JP2001352537A - Supervisory system by moving picture - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a supervisory system that can detect only falling rocks with high accuracy without being affected by a normal motion such as tree nodding in the case of using a moving picture to supervise e.g. falling rocks. SOLUTION: A moving picture in a supervisory area is received (S2), and the motion vector of each pixel between two consecutive frame images is calculated (S3-S5). The calculation of the motion vector uses a parameter (α) denoting the degree of spatial smoothness of the motion vector. The parameter (α) is selected for each place depending on the environmental elements of the supervisory area (e.g. gradient of slope, direction of gradient, number of trees, wind velocity, wind direction and amount of rainfall) (S1, setting of α0-α3, and γ) and the parameter is adjusted depending on changes in the environmental elements at any time (S10, correction of γ). Then, by extracting a motion vector in matching with a moving direction condition (θ) of falling rocks among motion vectors that are calculated, only a motion of the falling rocks is detected. The moving direction condition (θ) of falling rocks is selected for each place depending on the environmental elements of the supervisory area (S1, setting of θ1-θ4, β1, β3, k) and adjusted at any time depending on changes in the environmental elements (S10, correction of k).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for monitoring the movement of an object by using a moving image obtained by photographing the object, and is particularly suitable for outdoor monitoring of, for example, a landslide or a falling rock. .

[0002]

2. Description of the Related Art Conventionally, monitoring of landslides, falling rocks, and intruding objects has been performed by processing detection signals from contact sensors and moving images from video cameras. Monitoring with a contact sensor stops at only a specific point where the sensor is installed, whereas monitoring with a moving image from a video camera has the advantage that the entire shooting area can be monitored.

[0003] Conventional monitoring using a moving image is based on a difference image analysis technique of calculating a difference between two consecutive frames.
It is trying to detect abnormal movements such as landslides, falling rocks and intruding objects.

[0004]

Outdoors, there are various natural normal movements such as trees swaying, the weather changing, and birds flying. Accurately distinguishing these normal movements from the abnormal movements described above,
It is difficult with conventional difference image analysis. Therefore, conventional monitoring using moving images is not suitable for outdoor monitoring.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a monitoring technique based on a moving image which can accurately detect only a desired movement even in an area where normal movements naturally exist, such as outdoors. is there.

[0006]

According to a first aspect of the present invention, a surveillance system for analyzing a moving image of a monitoring area to determine the presence or absence of an object performing a desired movement in the monitoring area is provided. A motion vector of each pixel in a frame is calculated from two temporally different images included in the image. When calculating the motion vector, a parameter indicating the degree of spatial smoothness of the motion vector is calculated. Movement vector calculation means configured to be used; object detection means for determining the presence or absence of an object performing a target movement based on the movement vector calculated by the movement vector calculation means; and spatial smoothing of the movement vector. Parameter setting means for setting a parameter indicating the degree of the degree according to the environmental element of the monitoring area.

[0007] This surveillance system, when calculating a motion vector of each pixel in a frame from two temporally different images included in a moving image of a surveillance area, first,
After setting a parameter indicating the degree of spatial smoothness of the movement vector according to the environmental element of the monitoring area, the movement vector is calculated using the parameter.
Thereby, it is possible to accurately obtain the movement vector of the object performing the desired movement according to the environment of the monitoring area. Then, based on the movement vector, it is determined whether or not there is an object performing a target movement, so that the target object can be accurately detected.

[0008] Here, the environmental factors that are the basis for determining the parameters are, for example, in the case of monitoring rockfall in a mountain area as in a preferred embodiment described later, the inclination direction and gradient of the slope of the mountain area. , The number of trees, the exposure ratio of the rock surface, weather factors such as wind speed, wind direction, and the amount of rain and snow, the distance from the camera to the location, and the like. These environmental factors are the moving direction, speed and complexity of the falling rocks, the nature of the disturbing movements such as the rocking of trees and the movement of rain and snow on the video captured by the camera. Affects the correlation between falling rocks and the movement of objects other than falling rocks, and consequently affects the degree of continuity of movement of objects in the monitoring area (that is, the spatial smoothness of the movement vector). . Therefore,
By setting a parameter indicating the degree of spatial smoothness of the motion vector in accordance with such environmental factors, a motion vector that favorably reflects the motion of the target object can be obtained. For example, when there are many trees or when the wind is strong, setting the parameter indicating the degree of spatial smoothness of the movement vector to a large value will cause the movement of the disturbance that the trees sway in the same direction due to the wind. Since it is difficult to be reflected in the vector, the movement of the falling rock for the purpose of monitoring can be easily distinguished from the swinging motion of the tree, and the monitoring accuracy can be improved.

In a preferred embodiment, when setting the parameters, the monitoring areas are environmentally different (for example, they differ in the direction of inclination of the slope of the mountain, the degree of inclination, the number of trees, etc.).
It is divided into a plurality of areas, and optimal parameters are set for each area.

Furthermore, in a preferred embodiment, for environmental elements that change from moment to moment (for example, weather such as wind speed and wind direction), the parameters are adjusted according to the changes in the environmental elements. .

According to a second aspect of the present invention, a surveillance system for analyzing a moving image of a monitoring area to determine the presence or absence of an object performing a desired movement in the surveillance area differs in time included in the moving image. A moving vector calculating means for calculating a moving vector of each pixel in the frame from the images of the two frames, and a moving vector which satisfies a predetermined moving direction condition is detected from the moving vectors calculated by the moving vector calculating means. Accordingly, there are provided a target detecting means for judging the presence or absence of an object performing a desired movement, and a moving direction condition setting means for setting a moving direction condition according to an environmental element of the monitored area.

When detecting the presence or absence of an object performing a desired movement based on a calculated movement vector, the monitoring system sets a movement direction condition in accordance with an environmental element of a monitoring area, and then sets the movement direction condition. By detecting a movement vector that satisfies the direction condition, the presence or absence of an object that moves in a desired manner is determined. Here, the environmental element of the monitoring area is specifically as described above, and affects the moving direction of the monitoring target. Therefore, by setting the moving direction condition according to the environmental element and extracting the motion that meets the moving direction condition, the motion of the disturbance other than the monitoring target is filtered, and only the motion of the monitoring target is accurately detected. can do. For example, in rockfall monitoring, the range of rockfall movement direction changes depending on the maximum slope direction, slope, and wind speed of the slope, so the range of rockfall movement direction is set according to the maximum slope direction, slope, and wind speed. By extracting only the movement vector whose movement direction falls within that range,
Without being affected by disturbances such as shaking trees,
Only the target rock fall can be detected with high accuracy.

In a preferred embodiment, the surveillance area is divided into a plurality of environmentally different areas (for example, different in the direction of inclination of a mountain slope, the degree of slope, the number of trees, and the like). Optimal parameters are set.

Furthermore, in a preferred embodiment, the moving direction condition is adjusted according to the change of the environmental element (for example, weather such as wind speed and wind direction) which changes every moment. .

[0015]

FIG. 1 shows a configuration of a monitoring system according to an embodiment of the present invention. The following description will be given assuming that the monitoring system of this embodiment is intended to detect a falling rock such as a mountain.

The surveillance system includes a video camera 1 for photographing a place (surveillance area) where a falling rock may occur and outputting a moving image of the surveillance area, an anemometer 3 and an anemometer installed in the surveillance area. 5, a display device 9 for displaying a moving image from the video camera 1 on a screen, and capturing and processing moving image data from the video camera 1, wind speed data from the anemometer 3, and wind direction data from the anemometer 5. Vector processing device 11 for obtaining a moving vector (optical flow) of an object in the monitoring area and detecting an abnormal moving object (that is, falling rock) in the monitoring area
And an alarm transmitting device 17 for transmitting an alarm when the moving vector processing device 11 detects an abnormal moving object. The motion vector processing device 11 is typically configured using a computer, and executes a monitoring program,
Obtain the movement vector (optical flow) of the object in the monitoring area to find an abnormal moving object (ie,
Every time a rockfall is detected, a monitoring log file indicating that fact is created and stored in the monitoring log file database 13, and an abnormal moving object is detected from the moving image data of the monitoring area. The frame at the time (monitoring still image) is extracted and stored in the monitoring still image database 15.

FIG. 2 shows an example of a scene of the surveillance area viewed from the video camera 1.

In the example shown in FIG.
There are three mountains, and this monitoring area 100 is divided into A to E into a plurality of areas that differ in the nature of movement such as falling rocks and trees. The area A is an empty area. Regions B and E
Is an area of forest covered by trees. The region C is a steep cliff region where the rock surface is exposed at a higher slope. The area D is a gentle cliff area having a gentle slope and an exposed rock surface. In these areas A to E, the movement characteristics of objects (rocks, trees, clouds, etc.) existing in the areas (moving direction of falling rocks, degree of dispersion of moving directions, variation in movement of different trees and rocks, etc.) ) Varies by region. Therefore, when performing a process of calculating a motion vector and detecting an abnormal moving object, the motion vector processing device 11 assigns different optimal values for each region to various parameters used in the process, as described in detail later. Will be set. In addition, the nature of the movement of the object also changes depending on the wind speed and direction. Therefore, the movement vector processing device 11 adjusts the processing parameters according to the wind speed and the wind direction, as described later in detail.

Hereinafter, the processing performed by the movement vector processing device 11 will be described in detail.

This process analyzes where each pixel of the previous frame has moved in the next frame between each two consecutive frames of the moving image coming from the camera 1, and calculates a motion vector (optical flow) in the image. ) And normal motions in the image (for example, swaying of trees due to wind and the flow of clouds, etc.) and abnormal motions (in this embodiment, falling rocks in this embodiment) And an abnormality detection process of detecting only an object that moves abnormally (that is, a falling rock).

The optical flow estimating process in the preceding stage is performed by a method in which an improvement according to the present invention is added to a well-known regularization method introduced in Japanese Patent Application Laid-Open No. 9-297851 or the like. Here, the optical flow estimation processing by the regularization method means that the coordinate point (x, y) of each pixel on the image of a certain frame and its movement vector, that is, the optical flow (u, v), An analysis method using the following equation when the spatial brightness gradient in the frame image is (Ix, Iy) and the brightness gradient between the frame and the next frame is It. It is.

[0022]

(Equation 1)

However, the unknowns u and v cannot be estimated from this equation alone, and other constraints are required. About this, for example, "Determining Optical Flow"
(Artificial Intelligence 17 pp. 185-203 (1981))
Utilize the technology described in. In this technology,
The two assumptions of “the moving object in the image is a rigid body” and “the optical flow distribution in the neighboring area in the image is smooth” are respectively expressed by evaluation functions, and these two evaluation functions are expressed by using the relaxation method. The optical flow is estimated by minimizing the sum of Specifically, using α as a regularization parameter, u and v that minimize the following equation (2) are obtained by repeated calculation.

[0024]

(Equation 2) That is, on the right side of the equation (2), Ea is an evaluation function reflecting the rigidity of the moving object, and the moving object in the image is a completely rigid body (that is, an object that does not deform and does not change its own luminance). The closer the distance is, the more easily the expression (1) is satisfied, and Ea approaches zero. Eb is an evaluation function reflecting the smoothness of the spatial distribution of the optical flow, and the spatial distribution of the optical flow is completely smooth (that is, the optical flow does not change spatially, that is, any pixel in the image Eb is closer to zero as the state moves closer to the same direction in the same direction). In short, the right side of the equation (2) is a function that integrally evaluates the rigidity of the moving object and the smoothness of the spatial distribution of the optical flow in the integration region of the image, and the above two assumptions in the integration range region The more the “moving object in the image is a rigid body” and “the optical flow distribution in the near area in the image is smooth”, the smaller the value on the right side of the expression (2). In other words, if the above two assumptions are satisfied in the vicinity of a certain point in the image, u, v that minimizes the function of equation (2) in that vicinity are estimated as valid optical flows at that point. Is done.

Here, the regularization parameter α is “the smoothness of the optical flow distribution” on the right side of the equation (2).
Reflects the relative weight of That is, as α is set to be larger, the smoothness of the optical flow distribution is emphasized. Therefore, the regularization parameter α should be set appropriately in accordance with the image to be analyzed, in relation to the complexity of various objects present in the image, the nature of the motion, the magnitude of the motion, and the like.

Also, in the present embodiment, Japanese Patent Application Laid-Open No. 9-297
As in the case of No. 851, the evaluation function of the above-mentioned equation (2) is not used as it is, but in order to consider the error of the above-mentioned equation (1), the vicinity of the coordinate point (x, y) is taken into consideration. Introducing the variance σ2 of the value on the left side of equation (1) into equation (2),
U and v that minimize the evaluation function of the following equation (3) are obtained by an iterative operation using the relaxation method.

[0027]

(Equation 3)

After the motion vector for all pixel coordinates (x, y) in the frame image, that is, the optical (u, v) is obtained by the above-described method, the motion vectors are analyzed, and the From the vectors, only those that can be regarded as abnormal movements of the object (that is, falling rocks) are detected. As a method, paying attention to the fact that the moving direction of the falling rock is limited to a specific direction range (for example, a predetermined angle range centered on the maximum slope direction of the slope), only the movement moving in the specific direction range is considered. Is extracted. In this case, the maximum inclination direction and the width at which the direction of the falling rock varies depending on the slope, so that the above-described specific direction range is optimally set for each of the regions A to E illustrated in FIG. In addition, the width in which the moving direction of the falling rock is dispersed also changes depending on the wind speed and the wind direction (for example, as the wind speed in the horizontal direction increases, the degree to which the falling direction of the falling rock shifts from the maximum inclination direction increases). The above specific direction range is also adjusted according to the wind direction. With such a method, normal movements such as swaying of trees due to the wind and the like and abnormal movements such as falling rocks are distinguished from the movement vectors in the image. as a result,
If abnormal movement can be detected, it is determined that a falling rock has occurred,
If it cannot be detected, it is determined that no rockfall has occurred.

FIG. 3 is a flowchart showing a specific flow of the processing performed by the movement vector processing device 11 described above.

When the monitoring operation is started, the movement vector processing device 11 initially sets parameters used in the processing (S1). The parameters set here include:
Number of repetitions N of calculation in the relaxation method, regularization parameter reference value αi, rockfall movement direction reference value θi, rockfall movement direction width reference value βi, and regularization parameter adjustment coefficient γ and rockfall for each of areas A to E in the monitoring area There is a moving direction adjustment coefficient k.

FIG. 4 shows an example of the regularization parameter reference value αi, rockfall moving direction reference value θi, and rockfall moving direction width reference value βi, which are initially set for each of the regions A to E.

The regularization parameter reference value αi gives a standard value of the regularization parameter α used in the processing of each of the areas A to E. In the example of FIG. 4, α0>α1>α2> α
When there is a magnitude relationship of 3, the largest value α0 is set in the sky region A, the second largest value α1 is set in the forest regions B and E, and the third largest value α2 is set in the gentle cliff region D. The smallest value α3 is set in the steep cliff area C. This reflects the following degree of smoothness (continuity) of the optical flow expected in each of the regions A to E. That is, in the sky area A, the cloud only moves at a very low speed in one direction without largely changing its shape, so that the degree of smoothness (continuity) of the optical flow will be the largest. In forest areas B and E, the degree of smoothness (continuity) of the optical flow will be the second largest since many trees often sway in the same direction at a low speed in the wind. Further, in the gentle cliff region D and the steep cliff region C, the stable rock is immobile, and only the unstable rock moves as a falling rock, so that the degree of smoothness (continuity) of the optical flow is low. Among them, in the gentle cliff area D, every falling rock falls at a moderate speed in a similar direction along the maximum gradient direction, while in the steep cliff area C, the falling rock falls at a high speed. The rocks will make complex movements, such as changing directions, crushing, and other falling rocks, and will have the lowest degree of smoothness (continuity) of the optical flow in the steep cliff region E.

The falling rock moving direction reference value θi is defined as
The standard value of the falling direction of the falling rock in E is given, and is basically the maximum gradient direction of the slope in the area. Therefore, in the example of FIG. 4, the rockfall moving direction reference value θi is not set in the sky area A where no rockfall occurs, and the maximum slope directions θ1 to θ4 of the respective slopes are set in the other areas B to E. ing. The rockfall moving direction width reference value βi gives a standard value of a width in which the falling direction of the rockfall in each of the areas A to E deviates from the rockfall moving direction reference value θi. In the example of FIG.
If the degree <β2 <β1 <10 degrees, the smaller values β1 are set for the forest areas B and E and the gentle cliff area D where the rockfall will move monotonously, and The larger value β2 is set for the steep cliff region C that will be formed.

FIG. 5 shows the regularization parameter α of each of the regions A to E to be set as a result of the initialization of the regularization parameter reference value αi and the regularization parameter adjustment coefficient γ of each of the regions A to E. Is shown.

As shown in FIG. 5, the regularization parameter α for each of the regions A to E is set as the sum of the regularization parameter reference value αi and the regularization parameter adjustment coefficient γ for each of the regions A to E. The regularization parameter adjustment coefficient γ is a value much smaller than the regularization parameter reference value αi, but as will be described later, the value is adjusted according to the wind speed and the wind direction at any time after the initial setting. The regularization parameter α of each of the regions A to E is adjusted according to the wind speed and the wind direction.

FIG. 6 shows each area A to be set as a result of the initial setting of the rockfall moving direction reference value θi, rockfall moving direction width reference value βi, and rockfall moving direction adjustment coefficient k of each of the areas A to E. ~ E falling rock moving direction width β and falling rock moving direction range θ
Is shown.

As shown in FIG. 6, the rock moving direction width β of each of the regions A to E is the rock falling moving direction width reference value β of each of the regions A to E.
i is set as a product of the rockfall moving direction adjustment coefficient k, and the rockfall moving direction range?
Are set to the left and right (plus / minus) around the rockfall moving direction reference value θi of the range A to E by the width β of the rockfall moving direction of each of the regions A to E (that is, the ranges are plus and minus). Then, the falling rock moving direction range θ of each of the regions A to E is used as a filter for extracting only those indicating the falling rock from the movement vectors. Here, the rockfall moving direction adjustment coefficient k is a value within the range of 1.0 to 2.0, and as described later, the value is adjusted according to the wind speed and the wind direction as needed after initial setting. As a result, the falling rock moving direction range θ of each of the regions A to E is adjusted according to the wind speed and the wind direction.

Referring again to FIG. After initializing the above-described parameters in step S1, the motion vector processing device 11 inputs the image I (x, y, t) of the current frame from the camera 1 and the input frame image I (x, y, t).
For (x, y, t), x,
Brightness gradient images Ix (x, y, t) and Iy in the y and t directions
(X, y, t) and It (x, y, t) are generated (S
2). Here, x and y indicate the coordinate values on the horizontal axis and the vertical axis of the image space, and t indicates the coordinate value on the time axis.

[0039]

(Equation 4)

Next, the motion vector processing device 11 calculates x,
From the luminance gradient images in the y and t directions, a horizontal flow u and a vertical flow v that minimize the aforementioned equation (3)
Are determined by the relaxation method. Specifically, the horizontal flow u and the vertical flow v shown by the following equation (5) are calculated by N repetitions of the relaxation method (S3 to S3).
5).

[0041]

(Equation 5) Here, the horizontal and vertical flows u and v at arbitrary spatial coordinates (i, j) of the image are (uij,
vij), but in equation (5), (u,
v). In the equation (5), the subscripts n and n + 1 attached to the upper right of the flows u and v are the flow u obtained in the n-th and n + 1-th calculations during the repetitive calculations up to N times, respectively. , V.

When the final movement vector, that is, the optical flow (uij, vij) is obtained by N repetitive calculations, the movement vector processing device 11 next calculates the areas B to E in which rockfalls may occur. The movement vector (uij, vij) of each region is analyzed to check for the presence or absence of an object that has moved in a direction falling within the falling rock movement direction range θ of each region (S6). As a result, if there is a moving object that satisfies the condition, the motion vector processing device 11 determines that rockfall has occurred, causes the alarm transmitting device 17 to transmit an alarm (S7), and when and where the rockfall occurs. A monitoring log file indicating whether or not a frame has occurred and a frame image (monitoring still image) file at that time are created and stored in the respective databases 13 and 15 (S8).

Thereafter, the movement vector processing device 11 fetches wind speed data and wind direction data from the anemometer 3 and the anemometer 5, and corrects the adjustment coefficients γ and k based on the data (S9). 7 and 8 show examples of rules for correcting the adjustment coefficients γ and k. As shown in FIG. 7, the regularization parameter adjustment coefficient γ is corrected to a larger value as the wind speed increases, and to a smaller value as the wind speed decreases. this is,
As the wind speed increases, the degree of movement of the object in the same direction due to the wind increases, and the smoothness of the optical flow increases. This means that the regularization parameter α is set to a larger value. As shown in FIG. 8, the falling rock moving direction adjustment coefficient k increases the horizontal (horizontal) wind speed (for example, the east-west wind speed when the camera 1 is facing north) as viewed from the camera 1. If the horizontal wind speed decreases, the value is corrected to a lower value. This means that the higher the wind speed in the horizontal direction,
Since the direction in which the falling direction of the falling rock shifts from the maximum inclination direction of the slope due to the wind increases, this means that the falling rock moving direction range θ is set to a larger range.

After correcting the adjustment coefficients γ and k according to the wind speed and the wind direction in this way, the movement vector processing device 11
Returns to step S2, inputs a new frame image, and repeats the same processing. The above processing is repeated until termination of the monitoring operation is commanded (S10).

The embodiment of the present invention has been described above.
This is merely an illustrative example of the invention. Therefore,
The present invention can be implemented in various aspects other than the above embodiment. For example, the present invention can be applied to various monitoring purposes such as detection of intrusion of a suspicious person and monitoring of an indoor object as well as outdoor rock fall detection. The method of setting and adjusting the regularization parameter and the moving direction range may be different depending on the monitoring purpose and the monitoring target. Even in the same rock fall monitoring as in the above embodiment, the parameters are adjusted according to not only the wind speed and the wind direction, but also the rainfall amount and the snowfall amount, and different values are set as the adjustment coefficients γ and k for each region. Such,
Various variations can be employed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a monitoring system according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing an example of a scene of a monitoring area viewed from a video camera 1.

FIG. 3 is a flowchart of a process performed by a movement vector processing device 11;

FIG. 4 is an explanatory diagram showing an example of a regularization parameter reference value αi, a rockfall moving direction reference value θi, and a rockfall moving direction width reference value βi which are initially set for each of regions A to E;

FIG. 5 is a regularization parameter α set for each of areas A to E
FIG.

FIG. 6 is a moving direction width β of a rockfall set for each of the regions A to E.
FIG. 6 is an explanatory diagram showing an example of a range θ of movement.

FIG. 7 is an explanatory diagram showing an example of a rule for correcting a regularization parameter adjustment coefficient γ according to a wind speed.

FIG. 8: Rockfall moving direction adjustment coefficient k according to horizontal wind speed
FIG. 4 is an explanatory diagram showing an example of a rule for correcting the.

[Explanation of symbols]

 Reference Signs List 1 video camera 3 anemometer 5 anemometer 9 display device 11 movement vector processing device 13 monitoring log file DB 15 monitoring still image DB 17 alarm transmission device

Claims (8)

[Claims] 1. A monitoring system for determining the presence or absence of an object performing a desired movement in a monitoring area by analyzing a moving image of the monitoring area, comprising: A moving vector calculating unit configured to calculate a moving vector of each pixel in a frame, wherein the calculating the moving vector uses a parameter indicating a degree of spatial smoothness of the moving vector. Based on the movement vector calculated by the movement vector calculation means, target detection means for determining the presence or absence of the object performing the desired movement, and a parameter indicating the degree of spatial smoothness of the movement vector And a parameter setting means for setting according to environmental factors of the monitoring area. 2. The monitoring system according to claim 1, wherein the parameter setting unit divides the monitoring area into a plurality of environmentally different areas and sets the parameter for each area. 3. The monitoring system according to claim 1, wherein the parameter setting unit adjusts the parameter according to a change in a predetermined environmental element in the monitoring area. 4. A monitoring system for determining the presence or absence of an object moving in a target area in a monitoring area by analyzing a moving image in the monitoring area, comprising: A moving vector calculating unit that calculates a moving vector of each pixel in a frame; and detecting a moving vector that satisfies a predetermined moving direction condition from the moving vectors calculated by the moving vector calculating unit. A surveillance system comprising: a target detection unit for judging the presence or absence of a moving object; and a movement direction condition setting unit for setting the movement direction condition according to an environmental element of the monitoring area. 5. The monitoring system according to claim 4, wherein the moving direction condition setting means divides the monitoring area into a plurality of environmentally different regions, and sets the moving direction condition for each of the regions. 6. The monitoring system according to claim 4, wherein the moving direction condition setting means adjusts the moving direction condition according to a change in a predetermined environmental element in the monitoring area. 7. A method of analyzing a moving image in a monitoring area to determine the presence or absence of an object performing a desired movement in the monitoring area. Calculating a motion vector of each pixel in the motion vector, wherein, when calculating the motion vector, using a parameter indicating a degree of spatial smoothness of the motion vector; and A step of determining the presence or absence of the object performing the desired movement based on a movement vector; and a step of setting a parameter indicating a degree of spatial smoothness of the movement vector in accordance with an environmental element of the monitoring area. And a computer-readable recording medium carrying a program for causing a computer to execute the above. 8. A method for analyzing a moving image in a monitoring area to determine whether or not there is an object performing a desired movement in the monitoring area. Calculating a motion vector of each pixel in the above, and detecting a motion vector that satisfies a predetermined moving direction condition from the calculated motion vectors to determine the presence or absence of the object performing the desired motion. A computer-readable recording medium carrying a program for causing a computer to execute the following steps: setting the moving direction condition according to an environmental element of the monitoring area.