JP2012235355A - Network camera system and method of controlling the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically update setting information concerning camera setting when network cameras are replaced.SOLUTION: In a network camera system having an imaging device, and a client device which receives video data from the imaging device via a network and displays the video data on a display part, feature information is extracted on the basis of the video data acquired by the imaging device, the extracted feature information is stored in a storage medium, the feature information stored in the storage medium and newly extracted feature information are compared, and setting information containing at least one of a preset position, a masking position, and a motion detection position can be corrected on the basis of the comparison result, thereby allowing displacement to be corrected automatically when the network cameras are replaced.

Description

  The present invention relates to a network camera system and a control method for the network camera system, and more particularly to a technique suitable for use when replacing a network camera.

  2. Description of the Related Art Conventionally, a network camera system using a surveillance camera includes a camera that distributes acquired video to a network and a client machine that receives camera video via a network at a remote location and makes it visible on a display device. In addition, the client machine makes video visible and performs various camera settings.

  For example, settings related to image quality such as camera resolution, white balance, exposure, and shutter speed, and settings related to network distribution such as IP address, compression method, and number of distribution pixels are made. In addition, preset information, private masking, position information for motion detection, etc. are set. Therefore, when the camera is replaced due to a failure or the like, it is necessary to perform the same setting as before on the client machine so that the same setting state is obtained.

In Patent Document 1, subject information at a shooting position acquired in advance is stored in a recording device, and subject information newly acquired by a replaced camera is stored in subject information stored in the recording device. Compare with And the camera system which notifies a camera when it corresponds and alert | reports with a camera is disclosed.
Patent Document 2 discloses a monitoring system in which masking information is stored in a storage medium that can be attached to and detached from the camera, and masking is performed based on the masking information stored in the storage medium.

JP 2010-21920 A JP 2004-364030 A

  However, in the prior art disclosed in the above-mentioned Patent Document 1, it is necessary to compare subject information photographed by the camera before replacement with subject information photographed by the camera after replacement while trial and error. . For this reason, there is a problem that it takes a lot of time to specify the photographing position. Further, when there are a plurality of shooting positions as in the preset position, it is necessary to compare the shooting positions for each preset position. When there are many preset positions, it takes a very long time.

  In the prior art disclosed in Patent Document 2, the masking position is stored in a removable storage medium so that masking is possible even after the camera is replaced. However, when the installation position of the replaced camera is deviated, the masking position is deviated from a desired privacy area, and there is a problem that a meaningless position is masked.

In view of the above-described problems, a first object of the present invention is to automatically update setting information related to camera settings when replacing a network camera.
It is a second object of the present invention to make it possible to easily correct the position when a position shift occurs during the replacement of the network camera.

  The network camera system of the present invention is a network camera system including a photographing device and a client device that receives video data from the photographing device via a network and displays the data on a display unit. An input means for inputting setting information including at least one of a masking position and a motion detection position, and the imaging device captures an imaging range including all or a part of the range indicated by the setting information. A control unit configured to extract feature information based on video data acquired by the imaging device, store the extracted feature information in a storage medium, and set the imaging device based on the setting information; Means for comparing the feature information stored in the storage medium with the newly extracted feature information; and the ratio Based on the comparison result of the means characterized by comprising a correction means for correcting the setting information.

ADVANTAGE OF THE INVENTION According to this invention, the network camera system which can update the setting information which concerns on a camera setting automatically at the time of replacement | exchange of a network camera can be provided.
In addition, according to another feature of the present invention, it is possible to provide a network camera system capable of easily correcting a position when a positional deviation occurs when replacing the network camera.

It is a block diagram which shows the structure of the system in 1st Embodiment. It is a block diagram of a control part in a 1st embodiment. It is an image figure which shows the example of a display before and after camera replacement in 1st Embodiment. It is a flowchart explaining operation | movement of the system in 1st Embodiment. It is an image figure which shows the differential example of a Sobel filter. It is a figure which shows the example of a threshold value. It is an image figure which shows an example of the edge pixel line in 1st Embodiment. It is a figure which shows the example of an edge coordinate in 1st Embodiment. It is a figure which shows an example of the preset coordinate in 1st Embodiment. It is a block diagram which shows the structure of the system in 2nd Embodiment. It is an image figure of the example of a picture display after camera exchange in a 2nd embodiment. It is a flowchart explaining operation | movement of the system in 2nd Embodiment. It is a figure which shows an example of the edge coordinate in 2nd Embodiment. It is a figure which shows the external appearance of the network camera in 3rd Embodiment. It is a block diagram which shows the structure of the system in 3rd Embodiment. It is a flowchart explaining operation | movement of the system in 3rd Embodiment.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.
(First embodiment)
The network camera system according to the first embodiment of the present invention will be described below with reference to FIGS.
FIG. 1 is a block diagram illustrating an example of a system configuration in the present embodiment. In FIG. 1, reference numeral 1 denotes a network camera as a photographing apparatus, which distributes photographed video to a client machine 2 via a network 11. The client machine 2 receives and displays the video sent from the network camera 1.

  The network camera 1 includes an optical system 6, an image sensor 3, a control unit 4, an SD card 5, a mechanical mechanism 7 such as zoom, pan, and tilt, a drive unit 9 that drives zoom, pan, and tilt, and a network for distributing video. Network circuit 10. In the following description, the pan, tilt, and zoom mechanical mechanisms are abbreviated as PTZ mechanisms 7.

The optical system 6 forms an image of light incident from the photographing direction on the image sensor 3. The image sensor 3 photoelectrically converts an optical image from the shooting direction.
The control unit 4 performs image quality control, compression processing for network distribution, control of the PTZ mechanism 7 and the like based on the photoelectrically converted video, writing control to the SD card 5, video contour extraction processing, and the like. . The SD card 5 is used for storing image data contour parameters, contour extraction results, preset positions, masking positions, motion detection positions, zoom information, and the like necessary for processing by the control unit 4. In this embodiment, it is detachably attached to the network camera 1.

  The PTZ mechanism 7 is a mechanical mechanism composed of a motor, a gear, an HP sensor for detecting a mechanical home position, an encoder for detecting a rotational position, and the like that direct the optical system 6 to a preset position. The HP sensor and the encoder each include a mechanical mechanism that rotates in the pan direction and a mechanical mechanism that rotates in the tilt direction.

  The client machine 2 may be a normal PC (personal computer), an input unit 14 such as a keyboard and a mouse, a display unit 12 such as a liquid crystal display, a control unit 13 such as a CPU, a network circuit 15 for performing network communication, and a hard disk. And the like.

FIGS. 3A and 3B are image diagrams showing an example of what happens to the preset position when the network camera 1 is replaced.
In FIG. 3A, reference numeral 61 denotes an entire image. This is an image when the optical system 6 is directed toward the home position 62 detected by the HP sensor provided in each of the pan and tilt mechanisms of the network camera 1 and the zoom is set to the maximum wide.

  Reference numerals 63 and 65 are preset positions 64 and 66 set by the user using the input unit 14 of the client machine 2, and indicate a shooting range in which the zoom is designated by the user. As will be described later, the preset positions 64 and 66 are set by coordinate information and stored in the SD card 5 as setting information of the network camera 1. For example, the shooting range 63 specified by the preset position 64 is a shooting range for shooting a car traveling on a road, and is used for detecting a car and detecting whether the speed of the car is overspeed. The photographing range 65 designated by the preset position 66 is used for photographing the number plate and recognizing the number when there is a vehicle overspeeding. As another example, a car in which a signal is ignored in the shooting range 63 is detected, and similarly, it is used for shooting a license plate in the shooting range 65. As described above, there are various ways to set a plurality of preset positions. For example, you may set so that the imaging | photography range containing all or one part of the range which setting information shows may be image | photographed.

FIG. 3B shows an example in which the network camera is replaced.
The new network camera is an example installed at a position shifted to the right with respect to the network camera before replacement, and the home position is 72. For this reason, the entire image 71 is taken in the rightward direction compared to the image before the replacement.

  The preset positions should be set to positions 75 and 78, which are positions originally set by the network camera 1 before replacement. However, the network camera 1 is replaced and the entire camera has shifted to the right, and the preset positions stored as relative positions from the home position 72 are positions 73 and 76.

  In this embodiment, the control of the control unit 4 of the network camera 1 detects the feature of the video, and the preset positions 73 and 76 of the replaced network camera can be automatically changed to 75 and 78. 3 (a) and 3 (b), the preset position has been described. However, the mask position for masking the private area and the position of the video analysis range such as the moving object detection position change when the network camera is replaced. End up. By using the technique of the present embodiment, it is possible to automatically move the mask position, the image analysis range, and the like.

FIG. 2 is a block diagram illustrating an internal configuration example of the control unit 4 illustrated in FIG. 1. The same components as those in FIG. 1 are given the same numbers.
In FIG. 2, a video signal that is an analog signal obtained from the image sensor 3 is digitized by the input unit 21, and white balance and brightness adjustment are performed by the image quality control unit 22. In the present embodiment, the image quality control unit 22 outputs a video signal represented in a YCbCr format that represents a color space with a luminance signal and a dye signal.

  The video signal whose image quality has been adjusted is converted from the YCbCr format to the RGB format by the compression processing unit 23, and is converted into JPEG, MPEG, H.264, or the like. Video data is generated by compression using a compression method such as H.264. The compressed video data is packetized by the network processing unit 24 in accordance with a protocol such as TCP or UDP. The packetized video data is output to the network circuit 10 and transmitted to the client machine 2 via the network 11 of FIG.

  In this embodiment, edge detection is performed based on video data, and video comparison is performed based on the detected edge characteristics. In this embodiment, a canny filter is used for edge detection. The Canny filter is a filter that detects a thin line edge by combining a Gaussian filter and a Sobel filter, and detects a straight line or a curve as an edge by using upper and lower threshold information to increase the connectivity of the edge. It is. In the present embodiment, as shown in FIG. 2, the Canny filter includes a Gaussian filter 25, a Sobel filter 26, and a hysteresis threshold processing unit 27.

  The video signal in the YCbCr format whose image quality has been adjusted by the image quality control unit 22 is transmitted to the Gaussian filter 25. The Gaussian filter 25 is a filter that performs smoothing to smooth the change in the gray value of the image, and removes the noise component using only the Y component that is the luminance signal of the YCbCr format video signal. By using only the Y component, processing is performed on a monochrome image.

  Edge extraction is performed by the Sobel filter 26 based on the video signal of the black and white image from which the noise component has been removed by the Gaussian filter 25. The Sobel filter 26 is a kind of differential filter, and detects a change in the luminance of the video signal by differentiating. The Y component in the YCbCr format is represented by 8 bits, and the brightness value is expressed by a value from 0 to 255.

FIG. 5 is a diagram showing an example of differentiation in the Sobel filter 26.
FIG. 5A shows an example of a sharp luminance change edge in which the evaluation pixel n has a luminance value of 20 and the evaluation pixel n + 1 has a luminance value of 100. The luminance value at the evaluation pixel is
Y = 80X + 20, and when differentiated, ΔY = 80.

In FIG. 5B, the evaluation pixel n has a luminance value of 20, the evaluation pixel n + 1 has a luminance value of 60, and the evaluation pixel n + 2 has a luminance value of 100. The example of the edge of a gentle brightness | luminance change is shown compared with Fig.5 (a). The luminance value at the evaluation pixel is
Y = 40X + 20, and when differentiated, ΔY = 40.

  In the actual Sobel filter 26, smoothing is performed in the direction in which edges are detected and in the orthogonal direction, and small edges are deleted. Based on the edge value obtained by the Sobel filter 26, the hysteresis threshold processing unit 27 performs edge value selection using two types of threshold information. If the edge value is equal to or greater than the upper threshold value, it is selected as an edge pixel, and if it is equal to or lower than the lower threshold value, it is selected as not being an edge pixel. If it is less than the upper threshold and greater than or equal to the lower threshold, if the adjacent pixel is an edge pixel, it is selected as an edge pixel. Otherwise, it is classified as not an edge pixel.

In order to avoid confusion, the wording relating to the edge will be described below. Edge value: A differential value of each pixel obtained by the Sobel filter 26.
Edge pixel: A pixel whose edge value is obtained by the Sobel filter 26.
Edge pixels greater than or equal to threshold: edge pixels extracted by the hysteresis threshold processing unit 27 as being greater than or equal to the threshold.
Edge pixel line: A series of straight or curved pixels in which edge pixels equal to or greater than a threshold are connected.
Edge coordinates: The coordinates of the end of the edge pixel line and the refraction point (bent point).

  The RAM 30 is a volatile storage medium, and the edge pixel line for which the selection of edge pixels equal to or higher than the threshold is completed by the hysteresis threshold processing unit 27 is stored in the RAM 30 under the control of the CPU 29.

The CPU 29 is a central processing unit with a built-in flash memory. The following processing is executed by executing the microinstruction stored in the flash memory.
Based on the preset position and zoom command received from the client machine 2, a PTZ mechanism control signal is transmitted to the drive unit 9 in order to change the shooting direction. Further, control is performed to save the edge pixel line obtained by the Canny filter from the hysteresis threshold processing unit 27 to the RAM 30.

  Also, the upper and lower threshold information for edge detection stored in the SD card 5 is transferred to the hysteresis threshold processor 27. Further, based on the edge pixel line stored in the RAM 30, a storage process is performed in which edge coordinates are obtained and written to the SD card 5.

  FIG. 4 is a flowchart for explaining the processing procedure for correcting the preset position when the preset position changes when the network camera is exchanged under the control of the CPU 29, and the processing for storing the edge coordinates in the SD card. It is. Specifically, this is a process performed by the installer in a state where the network camera is installed, the SD card is inserted, and the power is turned on. When the network camera is replaced, the process is started after the SD card is taken out from the old network camera and attached to the new network camera, and then the power is turned on.

  In FIG. 4, initial setting is performed in S401. The PTZ mechanism 7 shown in FIG. 1 is driven to set the pan position and tilt position to the home position. Next, in step S402, the zoom is set to the wide end to acquire an image. As a result, the entire image 61 or 71 shown in FIGS. 3A and 3B is acquired.

  Next, in S403, it is determined whether or not the flag F1 stored in the SD card 5 is 1. If F1 = 1, this indicates that the network camera 1 has been replaced, and edge information has already been stored in the SD card 5, and the processing from S420 is performed. If F1 = 0, it indicates that a network camera has been newly installed, and the processing from S404 is executed.

S404 to S418 are processes when a network camera is newly installed.
In S404, the value of the variable X is set to 1.
In S405, the edge value of each pixel is acquired using the Gaussian filter 25 and the Sobel filter 26 of FIG.
In S406, using the upper threshold Na and the lower threshold Nb corresponding to the variable X, the hysteresis threshold processing unit 27 extracts edge pixels that are equal to or greater than the threshold.

  FIG. 6 is a diagram illustrating the relationship between the variable X, the upper threshold Na, and the lower threshold Nb in the present embodiment. For example, when X = 1, since the upper limit threshold value Na is 70, the differential value that is the edge value shown in FIG. 5A is 80, and is extracted as an edge pixel. The edge value shown in FIG. 5B is 40, and is not extracted as an edge pixel.

  When X = 1, the lower limit threshold Nb is 35. When adjacent pixels are detected as edge pixels, even if edge information = 40 shown in FIG. 5B is 40> 35, they are extracted as edge pixels.

  FIG. 7A illustrates an example of video display. There are 90, 91, 92, 93, 94, etc. as linear edge pixel lines. Reference numerals 90 and 91 denote wall surfaces of the building, and there is little difference from the brightness of the surrounding sky. The edge information of each pixel is 40, and it is not extracted as an edge pixel line from the upper limit threshold Na = 70 when X = 1. 92 and 93 are black poles for signals, and there is a large difference in luminance between the surrounding roads. The edge information of each pixel is 70 or more, and is extracted as an edge pixel line from the upper limit threshold value Na = 70 when X = 1.

Returning to the flowchart of FIG.
In S407, using the threshold values Na and Nb, it is determined whether or not there is an edge pixel line having a straight line whose extracted edge is longer than the constant K. If there is an edge pixel line longer than K as a result of determination, the processing after S413 is performed. If there is no edge pixel line longer than K as a result of determination, the processing after S408 is performed.

  In the example of FIG. 7A, 95 represents the length of the constant K. Reference numeral 94 denotes an edge pixel line on the back of the black car, which is a straight edge pixel line shorter than the constant K, and is not extracted as an edge pixel line. Since 92 and 93 are straight lines and are longer than the constant K, they are extracted as long edge pixel lines. In the following description, an edge pixel line longer than the constant K will be described as an extracted edge pixel line.

In S408, the variable X is incremented by "+1", and in S409, it is determined whether or not the variable X is 9.
S406 to S409 is a loop process that counts up from “X = 1” and is executed until “X = 9”. Each time the variable X is counted up, the upper limit threshold value Na and the lower limit threshold value Nb shown in FIG. 6 are changed and the detection of the extracted edge pixel line is repeated.

  The loop processing from S406 to S409 is repeated, and even if X = 8 and the upper limit threshold value Na = 35, if the extracted edge pixel line cannot be detected by the determination of S407, the process proceeds to S410 by the determination of S409. For example, a case where the photographed image is a blue sky or a video showing the sea can be considered.

  S410 is an error process for notifying the user that there is no extracted edge pixel line. Although not shown in the block diagram of FIG. 1, processing such as turning on an error lamp and sounding a buzzer notifying that an error has occurred is performed under the control of the control unit 4.

  In S411, the flag F1 stored in the SD card 5 is set to “0” to indicate that the edge coordinates are not stored in the SD card 5. In step S412, the client machine is notified of the end of the edge detection process and the process ends.

S413 to S118 indicate processing when the extracted edge pixel line is detected in the determination of S407.
In S413, the upper limit and lower limit threshold values Na and Nb are stored in the SD card 5. This is because when the network camera is replaced, the newly installed network camera uses the upper and lower thresholds Na and Nb.
In step S414, edge coordinates are obtained.

FIG. 7B shows an example of edge pixel lines extracted by the processing from S404 to S409.
In FIG. 7B, 160 shows an entire screen from which edge pixel lines are extracted. In this embodiment, edge pixel lines are extracted from an HD image of 1280 pixels in the horizontal direction and 720 pixels in the vertical direction. End points and refraction points which are end points of the pixel lines 150 and 155 are extracted.
The end points of the edge pixel line 150 are 151, 152, and 153.
The end points of the edge pixel line 155 are 156, 157 and 158.

  FIG. 8A shows an example of the edge coordinates of the end points of the edge pixel line. If the upper left corner of FIG. 7 is X coordinate = 1 and Y coordinate = 1, the edge coordinates of the end points of the edge pixel lines are the values shown in FIG.

  In S415, the edge coordinates are stored in the SD card 5. As shown in FIG. 8A, the edge coordinates are represented by numerical data. Each edge has 2 bytes for X coordinates and 4 bytes for 2 bytes for Y coordinates, and FIG. 8A has 6 end points. In the example shown, it can be represented by 24 bytes in total. In the example described in Patent Document 1, images are compared, and in order to store the images to be compared, several hundred kilobytes to several megabytes are required. In the present embodiment, one feature of the present embodiment is that edge pixel line information that enables image comparison can be stored with a small number of bytes.

In S416, “1” is set in the flag F1, indicating that the upper and lower thresholds Na and Nb and edge coordinates for edge pixel line extraction are stored in the SD card 5.
In S417, the video data is transmitted to the client machine 2 of FIG. Since the image transmitted from the network camera 1 has the zoom magnification set to the wide end in S402, the entire image as shown in FIG.

In S418, preset coordinates and zoom magnification are set and stored in the SD card 5.
The preset coordinates and zoom magnification are set according to information input by operating the input unit 14 of the client machine 2 in FIG. 1 by, for example, a security guard in the security room. The preset preset coordinates and zoom magnification are transmitted from the client machine 2 to the network camera 1, and the network camera 1 stores the received preset coordinates and zoom magnification in the SD card 5.

  FIG. 8A shows preset coordinates and zoom magnifications at the preset positions 64 and 66 in the entire image shown in FIG. The horizontal position is 1280 pixels, the vertical direction is 720 pixels, the coordinates of the upper left corner are (1, 1), and the preset position 64 is (980, 532), and the preset position 66 is (757, 692). After the process of S418, the process proceeds to S412 to notify the client machine 2 of the end of the edge detection process and the process ends.

The processing after S420 is processing when F1 = 1 is determined in S403 and the network camera is replaced.
In S420, the edge value of each pixel is acquired using the Gaussian filter 25 and the Sobel filter 26 shown in FIG.
In S421, using the upper threshold Na and the lower threshold Nb stored in the SD card 5, the hysteresis threshold processor 27 extracts edge pixels equal to or greater than the threshold. Also, edge pixel lines longer than the constant K are extracted.

In S422, the edge coordinates of the edge pixel line extracted in S421 are obtained.
In S423, it is determined whether or not an edge pixel line can be extracted by the processes in S421 and S422, and edge coordinates can be obtained. If the edge coordinates cannot be obtained, for example, when the shooting direction of the exchanged network camera is facing the sky or the sea, the edge cannot be detected, and the process proceeds to S424.

  S424 is error processing. Although not shown in the block diagram of FIG. 1, the error lamp is turned on under the control of the control unit 4. Perform processing such as sounding a buzzer to notify that an error has occurred. After completing the error process in S424, the process proceeds to S412 to notify the client machine 2 that the edge detection process is completed, and the process ends.

  On the other hand, if the edge coordinates can be obtained in the determination in S423, the process proceeds to S425, and the edge coordinates obtained in S422 and the edge coordinates stored in the SD card 5 in S415 are compared. Based on the comparison result in S425, it is determined in S426 whether or not the edge coordinate difference is large. In the present embodiment, based on the comparison result, it is determined that there is no difference if there is a difference of two pixels in both the X-axis direction and the Y-axis direction, and the process proceeds to S412 to indicate that the edge detection process is completed. And exit.

  If the difference in edge coordinates is large in the determination in S426, the process proceeds to S427, and the detected edge coordinates are stored in the SD card 5. FIG. 8B shows edge coordinates in the case where the exchanged network camera is installed with the imaging direction shifted to the left as shown in FIG. 3A as shown in FIG. 3B. As shown in FIG. 8, the X coordinates of the edge pixel lines 1 and 2 in FIG. 8B are 300 dots fewer than the edge pixel lines 1 and 2 in FIG. 8A, and the difference in S426 is large. .

  Next, in S428, the preset coordinates are corrected based on the difference in edge coordinates. The preset coordinates in the example of FIG. 3A are stored in the SD card 5 as shown in FIG. 9A, and the preset position is corrected by 300 dots, which is the difference in edge coordinates. FIG. 9B shows the preset coordinates after correction.

Next, the preset coordinates corrected in S429 are stored in the SD card 5, and the process proceeds to S412 to notify the client machine 2 of the end of the edge detection process. The first embodiment has been described above.
In the present embodiment, if the edge coordinates are not detected in S423, error processing is performed in S424. However, when the edge coordinates are not detected, the upper limit threshold value Na and the lower limit threshold value Nb shown in FIG. 6 may be gradually lowered to make it easier to detect the edge coordinates and reduce the frequency of error.

  In this embodiment, the zoom is set to be wide in S402, and a video as an entire image is acquired. As another example, using the PTZ mechanism 7 shown in FIG. 1, a panorama image that is the entire image of the shooting range of the network camera is generated, and edge pixel lines are extracted based on the generated panorama image. May be.

  In this embodiment, image features are extracted using edge detection using the Canny method. However, edge detection using other methods may be used, and image features are extracted using luminance distribution information. May be. Further, the feature information of the image may be extracted by combining the luminance distribution detection and the edge detection. Furthermore, in the present embodiment, the embodiment has been described in which the edge coordinates and the preset coordinates are stored in the SD card 5, but may be stored in the client machine 2 or a recording server (not shown).

(Second Embodiment)
Hereinafter, a second embodiment of the network camera system will be described with reference to FIGS.
FIG. 10 is a block diagram showing a schematic configuration of the network camera system of the present embodiment. The same blocks as those in the first embodiment are given the same numbers.
In FIG. 10, reference numeral 201 denotes a rotation drive mechanism, which is a mechanism that rotates the photographing angle of a lens unit that includes the optical system 6 and the image sensor 3. When the network camera 1 is mounted on an inclined ceiling or vertical wall, the lens unit is rotated so that the image can be acquired horizontally.
In the present embodiment, when the network camera 1 is mounted on an inclined ceiling or vertical wall, the angle of the mounting surface of the network camera 1 and the mounting direction of the network camera 1 are detected. Then, the rotation mechanism 201 is automatically controlled from the results of angle detection and direction detection, so that the photographing direction can be optimized.

  FIG. 11 is a diagram illustrating an example of an image when the network camera attached in a horizontal direction such as a ceiling is vertically attached to a wall or the like when the network camera photographed in FIG. 7A is replaced. . In FIG. 11, 202 and 203 are edge pixel lines extracted by the control unit 4 of FIG. 10, and are extracted in a shape rotated by 90 degrees compared to the edge pixel lines 92 and 93 of FIG. .

FIG. 12 is a flowchart showing a processing procedure after replacing the network camera in the present embodiment.
In the present embodiment, since S401 to S422 shown in FIG. 4 in the first embodiment are the same processing, they are omitted in FIG. Also, the same numbers are assigned to the same processing steps as those in the flowchart shown in FIG.

In S423 of FIG. 12, if the replaced network camera detects that there is an edge coordinate, the process proceeds to S425, and the obtained edge coordinate is compared with the edge coordinate stored in the SD card 5.
Based on the comparison result in S425, it is determined in S426 whether or not the edge coordinate difference is large. If the difference is large, the process proceeds to S1210, and the difference between the X and Y directions of the edge coordinates is calculated.

  FIG. 13A is a diagram illustrating the edge coordinates of the edge pixel lines 150 and 155 in FIG. 7B, which are the same values as in FIG. 8A. Both edge pixel lines 1 and 2 represent pixel lines that are long in the Y direction. FIG. 13B is a diagram illustrating the edge coordinates of the edge pixel lines 202 and 203 in FIG. 11, and both the edge pixel lines 1 and 2 represent edge pixel lines that are long in the X direction.

  FIGS. 13C and 13D are diagrams in which the difference between the edge coordinates in the X and Y directions, which is the processing of S1210, is calculated. The original image of FIG. 11 is an image inclined by 90 degrees compared to the image shown in FIG. 7B, and naturally the difference values in the X direction and the Y direction are opposite values. Yes.

  In S1211 of FIG. 12, the difference between the X direction and the Y direction is compared to determine whether or not the difference is large. If they are similar, the mounting directions are considered to be the same, and the process proceeds to S427 to S429 as in the first embodiment.

  If the difference is large in S1211, the process proceeds to S1212 and the comparison is made with the X and Y directions reversed. In the example shown in FIGS. 13C and 13D, if reversed, the difference is substantially the same, and the process proceeds to S1214 by the determination in S1213. In S1214, the rotation angle 201 is rotated by 90 degrees using the rotation mechanism 201 of FIG.

  On the other hand, if it is determined in S1213 that the difference is large, the network camera is attached obliquely, and the process proceeds to S1215. In S1215, the inclination angle of the edge pixel line is calculated, and in S1216, the imaging angle is rotated using the rotation mechanism 201 shown in FIG.

Since the shooting angle has been rotated in S1214 and S1216, the shot video has changed. Therefore, the edge coordinates are reset in S1217, and thereafter, the processing from S427 is performed as in the first embodiment.
The second embodiment has been described above. In this embodiment, the installation state of the network camera is detected based on the feature of the edge pixel line, and the photographing angle can be automatically corrected using the rotation mechanism 201.

(Third embodiment)
A network camera system according to the third embodiment of the present invention will be described below with reference to FIGS.
FIG. 14 is an image diagram of the network camera in the present embodiment.
In FIG. 14, reference numerals 301, 302, and 303 denote LEDs. When replacing the network camera, if there is an installation position shift, it is for informing the installation operator that the position shift has occurred. It is possible to prompt them to be in a position.

  The LED 301 is an LED that detects that the installation position of the network camera is shifted in the right direction and notifies the installation worker to rotate in the left direction. The LED 302 is an LED that detects that the installation position of the network camera is shifted to the left and notifies the installation operator to rotate in the right direction. The LED 303 is an LED that detects the displacement of the installation position and informs the installation worker that the displacement is no longer present by constantly lighting when the displacement is lost by flashing and lighting. is there.

FIG. 15 is a block diagram showing a schematic configuration of the network camera system in the present embodiment. The same components as those in the first embodiment are given the same numbers, and description thereof is omitted.
In FIG. 15, reference numerals 301, 302, and 303 denote LEDs described with reference to FIG. The LED lighting circuit 304 is a circuit that lights the LEDs 301, 302, and 303 under the control of the control unit 4.

  In the network camera 1 according to the present embodiment, the pan / tilt structure for changing the photographing direction is a structure for manually changing, and does not have a driving unit. In the second embodiment, the rotation provided is also changed manually. Reference numeral 307 denotes a punchle that is manually changed and a rotation structure. Reference numeral 306 denotes a zoom mechanism. Reference numeral 305 denotes a drive unit that electrically drives the zoom mechanism 306 under the control of the control unit 4.

FIG. 16 is a flowchart showing a processing procedure after replacing a network camera in the present embodiment.
In the present embodiment, S401 to S422 shown in FIG. 4 in the first embodiment are the same processing, and are omitted in FIG. The same processing steps as those in the flowchart shown in FIG.

In the flowchart of FIG. 16, in S423, if the edge coordinates are not stored in the SD card 5 in the replaced network camera, an error state occurs. In order to indicate the error state to the installation worker, the process proceeds to S1601, and the LEDs 301, 302, and 303 are blinked.
In S1602, a lighting time is secured by a 10-second timer, and in S1603, the LEDs 301, 302, and 303 are turned off. Thereafter, the process proceeds to S412 to notify the client machine of the end of the edge detection process and the process ends.

  If the edge coordinates are stored in the SD card 5 in the replaced network camera in S423, the process proceeds to S425, and the obtained edge coordinates are compared with the edge coordinates stored in the SD card 5. Based on the comparison result in S425, it is determined in S426 whether or not the edge coordinate difference is large.

  As a result of the determination, if the difference is small, the process proceeds to S1604, and the LED 303 is turned on to notify the installation operator that the installation direction of the replaced network camera is almost the same as before the replacement. Thereafter, the process proceeds to S412, where the client machine is notified of the end of the edge detection process and the process ends.

  As a result of the determination in S426, if the difference in edge coordinates is large, the process proceeds to S1605, where it is determined whether or not there is a shift in the right direction. If it is shifted to the right, the process proceeds to S1606 and the LED 301 blinks. Next, proceeding to S1607, it is determined whether or not there is a shift to the left. As a result of the determination in S1607, if it is shifted to the left, the process proceeds to S1608 and the LED 302 blinks. In the present embodiment, the process is merely blinking the LEDs 301 and 302. However, as the other process, the blinking period is lengthened as the difference is larger, or the blinking period is shortened as the difference becomes smaller. You may do it.

  In the video examples of FIGS. 3A and 3B shown in the first embodiment, the video of the network camera before replacement is the left of the video of FIG. 3B compared to FIG. 3A. It's off. This indicates that the installation position of the network camera is shifted to the right. As a result of the determination in S1605, the process proceeds to S1606, and the LED 301 blinks. By repeating S425 to S1608, the installation operator can clearly indicate the installation direction while viewing the LEDs.

The third embodiment has been described above.
As a modification of the above-described embodiment, the rotation angle adjustment can be assisted by combining with the second embodiment. When the network camera attached to the ceiling of the horizontal plane is replaced, when the camera is attached to an inclined ceiling or a vertical wall, an indication of the correction direction of the rotation angle can be manually performed by adding an LED. In the present embodiment, the installation direction is clearly indicated by the LED. However, as another embodiment, the installation direction may be notified by a voice guide device that performs voice output.

  As mentioned above, although preferable embodiment of this invention was described, this invention is not limited to these embodiment, A various deformation | transformation and change are possible within the range of the summary.

(Other embodiments)
The present invention can also be realized by executing the following processing. That is, software (computer program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various computer-readable storage media. Then, the computer (or CPU, MPU, etc.) of the system or apparatus reads out and executes the program.

1 network camera, 2 client machine, 5 SD card, 14 input unit, 25 Gaussian filter, 26 Sobel filter, 27 hysteresis threshold processing unit, 29 CPU

Claims (12)

In a network camera system having an imaging device and a client machine that receives video data from the imaging device via a network and displays the data on a display unit,
The client machine includes input means for inputting setting information including at least one of a preset position, a masking position, and a motion detection position,
The imaging device extracts feature information based on video data acquired by the imaging device that captures an imaging range including all or part of the range indicated by the setting information, and the extracted feature information is stored in a storage medium. Storing control means for setting the photographing apparatus based on the setting information,
The control means includes comparison means for comparing feature information stored in the storage medium with newly extracted feature information, and correction means for correcting the setting information based on a comparison result of the comparison means. A network camera system characterized by that. The control means comprises an edge detection means and a luminance distribution detection means,
The network camera system according to claim 1, wherein the feature information is coordinate information extracted by the edge detection unit or the luminance distribution detection unit.   3. The edge detection unit performs edge detection using upper and lower threshold information in order to increase edge connectivity, and the control unit stores the threshold information in the storage medium. The network camera system described in 1.   The network camera system according to claim 1, wherein the storage medium is a storage medium that is detachable from the photographing apparatus.   5. The video data used when the control means extracts the feature information is video data when the zoom of the photographing apparatus is set to a wide end. The network camera system described.   5. The video data used when the control unit extracts the feature information is a panoramic image that is an entire image of a shooting range of the shooting device. 6. Network camera system. The imaging device includes a rotation driving unit, and the control unit includes an angle detection unit that detects an angle of a mounting surface of the imaging device,
The network according to claim 1, wherein the control unit controls the rotation driving unit based on an angle detected by the angle detection unit based on a comparison result of the comparison unit. Camera system. In a network camera system composed of a photographing device and a client machine that receives video data from the photographing device via a network and displays the data on a display unit,
The client machine includes input means for inputting setting information including at least one of a preset position, a masking position, and a motion detection position,
The imaging device extracts feature information based on a video acquired by the imaging device that captures a shooting range including all or a part of the range indicated by the notification unit and the setting information, and the extracted feature information Control means for storing in a storage medium and setting the imaging device based on the setting information;
The control means includes comparison means for comparing feature information stored in the storage medium with newly extracted feature information, and controls the notification means based on a comparison result of the comparison means. Network camera system.   The network camera system according to claim 8, wherein the notification unit is an LED or a voice guide device that performs voice output. The control means has a direction detection means for detecting the mounting direction of the photographing device,
10. The network according to claim 8, wherein the control unit specifies the direction in which to correct the mounting of the photographing apparatus by controlling the notification unit based on the mounting direction detected by the direction detecting unit. 11. Camera system. In a control method of a network camera system having a photographing device and a client machine that receives video data from the photographing device via a network and displays the data on a display unit,
The client machine includes an input step of inputting setting information including at least one of a preset position, a masking position, and a motion detection position,
The imaging device extracts feature information based on video data acquired by the imaging device that captures an imaging range including all or part of the range indicated by the setting information, and the extracted feature information is stored in a storage medium. Storing a control step of setting the photographing apparatus based on the setting information,
The control step includes a comparison step of comparing feature information stored in the storage medium with newly extracted feature information, and a correction step of correcting the setting information based on a comparison result of the comparison step. A network camera system control method. In a control method of a network camera system including an imaging device having a notification unit and a client machine that receives video data from the imaging device via a network and displays the data on a display unit,
The client machine includes an input step of inputting setting information including at least one of a preset position, a masking position, and a motion detection position,
The imaging device extracts feature information based on the video acquired by the imaging device that captures an imaging range including all or part of the range indicated by the setting information, and stores the extracted feature information in a storage medium And a control step of setting the photographing device based on the setting information,
The control step includes a comparison step of comparing the feature information stored in the storage medium with the newly extracted feature information, and controls the notification unit based on a comparison result of the comparison step. Network camera system control method.

Images