JP2001169270A - Image supervisory device and image supervisory method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image supervisory device and an image supervisory method that can automatically decide an image change due to intrusion of a person or a thing being a substantial detection object from an image change due to fluctuation in an environmental condition, a camera-shake and an automatic adjustment function on the basis of a statistic distribution of a feature quantity obtained from the image and automatically estimate a weather and a shining status so as to reduce the opportunities of misdetection due to other conditions than the intrusion of a person or a thing thereby considerably enhancing the detection performance. SOLUTION: The image supervisory device comprises an image entry section 11 that enters an image obtained from an optional image pickup unit 101, a characteristic distribution calculation section 12 that obtains a statistic distribution of an optional feature quantity of the image entered by the image entry section 11, an image change tracing section 13 that traces a changing state of the image from the characteristic distribution obtained by the characteristic distribution calculation section 12, and a result decision section 14 that decide the intrusion of a person or a thing being a detection object from the tracing result of the changing state obtained by the image change tracing section 13.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention
The present invention relates to an image monitoring apparatus and an image monitoring method for automatically monitoring a desired target (area) by using an imaging unit such as an LTE (Television) camera.

[0002]

2. Description of the Related Art Image processing technology for automatically detecting intrusion of a suspicious person or a vehicle from an image in a monitoring area obtained by an imaging means such as an ITV camera has been used mainly in the security field. Therefore, there was an effect of cost reduction, such as a reduction in the length of monitoring work performed by the monitoring personnel and a reduction in the number of personnel required for monitoring.

[0003] As an image processing method for detecting a change in an image in such an automatic monitoring apparatus, a method of detecting a change between a plurality of images (frames) continuously photographed in time series or a background image created in advance and a current image is used. There is a method of obtaining a difference between the image and the image.

As a result of detecting a change in an image by calculating a difference between a plurality of screens as described above, if an arbitrary change is found in the monitoring area (in many cases, a large change is made), monitoring is performed. It is a conventional image monitoring device to issue a warning signal on the assumption that some kind of intrusion or abnormality has occurred in the area.

However, when a simple criterion such as issuing a warning when a large image change occurs in the monitoring area is used, an image change due to time fluctuations such as weather and sunshine is determined to be “intrusion detected, warning issued. Was detected incorrectly.

[0006] Also, the AGC of the camera itself, which is an imaging device,
(Auto Gain Control) and other automatic adjustment functions could not correctly detect the position of an intruder.

Further, depending on the installation conditions of the camera, the image changes when a vehicle such as an automobile or a train passes near the installation site of the camera, or when the camera itself vibrates due to strong wind or the like. In some cases, it was incorrectly detected as "intrusion."

[0008] Due to such erroneous detection, only the image change due to the intrusion of a person or an object, which is the original detection purpose, cannot be correctly detected.

Further, as environmental conditions for photographing, there are weather, such as sunny, cloudy, rain, or fog, and sunshine, such as day or night. These environmental conditions are factors that greatly affect the detection performance of the automatic monitoring device. It is. Despite these factors, various conditions could not be determined just by photographing with a camera.

Therefore, in order to reduce the malfunction of the automatic monitoring device and improve the performance, the information on the weather and the sunshine condition is obtained by relying on other sensors or other means such as manual input operation to detect the automatic monitoring device. There is no other choice but to set the conditions to the optimal style at that time, and there is a problem that labor and installation costs are high.

[0011]

In the conventional image monitoring apparatus described above, since the detection method and the judgment standard of the image change are simple, fluctuations in environmental conditions such as weather and sunshine conditions and vibrations of the camera itself are performed. And the automatic adjustment function cannot correctly detect only intruders. In addition, there is a problem that environmental conditions that greatly affect the detection performance can be obtained only by relying on other means.

In view of the foregoing, the present invention has been made in view of the above-mentioned conventional problems. Based on the statistical distribution of feature amounts obtained from an image, it is possible to reduce the image change caused by the intrusion of a person or an object, which is the original detection object, and the By automatically judging changes in conditions, image changes due to camera vibration and automatic adjustment functions, and automatically estimating the weather and sunshine conditions, it reduces false detections other than intrusion of people or objects, and detects It is an object of the present invention to provide an image monitoring device and an image monitoring method that significantly improve performance.

[0013]

In order to achieve the above object, an image monitoring apparatus according to the present invention comprises: an image input unit to which an image obtained from an arbitrary image pickup device is input; A feature distribution calculation unit for calculating a statistical distribution of an arbitrary feature amount of the image, an image change calculation unit for calculating a change state of an image from a change in the feature distribution obtained by the feature distribution calculation unit, Consisting of a result determination unit that distinguishes between an image change caused by a change in environmental conditions or shooting conditions and an image change caused by a change in the environmental conditions or shooting conditions other than an image change based on the calculation result of the change state obtained by the calculation unit Is done.

In the image monitoring method according to the present invention, there is provided an image inputting step of inputting an image obtained from an arbitrary image pickup device, and a feature distribution calculating step of obtaining a statistical distribution of an arbitrary characteristic amount of the input image. And an image change calculation step of calculating a change state of an image from the feature distribution obtained in the feature distribution calculation step; and an environmental condition or a shooting condition from the calculation result of the change state obtained in the image change calculation step. And a determination step of distinguishing between an image change caused by a change in the image condition and an image change not caused by a change in the environmental condition or the photographing condition.

Also, in a computer-readable recording medium on which a program for operating a computer according to the present invention is recorded, an image input function for inputting an image obtained from an arbitrary image pickup means is provided. A feature distribution calculation function for calculating a statistical distribution of an arbitrary feature amount of an image; an image change calculation function for calculating a change state of an image from the feature distribution obtained by the feature distribution calculation means; From the calculation result of the change state obtained by the above, and a result determination function of determining the intrusion of a detection target person or object. It is characterized in that an image change due to a change and an image change due to the intrusion of a detection target person or object are distinguished and detected.

Further, various information obtained by each function can be recorded on a recording medium such as an optical disk or a magnetic recording medium, and can be appropriately displayed on a display means such as a display.

[0017]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of an image monitoring apparatus according to a first embodiment of the present invention. FIG. 2 is a flowchart of the image monitoring method according to the first embodiment of the present invention.
FIG. 4 is a diagram illustrating a setting example of a detection area in an image, FIG. 4 is an explanatory diagram of a feature distribution calculating method, and FIG. 5 is an explanatory diagram of a feature distribution calculated at two times under different conditions. FIG. 6 is an explanatory diagram of the feature distribution calculation when the automatic adjustment function of the imaging apparatus is activated, and FIG. 7 is an explanatory diagram of a method of correcting an image using the result from the image change calculating unit. FIG. 4 is an explanatory diagram when a background difference process is performed on an image after correction.

In the image monitoring apparatus, an image input unit 11, a feature distribution calculation unit 12, an image change tracking unit 13 (image change calculation unit), and a result determination unit 14 are cascaded.

An arbitrary image pickup device 101 such as an ITV camera is connected to the image input unit 11. In addition, the result determination unit 14 includes a display device 102 such as a display that displays the determination result, and a communication device 103 that outputs (transmits) the determination result to the outside of the image monitoring device. still,
The communication device 103 can transmit information to a predetermined place wirelessly or online.

The operation of the embodiment having such a configuration will be described.

(1) A predetermined monitoring area is imaged by the imaging device 101. The captured image is sent to the image input unit 11 (image input step).

(2) The video signal continuously output from the imaging device 101 is sequentially A / D-converted in the image input unit 11. The converted video signal is stored as a digital image or an image sequence in a storage unit provided in the image input unit 11. Further, it outputs the pixels of the entire image at an arbitrary time or the pixels of an arbitrary area at an arbitrary time to the feature distribution calculating means 2 at the subsequent stage (image input step).

(3) Based on information output from the image input unit 11, a feature amount preset for an arbitrary time or an arbitrary region of an image or an image sequence stored in the image input unit 11 is calculated, and its statistical value is calculated. (A feature distribution calculation step).

Here, the characteristic amount indicates arbitrary information that can be obtained in each pixel or an arbitrary region. For example, from simple values such as lightness and color information at each pixel, the result of differentiation or integration by space or time, the result of superposition of an arbitrary image filter, statistics such as average or variance, etc., and the difference from the background image created in advance Processing, such as the result of performing any object detection or region division processing, and the fillet diameter and area of the region selected from the result, the amount of motion, etc., any scalar amount or vector amount that can be calculated from the image,
Furthermore, it is possible to use an arbitrary combination of feature values.

In the present embodiment, the difference between the brightness value at each time and each pixel of the image and the brightness value at each pixel in the background image created beforehand is determined by the feature amount. Used as

That is, based on a method of detecting an image change by a difference process between a captured image and a background image, the image is composed of lightness information of each pixel (shade image), It is assumed that the background image is created by the following.

Any method can be used as a method of creating the background image. For example, (a) one time (frame) or an image preceding by a specific time is used as a background image as it is (in this case, a continuous difference method). (A) It is known that there is no intrusion of a person or an object. One image is extracted from the image sequence and used. (C) The image is created by performing any operation such as calculating the brightness or the average or a weighted sum of any feature amount at each pixel of the image sequence. By performing an arbitrary operation such as applying a median filter or a Kalman filter to the brightness of each pixel or a time series of an arbitrary feature amount in an image sequence (it may be unknown whether there is an intrusion or not). That is.

The background image created in this manner may be updated according to the time, or the background image created once may not be used and updated.

Now, the pixel (x,
The brightness value at y) is I (x, y, t), and the brightness value of the pixel at the same position in the background image at that time is B (x, y, t) (the background image is updated temporally). In this case, if t changes, the background image B also changes).

A difference value D (x,
y, t) is obtained by the following equation (1).

[0032]

(Equation 1) (1) Here, the value D representing the difference from the background may be obtained by the above-described subtraction, or may be obtained by dividing by the brightness value B of the background pixel to obtain the change ratio with respect to the background. , By any operation.

The feature distribution F (t) at time t is calculated by obtaining a feature amount for each pixel (x, y) belonging to the detection area R set in the image, and obtaining this statistical distribution. Is done. The detection region R may be the entire image or may be set to any shape as shown in FIG.

Here, if the feature distribution F is simply obtained as a histogram of the feature amount D, the graph of FIG. 4 is obtained.
The case where a feature distribution is obtained as a histogram is described as follows.

[0035]

(Equation 2) It is.

In this way, a feature distribution F (t) at an arbitrary time t is obtained.

In FIG. 4, the detection area is set to be rectangular, and an image is shown in the detection area. In addition, since the captured image has no change compared to the background image, the feature distribution F has a shape that is close to a normal distribution with a single peak and an average of 0.

(4) In the image change tracking unit 13, the arbitrary time t calculated by the feature distribution calculation unit 12 as described above
Is analyzed, and the result is transmitted to the next result result determination unit 14 (image change calculation step).

Now, as shown in FIG. 5, two times t1, t2
The process in the image change calculation unit 13 will be described by taking as an example the case where the characteristic distributions F (t1) and F (t2) are obtained.

In FIG. 5, at time t1, there are no intruders such as persons and vehicles, and there is almost no change from the background. On the other hand, at time t2, there is a change in the sunshine intensity such as clearing of the clouds and the like, and there is an environmental change such that the entire screen becomes brighter, and a person (brighter than the background) enters. .

At time t1, since there is almost no difference (brightness difference) between the captured image at time t1 and the background image, the characteristic distribution F (t1) has a single-peak distribution having an average of 0 as in FIG. Become.

On the other hand, the feature distribution obtained at time t2
In F (t2), since the brightness value of each pixel in the background region other than the person becomes uniformly large (brighter), the average value of the distribution coming from the background portion becomes larger than 0 (p22), and the distribution by the person region Can be reduced to a portion darker than the background (p21), and the feature distribution F (t2) is obtained as a bimodal combination of these two distributions.

Here, an image change is detected by a conventional simple difference processing between a captured image and a background image.

That is, when the change is detected based on the magnitude of the value D at each pixel, the change in the brightness of the background itself is larger than the change in the brightness by the intruding person at the time t2, and thus the determination is made based only on the magnitude of the change. In such a case, not only the intruder but also the entire background is erroneously detected as an image change.
Only intruders cannot be detected correctly.

In order to avoid such an adverse effect on the background difference, the image change calculation unit 13 calculates the peak (peak) of the characteristic distribution due to the background region at each time, thereby obtaining the image change due to the background change and the intrusion of the object. It is distinguished from image change.

First, the peak of the characteristic distribution at each time is obtained. Since there is only one peak at time t1, a peak can be easily obtained. However, at time t2, since there are two (plural) peaks, it is necessary to distinguish each peak.

As a method of obtaining peaks by distinction, there is a method in which the maximum value in the feature distribution is most simply regarded as a peak. In addition, there is a method such as an EM algorithm that approximates a complex distribution by a sum of a plurality of normal distributions (see References 1 and 2).

[Reference 1] A. Dempster, N. Laird, an
d D. Rubin, Maximum likelihood from incomplete dat
a via the EM algorithm, Journal of the Royal Stati
sticalSociety, 39 (Series B): 1-38, 1977. [Ref. 2]… N. Friedman, S. Russell, Image Segmen
tation in video sequences: A probabilistic approac
h, Proc.Thirteenth Conf.on Uncertainty in Artific
ial Intelligence (UAI), 1997 Using this method, a normal distribution p with mean μ and standard deviation σ
(X) is represented as follows,

(Equation 3) (3) Parameters such as the mean, standard deviation, and weight of the normal distribution corresponding to p21 and p22 are represented by μ21, σ21, w21, and μ2.
2, σ22, w22, the feature distribution F (t2) can be approximated by a weighted sum of two normal distributions as follows:
1 and p22 can be distinguished.

[0049]

(Equation 4) (4) Here, in order to correctly detect only the human intrusion area at the time t2, it is necessary to determine which of the plural (two in this example) distributions is caused by the background area.

For this purpose, the distribution close to the distribution by the background region at the previous time t1 (in this example, there is only p1) is determined, and it is determined that the close distributions are distributions by the background region. , The distribution of the background area at each time is calculated. As the distribution closeness S, for example, the Euclidean distance between parameters in each distribution can be calculated as follows.

[0051]

(Equation 5) ... (5) As the proximity S of the distribution, not only the distance between parameters but also
It is possible to calculate by any calculation using each distribution.

In this way, the closeness S is obtained by p21 and p22, and it can be seen that the distribution closer to p1 is the distribution due to the background area (p22 in this example; the downward arrow on the right side in FIG. 5).

As described above, the image change tracking section 13 tracks the distribution of the background area at each time, and sends the result to the next result determination section 14.

In the embodiment described above, the case where there is an environmental change in the background has been described. However, an image may be changed by an automatic adjustment function such as AGC of an image pickup means such as a camera.

In FIG. 6, the characteristic distribution sequences F (t1) to F (t4) at the respective times t1 to t4 are obtained and described on the right side (in the figure) of the captured image. This shows a case in which the brightness of the entire image changes due to the intrusion of a white vehicle into the detection area despite the environmental change (change of the imaging area).

At time t1 before intrusion, there is no significant change between the photographed image and the background image, so that the feature distribution F (t
1) is obtained as a single-peak distribution similarly to FIG.

Subsequently, when the vehicle enters as time t2, t3, and t4 elapse, the vehicle is white, and the imaged area (pixels of the vehicle) is brighter than the background. Further, since it is regarded as a large area, another peak (peak) starts to be formed in a plus portion of the feature distribution and gradually increases, so that a bimodal feature distribution is obtained.

At this time, the invasion of a bright (white) vehicle area causes the AGC function of the camera to work, thereby correcting the brightness in a direction to darken the entire image. It will be dark. In the case where the automatic adjustment function of the camera works as described above, the conventional monitoring device cannot correctly detect only the intruding object from the image change, but by using the image change tracking unit 13 of the present invention, Even if the background fluctuates, the feature distribution is tracked over time (downward arrow in the graph of FIG. 6), and a result determination is performed to explain the details in the next step. An area can be detected separately.

(5) Finally, the result determination unit 14 performs image processing using the image change tracking information obtained by the previous image change tracking unit 13 in chronological order, and then performs image processing. It is determined whether a person, an object, or the like is invading inside (determination step).

The result determination in FIG. 5 (when there is a background change and a person invasion at time t2) will be described.

Since the image distribution calculation means 3 tracks the feature distribution of the background area (p1 → p22 in FIG. 5), by using this information to correct the image at time t2, the brightness due to the background fluctuation is obtained. (Feature) Removes the effect of the change.

That is, the distribution p2 of the background area at time t2
When the background variation is corrected by subtracting the brightness value of each pixel of the image by this value using the average value μ22 of 2, the brightness of the background portion becomes the same as the background image created in advance as shown in the lower part of FIG. Can be the same. At this time, the distribution of the background region is p22 ', and the average value is also μ22'. The image at the time t2 before the correction is as shown in the middle part of FIG.

Here, as a method of correcting an image, as shown in the present embodiment, when the difference D between the captured image and the background image is used as the feature amount, the average value of each pixel is equal to the average value of the background distribution. It can also be subtracted from the brightness value, and when using other feature amounts such as the change ratio between the brightness value of the imaged pixel and the brightness value of the background pixel, the image is obtained by an arbitrary method according to the feature calculation method. Can be corrected.

If the difference processing with the background image is performed without correcting the image, the entire background may be detected as an image change as described above. However, if the difference processing with the background image is performed using the corrected image as in the present embodiment, the background change is not detected as shown in FIG. 8, and only the area of the intruding object is detected. Will be.

In this way, even if the background image changes due to environmental changes such as changes in sunlight or the automatic adjustment function of the camera, only the intrusion of a person or object, which is the original detection purpose of the monitoring device, is correctly detected. Erroneous operation such as erroneously issuing a warning to the user can be greatly reduced.

Next, the configuration and operation of an image monitoring method according to a second embodiment of the present invention will be described with reference to FIG.

The feature of the second embodiment is that, by estimating changes in environmental conditions and imaging conditions from the result of the difference processing, it is determined whether or not the detected image change is a detection purpose, thereby improving reliability. It is to make it.

FIG. 9 is a characteristic distribution diagram calculated from images captured under various conditions.

When a change in an image is detected by a difference process between a captured image and a background image to detect the intrusion of a person or an object, if the captured image has hardly changed from the background image, The characteristic distribution (here, as in the previous examples, the case where the brightness difference from the background image is used as the characteristic amount will be described) is a single-peak normal distribution having an average value of 0 (the upper part of FIG. 9). 9), when the brightness of the entire captured image changes (for example, becomes darker) due to only environmental fluctuations such as sunlight fluctuations, the average value shifts from 0 to, for example, the right side while maintaining the shape of the characteristic distribution (the middle part of FIG. 9). ).

However, in the upper and middle diagrams of FIG. 9, it is assumed that there is no intruder, and the feature distribution is equal to the feature distribution of the background region.

Further, when the installation conditions of the imaging means such as a camera are poor and a vehicle such as an automobile or a train passes nearby, or when the camera itself vibrates due to the influence of a strong wind or the like, the image is taken in the image. When the position of the background object shifts, the variance increases while the average value of the feature distribution remains the same.

From these facts, conversely, from the change in the characteristic distribution obtained in the photographed image, the change in the environmental condition or the photographing condition at that time can be estimated from the change in the characteristic distribution.

For example, when only the average value of the feature distribution in the background area changes, environmental conditions such as fluctuations in sunlight intensity change. When only the variance changes while the average value remains unchanged, the camera shakes. For example, it can be determined that the shooting condition has changed.

As described above, when an image change is detected by image processing such as background subtraction, a change in environmental conditions and photographing conditions at that time is estimated from the characteristic distribution.
This can be used as an index for determining whether or not the detected image change is truly due to the intrusion of a person or an object, or whether it is due to a change in environmental conditions or imaging conditions.

Further, when it is determined that a change in environmental conditions or photographing conditions has occurred, it is possible to greatly reduce malfunctions of the image monitoring apparatus, for example, by not giving a warning to the user. it can.

Next, the configuration and operation of the image monitoring method of the present invention will be described with reference to FIG.

The feature of the third embodiment is that the weather and the sunshine state at the time when the image is photographed are estimated from the result of the difference processing, and the reliability of the extraction of the object as a true detection purpose is improved. It is.

FIG. 10 shows various weather and sunshine conditions (time).
FIG. 6 is a feature distribution diagram calculated in FIG.

As described in the first embodiment, there is a method in which the image change calculating means 3 can approximate a feature distribution having a complicated shape by a sum of a plurality of normal distributions. By distinguishing each peak, the weather and the sunshine state at the time of capturing the image can be estimated.

The upper diagram in FIG. 10 shows an example of the characteristic distribution when the vehicle is invading in the image (it is assumed that there is no background fluctuation), but as described in the first embodiment, When there is an intrusion, a multimodal feature distribution is obtained in which two (plural) distributions of a peak due to the background region and a peak due to the intruding object region are added.

In the upper diagram, it is assumed that the weather is cloudy. However, as shown in the middle diagram of FIG. 10, when the weather is fine, the “shadow” of the intruding object is reflected in the image. The distribution due to the shadow region appears at a place darker than the distribution due to the background in the feature distribution.

When the time changes and it becomes nighttime, the vehicle turns on the headlights, so that the distribution of the reflection area of the headlights appears brighter than other distributions.

Further, although not shown, if the weather is rainy or snowy, many small image changes due to raindrops or the like appear, so that the characteristic distribution of the observed background region can be compared with when the feature distribution is fine or cloudy. Therefore, the nature of the feature distribution may change, for example, the variance is greatly observed.

By measuring the peaks and properties of a specific feature distribution in this manner, information on the time zone (sunshine condition) such as whether the time when the image was taken is daytime or nighttime, It is possible to estimate weather conditions such as whether the imaging area is sunny, cloudy, rainy, snowy, or fogy.

Further, the weather and the sunshine state can be estimated from the photographed image without depending on other various sensors or means such as manual input operation of the imaging environment, and the conditions of the image monitoring apparatus can be optimized based on the estimation result. By setting (such as raising or lowering the sensitivity of the camera or resetting the detection area), malfunctions can be greatly reduced and the reliability of detection of the detection target object can be improved.

The present invention is not limited to the above embodiment,
It goes without saying that various modifications can be made without departing from the spirit of the invention. For example, by providing a plurality of imaging devices, the reliability of specifying a detection target can be improved.

[0087]

As described above, according to the present invention, the detection performance of the image monitoring device can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram of a first embodiment of an image monitoring apparatus according to the present invention.

FIG. 2 is a flowchart of a first embodiment of the image monitoring method of the present invention.

FIG. 3 is a diagram showing an example of setting a detection area in an image.

FIG. 4 is an explanatory diagram of a feature distribution calculation method.

FIG. 5 is an explanatory diagram of a characteristic distribution calculated at two times under different conditions.

FIG. 6 is an explanatory diagram of calculation of a feature distribution when an automatic adjustment function of an imaging device is activated.

FIG. 7 is an explanatory diagram of a method of correcting an image using a result from an image change calculation unit.

FIG. 8 is an explanatory diagram in the case where background subtraction processing is performed on a corrected image.

FIG. 9 is a feature distribution diagram calculated from images taken under various conditions for explaining a second embodiment of the image monitoring method of the present invention.

FIG. 10 is a feature distribution diagram calculated in various weather and sunshine states for explaining a third embodiment of the image monitoring method of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Image input part 12 Feature distribution calculation part 13 Image change tracking part (image change calculation part) 14 Result determination part 101 Imaging device 102 Display device 103 Communication device

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference) GG22 GG23 GG30

Claims (8)

[Claims] An image input unit to which an image obtained from an arbitrary image pickup device is input; a feature distribution calculating unit for obtaining a statistical distribution of an arbitrary feature amount of the image input by the image input unit; An image change calculation unit that calculates a change state of an image from a change in the feature distribution obtained by the distribution calculation unit; and a calculation result of the change state obtained by the image change calculation unit. An image monitoring device, comprising: a result determination unit that distinguishes between an image change and an image change other than an image change due to a change in environmental conditions or imaging conditions. 2. The method according to claim 1, wherein said image change calculating section estimates changes in environmental conditions and photographing conditions by calculating changes in properties of the characteristic distribution obtained by said characteristic distribution calculating section. 2. The image monitoring device according to 1. 3. The image monitoring apparatus according to claim 1, wherein the image change calculating unit estimates weather and sunshine state from a specific property of the feature distribution obtained by the feature distribution calculating unit. . 4. The image monitoring method according to claim 1, wherein the image change not caused by the change in the environmental condition or the photographing condition is an intrusion of a detection target person or object into the image. . 5. An image inputting step of inputting an image obtained from an arbitrary image pickup means; a feature distribution calculating step of obtaining a statistical distribution of an arbitrary feature amount of the input image;
An image change calculation step of calculating a change state of an image from the feature distribution calculated in the feature distribution calculation step; and a change in environmental conditions and photographing conditions based on the calculation result of the change state calculated in the image change calculation step. Image change due to
A determining step of distinguishing between an image change not caused by a change in an environmental condition or a photographing condition. 6. The image monitoring method according to claim 5, wherein the image change not caused by a change in the environmental condition or the photographing condition is an intrusion of a detection target person or object into the image. . 7. The image monitoring method according to claim 5, wherein a change in an environmental condition or a photographing condition is estimated by calculating a change in a property of the characteristic distribution obtained in the image change calculating step. . 8. The image monitoring method according to claim 5, wherein the weather and sunshine state are estimated from specific characteristics of the characteristic distribution obtained in the characteristic distribution calculating step.