JP5421597B2 - Image monitoring device - Google Patents

Description

The present invention relates to an image surveillance equipment using the zoom pivoting camera to the imaging unit.

  In an image monitoring apparatus that performs image processing on a video acquired from an imaging unit and detects an abnormality in the video, a highly reliable monitoring result cannot be obtained if the installation conditions of the imaging unit change. Therefore, it is necessary to detect and correct a change (shift) in the installation conditions of the imaging unit. Here, the installation conditions of the imaging unit are the focal length of the lens related to the optical system of the imaging unit, and the vertical and horizontal angles that determine the shooting direction of the imaging unit. These camera parameters may change depending on causes such as maintenance work of the imaging unit, defective / insufficient strength of the fixed unit, or natural phenomena such as earthquakes and strong winds.

  In the conventional image monitoring apparatus, the change in the installation angle of the imaging unit is calculated from the apparent change amount of the image, and the deviation of the installation angle of the imaging unit is corrected using this value. As such an example, in the image measuring device presented in Patent Document 1, whether or not there is a shift between the current image and the image stored in advance, mainly for the fixed background image portion. An image comparison unit for examining the image is provided. In this example, when there is a deviation, the installation angle resetting unit obtains the change amount of the installation angle of the imaging device from the change amount on the image, and sets the installation angle of the imaging device stored in the installation condition storage unit. I try to fix it.

JP-A-4-32703

  The problem to be solved is that in the case of a zoom swivel camera in which the imaging unit acquires images while changing the focal length according to the subject, the focal length differs depending on the monitoring direction, and the zoom amount is not constant. In order to detect a change in the installation angle of the image pickup unit, it is necessary to enlarge or reduce the image, and the image processing becomes complicated.

The present invention has been made to solve the above-described problems. In an image monitoring apparatus using a zoom turning camera as an imaging unit, the focal length of the lens of the imaging unit, the vertical angle of the imaging unit, and the right and left It is an object of the present invention to obtain a highly reliable image monitoring apparatus capable of correcting a deviation of a direction angle with high accuracy .

An image monitoring apparatus according to the present invention is installed so as to be turnable in a predetermined direction, and an imaging unit that acquires an image while changing a focal length according to an imaging target, and an installation condition of the imaging unit with respect to the image acquired by the imaging unit An image processing unit that performs image processing using the image and detects an abnormality in the video, and a reference point storage unit that stores coordinates at the time of setting three or more reference points arbitrarily set in the video acquired by the imaging unit A reference point acquisition unit that acquires the current coordinates of the reference point from the video including the reference point acquired by the imaging unit, and the current coordinates of the reference point acquired by the reference point acquisition unit and the reference point storage unit When the reference point comparison unit determines whether there is a difference of a predetermined amount or more and the reference point comparison unit determines whether there is a difference of a predetermined amount or more. , the least among the three or more reference points reference point acquiring unit acquires And current coordinates of the two reference points, the corresponding at least two of the reference point setting when the coordinates and installation conditions of the image pickup unit using a stored in the reference point storage unit calculates, the calculation result Is provided to the image processing unit, and an installation condition correction unit is provided as an installation condition used for image processing of the image processing unit.

According to the image monitoring apparatus according to the present invention, three or more reference points are arbitrarily set in the video acquired by the imaging unit, and the imaging unit is installed from the current coordinates of the reference points and the coordinates at the time of the setting. Since the conditions are calculated, the deviation in the installation conditions of the imaging unit can be corrected with high accuracy .

FIG. 1 is a block diagram showing the configuration of the image monitoring apparatus. (Example 1) FIG. 2 is a diagram illustrating an example of setting reference points in the image monitoring apparatus. (Example 1) FIG. 3 is a diagram illustrating reference points of the image monitoring apparatus. (Example 1) FIG. 4 is a diagram for explaining how to set the coordinate axes. (Example 1) FIG. 5 is a block diagram showing the configuration of the reference point acquisition unit of the image monitoring apparatus. (Example 2) FIG. 6 is a diagram for explaining how to create a new reference point of the image monitoring apparatus. (Example 2) FIG. 7 is a block diagram showing the configuration of the image monitoring apparatus. (Example 3) FIG. 8 is a diagram illustrating an example of a reference point evaluation method by the reference point evaluation unit of the image monitoring apparatus. (Example 3) FIG. 9 is a block diagram showing the configuration of the image monitoring apparatus. Example 4 FIG. 10 is a block diagram showing the configuration of the image monitoring apparatus. (Example 5) FIG. 11 is a block diagram showing the configuration of the image monitoring apparatus. Example 6

  An image monitoring apparatus according to the present invention acquires an image to be monitored using a zoom turning camera in an imaging unit, for example, road monitoring including traffic flow measurement for measuring a moving speed and a distance of a vehicle, a specific It is installed for the purpose of monitoring suspicious persons to prevent suspicious persons from entering the management area. However, the use of the image monitoring apparatus according to the present invention is not particularly limited. Hereinafter, Examples 1 to 6 of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram illustrating the configuration of the image monitoring apparatus according to the first embodiment of the present invention. The imaging unit 1 is a zoom turning camera that is installed so as to be able to turn in a predetermined direction and acquires an image while changing a focal length in accordance with an imaging target. The imaging unit 1 includes a zoom camera body that is an imaging unit capable of changing a focal length, and a swivel base that is a driving unit that rotates the imaging unit. There are two swivel axes of the swivel base, and the first axis is an axis where the camera body is fixed to a support column or a structure, and the axis that swings the line of sight to the left and right. The second axis is an axis that swings the line of sight up and down. The rotation axis and the rotation angle will be described later in detail.

  The image processing unit 2 stores the installation conditions of the imaging unit 1, performs image processing on the video acquired by the imaging unit 1 using the installation conditions of the imaging unit 1, and detects abnormalities in the video. . The installation conditions of the imaging unit 1 are camera parameters including the focal length of the lens of the optical system of the imaging unit 1, the vertical angle that determines the shooting direction of the imaging unit 1, and the horizontal angle. The image processing unit 2 continuously monitors the monitoring target.

  When the image processing unit 2 detects any abnormality in the video, the abnormality reporting unit 3 receives an abnormality signal from the image processing unit 2 and reports this to the user (administrator) of the image monitoring apparatus. As a method for the abnormality reporting unit 3 to report the abnormality to the user, an alarm sound (including an audible warning for leaving in the case of suspicious person intrusion, etc.), a method of flashing a lamp, an abnormality signal to a server such as a management center, etc. The method according to the installation purpose of the image monitoring apparatus is selected.

  The reference point storage unit 4 stores coordinates at the time of setting three or more reference points arbitrarily set in the video acquired by the imaging unit 1. The reference point storage unit 4 stores more reference points than are actually required, including spare reference points (see the fifth embodiment). A reference point setting example will be described later with reference to FIG.

  The reference point acquisition unit 5 acquires the coordinates of the reference point in the current video from the video including the reference point acquired by the imaging unit 1. Further, the reference point comparison unit 6 compares the current coordinates of the reference points acquired by the reference point acquisition unit 5 with the coordinates at the time of setting stored in the reference point storage unit 4, and exceeds a predetermined amount. It is to judge whether there is a difference.

  Further, the camera parameter correction unit 7 which is an installation condition correction unit has the current coordinates of at least two reference points acquired by the reference point acquisition unit 5 and the setting coordinates stored in the reference point storage unit 4. The current camera parameter is calculated from the coordinates, and the calculation result is transmitted to the image processing unit 2 and reflected in the image processing. In the first embodiment, when the reference point comparison unit 6 determines that there is a difference of a predetermined amount or more, the camera parameter correction unit 7 calculates camera parameters. A calculation method in the camera parameter correction unit 7 will be described later.

Next, an installation condition correction method for the image monitoring apparatus according to the first embodiment will be described. First, as a first step, three or more reference points are arbitrarily set in the video acquired by the imaging unit 1, and the coordinates at the time of setting are stored in the reference point storage unit 4. An example of setting reference points is shown in FIG. In FIG. 2, R ₁ , R ₂ , R ₃ , R ₄ are reference points set in the video 1 a acquired by the imaging unit 1. Thus, when setting the reference point, the position is clear, such as the characteristic part of the structure in the image and the corner of the road, and it is difficult to move due to natural phenomena such as earthquakes and strong winds, Those that are unlikely to be moved or removed artificially are selected.

In addition, since the set reference points may become unusable due to a change in the situation, a number of reference points R ₁ , R ₂ ,... R _n , including spare reference points, as shown in FIG. It is desirable to set R _{n + 1} in advance. As described above, in the image 1b at the predetermined focal length and the turning position of the imaging unit 1, the coordinates (x _n , y _n ) on the screen when the reference point R _n is set are determined, and the reference point is stored together with the camera parameters. Stored in part 4. Further, the coordinates of the reference point obtained from the map data in the world coordinate system are also stored as the coordinates at the time of installation.

  The coordinates of the reference points stored in the reference point storage unit 4 are not shifted in the installation conditions of the imaging unit 1 and are between the world coordinate system and the screen coordinate system (the coordinate system on the image, which will be described later in detail). Can be converted to each other, so you can store any 3 or more of the 3 coordinate values in the world coordinate system and 2 coordinate values in the screen coordinate system together with the installation conditions of the imaging unit 1 That's fine. As a coordinate value of the world coordinate system at the time of installation, a value obtained by calculation from the coordinate value of the screen coordinate system and the installation conditions of the imaging unit 1 may be stored.

  Next, as a second step, the reference point acquisition unit 5 acquires the coordinates of the reference point in the current video from the video including the reference point acquired by the imaging unit 1. The coordinates of the reference point in the current video are the same as the coordinates at the time of setting stored in the reference point storage unit 4 if the camera parameters of the imaging unit 1 have not changed since the time of setting. Even if the reference point acquisition unit 5 fails to acquire a reference point, in the first embodiment, since three or more reference points are prepared, the reference point acquisition unit 5 has at least two reference points. Can be obtained. When the camera parameter has changed since the reference point was set, the coordinates on the image when the camera parameter is a value at the time of setting the reference point are obtained by the method described later, and this is set as the current coordinate.

  Next, as a third step, the reference point comparison unit 6 sets the coordinates in the current video and the coordinates at the time of setting stored in the reference point storage unit 4 for the reference points acquired by the reference point acquisition unit 5. And determine whether there is a difference of a predetermined amount or more. That is, when there is a difference of a predetermined amount or more between the current coordinates of the reference point and the coordinates at the time of setting, the reference point changes (shifts) to at least one of the focal length, the vertical angle, and the horizontal angle of the lens of the imaging unit 1. ). At this time, it is not determined that only the coordinate coincidence is “no deviation”, but a determination criterion in consideration of an error due to the resolution of the camera is set in advance, and the determination is performed based on the determination criterion. It may be.

  When it is determined in the third step that the camera parameter of the imaging unit 1 has changed, the camera parameter correction unit 7 determines from the coordinates in the current video of at least two reference points and the coordinates at the time of setting them. The camera parameters of the imaging unit 1 are calculated. The calculation result is transmitted to the image processing unit 2. The image processing unit 2 executes image processing using the new camera parameters obtained by the camera parameter correction unit 7 and continues the monitoring operation.

Next, a camera parameter calculation method in the camera parameter correction unit 7 will be described. First, how to obtain the coordinate axes will be described with reference to FIG. As shown in FIG. 4A, the coordinates of the actual point are represented, and the world coordinate system indicated by O _w −X _w Y _w Z _w is a right-handed system. Further, the screen coordinate system representing the coordinates of the points on the video has a point (x _f , y _f , z _f ) that is the installation position of the imaging unit 1 in the world coordinate system as the origin O _s, and the direction in which the imaging unit 1 faces. the linear and Z _s axis is on Y _w axis and flush in the Z _s axis and the world coordinate system, an axis perpendicular to the Z _s axis toward the upper side in the Y _w axis and Y _s axis . The O _s -X _s Y _s Z _s coordinate system is a left-handed system.

In the following, the coordinates of a point on the O _w −X _w Y _w Z _w coordinate system are represented by adding a subscript “ _w ”, and the coordinates of the point on the O _s −X _s Y _s Z _s coordinate system are represented by the subscript “ “ _s ” is added. For example, the point O _s is expressed as (x _f , y _f , z _f ) _w or (0, 0, 0) _s . Further, as shown in FIG. 4B, α is an angle of the Y _s Z _s plane with respect to the Y _w Z _w plane when viewed from the Y _w axis direction, and the X _w axis rotates so as to overlap the Z _w axis. The direction to do is positive. Further, as shown in FIG. 4C, β is an angle of the X _s Z _s plane with respect to the X _w Z _w plane when viewed from the X _s axis direction, and rotates so that the Y _s axis overlaps the Z _s axis. The direction to do is positive. The point O _s (x _f , y _f , z _f ) _w is a position in the actual coordinates of the imaging unit 1.

Between the actual coordinates (x _w , y _w , z _w ) _w of the reference point and the coordinates (x _s , y _s , z _s ) _s of the point on the image, there is a relationship of the following formula 1. It is known. In Equation 1, h is the focal length of the optical system of the imaging unit 1, and k is the size of the image sensor. α and β are rotation angles around the Y _w and X _s axes of the imaging unit 1. Note that the video is usually represented by two-dimensional coordinates, but in Equation 1, it may be considered that z _s = 0.

(Equation 1)
_{(X w -x f) / V} x = (y w -y f) / V y = (z w -z f) / V z

However, in Formula 1, V _x , V _y , and V _z are as follows.
_{V x = k · x s ·} cosα + k · y s · sinα · sinβ
+ H ・ sinα ・ cosβ
_{V y = k · y s ·} cosβ-h · sinβ
V _z = k · x _s · sin α−k · y _s · cos α · sin β
-H · cos α · cos β

Here, the position (x _f , y _f , z _f ) _w of the imaging unit 1 and the size k of the imaging element are fixed and do not change. The camera parameters that may change are the focal length h of the lens that changes depending on the zoom amount, the vertical angle of the imaging unit 1, and the horizontal angle β. When one reference point is determined, its actual coordinates (coordinates at the time of setting) and coordinates on the video are determined, and two mathematical formulas independent of the mathematical formula 1 are determined. Now, since there are three parameters, h, α, and β, if two reference points are determined, four independent mathematical expressions can be determined, and h, α, and β can be calculated and obtained. it can.

A calculation method for obtaining h, α, and β from two reference points will be specifically described. The actual coordinates (setting coordinates) of the reference point R ₁ are (x1 _w , y1 _w , z1 _w ) _w, and the coordinates on the video are (x1 _s , y1 _s , z1 _s ) _s . The actual coordinates (setting coordinates) of the reference point R ₂ are (x2 _w , y2 _w , z2 _w ) _w, and the coordinates on the video are (x2 _s , y2 _s , z2 _s ) _s . By substituting the actual coordinates of these reference points R ₁ and R ₂ and the coordinates on the video into the above-described equation 1, the following equations 2 and 3 are obtained.

(Equation 2)
_{(X1 w -x f) / V1} x = (y1 w -y f) / V1 y = (z1 w -z f) / V1 z

However, in Equation _{2, V1 x, V1 y,} V1 z are as follows.
V1 _x = k · x1 _s · cos α + k · y1 _s · sin α · sin β
+ H ・ sinα ・ cosβ
V1 _y = k · y1 _s · cos β−h · sin β
V1 _z = k · x1 _s · sin α−k · y1 _s · cos α · sin β
-H · cos α · cos β

(Equation 3)
_{(X2 w -x f) / V2} x = (y2 w -y f) / V2 y = (z2 w -z f) / V2 z

However, in Formula 3, V2 _x , V2 _y , and V2 _z are as follows.
V2 _x = k · x2 _s · cos α + k · y2 _s · sin α · sin β
+ H ・ sinα ・ cosβ
_{V2 y = k · y2 s ·} cosβ-h · sinβ
_{V2 z = k · x2 s ·} sinα-k · y2 s · cosα · sinβ
-H · cos α · cos β

  From Expression 2, the following three expressions (Expression 4, Expression 5, and Expression 6) are obtained.

(Equation 4)
_{(X1 w -x f) / V1} x = (y1 w -y f) / V1 y

(Equation 5)
( _{X1 w} −x _f ) / V1 _x = (z1 _w −z _f ) / V1 _z

(Equation 6)
_{(Y1 w -y f) / V1} y = (z1 w -z f) / V1 z

  Further, from the formula 3, the following three formulas (formula 7, formula 8, formula 9) are obtained.

(Equation 7)
_{(X2 w -x f) / V2} x = (y2 w -y f) / V2 y

(Equation 8)
_{(X2 w -x f) / V2} x = (z2 w -z f) / V2 z

(Equation 9)
_{(Y2 w -y f) / V2} y = (z2 w -z f) / V2 z

  Next, two mathematical formulas are freely selected from the three mathematical formulas (Mathematical Formula 4, Formula 5, and Formula 6) obtained from Formula 2. Similarly, two mathematical formulas are freely selected from the three mathematical formulas (Formula 7, Formula 8, Formula 9) obtained from Formula 3, and a total of four independent formulas are obtained. By solving these four mathematical expressions, h, α, and β are obtained. The difference between the new camera parameters h, α, and β obtained here and the camera parameters recognized by the image monitoring apparatus at the time of capturing the image used to calculate the camera parameters is the amount of change. Thus, if at least two reference points are determined, the reference point acquisition unit 4 acquires the coordinates of the two reference points from the current video, and the camera parameters h, α based on the coordinates at the time of setting. , Β can be calculated.

  Furthermore, in the first embodiment, two or more reference points are set in advance, thereby securing two reference points necessary for calculation in the camera parameter correction unit 7. If more reference points are set than necessary, even if the reference point acquisition unit 5 fails to acquire one reference point, it is only necessary to acquire another reference point, which may make calculation impossible. Absent. In addition, since the result can be derived from different calculation processes by using three or more reference points, the effect of generating redundancy in the calculation can be obtained. Therefore, the calculation in the camera parameter correction unit 7 is executed stably, and the result is highly accurate and reliable.

  Further, the camera parameter may be recalculated periodically, for example, every hour without installing the reference point comparison unit 6. When the camera parameter is calculated, the difference between the value obtained by the calculation and the camera parameter recognized by the image recognition apparatus at the time of capturing the image used to calculate the camera parameter is also stored. By doing so, it is possible to grasp at which point the camera parameter deviation occurs.

  As described above, according to the first embodiment, three or more reference points are set in the image monitoring apparatus using the zoom turning camera in the image pickup unit 1, and the image is picked up from the deviation of the coordinates of these reference points. Since the focal length h that may vary among the camera parameters of the unit 1, the vertical angle of the imaging unit 1, and the horizontal angle β can be obtained by calculation, the deviation of these camera parameters is highly accurate. Thus, it is possible to obtain a highly reliable image monitoring apparatus that can be corrected by the above.

  Moreover, according to the installation condition correction method for the image monitoring apparatus according to the first embodiment, the camera parameters of the imaging unit 1 are calculated from the coordinates of the reference point in the current video and the coordinates at the time of setting. Therefore, compared with the conventional method of detecting a change in the camera parameter of the image pickup unit from the image shift, image enlargement or reduction is unnecessary, so that the image processing is not complicated, and the camera parameter shift of the image pickup unit 1 is reduced. Correction can be easily performed with high accuracy.

  FIG. 5 is a block diagram illustrating the configuration of the reference point acquisition unit of the image monitoring apparatus according to the second embodiment of the present invention. In FIG. 5, the same or corresponding parts as those in FIG. The reference point acquisition unit 5 of the image monitoring apparatus according to the second embodiment creates a new reference point from two or more acquired reference points, and makes a new reference based on the current coordinates of the two or more reference points. A reference point creation unit 5a for obtaining the current coordinates of the point is provided. Since the other configuration of the image monitoring apparatus according to the second embodiment is the same as that of the first embodiment (FIG. 1), description thereof is omitted.

  In the first embodiment, the camera parameter correction unit 7 performs the calculation using the coordinates in the current video of the reference point acquired by the reference point acquisition unit 5. However, the position of the reference point is originally acquired as a coordinate value rounded to the resolution of the video image expressed by an infinite digit number. For this reason, rounding errors occur depending on the resolution of the video, and the calculation result may differ slightly depending on which reference point is used. In addition, it is not easy to acquire coordinates on the video in more detail than the resolution because advanced image processing is required.

  Therefore, in the second embodiment, the reference point creation unit 5a is provided in order to improve the accuracy of the reference points. In this reference point creation unit 5a, a new reference point is created from two or more existing reference points acquired by the reference point acquisition unit 5, and based on the current coordinates of the two or more existing reference points, The current coordinates of the new reference point are obtained. Thereby, the rounding error of the coordinate of a some reference point is averaged, and the influence of a rounding error can be made small. Note that the coordinates at the time of setting a new reference point are obtained by calculation from the coordinates at the time of setting an existing reference point used when creating a new reference point.

  Further, the reference point comparison unit 6 compares the current coordinates with the coordinates at the time of setting for the new reference point created by the reference point creation unit 5a, and determines whether there is a difference of a predetermined amount or more. Further, the camera parameter correction unit 7 performs calculation for obtaining the camera parameters h, α, and β using the new reference point.

An example of a new reference point creation method in the reference point creation unit 5a will be described with reference to FIG. In FIG. 6, R ₁ , R ₂ ,... R _n , R _{n + 1} are existing reference points set in advance in the image 1c at a predetermined turning position of the imaging unit 1, and N ₁ , N ₃ , N _n is a new reference point created by the reference point creation unit 5a. In this example, the existing two reference points R _n, obtains a middle point N _n of R n _{+ 1,} it is to use the middle point N _n as a new reference point.

The camera parameters are calculated using this new reference point by the following procedure.
(A) This new reference point is a midpoint between two reference points on the current image (screen coordinate system), but is not necessarily a midpoint in the world coordinate system. In the world coordinate system of the new reference point using the current camera parameters, the above reference (1) and the constraint that a new reference point exists on the line segment connecting the two points also in the world coordinate system. Calculate the coordinates of.
(B) Expressions (2) and (3) are obtained from the calculated coordinates in the world coordinate system and the coordinates of the new reference point in the current screen coordinate system, and the camera parameters are calculated in the same manner as described above. . In some cases, (A) and (B) may be repeated a predetermined number of times or until the camera parameters do not change.

  According to the second embodiment, in addition to the same effects as in the first embodiment, the reference point acquisition unit 5 includes the reference point creation unit 5a, so that, for example, the existing p reference points that are initially set can be obtained. In addition, when those coordinates are averaged to create one new reference point, the rounding error of the new reference point can be expected to be 1 / p with respect to the original rounding error. Therefore, the accuracy of the reference point is higher than in the first embodiment, and the deviation of the camera parameter can be corrected with higher accuracy.

  FIG. 7 is a block diagram illustrating the configuration of the image monitoring apparatus according to the third embodiment of the present invention. In FIG. 7, the same or corresponding parts as in FIG. As shown in FIG. 7, the image monitoring apparatus according to the third embodiment has a configuration in which a reference point evaluation unit 8 is added to the image monitoring apparatus (FIG. 1) according to the first embodiment. The reference point evaluation unit 8 evaluates the accuracy of the reference point acquired by the reference point acquisition unit 5 and selects a highly reliable reference point. The camera parameter correction unit 7 calculates camera parameters using the reference points evaluated to have been accurately acquired by the reference point evaluation unit 8.

  Either in the case of using an existing reference point set in advance as in the first embodiment or in the case of creating a new reference point from two or more existing reference points as in the second embodiment If the reference point cannot be obtained accurately from the above, the accuracy of the calculation result by the camera parameter correction unit 7 is low. In other words, the accuracy of the reference point is high reliability, but the degree is influenced by the accuracy of image processing for recognizing the reference point.

  In the image processing for recognizing the reference point, the reference point acquisition unit 5 may fail to detect and recognize the reference point depending on the lighting conditions and environmental conditions such as dirt on the lens of the imaging unit 1. If a place that is not a reference point is mistakenly detected as a reference point and is used for calculation in the camera parameter correction unit 7, the obtained calculation result is incorrect. Therefore, in the third embodiment, the reference point evaluation unit 8 evaluates the accuracy (reliability) of the reference point acquired by the reference point acquisition unit 5 and detects an unusable reference point that has not been acquired correctly. I tried to do it.

An example of a reference point evaluation method by the reference point evaluation unit 8 of the image monitoring apparatus according to the third embodiment will be described with reference to FIG. In FIG. 8, R ₁ , R ₂ ,... R _n , R _{n + 1} are reference points set in advance in the image 1d at a predetermined turning position of the imaging unit 1, and E ₃ is a reference point acquisition unit. 5 shift occurs for some reason when acquiring the coordinates of the reference points R _3, it shows the acquired location as the coordinate of the reference point R _3. Further, the horizontal direction of the video 1d is defined as the X direction, and the vertical direction is defined as the Y direction.

  In the example illustrated in FIG. 8, the reference point evaluation unit 8 uses the fact that the information acquired by the reference point acquisition unit 5 is video information, so that the reference point is accurately determined from the arrangement order of the reference points in the X direction and the Y direction. It is to evaluate. That is, since the position of the imaging unit 1 is unchanged, even if the apparent size or direction of the object in the monitoring target video 1d changes, the positional relationship between them does not change.

The reference points stored in the reference point storage unit 4 are arranged in the order of R ₁ → R ₂ → R ₄ → R _{3 in} the X direction, and R ₂ → R ₃ → R ₁ → R _{4 in the} Y direction. Are arranged in this order. Here, when the reference point acquisition unit 5 fails to acquire a certain reference point and the coordinates are greatly shifted and the arrangement order of either the X direction or the Y direction is changed, the reference point evaluation unit 8 It is determined that the reference point where has changed has not been acquired correctly, and cannot be used.

For example, if the reference point acquisition unit 5 fails to acquire the reference point R ₃ for some reason and erroneously acquires the location of E ₃ as the reference point, the arrangement order in the X direction does not change, but the Y direction Are arranged in the order of R ₃ (E ₃ ) → R ₂ → R ₁ → R ₄ , and the order of R ₂ and R ₃ is switched. Thus, the reference point evaluation unit 8 determines that one of R ₂ and R ₃ are not able to accurately acquire and unusable in R ₂ and R ₃ the reference point comparator 6 and the camera parameter correcting section 7 can do.

  According to the third embodiment, in addition to the same effects as in the first embodiment, the camera parameter correction unit 7 determines the accuracy of the reference point acquired by the reference point acquisition unit 5 by providing the reference point evaluation unit 8. Since the evaluation can be performed before use in the calculation, the accuracy of the reference point used in the calculation becomes higher, and the deviation of the camera parameter can be corrected with higher accuracy.

  FIG. 9 is a block diagram illustrating the configuration of the image monitoring apparatus according to the fourth embodiment of the present invention. In FIG. 9, the same or corresponding parts as those in FIGS. 1 and 7 are denoted by the same reference numerals, and description thereof is omitted. As shown in FIG. 9, the image monitoring apparatus according to the fourth embodiment has a configuration in which a reference point defect notification unit 9 is added to the image monitoring apparatus according to the third embodiment (FIG. 7).

  In the fourth embodiment, when there is a reference point that is evaluated not to be accurately acquired by the reference point evaluation unit 8, the reference point defect notification unit 7 that reports this to the user is provided. As a result, when the reason why the reference point acquisition unit 5 cannot accurately acquire the reference point is not a malfunction of the image processing but a change in the actual situation, the reference point information stored in the reference point storage unit 4 is stored. The user can be requested to update in accordance with the actual situation. This allows the user to change the reference point setting when the structure that has been set as the reference point has been removed, or when it has become difficult to acquire a reference point due to the construction of a new structure. can do.

  According to the fourth embodiment, in addition to the same effects as those of the first and third embodiments, the reference point defect reporting unit 9 reports the reference point malfunction to the user, so that stable monitoring can be continuously performed. A highly reliable image monitoring apparatus can be obtained.

  FIG. 10 is a block diagram illustrating the configuration of the image monitoring apparatus according to the fifth embodiment of the present invention. In FIG. 10, the same or corresponding parts as those in FIGS. 1 and 7 are denoted by the same reference numerals and description thereof is omitted. As shown in FIG. 10, the image monitoring apparatus according to the fifth embodiment has a configuration in which an alternative reference point acquisition unit 10 is added to the image monitoring apparatus according to the third embodiment (FIG. 7).

  In the fifth embodiment, when there is a reference point that is evaluated not to be accurately acquired by the reference point evaluation unit 8, a new reference point is selected from the spare reference points stored in the reference point storage unit 4. An alternative reference point acquisition unit 10 for acquisition is provided. In the reference point storage unit 4, in addition to the reference point used first, a spare reference point is stored in advance. When the reference point evaluation unit 8 evaluates that a certain reference point is unusable, the alternative reference point acquisition unit 10 acquires a spare reference point in place of the reference point from the reference point storage unit 4 and creates a new reference point. Replenish as. When selecting a new reference point, a reference point close to a reference point evaluated as unusable may be selected, or a priority order of spare reference points is set in advance. You may make it replenish according to. The subsequent processing is the same as in the first embodiment.

  According to the fifth embodiment, in addition to the same effects as the first and third embodiments, an alternative reference point acquisition unit that acquires a new reference point instead of a reference point that is evaluated as unusable by the reference point evaluation unit 8 Therefore, it is possible to prevent the reference point comparison unit 6 and the camera parameter correction unit 7 from not operating due to a shortage of usable reference points due to a long-time operation. In addition, since more than necessary reference points that can be used for calculation in the camera parameter correction unit 7 can always be secured, a highly reliable image monitoring apparatus that can continuously perform stable monitoring can be obtained.

  FIG. 11 is a block diagram illustrating the configuration of the image monitoring apparatus according to the sixth embodiment of the present invention. In FIG. 11, the same or corresponding parts as those in FIGS. 1 and 10 are denoted by the same reference numerals, and description thereof is omitted. As shown in FIG. 11, the image monitoring apparatus according to the sixth embodiment has a configuration in which an operation failure notification unit 11 is added to the image monitoring apparatus according to the fifth embodiment (FIG. 10).

  In the sixth embodiment, the alternative reference point acquisition unit 10 acquires a new reference point from the spare reference points stored in the reference point storage unit 4, and there is no spare reference point that is not used. That is, it is provided with an operation failure notifying unit 11 for notifying the user when the subsequent replenishment is impossible. The user who has received the report immediately sets a new reference point and stores it in the reference point storage unit 4.

  According to the sixth embodiment, in addition to the same effects as those of the fifth embodiment, when there is no spare reference point that is not used in the reference point storage unit 4, the user is informed that the subsequent replenishment cannot be performed. Since the operation failure notification unit 11 is provided, it is possible to request the user to set a new reference point when the number of reference points that cannot be used increases due to environmental changes. As a result, it is possible to prevent the reference point comparison unit 6 and the camera parameter correction unit 7 from operating, and it is possible to continuously perform stable monitoring and obtain an image monitoring apparatus that is easy to maintain.

  The image monitoring apparatus according to the present invention can be used as an image monitoring apparatus that performs road monitoring, suspicious person monitoring, and the like using a zoom turning camera.

  DESCRIPTION OF SYMBOLS 1 Imaging part, 1a, 1b, 1c, 1d image | video, 2 Image processing part, 3 Abnormality notification part, 4 Reference point memory | storage part, 5 Reference point acquisition part, 5a Reference point creation part, 6 Reference point comparison part, 7 Camera parameter Correction part, 8 reference point evaluation part, 9 reference point trouble report part, 10 alternative reference point acquisition part, 11 operation trouble report part.

Claims (7)

An imaging unit that is installed so as to be able to turn in a predetermined direction and acquires an image while changing a focal length according to an imaging target,
An image processing unit that performs image processing on the video acquired by the imaging unit using an installation condition of the imaging unit and detects an abnormality in the video;
A reference point storage unit for storing coordinates at the time of setting three or more reference points arbitrarily set in the video acquired by the imaging unit;
A reference point acquisition unit that acquires the current coordinates of the reference point from an image including the reference point acquired by the imaging unit;
Compares the current coordinates of the reference point acquired by the reference point acquisition unit with the coordinates at the time of setting the corresponding reference point stored in the reference point storage unit, and whether there is a difference of a predetermined amount or more Reference point comparison part to determine whether
When the reference point comparison unit determines that there is a difference of a predetermined amount or more, the current coordinates of at least two of the reference points among the three or more reference points acquired by the reference point acquisition unit, corresponding stored in the reference point storage unit by using the coordinates of the setting time of at least two of the reference points to calculate the installation conditions of the imaging unit, and sends the calculation result to the image processing unit, An image monitoring apparatus comprising: an installation condition correction unit as an installation condition used for image processing of the image processing unit.   The image monitoring device according to claim 1, wherein the installation condition of the imaging unit includes a focal length of a lens related to an optical system of the imaging unit, an angle in the vertical direction of the imaging unit, and an angle in the horizontal direction. An image monitoring apparatus characterized by including camera parameters.   The image monitoring apparatus according to claim 1, wherein the reference point acquisition unit creates a new reference point from the acquired two or more reference points, and includes the current coordinates of the two or more reference points. And a reference point creation unit for obtaining current coordinates of the new reference point.   The image monitoring apparatus according to claim 1, further comprising a reference point evaluation unit that evaluates accuracy of the reference point acquired by the reference point acquisition unit, wherein the installation condition correction unit is the reference point evaluation unit. An image monitoring apparatus that calculates an installation condition of the imaging unit by using a reference point that is evaluated to have been accurately acquired. The image monitoring apparatus according to claim 4 , further comprising: a reference point failure notification unit that notifies a user of a reference point that is evaluated as not being accurately acquired by the reference point evaluation unit. An image monitoring apparatus comprising: The image monitoring apparatus according to claim 4 , wherein when there is a reference point that is evaluated as not accurately acquired by the reference point evaluation unit, a spare reference point stored in the reference point storage unit An image monitoring apparatus, further comprising: an alternative reference point acquisition unit that acquires a new reference point from the above. The image monitoring apparatus according to claim 6 , wherein the alternative reference point acquisition unit acquires a new reference point from the spare reference points stored in the reference point storage unit and is not used. An image monitoring apparatus, further comprising an operation failure notifying unit for notifying the user of the reference point when the reference point does not exist.

Images