JP3294468B2 - Object detection method in video monitoring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video surveillance system using a television camera, and more particularly to a video surveillance system suitable for a case where extremely high reliability is required, such as detection of an object entering a dangerous area. In the object detection method .

[0002]

2. Description of the Related Art In recent years, a video surveillance device using a television camera has been widely and generally used. However, in such a surveillance system, not only manned surveillance by an image monitor but also within a surveillance field of view is required. A system that automatically detects an intruding object such as a human or a car from an image signal and obtains a predetermined notification or alarm is required. Therefore, it is necessary to detect an object by signal processing.

By the way, as a prior art of such a method of detecting an object by processing an image signal, a latest image of sequentially input images is compared with an old image, and a difference in luminance value is obtained for each pixel. A method of detecting an area having a large difference value as an object is known and widely used.

FIG. 8 shows a basic flow of this prior art method . That is, in this prior art, first, processing 8
At step 1, an image is captured. Then, at step 82, a difference between luminance values is obtained. The difference is binarized at step 83. Finally, at step 84, the binarized data is determined by a predetermined threshold. An object is detected.

However, if the difference is simply obtained as in the prior art, as shown in FIG. 9A, a partially missing portion 91 occurs in the object 90 to be detected, and the problem of erroneous detection is caused. There is.

However, the above-mentioned problem of the prior art can be solved by applying an expansion / contraction operation.

As an application example of the moving object detection method by the difference processing, for example, published in July 1994,
Ueda described in “O plus E” No.176, pp.122-136
There is a paper entitled "Intelligent Video Handling Using Image Recognition Technology" by another one.

The method for detecting a moving object according to this paper is such that an object is detected from three images which are consecutive in time, and as shown in FIG. , 112, and 113 as input images, calculate the luminance difference between the image 111 and the image 112, and the luminance difference between the image 112 and the image 113, binarize the image, and perform the dilation / contraction operation to obtain the image 114 and the image 115. obtain. Next, the AND of these binarized images 114 and 115
The common part of each image is obtained by the processing, and the image 116 of the object is obtained.
You get

However, according to this method , FIG.
(B) False detection 93 of the background 92 due to illumination fluctuation. (c) If the entering object emits light, an erroneous detection 95 due to the light-reflecting area 94. (d) False detection 97 by the shadow area 96 of the entered object.
As a result, there is a risk that the object may be erroneously detected as being present even though the object is not actually present in the image visual field.

[0010]

In the prior art, as described above, it cannot be said that sufficient consideration is given to the point of erroneous detection of an object, and as shown in FIG. There is a problem of false detection 93 due to the background, and FIG. 9 (c)
As shown in FIG. 9, when an object emits light such as a headlight of a car, the problem of erroneous detection 95 due to the light-reflecting area 94 or the intrusion as shown in FIG. There is a problem of false detection 97 due to the shadow area 96 of the object.

The problem of erroneous detection can be almost fatal, especially when applied to monitoring of objects such as pedestrians and vehicles entering a railroad crossing.
There is a disadvantage that the range of application is limited.

A first object of the present invention is to make it possible to easily identify only an image of an object present in a screen of an image signal, and to surely detect an object entering a field of view being monitored. The object of the present invention is to provide an object detection method in a video monitoring device.

A second object of the present invention is to provide a video surveillance device which can reliably detect a portion where the temperature differs locally in the field of view being monitored. It is to provide a method .

[0014]

First, according to a first aspect of the present invention, the first object is to provide a basic processing of the conventional object detection method shown in FIG. 8 as shown in FIG. First, a code information storage step 13, a decision processing step 14 for branching according to the code, and a first binarization processing step 1
This is achieved by adding 5 and the second binarization processing step 16 and making the thresholds of the first binarization processing step 15 and the second binarization processing step 16 different. The numbers in parentheses indicate steps having the same processing contents in other figures.

Next, according to a second aspect of the present invention, the first object is, as shown in FIG. 2, the similarity calculating step 24 in the configuration of the first aspect of the present invention shown in FIG. And the first for performing threshold determination by adding similarity information to difference value information
And the second binarization processing steps 26 and 27 are added.

According to the third aspect of the present invention, the first object is, as shown in FIG. 3, a judgment processing step of classifying into one of the two regions 37 and 38 according to the degree of similarity. Is added, and the region 37 and the region 38 are detected separately.

According to a fourth aspect of the present invention, the second object is to use an infrared imaging device as the imaging means in the configuration of the second invention shown in FIG. As shown in FIG. 6, this processing is achieved by adding an evaluation processing step 67 of the similarity between the local luminance distribution shape of the input image and the background image.

According to the fifth aspect of the present invention, the second object is that, in the configuration of the second aspect of the present invention shown in FIG. 2, an infrared imaging device is used as the imaging means, and the sign is negative. As shown in FIG. 7, the processing in this case is achieved by adding an evaluation processing step 77 of the similarity between the local luminance distribution shape of the input image and the background image.

[0019]

In FIG. 1, when an image is fetched from a television camera in an image input processing step 11, a difference processing step 12 calculates a difference between a luminance value of the fetched input image and a previously created background image. Is calculated. Next, in the sign information storage processing step 13, the sign of the luminance difference value is stored. Then, in the judgment processing step 14, the processing branches according to this sign, and either the first binarization processing step 15 or the second binarization processing step 16 is executed according to the positive or negative sign. You.

These first and second binarization processing steps 1
5 and 16 have different threshold values. As a result, accurate binarization processing can be obtained even when the luminance value change is positive and negative with respect to the background image and the characteristics are different.
7 becomes possible.

That is, when the change of the input image with respect to the background image becomes positive due to the image becoming brighter, for example, from cloudy to clear or illuminated by a light, the change is, for example, Shadows on objects,
When the image becomes darker and becomes negative,
The degree of change is different, and simply binarizing the absolute value of the difference as in the related art cannot cope with this degree of change.

However, in the method shown in FIG. 1, the threshold value is changed depending on the direction of change, that is, whether the threshold value is positive or negative. Therefore, it is possible to flexibly cope with the degree of the change amount and to perform accurate detection.

In FIG. 2, a processing step 24 for calculating the similarity between the local luminance distribution shapes of the input image and the background image is added to the processing of FIG. 1, and a first binarizing processing step 26 and a second binarizing processing step are performed. Since the similarity is included in the binarization condition in the binarization processing step 27, the degree of change can be flexibly dealt with, and more accurate detection can be performed.

Here, the similarity represents the correlation between the shape of the local luminance distribution in the input image and the background image, and in FIG. Even if the input image becomes dark as a whole as shown by 102 due to the change in illuminance of
Since the patterns such as the white line and the trajectory in 1 remain as they are, a local region of the background image 101, for example, the region 104
And the local area 105 in the input image 102 have a high degree of similarity.

On the other hand, as shown by 103 in the input image,
In the area where the object (vehicle) 106 exists, the object 1
06 blocks the background, no patterns such as white lines and trajectories remain in the background image.
And the region 106 have a low similarity.

Therefore, in FIG. 3, following the binarization processing step 35, an area evaluation step 36 based on the similarity is performed.
Is provided, it is possible to separate two regions having different properties, such as a region where the luminance of the background image has changed, and a region where the object is present.

Next, in FIG. 4, in the processing of FIG. 2, when the sign becomes positive, a judgment processing step 47 for evaluating the similarity of the local luminance distribution shape between the input image and the background image is added. As a result, as shown in FIG. 9C, the area 94 brightened by light from a light or the like.
Is present, the object detection 49 is obtained separately from the object, so that the object can be detected more accurately.

That is, as shown by an area 94 in FIG. 9 (c), the area which has become brighter due to a light or the like has a higher luminance value but the road surface pattern remains.
Similarity increases as in the case of the local regions 104 and 105 in FIG. Therefore, by using this similarity, the region 94 brightened by the light can be reliably distinguished from the object.

Next, in FIG. 5, when the sign becomes negative in the processing of FIG. 2, a judgment processing step 57 for evaluating the similarity of the local luminance distribution shape between the input image and the background image is added. Thus, as in the case of FIG. 9D, in the case where the region 96 darkened by the shadow exists, the object detection 59 can be obtained by identifying the region 96, so that the object can be detected more accurately.

That is, the area darkened by the shadow is
As shown by an area 96 in FIG. 9D, although the luminance value is low, the road surface pattern remains but the similarity is high as in the case of the local areas 104 and 105 in FIG.

Therefore, by using this similarity,
The object 96 can be reliably distinguished from the region 96 darkened by the shadow.

In FIG. 6, by adding an evaluation processing step 67 of the similarity between the local luminance distribution shape of the input image and the background image, it is possible to detect a region where the temperature is locally increased.

That is, in a region where the temperature is locally increased, a difference in luminance distribution is generated between the region and the region where the temperature is not increased, so that the similarity of the luminance distribution shape is low. For example, when applied to pyroclastic flow detection, the portion where the lava dome was formed by the pyroclastic flow locally sharply rises in temperature, resulting in lower similarity, but on the other hand, in clouds heated by the pyroclastic flow, The similarity is high because the temperature rises almost uniformly throughout.

Therefore, by providing the similarity evaluation processing step 67, it is possible to separate the heated cloud portion and the pyroclastic flow portion.

In FIG. 7, by adding an evaluation processing step 77 of the similarity between the local luminance distribution shape of the input image and the background image, it is possible to detect a region where the temperature has dropped locally.

That is, in a region where the temperature has locally decreased, a difference in luminance distribution occurs between the region and the region where the temperature has not decreased, so that the similarity of the luminance distribution shape is low. For example, when applied to pyroclastic flow detection, where the lava dome collapses, the heat source falls down, the temperature drops rapidly, and the similarity decreases,
On the other hand, in a cloud or the like heated by a pyroclastic flow, even if the heating by the heat source is reduced, the whole is almost uniformly cooled and the temperature is lowered, so that the similarity is high.

Accordingly, these can be separated and detected by the similarity evaluation processing step 77.

[0038]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an object detection method according to the present invention will be described.
This will be described in detail with reference to the illustrated embodiment.

In the embodiment described below, the present invention is applied to a video surveillance device for detecting a pedestrian entering a railroad crossing or an object such as a car, and a television camera is provided at a position where the entire railroad crossing can be monitored. Is installed, and an image signal obtained thereby is processed to detect an object.

FIG. 12 shows an example of a video monitoring apparatus to which an embodiment of the present invention is applied.
The image signal obtained by imaging the monitoring area (monitoring field) at 21 is
The data is input to the data bus 122 via the input I / F 123 and stored in the image memory 124.

On the other hand, the CPU 126 analyzes the image in the work memory 125 according to the program stored in the program memory 127, executes the object detection processing, and, depending on the result, outputs the alarm lamp 1 via the output I / F 128.
30 is turned on, and an image is displayed on the monitor 131 via the image output I / F 129.

FIG. 13 shows an embodiment of the object detection process by the CPU 126. The process according to this embodiment is a combination of the processes of FIGS. 4 and 5 and a background image updating process added thereto. is there.

When the processing in FIG. 13 is started, first, in an image input processing step 141, an image signal captured by the camera 121 is converted into image data f (x,
y). Next, in determination step 143, it is determined whether a new background image has been created. The generation of the background image is performed by the CPU 126 executing the processing shown in FIG. 14 in parallel. This FIG.
First, in processing step 161, the input image is captured and stored at an interval of 0.2 seconds for 20 frames.
(Remember.

Next, in processing step 162, for each of the pixels of the stored 20-frame image,
Calculate the median according to equation (1) and calculate the background image data r
Create (x, y).

[0045]

(Equation 1)

Here, med {} represents a median. In the case of equation (1), the same pixels are arranged in order of luminance value in an image of 20 frames, and the tenth value (intermediate value) is obtained. ) To extract the background image data r
(x, y), whereby a new background image 163 is obtained. Therefore, the background image 1
63 = image data r (x, y)

Returning to the processing of FIG. 13, when the creation of the background image has been completed, in the processing step 143, the background image so far is updated to the newly created background image, and then the next processing is performed. The process proceeds to step 144, but if it has not been completed, the process directly proceeds to the difference processing step 144, where the input image is divided into blocks of 2 × 2 pixels, and the entire image is divided into 160 × 120 block areas f
_{From n} (i, j), the average luminance Af of each block area is determined. Here, n is the number of the block, and i and j represent the positions of the pixels in the block.

Similarly, for the background image data r (x, y), 160 × 120 block areas r _n (i, i,
j), the average luminance Ar of each block area is calculated, and the difference value ε _n is calculated by the equation (2).

[0049]

(Equation 2)

[0050] At the storage processing step 145 subsequent code information, the code information S _n calculated in (3), and stores it.

[0051]

(Equation 3)

In the similarity calculation processing step 146, the similarity γ _n is calculated for each block by the equation (4).
Here, D _n is a neighborhood area of the n-th block, and [] represents the number of elements.

[0053]

(Equation 4)

Next, the determination processing step 147, according to the sign information S _n, the first binarization processing step 148
Alternatively, the flow branches to the second binarization processing step 152.

These processes are binarization processes using a threshold value.
First, the threshold value when the sign is positive, that is, processing step 1
The threshold value th ^{+ at} 48 is obtained by Expression (5). Here, K ⁺ is an adjustable parameter, and R ⁺ is a region where the sign is positive and no object exists.

[0056]

(Equation 5)

Where K ⁺ is an adjustable parameter,
R ⁺ is a region where the sign is positive and no object exists.

The threshold value when the sign is negative, that is,
Similarly, the threshold value th ~ in the processing step 152 is obtained by Expression (6).

[0059]

(Equation 6)

Where K ~ is an adjustable parameter,
R ~ is a region where the sign is negative and no object exists.

Similarity determination processing steps 149 and 15
In 3, the parameters T ⁺ and TＴ that can be adjusted for each are set as thresholds, and the similarity γ _n is determined. That is, first in the determination processing step 149, the similarity gamma _n is compared with a threshold T ^+, in a case where it exceeds the threshold value T ^+, in FIG. 9 (c), shown as region 94 that is reflected light Such an area 150 brightened by the light is ignored as being detected, and is classified as the object detection 151 only when the area becomes equal to or less than the threshold value T ⁺ .

Similarly, in the judgment processing step 153, the similarity γ _n is compared with a threshold value T ~, and if it exceeds the threshold value T ~, it is shown as a shadow area 96 in FIG. Such an area 154 darkened by a shadow is detected and ignored, and is classified as the object detection 155 only when the area 154 becomes equal to or less than the threshold value T ~.

Then, the CPU 126 detects the object 15
1 is obtained, and when the object detection 151 is obtained, the alarm lamp 1 is output via the output I / F 128.
By turning on 30 or displaying an image on the monitor 131 via the image output I / F 129, it is notified that an object has been detected.

Therefore, according to this embodiment, it is always possible to reliably detect only the object from the field of view.
For example, in FIG. 15, the input image 171 shows a case where there is a vehicle 171A stopped outside the railroad crossing at dusk and the shadow 171B thereof extends into the railroad crossing. If the image 171 is determined by the conventional method , as shown in the image 173, it is recognized that the object 173A that has entered the railroad crossing exists.

On the other hand, according to the embodiment of FIG. 13, on the one hand, the background image 17
2 is obtained and updated to a new background image about every 4 seconds. Next, in this case, since the shadow 171B, the sign information Sn is negative, so the process proceeds to the determination processing step 153, where the similarity between the background image 172 and the input image 171 is determined.

In the area where the vehicle 171A exists, since a part of the background is hidden, a high similarity cannot be obtained. Therefore, the area is classified as the object 154 in FIG. 13, and this area actually exists. In the area of the shadow 171B, the degree of similarity is high because the background is not hidden, and as a result, the area is classified as the area 154 in FIG.

Therefore, according to this embodiment, the image 174
As shown in FIG. 1, the vehicle 174A and the shadow 174B
3 can be clearly classified into the object 155 and the area 154, and it is possible to reliably detect only the object that has entered the railroad crossing,
In this case, it can be determined that there is no intruding object in the railroad crossing despite the presence of the shadow 174B.

Next, FIG. 16 shows a case where the input image 181 shows that the light-irradiated portion 181B of the motorcycle 181A before entering the railroad crossing has entered the road surface inside the railroad crossing. If the input image 181 is determined by the conventional method , as shown in the image 183, it is recognized that the object 183A that has also entered the railroad crossing exists.

On the other hand, according to the embodiment of FIG. 13, first, on the other hand, the background image 17
2 is obtained and updated to a new background image about every 4 seconds. Next, in this case, since the portion 181B is a light irradiation portion, the code information Sn becomes positive this time, and the process proceeds to the determination processing step 149 where the background image 172 is obtained.
And the similarity of the input image 181 are determined.

In the area where the motorcycle 181A is present, since a part of the background is hidden and a high similarity cannot be obtained, the area is classified as the object 151 in FIG. Can be identified as the presence of an object that has entered the area. On the other hand, in the area of the light-irradiated portion 181B, the similarity is high because the background is not hidden, and as a result, as the area 150 in FIG. It will be classified.

Therefore, according to this embodiment, the image 184
As shown in FIG. 13, the motorcycle 174A and its light-irradiated portion 184B can be clearly classified into the object 151 and the area 150 in FIG. 13, and it is possible to reliably detect only the object that has entered the railroad crossing. Also in this case, it can be determined that there is no intruding object in the railroad crossing despite the presence of the light irradiation portion 184B.

Next, another embodiment of the present invention will be described. In this embodiment, the present invention is applied to monitoring of a pyroclastic flow at an active volcano, and an infrared television camera is used as the camera 121 in the video monitoring device shown in FIG.
(Infrared imaging device) is used to obtain an input image in which the luminance is high in a portion where the temperature of the subject is high, and then the CPU 126 executes the processing shown in FIG. is there.

The processing steps 191 to 198 in FIG. 17 and the processing step 201 are, as shown in parentheses, respectively, the processing step 1 in FIG.
41 to processing step 148, and processing step 1
Using the same routine as 52, the process branches to one of the similarity determination processing step 199 and the similarity determination processing step 202 in accordance with the luminance in the area of the input image and the background image, that is, the sign of the temperature difference. Using the adjustable parameters T ⁺ , T ~ as thresholds for determination, and when the similarity is equal to or less than the threshold, the object 200 whose temperature is rising,
The object is classified into the descending object 203.

The operation of this embodiment will be described with reference to FIG. 18. FIG. 18 shows an image at the moment when the pyroclastic flow occurs as an input image 211, and the background image is 212.
It is assumed to be as shown in FIG.

According to this embodiment, an image 213 is obtained as a processing result, and a portion where the pyroclastic flow is present and the temperature is high and a portion where the pyroclastic flow flows and the temperature is lowered are identified. , You can check.

That is, in a region where the temperature is locally increased, a difference in luminance distribution occurs between the region and the region where the temperature has not increased, and the similarity of the luminance distribution shape is low. For this reason, the temperature of the lava dome formed by the pyroclastic flow is similar to that of the local area where the temperature rises rapidly and the similarity is low.On the other hand, the temperature is almost uniform throughout the cloud heated by the pyroclastic flow. , The similarity increases.

Therefore, the similarity evaluation processing step 199
Is provided, the pyroclastic flow portion can be clearly separated as the region 200 where the temperature has increased, even if there is a portion due to a heated cloud or the like.

On the other hand, in a region where the temperature has locally decreased, a difference in luminance distribution occurs between the region and the region where the temperature has not decreased, so that the similarity of the luminance distribution shape is low. As a result, the heat source falls down at the part where the lava dome has collapsed, and its temperature drops rapidly, resulting in a low degree of similarity.On the other hand, heat sources such as clouds heated by pyroclastic flows Even if the heating due to is reduced, the overall degree of cooling is substantially uniform and the temperature decreases, so that the similarity increases.

Therefore, the similarity evaluation processing step 202
Is provided, it is possible to clearly separate and detect the portion where the pyroclastic flow has flowed off as the region 203 where the temperature has decreased, even if there is a portion due to a cloud or the like whose temperature has decreased.

Note that the present invention is not limited to the above application examples,
It goes without saying that the present invention is also applicable to a monitoring system for intruders into a restricted area such as plant equipment.

[0081]

According to the present invention, the sign of the difference value is used for the problem of erroneous detection due to a change in the background due to a change in illuminance. Since the detection of the object can be accurately obtained by reducing the risk of receiving the pattern, and the preservation of the pattern is determined using the similarity of the luminance distribution shape, the illuminance change regardless of the presence of the object Object can always be reliably detected even when there is a region due to.

Therefore, according to the present invention, even when extremely high reliability is required such as at a railroad crossing, the present invention can be applied with confidence, and the applicable range of the video monitoring device can be sufficiently widened.

[Brief description of the drawings]

FIG. 1 is a flowchart for explaining the operation principle of the first invention.

FIG. 2 is a flowchart for explaining the operation principle of the second invention.

FIG. 3 is a flowchart for explaining the operation principle of the third invention.

FIG. 4 is a flowchart for explaining the operation principle of the fourth invention.

FIG. 5 is a flowchart for explaining the operation principle of the fifth invention.

FIG. 6 is a flowchart for explaining the operation principle of the sixth invention.

FIG. 7 is a flowchart for explaining the operation principle of the seventh invention.

FIG. 8 is a flowchart for explaining the operation principle of the conventional technique.

FIG. 9 is an explanatory diagram showing a problem of the related art.

FIG. 10 is an explanatory diagram of similarity in the present invention.

FIG. 11 is an explanatory diagram of object detection using a temporal difference according to the related art.

FIG. 12 is a configuration diagram illustrating an example of a video monitoring device to which an embodiment of the present invention has been applied.

FIG. 13 is a flowchart showing an embodiment in which the present invention is applied to detection of an approaching object at a railroad crossing.

FIG. 14 is a flowchart illustrating background image creation processing according to an embodiment of the present invention.

FIG. 15 is an explanatory diagram illustrating an example of an object detection operation according to an embodiment of the present invention.

FIG. 16 is an explanatory diagram illustrating an example of an object detection operation according to an embodiment of the present invention.

FIG. 17 is a flowchart showing an embodiment when the present invention is applied to pyroclastic flow generation monitoring.

FIG. 18 is an explanatory diagram showing an example of a detection operation according to an embodiment when the present invention is applied to pyroclastic flow generation monitoring.

[Explanation of symbols]

 121 Television camera (video input device) 122 Data bus 123 Input I / F 124 Image memory 125 Work memory 126 CPU 127 Program memory 128 Output I / F 129 Image output I / F 130 Warning lamp 131 Warning display monitor

Claims (13)

(57) [Claims] 1. A object detecting method for detecting an object entering the field of view of the imaging device, calculating an image input step, a difference value between the luminance values of the input image and the background image, the positive of the difference value A first binarization step of binarizing a difference value with respect to the sign of the second image ; a second binarization step of binarizing a difference value of the difference value with respect to a negative sign of the difference value; Step to calculate similarity with background image
And a sign is provided according to the sign of the difference value and the value of the similarity.
Object detection method characterized by distinguishing and detecting objects
Law . 2. An object detection method for detecting, by a computer, an object entering a field of view using an input image of a camera, comprising: an image input step; calculating a difference, and storing the code information of positive and negative of the difference value, and determining the similarity of the local luminance value distribution shape of the input image and the background image, depending on the sign of the difference value Branching, and performing threshold determination by adding similarity information to the difference value information;
An object detection method comprising the steps of:
Law . 3. The method according to claim 2 , further comprising the step of determining a region having a high similarity and a region having a low similarity of the local luminance value distribution shape of the input image and the background image as different regions. Object detection characterized by the following:
How . 4. The invention according to claim 2 , wherein an area in which the sign information of the difference value is positive and in which the similarity of the local luminance value distribution shape between the input image and the background image is high is included in the visual field. Object detection method characterized by adding a step of determining as an area brightened by a light
Law . 5. The method according to claim 2 , wherein, in the area where the sign information of the difference value is negative, an area in which the similarity of the local luminance value distribution shape between the input image and the background image is high , Object detection characterized by adding a step of determining as an area darkened by a shadow in a visual field
How . 6. The image input device according to claim 2 , wherein the image input device is an infrared camera, and a local luminance value distribution shape of the input image and the background image in a region where the sign information of the difference value is positive. Object detection characterized by adding a step of determining a region having a low similarity as a region where the temperature has risen significantly due to the pyroclastic flow.
How . 7. An image input device according to claim 2 , wherein the image input device is an infrared camera, and a local luminance value distribution shape of the input image and the background image is determined in a region where the sign information of the difference value is negative. An object detection method characterized by adding a step of determining a region having a low similarity as a region where the pyroclastic flow has dropped and the temperature has dropped significantly locally. 8. An image signal which is sequentially taken in frame units from a predetermined field of view by a predetermined image pickup means,
An object that enters the predetermined field of view based on image data obtained by comparing a past image signal and a latest image signal for each region including a plurality of pixels and binarizing a difference between their luminance values. in the object detection method in a video surveillance apparatus of a system for detecting the capture image signals of a predetermined number of frames up to the latest image signals being taken, a background image signal created to create a background image signals by their median calculation Means, comparing the background image signal and the latest image signal for each region composed of a plurality of pixels, a code storage means for detecting and storing the difference between the brightness values, including the positive and negative signs thereof, Sign determining means for determining the sign of the difference between the luminance values; first and second binarizing means having different threshold values; and a region comprising a plurality of pixels for the background image signal and the current image signal. And a similarity detecting means for comparing each of the luminance value distribution shapes and detecting a similarity between the luminance value distribution shapes, and detecting the object using the first binarizing means when the determination result by the sign determination means is positive. When negative, an object is detected using the second binarization unit, and according to the similarity detected by the similarity detection unit,
An object detection method for a video surveillance device, wherein the authenticity of an object detected by one of the first binarization unit and the second binarization unit is determined. 9. An image signal which is sequentially taken in a frame unit from a predetermined field of view by a predetermined imaging means,
An object that enters the predetermined field of view based on image data obtained by comparing a past image signal and a latest image signal for each region including a plurality of pixels, and binarizing a difference between their luminance values. in the object detection method in a video surveillance apparatus of a system for detecting the capture image signals of a predetermined number of frames up to the latest image signals being taken, a background image signal created to create a background image signals by their median calculation Means, comparing the background image signal and the latest image signal for each region composed of a plurality of pixels, a code storage means for detecting and storing the difference between the luminance values, including the positive and negative signs thereof, Sign determining means for determining the sign of the difference between the luminance values; first and second binarizing means having different threshold values; and a region comprising a plurality of pixels for the background image signal and the current image signal. And a similarity detecting means for comparing the similarity of each of the brightness value distribution shapes, and detecting the object using the first binarizing means when the determination result by the sign determination means is positive. And when negative, an object is detected using the second binarization means, and according to the similarity detected by the similarity detection means,
An object detection method for a video surveillance device, wherein an area detected by one of the first binarization unit and the second binarization unit is detected separately. 10. The invention according to claim 9 , wherein when the sign of the difference value is determined to be positive and an area in which the similarity is determined to exceed a predetermined value is detected, the area is determined to be an object. A method for detecting an object in a video surveillance device, wherein the method is configured to determine that the detection is not a detection based on a local increase in illuminance in the visual field, but a detection based on presence. 11. The invention according to claim 9 , wherein, when the sign of the difference value is determined to be negative and an area in which the similarity is determined to exceed a predetermined value is detected, the area is determined to be an object. An object detection method in a video surveillance device, which is configured to determine that detection is not due to local illuminance reduction but to detection due to presence. 12. The invention according to claim 9 , wherein an infrared image pickup device is used as said image pickup means, and an area in which the sign of said difference value is determined to be positive and said similarity is determined to be not more than a predetermined value is detected. An object detection method in a video surveillance device, which is configured to determine that there is a locally high temperature portion in the field of view when the image is viewed. 13. The invention according to claim 9 , wherein an infrared imaging device is used as said imaging means, and a region in which the sign of said difference value is determined to be negative and said similarity is determined to be equal to or less than a predetermined value is detected. An object detection method in a video surveillance device, which is configured to determine that there is a locally low temperature portion in the field of view when the temperature of the image monitoring device is low.