JP2000023142A - Image monitoring device - Google Patents

Abstract

(57) Abstract: Provided is an image monitoring device that performs high-precision binarization processing in a short time even if there are a plurality of monitoring areas having different sensitivity levels in an environment where illumination changes drastically. An image monitoring apparatus according to the present invention sets a plurality of monitoring areas on a monitoring screen and specifies a sensitivity level for each monitoring area. Create a screen in which the sensitivity level value is written for each area, calculate the minimum luminance / maximum luminance for each pixel of the input image and the reference image, calculate the conversion magnification corresponding to each pixel of the luminance value, and adjust the sensitivity level I do. A difference process is performed for each pixel between the input image and the reference image, and a difference / binarization is performed between the difference image and the luminance-adjusted binarization threshold screen to extract a change region. Is generated. In the current input image and the reference image, a monitoring target object is detected by performing grayscale pattern matching based on normalized correlation between the same integrated regions.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for monitoring an image of an object by using the image obtained by capturing an image of the object, and a plurality of monitoring areas having different sensitivity levels in an environment where illumination changes drastically. The present invention relates to an image monitoring apparatus that performs high-precision binarization processing in a short period of time even in image monitoring in which exists.

[0002]

2. Description of the Related Art As a first conventional technique for monitoring an object,
2. Description of the Related Art There is known a method in which an object is monitored by performing a binarization process on each of a plurality of regions having different sensitivity levels in an environment where illumination fluctuations are severe to extract a changed region.

As a second conventional technique for monitoring an object, there is known a method of calculating a binarization threshold automatically by calculating a density frequency distribution or the like of a difference image when illumination fluctuation is severe.

[0004]

In the above-described first prior art, if a complex background scene is subjected to binarization processing for each region by changing the sensitivity level in a plurality of regions, the processing time is reduced to the number of regions. There is a problem that it increases in proportion and online monitoring becomes impossible.

In the second prior art method, since the binarization threshold value is automatically calculated for each area because the sensitivity level is different, the processing time increases in proportion to the number of areas, and online monitoring becomes impossible. turn into. On the other hand, if the binarization threshold value is automatically calculated by calculating the density frequency distribution or the like for the entire screen in order to reduce the processing time, it is not possible to calculate an appropriate binarization threshold value due to one of the different luminance changes. There's a problem.

SUMMARY OF THE INVENTION An object of the present invention is to overcome the above-mentioned problems of the prior art, and to perform a high-precision binarization process in a short time even when a plurality of monitoring areas having different sensitivity levels exist in an environment where illumination changes drastically. To provide an image monitoring device that performs the following.

[0007]

According to the method of the present invention for achieving the above object, first, when different sensitivity levels are set for a plurality of monitoring areas, sensitivity level values corresponding to each monitoring area are written on a screen. Creates a binarization screen. Next, the input image and the reference image are compared for each pixel to create an image having the maximum luminance or the minimum luminance (maximum luminance / minimum luminance screen).
Using the maximum luminance / minimum luminance screen, the sensitivity level value is adjusted for each pixel of the binarization screen such that the dark part is lower than the set level value and the light part is higher than the set level value. It is characterized in that the threshold value of the conversion screen is adjusted.

On the other hand, a difference image is created by performing a difference for each pixel between the input image and the reference image. Binarization is performed at a predetermined threshold (fixed value) on a difference screen obtained by performing a pixel-by-pixel difference between the difference image and the screen on which the threshold of the binarization screen has been adjusted. It is characterized in that a change area is extracted.
A noise is removed from the extracted change area, and an integrated area is created by combining the change areas adjacent to each other on the binary image by a circumscribed rectangle or the like. Then, in the reference image and the input image, a tracking source area based on the integrated area is used as a template pattern in one image, and a search area is set to be substantially the same as the area corresponding to the tracking source area in the other image. The method is characterized in that shading pattern matching by normalized correlation is performed, a change area having a similarity based on a correlation value higher than a predetermined value is eliminated as a disturbance, and a change area having a low similarity is extracted and monitored as an object to be monitored. .

An apparatus according to the present invention that achieves the above object, includes an image pickup means such as an ITV camera installed in equipment, detects an object in a monitoring area by processing the image, and stores and notifies the detected object image. An input frame is captured by a monitoring device main body that performs display control and an object monitoring device that includes a monitor such as a display device that displays a detected object.
Input means for conversion, sensitivity area setting means for setting sensitivity levels of a plurality of areas, binarization screen creation means for creating a binarization screen according to the set sensitivity levels, maximum luminance for each pixel between input images / Binarization screen adjustment means for adjusting the threshold value of the binarization screen corresponding to the minimum luminance value, difference image creation means for performing a difference between the reference image and the input image, and binarization of the difference image Changing area extracting means for performing a difference from the screen adjusted by the threshold value by the screen adjusting means and further binarizing to extract a changing area; extracting area integrating means for integrating adjacent extracting areas into one integrated area; An object detecting means for detecting a change area having a similarity lower than a predetermined value as an object by performing a grayscale pattern matching by a normalized correlation process between the reference image and the input image, and storing the image when the object is detected, as necessary. Important Display control means for informing at when a and detection, characterized in that provided in the monitoring apparatus. In addition, the apparatus of the present invention that achieves the above object captures an image with an ITV camera or the like installed in equipment, processes the image to detect an object in a monitoring area, and stores and notifies the detected object image. In an object monitoring device including a monitoring device main body that performs control and a monitor such as a display device that displays a detected object, an image is captured by an ITV camera or the like installed in the facility, and the image is processed to determine an object in a monitoring area. An input frame is captured and A / D-converted by a monitoring apparatus body that performs display control for detecting and storing and detecting a detected object image and notifying the same and a monitor such as a display device that displays a detected moving object. Input means to
Sensitivity area setting means for setting sensitivity levels of a plurality of areas, binarization screen creation means for creating a binarization screen according to the set sensitivity levels, maximum luminance / minimum luminance values for each pixel between input images A binarizing screen adjusting unit for adjusting a threshold value of the binarizing screen in correspondence with the above, a difference image creating unit for performing a difference between the reference image and the input image, and a binarizing screen adjusting unit for the differential image. Changing area extracting means for performing a difference with the screen adjusted for the threshold value and further binarizing to extract a changing area; extracting area integrating means for combining adjacent extracted areas into one integrated area;
Object detection means for detecting a change area having a similarity lower than a predetermined value by performing grayscale pattern matching by normalized correlation processing between the reference image and the input image for each integrated area, and storing the image when the object is detected Display control means for notifying when a request is made as necessary or upon detection is provided in the monitoring apparatus main body.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of an image monitoring apparatus according to an embodiment of the present invention. The monitoring device according to the present embodiment includes a monitoring device main body 10 and a display device 8000. Alternatively, the monitoring device main body 20, the ITV camera 100, and the display device 8000 may be configured. The main body 20 may be realized by mounting an image processing board on a personal computer. The main body 10 may be realized as an intelligent camera in which a camera, an image processing board, and a personal computer are integrated.

In this embodiment, first, the ITV camera 100
When the camera captures an image of the monitoring target, the image input processing unit 500 performs A / D conversion of the captured frame image signal, CCD noise reduction processing, and the like. The area sensitivity setting unit 1000 sets a plurality of monitoring areas on a monitor screen, and further specifies a sensitivity level corresponding to each monitoring area. The binarization screen creation unit 1500 creates a screen in which a sensitivity level value is written for each area from the area set by the area sensitivity setting unit 1000 and the designated sensitivity level. Screen adjustment unit for binarization 2000
Calculates the minimum luminance or the maximum luminance of each pixel of the input image and the reference image, calculates the conversion magnification corresponding to each pixel of the luminance value, and adjusts the sensitivity level of the binarization screen.

The difference image creation unit 3000 performs difference processing for each pixel between the input image and the reference image. Change area extraction unit 4
000 performs a difference / binarization between the difference image of the difference image creating unit 3000 and the binarization threshold screen whose luminance has been adjusted by the binarization screen adjustment unit 2000, and extracts a change area. The extraction area integration unit 5000 generates an integration area in which neighbors are gathered in the extracted area. The object detection unit 6000 performs grayscale pattern matching based on the normalized correlation between the same integrated regions in the current input image and the reference image, determines the similarity of the normalized correlation, and excludes a region having a high similarity as a disturbance. , An area having a low similarity is detected as an object to be monitored.

A display control unit 7000 stores image data of a detected object, and displays information such as a detection position and information such as a detection date and time on a display device 8000 in real time or whenever a request is generated. .

An outline of the processing of the present invention will be described with reference to FIG. First, when monitoring and monitoring a monitoring target scene with the ITV camera 100, one monitoring target scene includes a plurality of important regions for which a higher sensing level is required and a plurality of non-critical regions for which the sensing level is desired to be reduced. Further, when monitoring is performed throughout the day and night, the brightness varies greatly. Furthermore, when there is a lighting device or the like in the monitoring range, the flickering of the lighting device causes local lighting fluctuation. If the process is repeated for a plurality of areas having different sensitivity levels, it takes time, and online monitoring cannot be performed. Therefore, different sensitivity levels are set (for example, manually set) for each monitoring area. First, a binarization screen 217 in which values corresponding to levels are written for each area
Create 0. On the other hand, the input image 2110 and the reference image 21
Screen Ma for which the minimum luminance or the maximum luminance for each pixel was obtained with 20
x / Min (B, F) 2140 is created. Next, Max / Min
(B, F) Conversion magnification 2160 according to luminance value of 2140
(For example, a function of magnification a = 0 to 2) is created and Max / M
The binarization screen 2170 is multiplied by the conversion magnification corresponding to each pixel of the luminance value of in (B, F) 2140. This allows
A binarization adjustment screen 2180 is created in which the sensitivity level set for each monitoring area is adjusted according to the luminance change of the image. Therefore, the input image 2110 and the reference image 212
After performing a pixel-by-pixel difference between the difference image 2130 in which the difference of each pixel is 0 and the binarization adjustment screen 2180, the binarization 2150 (a fixed threshold that does not extract noise) is performed with a predetermined threshold value. Value) to extract the change area.

Further, the extraction area integration means integrates the adjacent areas of the change area to create a circumscribed rectangular area, and the object detection means registers one of the current input image 2110 or the reference image 2120 as a template pattern. With respect to one of the images, a region substantially the same as the template pattern region is used as a search region, and the search region is subjected to light and shade pattern matching with the template pattern, and a change region in which the similarity by the correlation value is higher than a predetermined value is eliminated as a disturbance, A change area having a low similarity is detected as an object to be monitored.

As described above, since a plurality of monitoring regions are collectively collected to extract a changing region in one process, the threshold value of the binarization screen following the illumination fluctuation is also adjusted.
High-speed processing following illumination fluctuations becomes possible, enabling highly accurate online monitoring.

FIG. 3 shows an area sensitivity setting section 1 according to the present invention.
FIG. 4 is an explanatory diagram showing an example of setting a monitoring area of 000 and its sensitivity level. On the monitor displaying the monitoring target image captured by the ITV camera 100, an important area 1200 for which the sensitivity is to be increased is manually or automatically designated, and the sensitivity level is designated with high sensitivity (for example, a luminance value of 64). In addition, the area 1300 where no sensing is performed is manually or automatically designated, and the sensitivity level is designated without sensing (for example, a brightness value of 255). On the other hand, since the sensitivity of the other region 1100 may be reduced, the sensitivity level is set to the medium sensitivity (for example, the luminance value 1).
Specify in 2). Now, in the case of the entry / exit monitoring, the important area where the sensing is high is the door area, and the other area may be designated as the area where the sensing is low.

FIG. 4 shows an image input processing unit 5 according to the present invention.
FIG. 3 is a block diagram showing an example of the internal configuration of the 00. A / D
When the conversion unit 550 takes in an image captured by the ITV camera 100 and performs A / D conversion to create an input image 560Gi, the noise removal unit 570 of the input image Gi performs a smoothing process, a median filter process, or the like corresponding to the noise. To remove the input image 560 from the CCD F and the like.
Create i.

FIG. 5 is a block diagram showing an embodiment of the inside of the binarizing screen adjusting unit 2000 according to the present invention.
The binarization screen setting unit 2510 includes a region sensitivity setting unit 100
A value in which each sensitivity level value is set for each monitoring area designated by 0 is written on the screen to create a binarized threshold screen. The maximum / minimum luminance screen creation unit 2520 includes a current input image and a reference image (a background image or a previous input image may be used).
A maximum / minimum luminance screen in which the maximum luminance or the minimum luminance of each pixel is obtained is created. The binarization screen conversion unit 2530 obtains the pixel value (threshold screen value) of the binarization threshold screen created by the binarization screen setting unit 2510 by the maximum / minimum luminance screen creation unit 2520. The threshold is converted in accordance with the maximum / minimum luminance screen luminance. The conversion magnification is about 0 to 2 times, and the maximum luminance may be used as it is in the non-sensing area because the maximum luminance is used as the threshold value.

Thus, if the conversion magnification is specified as a non-linear (or linear) function corresponding to the luminance, the threshold value of the binarized threshold value screen can be arbitrarily changed for each pixel.

FIG. 6 is an explanatory diagram showing an embodiment of the conversion magnification of the screen conversion unit for binarization 2530 in the present invention.
The conversion magnification 224 is applied to the luminance value 2230 of the maximum / minimum luminance screen created by the maximum / minimum luminance screen creation unit 2520.
The magnification of 0 is a function 2160. For example, when the input image is dark, the luminance difference between the reference image and the input image is often small. Therefore, a smaller threshold value for binarization is more suitable for extracting a changing object. However, when the input image is bright, the luminance difference between the reference image and the input image is often large. Therefore, increasing the threshold value for binarization is more suitable for extracting a changing object. Therefore, when the luminance value 2230 (Max / Min (B, F)) of the maximum / minimum luminance screen is small, the function 2160 sets the conversion magnification of the pixel in the threshold part of the binarization to 1.
A non-linear function is set to be less than 0, and if it is larger, the conversion magnification of the pixel in the threshold part of the binarization is set to 1.0 or more.

FIG. 7 is an explanatory diagram showing an embodiment of the binarization threshold screen converted by the binarization screen conversion section 2530 in the present invention. Area sensitivity setting section 100 shown in FIG.
In the case of the monitoring area of 0 and the value of the sensitivity level, the value 6 of the high sensitivity area 2630 is set according to the luminance value 2230 of the maximum / minimum luminance screen created by the maximum / minimum luminance screen creating unit 2520.
Since the magnification differs for each pixel, 64 × 0.0-6
It varies in the range of 4 × 2.0. Similarly, the medium sensitivity region 26
For the value 12 of 10, 12 ×
It varies in the range of 0.0 to 12 × 2.0. However, the non-sensing area 2620 may remain at 255 where the sensing level is the highest luminance. That is, when the input image is dark, the sensitivity is lower than the specified sensitivity and the change is easy to extract even if the luminance difference is small. When the input image is bright, the input image is higher than the specified sensitivity and the binarization is difficult to extract noise. Create a threshold screen. This makes it possible to adjust the threshold value of the binarized screen for each pixel even in an environment in which there are a plurality of regions having different sensing levels and illumination fluctuations are severe, and the detection accuracy can be improved at high speed.

FIG. 8 shows a change area extracting section 4 according to the present invention.
000 is a block diagram showing an example of the inside of the 000. The binary image creation unit 4100 performs a difference / binarization between the difference image calculated by the difference image creation unit 3000 and the binarization threshold screen converted by the binarization screen conversion unit 2530 to generate a binary image. Create The binarization is performed with a fixed threshold, and a value obtained by adding about two to three gradations to the maximum noise may be used as the threshold. The small area exclusion unit 4200 excludes an isolated region having a small area of a noise level.

FIG. 9 shows an extraction area integration unit 5 according to the present invention.
000 is a block diagram showing an example of the inside of the 000. The label image creation unit 5010 labels the binary image created by the change area extraction unit 4000 to create a label image. The inter-label distance calculation unit 5020 calculates the center of gravity (center) of each label and calculates the distance between the centers of gravity of each label. The label integration determination unit 5030 determines whether or not the distance between the centers of gravity is within a predetermined distance, and sets a plurality of labels within the predetermined distance as the same object to be integrated. Integrated size calculation unit 50
40 calculates the size of the circumscribed rectangle of a plurality of labels to be integrated. The integrated size determination unit 5050 determines whether the size of the circumscribed rectangle of the plurality of labels to be integrated is equal to or larger than a predetermined size. The label integration unit 5060 includes the integrated size determination unit 5
050 assigns the same label number to a plurality of labels determined to be within the predetermined size.

FIG. 10 is an explanatory diagram showing one procedure for integrating extraction areas by the extraction area integration unit 5000 in the present invention. In step 5110, labeling of the binary image is performed, and when labels from 1 to Ln are attached, Ln becomes the total number of labels. Step 5 to repeat the total label number Ln
At 120, the label number (i) is initialized. Next, in step 5130, the label number is increased by one (i is increased).
Then, in step 5140, it is checked whether or not all the labels have been completed. If the processing is not to be ended, the processing after step 5150 is performed on the i-th label. Step 51
At 50, the center of gravity (center) coordinates are calculated for the i-th label. In step 5160, it is checked whether or not the distance between the i-th label and the center between the (i + 1) -th to Ln-th labels in the X and Y directions is within an allowable range. If not, the process returns to step 5130. If it is within the allowable range, it is checked in step 5170 whether or not the size of the circumscribed rectangle of the label in the allowable range is within a predetermined size range. If the circumscribed rectangle is not within the allowable range, the process returns to step 5130. If it is within the allowable range, at step 5180, all the labels within the allowable range are added to the i-th label, and the added label is deleted. When the i-th to Ln-th labels are sorted in ascending order in step 5190, the process returns to step 5130, and the processing is performed again from the i-th label. As a result, labels in which the distance and the size of the circumscribed rectangle are within the allowable range are integrated one after another.

FIG. 11 shows step 516 in the present invention.
It is explanatory drawing which shows one Example at the time of checking whether the distance from the center is an allowable range at 0. For example, when there are two labels, whether or not the label 5220 is within the allowable range of the label 5210 is determined by calculating the center coordinates (o2x, o2y) 5250 of the label 5220 from the center coordinates (o1x, o1y) 5240 of the label 5210. Ask. If the distance 5260 in the X direction and the distance 5270 in the Y direction between the coordinates (o1x, o1y) and the coordinates (o2x, o2y) are within the allowable range, they are integrated as the same label. The allowable range of the distance 5260 in the X direction and the allowable range of the distance 5270 in the Y direction are, for example, about 5 to 10 in the X direction and about 5 in the Y direction when the moving object is a vertically long person.
Alternatively, the X direction may be about 5 to 15, and the Y direction may be about 5 to 15. In any case, an appropriate setting may be made depending on the range to be integrated.

FIG. 12 shows an object detection unit 60 according to the present invention.
FIG. 3 is a block diagram showing an example of the internal configuration of the 00. The circumscribed rectangle calculation unit 6010 of the integrated area calculates a circumscribed rectangle of the integrated object. The correlation processing unit 6030 of the input image and the reference image (which may be a background image or the immediately preceding input image) uses the current input image and the reference image to set one of the circumscribed rectangular areas as a template pattern and The normalized correlation is performed on the image at a position substantially the same as the circumscribed rectangular area (the extended size is about ± 1 pixel). The similarity calculator 6050 between the input image and the reference image calculates the similarity based on the normalized correlation.

The similarity is calculated by

[0030]

(Equation 1)

By the normalized correlation processing. That is, a brightness difference is obtained by normalizing the brightness of the registered template pattern and the target image (an application of the light and shade pattern matching processing of the 3F-8 car number recognition system, the IPSJ 49th National Convention, late 1994) ), And performs the operation of (Expression 1) over the entire matching region to calculate the similarity.

When the similarity calculated by (Equation 1) is equal to or less than a predetermined value (for example, about 0.5 to 0.8), the object determining unit 6060 detects the object as an object and calculates the position information of the circumscribed rectangle. I do. Otherwise, it is excluded as a disturbance.

FIG. 13 is an explanatory diagram showing an embodiment in which an object is detected using the normalized correlation processing. The circumscribed rectangle 6210 or the object 624 of the integrated area of the object 6220 extracted by the difference between the input image 6300 and the reference image 6200
In the circumscribed rectangle 6230 of the integrated area 0, the input image 630
Either 0 or the reference image 6200 includes the extracted object, and the other does not include the extracted object. Now, when a normalization correlation is performed between the same circumscribed rectangular regions in the reference image and the input image of the detection object 6220, the object 6220 becomes
Since it is present in the input image but not in the reference image, the similarity is low. On the other hand, when the normalization correlation is performed between the same circumscribed rectangular regions between the reference image of the detected object 6240 and the input image, even if the object 6240 exists in the input image and exists in the reference image, there are few changed parts. , The similarity decreases.

From this, when the similarity is low, the object is a detection target object, and since the object with a high similarity is similar to the background and has little change, it is determined to be a disturbance. For example, disturbances such as flickering of lighting, shaking of houseplants, and flickering of edges of equipment can be excluded.

FIG. 14 shows an object detection unit 60 according to the present invention.
It is an explanatory view showing one procedure of processing of 00. Step 64
Reference numeral 10 registers, as a template pattern, a grayscale image of a circumscribed rectangular area calculated by the integrated area circumscribed rectangle calculation unit with respect to the current input image. A step 6420 sets a circumscribed rectangular area at substantially the same position as the step 6410 for the reference image as a pattern matching area. Step 6
A step 430 calculates a similarity by performing light and shade pattern matching with the pattern registered in step 6410. In step 6440, the calculated similarity is set to a predetermined value (about 0.5 to 0.5).
0.8) or more. If it is more than the specified value
In step 6540, since it is similar to the background, it is determined that a disturbance has occurred. If it is less than the predetermined value, step 6450 determines that the object is an object because it is not similar to the background. Step 6
A step 550 determines whether or not all the generated circumscribed rectangular areas have been completed. If not, the process returns to step 6410. FIG. 15 is an explanatory view showing one embodiment in which the detection result is displayed on the display device 8000 according to the present invention. Object 62
When detecting the number 20, the display control unit 7000 displays a circumscribed rectangular frame 6210 on the display device 8000 using the stored detection position of the detection object. When the display control unit 7000 controls the display of the detected object on the display device 8000, any method can be used as long as it is clearly visible to the observer that the detection has been performed in color or monochrome, and any display method can be used as long as the detected object can be clearly indicated. By displaying on the display device 8000 in this manner, for example, in the case of monitoring a person, the monitor person can grasp the detected person and the state of the person online using the display device 8000 as an image. If the display device 8000 is at a remote place, the fact that a person has been detected by a standard communication means such as RS-232C may be notified to a videophone or the like, and displayed on a remote display device.

According to the present invention, first, when monitoring and monitoring a scene to be monitored by an ITV camera, one monitored object scene includes, for example, an important region where a higher sensing level is required or an unimportant region where a lower sensing level is desired. Are mixed. In addition, when monitoring day and night, the brightness varies greatly. Furthermore, local illumination fluctuations also occur due to blinking of the illumination device and the like. If the process is repeated for a plurality of areas having different sensitivity levels, it takes time, and online monitoring cannot be performed. Therefore, when a different sensitivity level is set for each monitoring area, first, a binarization screen in which a value corresponding to the level is written for each area is created. On the other hand, a screen Max / Min (B, F) in which the minimum luminance or the maximum luminance of each pixel of the input image and the reference image is obtained is created. Next, a function of a conversion magnification (for example, magnification a = 0 to 2) according to the luminance value of Max / Min (B, F) is created, and Max.
The conversion magnification corresponding to each pixel of the luminance value of / Min (B, F) is multiplied by the binarization screen. As a result, even in an environment where there are a plurality of regions having different sensing levels and illumination fluctuations are severe, the threshold value is adjusted for each pixel by following the brightness of the input image.
Since a binarized screen is created, even if there are a large number of monitoring areas having different sensing levels, it is possible to extract an object by processing the entire screen regardless of the number of areas, so that processing can be speeded up and detection accuracy can be improved. effective.

[0037]

According to the present invention, there is provided an image monitoring apparatus capable of performing high-precision binarization processing in a short time even when a plurality of monitoring areas having different sensitivity levels are present in an environment in which illumination changes are severe. Can be provided.

[Brief description of the drawings]

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

FIG. 2 is an explanatory diagram showing a processing outline of the present invention.

FIG. 3 is an explanatory diagram showing an embodiment of a monitoring area of a sensitivity area setting unit and setting of its sensitivity level in the present invention.

FIG. 4 is a block diagram showing an embodiment of the inside of an image input processing unit according to the present invention.

FIG. 5 is a block diagram showing one embodiment of the inside of a binarizing screen adjustment unit in the present invention.

FIG. 6 is an explanatory diagram showing an embodiment of a conversion magnification of a binarization screen conversion unit according to the present invention.

FIG. 7 is an explanatory diagram showing one embodiment of a binarization threshold screen converted by a binarization screen conversion unit according to the present invention.

FIG. 8 is a block diagram showing one embodiment of the inside of a change area extraction unit according to the present invention.

FIG. 9 is a block diagram showing one embodiment of the inside of the extraction area integration unit in the present invention.

FIG. 10 is an explanatory diagram showing one procedure for integrating extraction regions by an extraction region integration unit according to the present invention.

FIG. 11 is an explanatory diagram showing one embodiment of checking whether or not the distance from the center is within an allowable range according to the present invention.

FIG. 12 is a block diagram showing one embodiment of the inside of the object detection unit in the present invention.

FIG. 13 is an explanatory diagram showing an embodiment in which an object is detected using the normalized correlation processing according to the present invention.

FIG. 14 is an explanatory diagram showing one procedure of processing of the object detection unit in the present invention.

FIG. 15 is an explanatory diagram illustrating an example in which a detection result is displayed on a display device according to the present invention.

[Explanation of symbols]

100: ITV camera, 500: image input processing unit, 10
00: region sensitivity setting unit, 1500: binarization screen creation unit, 2000: binarization screen adjustment unit, 3000: difference image creation unit, 4000: changing region extraction unit, 5000: extraction region integration unit, 6000 ... Object detection unit, 7000: display control unit, 8000: display device.

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Yoshiki Kobayashi 7-1-1, Omika-cho, Hitachi City, Ibaraki Prefecture Inside Hitachi, Ltd.Hitachi Research Laboratory Co., Ltd. Hitachi, Ltd. Video Information Media Division (72) Inventor Yasuyuki Kakuguchi 1410, Inada, Hitachinaka-shi, Ibaraki Pref. Hitachi, Ltd. Video Information Media Division, F-term (reference) GB01 HA18 5C084 AA01 AA06 BB06 BB31 CC19 DD11 GG56 GG57 GG78

Claims (3)

[Claims] 1. An image capturing section for capturing an image of a monitoring target, and a plurality of monitoring areas are set in an image captured by the image capturing section, and a binarizing screen in which a different sensitivity level is designated for each area is created. An adaptive binarization screen is created by adjusting the sensitivity level of the binarization screen according to the luminance of the captured image signal, and a difference between the difference image between the images and the adaptive binarization screen is calculated. An image monitoring device including a monitoring device main body that binarizes a difference image with an adaptive binarization screen. 2. The image monitoring apparatus according to claim 1, wherein the adaptive binarization screen performs luminance adjustment depending on a maximum luminance value or a minimum luminance value for each pixel of two frames for performing a difference between images. 3. An imaging section for imaging a monitoring target, a sensitivity area setting section for setting a plurality of monitoring areas on an image captured by the imaging section and designating a sensitivity level for each of the monitoring areas; A binarizing screen creating unit for creating a screen in which a sensitivity level value is written for each of the monitoring areas from the area set by the area setting unit and the designated sensitivity level; and an image captured by the imaging unit and a reference prepared in advance. A binarization screen adjustment unit that obtains a minimum luminance or a maximum luminance for each pixel with an image, calculates a conversion magnification corresponding to each pixel of a luminance value, and adjusts a sensitivity level of the binarization screen; A difference image creation unit that performs a difference process for each pixel between an image captured by an imaging unit and the reference image, and a binary image obtained by adjusting the brightness of the difference image of the difference image creation unit and the binarization screen adjustment unit Change by performing difference / binarization with screen for use A changing region extracting unit that extracts a region; an extracting region integrating unit that generates an integrated region in which neighborhoods are gathered in the region extracted by the changing region extracting unit; and a shading pattern based on a normalized correlation between the same integrated regions in the reference image. Performs matching, determines the similarity of the normalized correlation, excludes areas with high similarity as disturbances, and detects areas with low similarity as objects to be monitored. An image monitoring apparatus, comprising: a display control unit that stores image data of a detected object, and controls to display information such as a detection position and information such as a detection date and time in real time or whenever a request is generated.