution calculating unit that obtains a luminance difference for each pixel and also obtains a distribution of the number of pixels for each luminance difference, and a luminance difference distribution obtained by the luminance difference distribution calculating unit is a luminance difference portion of the maximum number of pixels Distribution state determination means for determining whether or not the line center is line-symmetrical, and when the distribution state determination means determines that the brightness difference distribution is asymmetric, the size of the asymmetric portion is obtained, and this asymmetry is determined. The judgment value calculating means for multiplying the size of the part by a predetermined coefficient according to the brightness difference at the point where the asymmetry occurs, and adding this multiplication value for the part where the asymmetry occurs, to the judgment value calculating means. Ri If the calculated value exceeds a predetermined value, the video signal luminance difference distribution variation sensing apparatus characterized by comprising anda change output means for outputting a predetermined signal as there changes in comparison screen.