JP2009100259A - Monitoring camera and image monitoring system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a monitoring camera and an image monitoring system capable of accurately analyzing image data of a place where abnormal condition is broken out. <P>SOLUTION: This monitoring camera 2 comprises a CPU 26 for dividing image data having m×n of pixels formed of horizontal m pixels and vertical n lines into a plurality of blocks, an image data processing section 23 for detecting existence of a condition change in each of the plurality of blocks, an image data compressing section 24 for performing digital compression to the image data, and a LAN connection section 25 for transmitting the compressed image data. When the number of block, of which condition changed, is less than a predetermined value, the image data is converted to image data having p×q of pixels formed of horizontal p (<m) pixels and vertical q (<n) lines, and thereafter, first digital compression is performed, and the image data is transmitted. When the number of block, of which condition changed, is a predetermined value or more, at least specified blocks including a block, of which condition change is detected, is converted to image data having pixels more than p×q and less than m×n, and thereafter, second digital compression is performed, and the image data is transmitted. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a monitoring camera and an image monitoring system that detect a change in state of an image capturing a monitoring area.

  A conventional image monitoring system compresses and transmits an image of a monitoring area as a still image in a normal state, and compresses and transmits as a moving image when an abnormal state is detected, thereby transmitting detailed data on the time axis only at the time of abnormality. Is reduced, and the line use cost is reduced (see, for example, Patent Document 1).

Japanese Patent Laid-Open No. 7-121789

  The image monitoring system described in Patent Document 1 has little effect on analyzing an abnormality occurrence location with higher accuracy because the image resolution of a moving image where an abnormality is detected is substantially the same as the resolution of a normal still image. .

  The present invention has been made to solve these problems, and an object of the present invention is to provide a surveillance camera and an image surveillance system capable of analyzing image data of an abnormality occurrence location with high accuracy.

  In order to solve the above-described problems, a surveillance camera according to the present invention includes a unit that divides image data of a monitoring area of m × n pixels composed of m pixels horizontally and n lines vertically into a plurality of blocks, and a plurality of blocks Means for detecting the presence / absence of a state change, a means for digitally compressing image data, and a means for transmitting compressed image data, and when the number of blocks of state change is less than a predetermined value, The data is converted into image data of p × q pixels consisting of p (<m) pixels horizontally and q (<n) lines vertically, and then first digitally compressed and transmitted, and the number of state change blocks is a predetermined value. At this time, only the specific block including at least the block whose state change is detected is converted into image data having more than p × q pixels and not more than m × n pixels, and then transmitted after being subjected to second digital compression. To.

  According to the first aspect of the present invention, when the number of state change blocks is less than a predetermined value, the image data is p × q consisting of p (<m) pixels horizontally and q (<n) lines vertically. When the number of blocks of state change is greater than or equal to a predetermined value after being converted into image data of pixels and transmitted, and only a specific block including at least a block in which the state change is detected is greater than p × q pixels and Since the second digital compression is performed after the image data is converted into image data of m × n pixels or less, the image data at the location where the abnormality has occurred can be analyzed with high accuracy.

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

  FIG. 1 is a configuration diagram of an image monitoring system 1 according to an embodiment of the present invention. As shown in FIG. 1, the image monitoring system 1 has a monitoring camera 2 that can shoot a monitoring area and perform network distribution, and a function of inputting image data captured by the monitoring camera 2 and performing recording and display processing. And a display device 4 that displays image data processed by the monitoring terminal device 3.

  The monitoring camera 2 includes a lens unit 21 that collects visible light, a CCD, a CMOS, and the like, and an image sensor 22 that converts visible light collected by the lens unit 21 into image data, and an input from the image sensor 22. The image data processing unit 23 that performs resolution adjustment of the image data and the state change detection process together with the CPU 26, and the image data input from the image data processing unit 23 is controlled by the JPEG (Joint Photographic Experts Group) system, for example, The image data compression unit 24 that performs digital compression, the LAN connection unit 25 that transmits / receives digital data to / from the monitoring terminal device 3, and controls the units 22 to 25, and has m pixels horizontally and n lines vertically Monitoring area of m × n pixels consisting of Dividing the image data into a plurality of blocks, and a CPU (Central Processing Unit) 26 Metropolitan for detecting the presence or absence of state change for each of a plurality of blocks.

  The monitoring terminal device 3 includes a LAN connection unit 31 that transmits and receives image data photographed and processed by the monitoring camera 2 and additional information of the image data through a transmission network such as a LAN, a CPU, and a memory. A control unit 32 that controls the constituent elements, a storage device 33 that records the received compressed image data, and a decoding process for the digitally compressed image data, and also rewrites the image data of the specific block unit for the image data The compressed data decompression unit 34 and a display driver 35 for displaying an image on the display device 4 are configured.

  Here, the lens unit 21 in the monitoring camera 2 is various according to the installation environment such as a fixed focal point having a focal length of 3 to 9 mm, a varifocal lens, and a telephoto lens having several tens of mm. In the present invention, these specifications are not specified.

  The image sensor 22 converts visible light collected by the lens unit 21 into image data for each pixel using a sensor array such as a CCD or CMOS, and converts the image data into a digital signal using a built-in AD converter. . The conversion pixel size is generally 640 pixels horizontally and 480 lines vertically compatible with VGA (Video Graphics Array), but recently, due to the demand for high-definition images and the progress of semiconductor processes, the horizontal 1280 pixels and vertical 1024 pixels. An image sensor corresponding to SXVGA (Super eXtended Video Graphics Array) of the line is also used. In the embodiment of the present invention, a monochrome image using a VGA quadruple density image sensor having 1280 pixels horizontally and 960 lines vertically will be described.

  The image data processing unit 23 temporarily stores one frame of 1280 pixels and 960 lines in the buffer memory provided inside the image data input from the image sensor 22 and is controlled by the image monitoring system 1. At least two types of processing are performed to detect a change in state from the setting of the resolution and the change in image data.

  Next, the necessity of resolution setting and the resolution setting process will be briefly described.

  In the image monitoring system 1 according to the embodiment of the present invention, the monitoring camera 2 and the monitoring terminal device 3 are directly connected, but it is also assumed that a plurality of monitoring cameras 2 are connected to the monitoring terminal device 3 via a LAN. . When the monitoring camera 2 and the monitoring terminal device 3 are directly connected under the condition where the LAN's effective transmission capacity is about 50 Mbps, even megapixel images compressed to about 25 Mbps can be exchanged at a high rate. However, when the number of monitoring cameras 2 increases, in order for the monitoring terminal device 3 to accurately acquire images transmitted from the respective monitoring cameras 2, it is necessary to reduce the resolution or the number of frames in order to reduce the amount of image data. is there. In the surveillance image system 1, it is necessary to reliably obtain image data at the moment of occurrence of an abnormality, so extremely low frame rate is inappropriate. Therefore, it is appropriate to set the resolution to VGA or QVGA (Quarter Video Graphics Array) with 320 × 240 pixels in normal times.

  2 and 3 are diagrams showing pixel arrangement of image data according to the embodiment of the present invention. In the embodiment of the present invention, the SXVGA resolution of 1280 × 960 pixels stored in the buffer memory of the image data processing unit 23 as shown in FIG. 2 is sampled only for odd-numbered pixels in both the line direction and the vertical direction, It outputs to the image data compression part 24 as VGA resolution of 640x480 pixels as shown in FIG. In FIG. 3, ● represents an effective pixel, and x represents a rejected pixel. Further, only odd-numbered pixels are sampled in FIG. 3, but only even-numbered pixels may be sampled. Another method is to generate 640 × 480 pixels with an average value obtained from four pixels on the left, right, top and bottom. High quality image.

  Since it is important for the image monitoring system 1 to reliably detect an abnormal state in the monitoring area, it is generally considered that the monitor constantly monitors the display device. However, since it is not known when an abnormal state occurs, it is easy to overlook it with a constant monitor alone. In the embodiment of the present invention, a state change is automatically detected from a change in image information, an alarm recording that records images before and after the occurrence of an abnormality for a predetermined period, and an image transmitted from the monitoring camera 2 to the monitoring terminal device 3 is performed. A trigger that switches conditions is generated.

  Next, a process for detecting a state change will be described in detail.

  The image data for one frame consisting of 1280 pixels × 960 lines temporarily stored in the buffer memory in the image data processing unit 23 is processed by the CPU 26 into an 80 × 60 block having 16 pixels × 16 lines as one block. Divide into areas. Then, the luminance signal value of the image data in each block is smoothed and temporarily stored in the current image register in the CPU 26 as smoothed image data for each block. Further, the CPU 26 has a previous image register, which stores the smoothed image data in the current image register and moves the previous smoothed image data stored in the current image register to the previous image register. The smoothed image data stored in the previous image register and the smoothed image data stored in the current image register are compared, and the screen changes from a normal state as shown in FIG. 4 to an abnormal state as shown in FIG. In such a case, it is determined that the state change is “present” for a block having a luminance signal change greater than or equal to a predetermined value, and a flag “1” is set. The above method is used to check whether or not there is a state change for all 80 × 60 blocks, and the state change is “present” for image data in which the consecutive frequency and absolute number of the flag “1” block exceed a predetermined value. A detection signal is generated.

  In the embodiment of the present invention, 16 pixels × 16 lines are divided into 80 × 60 areas with one block, but this is important for filtering detection of excessive state changes due to edge noise of image signals and small fluctuations in the natural world. It is a thing. In addition, if the block size that is the target for detecting the presence or absence of a state change is set to be the same or an integer multiple of the information source encoding unit at the time of image data compression, the compressed image data and the state change detection target Correlation with a block can be obtained. In the embodiment of the present invention, a JPEG (Joint Photographic Experts Group) method is adopted for image compression, and compression encoding is a DCT (Discrete Cosine Transform) encoding method in which 8 pixels × 8 lines are one block. . Accordingly, the block size (16 × 16) that is the target for detecting the presence / absence of the state change is an integral multiple of the information source coding unit (8 × 8) at the time of image data compression, and an image composed of 16 × 16. The data is associated with a DCT encoded block (8 × 8) when the image is compressed with VGA resolution.

  The image data compression unit 24 compresses the amount of information of the image data input from the image data processing unit by a JPEG method that employs a DCT encoding method or an entropy encoding method. Image data for which no change in state is detected by the CPU 26 is input to the image data compression unit 24 as VGA resolution image data of 640 × 480 pixels, and the information amount of the image data is compressed to 30K to 40K bytes and the LAN connection unit. To 25.

  The LAN connection unit 25 processes the additional information such as the compressed image data input from the image data compression unit 24 and the state change detection signal detected by the CPU 26 according to a predetermined LAN protocol, and performs LAN connection of the monitoring terminal device 3. It transmits to the part 31.

  The monitoring terminal device 3 is connected to the monitoring camera 2 via a transmission line via a LAN connection unit 31, performs decoding processing on the digitally compressed image data to display an image, and changes the state of the image data An image is displayed by replacing the image data of the specific block portion with “Yes”.

  The control unit 32 includes a CPU, a memory, and the like, and processes compressed image data and additional information input from the LAN connection unit 31. The storage device 33 records the received compressed image data. The compressed data decompression unit 34 performs a decoding process on the compressed image data and returns the image data to VGA resolution. The image data decoded by the compressed data decompression unit 34 is displayed on the display device 4 by the display driver 35. The compressed data decompression unit 34 has a video memory of several megabytes in order to process the decoded image data, and the VGA resolution of the state change “none” screen received immediately before in the video memory. An immediately preceding VGA image storage area for storing image data is provided.

  Next, a situation where a state change of the monitoring area is detected and an operation when the state change “exist” screen is detected in the monitoring camera 2 will be described.

  The image monitoring system 1 has a system for quickly and accurately grasping abnormalities through constant monitoring by a monitoring person, and an abnormal state that is largely projected on the screen can be easily found, but an unmanned site as shown in FIG. In many cases, the intruders appearing in the distance are often overlooked, and even if they can be found, it is difficult to identify a person. By detecting an abnormality in the state detection process according to the embodiment of the present invention and issuing an alarm, the frequency of oversight by the supervisor is greatly improved, but there is a problem that the abnormal object is small and cannot be identified.

  In the embodiment of the present invention, it is possible to identify a small abnormal object by increasing the resolution of an area including the abnormal object. As shown in FIG. 5, when a state change occurs due to an intruder (abnormality) in an unmanned site, out of all 80 × 60 blocks in which one block of 16 pixels × 16 lines of image data is a state change detection block unit The detection flag “1” is set for the block in which the state change occurs. As described above, the state change detection block is associated with the DCT coding block of the image data compressed with the VGA resolution when the state change is “none”.

  In the image data processing unit 23 of the surveillance camera 2, as shown in FIG. 6, a hatched block including a block in which the detection flag “1” of the screen where the state change “exist” is detected is set as an image of the specific area, The image data is compressed and output to the image data compression unit 24 in association with the maximum resolution SXVGA pixel.

The image data output from the image data processing unit 23 has the information amount of the unit area quadrupled by changing the resolution from VGA to SXVGA. However, since the compression processing is limited to a specific area, It becomes smaller than the image data amount of VGA resolution when the change is “none”. Accordingly, since the image data compression unit 24 can lower the compression rate for the specific area image with the SXVGA resolution than the image data with the VGA resolution, it is possible to generate higher-quality compressed image data in combination with the resolution. is there. That is, when there is m × n pixel monitoring area image data consisting of m pixels horizontally and n lines vertically, the state change is “none”.
When the number of state change blocks is less than a predetermined value, the image data is digitally compressed after being converted into p × q pixel image data consisting of p (<m) pixels horizontally and q (<n) lines vertically. When the first digital compression is performed and the state change is “present” (the number of state change blocks is equal to or greater than a predetermined value), at least only specific blocks including the block in which the state change is detected are larger than p × q pixels and m × n. Digital data compression (second digital compression) is performed after conversion to image data of pixels or less.

  The CPU 26 generates additional information for replacing the image data of the specific area subjected to image compression in association with the image data area of the VGA resolution when the state change is “none”. The generated additional information is associated with the compressed image data and transmitted from the LAN connection unit 25 to the monitoring terminal device 3 via a transmission network such as a LAN.

  The compressed image data and additional information transmitted from the monitoring camera 2 are received from the LAN connection unit 31 of the monitoring terminal device 3 and stored in the storage device 33 by the control unit 32 or decrypted by the compressed data decompression unit 34. Do.

When the state change is “none”, the compressed data decompression unit 34 returns the image data to VGA resolution, and the display driver 35 displays the image on the display device 4. That is, the first digitally compressed image data is decoded to display the first image. On the other hand, when the status change is “Yes”
The compressed data decompression unit 34 returns the image data to SXVGA resolution, and the additional data is added to the immediately preceding VGA resolution image data stored in the video memory (previous VGA image storage area) in the compressed data decompression unit 34. After replacing the obtained image data of the specific area portion, the display driver 35 displays an image on the display device 4. That is, the second digitally compressed image data of the specific block portion is decoded and replaced with the corresponding portion of the first image display to display the second image.

  As described above, the resolution of the specific area captured when the monitoring area is abnormal is four times that of the normal area, and the compression rate of the image data is lower than that of the normal area, so that high-definition image display is possible. Therefore, it is possible to analyze the image data with high accuracy by replacing the specific area at the time of abnormality with the normal image data and displaying the image, and displaying the enlarged image of the specific area where the abnormality has occurred. It is possible to confirm the detection of an abnormal state.

  In the embodiment of the present invention, a black and white image has been described. However, color image data based on RGB and color difference component (Y, Cb, Cr) signals are also classified in association with a luminance signal, and a specific area is defined. If the same processing is performed after setting, a high-definition display of the abnormal area can be obtained.

1 is a configuration diagram of an image monitoring system according to an embodiment of the present invention. It is a figure which shows the pixel arrangement | positioning of the image data by embodiment of this invention. It is a figure which shows the pixel arrangement | positioning of the image data by embodiment of this invention. It is a figure which shows the state change detection map at the time of no state change by embodiment of this invention. It is a figure which shows the state change detection map at the time of a state change by embodiment of this invention. It is a figure which shows the specific area image at the time of the state change by embodiment of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Image monitoring system, 2 Monitoring camera, 3 Monitoring terminal device, 4 Display apparatus, 21 Lens unit, 22 Image sensor, 23 Image data processing part, 24 Image data compression part, 25 LAN connection part, 26 CPU, 31 LAN connection part 32 Control unit 33 Storage device 34 Compressed data decompression unit 35 Display driver.

Claims (4)

Means for dividing the image data of the monitoring area of m × n pixels composed of m pixels horizontally and n lines vertically into a plurality of blocks;
Means for detecting the presence or absence of a state change for each of the plurality of blocks;
Means for digitally compressing the image data;
Means for transmitting the compressed image data;
With
When the number of blocks of the state change is less than a predetermined value, the image data is converted into image data of p × q pixels including p (<m) pixels horizontally and q (<n) lines vertically. When digitally compressed and transmitted, and the number of state change blocks is greater than or equal to a predetermined value, only specific blocks including at least the block in which the state change is detected are converted to image data of more than p × q pixels and less than m × n pixels And a second digital compression before transmission.   The surveillance camera according to claim 1, wherein the second digital compression has a lower compression rate than the first digital compression.   The monitoring camera according to claim 1, wherein a block size that is a target for detecting presence / absence of a state change is the same as or an integer multiple of an information source coding unit at the time of image data compression. A surveillance camera according to claim 1;
A monitoring terminal device connected to the monitoring camera via a transmission line;
With
The monitoring terminal device decodes the first digitally compressed image data and displays a first image, and also decodes the second digitally compressed image data of the specific block unit and An image monitoring system, wherein a second image is displayed in place of a corresponding portion of the first image display.

Images