JPH0981733A - Moving body detecting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately detect the movement state of a moving body which has various luminance levels even under a condition wherein ambient lightness changes by providing a difference quantity calculating means, 1st and 2nd multiplying means, an image data update means, a multi-valued data converting means, a multiplier switching means, and a monitor display means. SOLUTION: A subtracter 5 performs subtraction in pixel units between 1st image data read out of a 1st image memory 1 by a 1st data processing means 11 and 2nd image data read out of a 2nd image memory by a 2nd data processing means 12, and the calculation result is written in pixel units in a 3rd image memory 3 by a 3rd data processing means 13. A binarization, comparison, and decision unit 6 compares the 3rd image data read out of the 3rd image memory 3 with a specific set threshold and make a decision. A switching adder 7 performs multiplication processing for the 1st and 2nd image data at the same time and adds the multiplication results to each other.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving body detecting device for detecting a moving body and a moving direction and a moving distance of the moving body.

[0002]

2. Description of the Related Art In recent years, ordinary NTSC (National Telev
The development and commercialization of a device for inputting a video signal and performing signal processing to detect a motion of a moving body, recognize a shape of the moving body, and the like are actively performed. For example, for the purpose of intrusion monitoring, attach a camera in one corner of the building entrance and exit lobby,
There is a device that detects an intruder by analyzing an image captured by the camera and tracks the intruder. In such a device, generally, a background image to be referred to is created in advance, and a moving object is detected by detecting a difference amount between the background image and the captured image. There is.

Regarding how to make a background image in that case,
There are the following two methods. The first method basically considers that the background image does not change, and refers to the image captured in the initial state where there is no moving body such as a person as the background image. On the other hand, the second method considers that the background image also changes to some extent, and updates the background image to be referenced by adding a certain amount, which is a percentage of the repeatedly captured image, to the background image one frame before. I am supposed to do it.

In a surveillance system for monitoring an intruder over time, there are many situations where the brightness of the spot changes moment by day, and under such circumstances, The background image to be referenced naturally changes. Therefore, it can be said that the second method can more accurately detect the moving body under a monitoring situation in which the background image changes from moment to moment.

[0005]

However, when a background image is created by the second method, as described above, a certain amount of a percentage of the images repeatedly captured is updated by adding it to the background image one frame before. Therefore, if the brightness level of the moving object has a level (extremely bright or dark) farther than that of the background image, the background image to be updated is significantly changed. I will let you.

For example, when a high-luminance object emitting a flashlight comes in, the image captured at that time is updated as a background image for the image after the intruder has moved. As a result, the relevant portion of the updated background image is an image having a high brightness level that is significantly different from the background images up to that time when there was no intruder. As a result, when the background image is used as the next reference image, it is erroneously detected as if a new intruder were present at the position before the movement of the intruder. Become. Further, in such a case, it is naturally impossible to grasp the exact position of the intruder, and it is recognized that the intruder is near the intermediate position between the position before the movement and the position after the movement. There is also a risk that the intruder will not be able to detect an accurate position after the movement of the intruder.

The present invention has been made in view of the above circumstances, and it is possible to accurately detect the moving state of a moving body having various brightness levels even in a situation where the brightness of the surroundings changes. It is an object of the present invention to provide a possible moving body detection device.

[0008]

In order to achieve the above-mentioned object, the invention described in claim 1 uses a sampled first image data and a background image referred to for detecting a moving object. A difference amount calculating means for calculating a difference amount from the stored second image data one frame before, and the first
First multiplication means for multiplying the image data of 1 by a first multiplier, second multiplication means for multiplying the second image data by a complement of 1 of the first multiplier, Adding the first product obtained by the first multiplication means and the second product obtained by the second multiplication means,
An image data updating unit that updates the second image data by using the added value and a difference amount calculated by the difference amount calculating unit are subjected to threshold value determination, and the determination result is converted into multivalued data including binary data. According to the multivalued data conversion means and the multivalued data obtained by the multivalued data conversion means, the first multiplier in the multiplication by the first multiplication means is switched to the second multiplier, and the second multiplication is performed. Multiplier switching means for switching the multiplier by means to the complement of 1 of the second multiplier, and monitor display means for displaying the moving state of the object on the monitor based on the multivalued data obtained by the multivalued data conversion means. It is characterized by having.

According to a second aspect of the present invention, the difference amount calculating means in the moving object detecting apparatus according to the first aspect corresponds to pixels corresponding to each other between the first and second image data sampled in frame units. The feature is that the difference amount is calculated for each and the absolute value is obtained. Further, in the invention described in claim 3, the first multiplier multiplied by the first multiplication means in the moving object detecting device according to claim 1 or 2 is expected in a state where the moving object is not detected. A second multiplier which is a predetermined value set in consideration of the degree of influence of the second image data on the first image data under a change over time, and which is multiplied by the first multiplication means. Considers the degree of influence of the second image data on the first image data when a moving object having an estimated predetermined brightness level is detected under an expected change over time. It is characterized in that it is a predetermined value set by.

Further, according to the invention described in claim 4, the first multiplier in the moving object detecting device according to claim 3 is set to a numerical value of 0, which is set when the second image data does not change with time. A positive numerical value smaller than 5, a second multiplier is a positive numerical value smaller than the first multiplier, and a ratio with the first multiplier is the multi-valued data conversion. It is characterized in that the means is a predetermined number of positive numerical values determined according to a threshold level used for threshold determination of the difference amount.

According to a fifth aspect of the present invention, in the case where the image data updating means in the moving object detecting apparatus according to the first or fourth aspect further completes the updating of the second image data. It is characterized in that it is provided with an image data acquisition instruction means for instructing acquisition of the next image data. In the invention according to claim 6, the threshold value used by the multi-valued data conversion means in the moving object detection device according to claim 1 or 5 in the threshold value determination of the difference amount is the second image data. It is characterized in that it is set as a difference amount value of several steps necessary to detect a moving body exhibiting several levels of brightness levels estimated corresponding to the change with time.

According to a seventh aspect of the present invention, in the multi-valued data obtained by the multi-valued data converting means, the multi-valued data is moved by the multiplier switching means in the moving object detection apparatus according to the first or sixth aspect. When the data corresponds to the case where the body is not detected, the first multiplier is switched to be used, and the multivalued data obtained by the multivalued data conversion means is converted into the multivalued data corresponding to the brightness level of the moving body. When it is value data, it is characterized in that it is switched so that the corresponding second multiplier is used according to the multivalued data.

According to the above configuration, the first image data and the first image data are based on the information on the difference amount between the first image data which is the sampling image data and the second image data which is the background image data. The mixing ratio with the image data of 2 can be changed. In that case, when the difference amount exceeds the set predetermined threshold value, the mixing ratio of the first image data is increased, and when the difference amount is less than or equal to the predetermined threshold value, the mixing ratio of the first image data is decreased. .

As a result, even when an intruder having a significantly different brightness level from the background image is captured, the adverse effect of the image of the intruder on the background image can be suppressed as much as possible.
Further, even if the intruder further moves, there is no possibility of erroneously detecting the state before the move as a new intruder, and the position after the move of the intruder can be accurately grasped. .

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be specifically described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration example of a moving body detection device according to a first embodiment of the present invention. The moving object detecting device is provided with the first to fourth four image memories 1 to 4.

Here, the image memory (also referred to as a frame memory) is a large-capacity IC memory having one storage location for each pixel on the television screen, and images for one screen sent one after another. The data is written in real time, and the image data on the memory is read out in real time. The image memory capacity is
It is determined by the number of pixels forming one frame and how many bits each pixel is expressed. Here, the number of pixels forming one frame is 512 × 512 = 262,144, and in the first to third three image memories 1 to 3, one pixel is represented by 8 bits (that is, 2 ⁸ = In the fourth image memory 4, one pixel is represented by 2 bits. Therefore, the capacities of the first to third image memories 1 to 3 are 512 × 512 × 8, respectively.
= 2,097,152 bits, which is about 2.1 Mbits. The capacity of the fourth image memory 4 is 5
12 × 512 × 2 = 524,288 bits, which is about 5
It is 25K bits.

In order to increase the resolution, the number of pixels may be set to 512 × 1024, and one pixel is expressed by 24 bits in order to enhance the color expression power (that is, RG).
Each of B is represented by 8 bits). Next, the configuration of the moving body detection apparatus will be sequentially described according to the flow of signals processed by the moving body detection apparatus (indicated by an arrow in the drawing). It should be noted that each functional block in this moving object detection device operates in units of one frame of an input video signal.

The A / D converter 10 is supplied with an analog video signal of NTSC system imaged by a surveillance camera or the like (not shown), and the sampled video signal in pixel units is A / D converted to 1 The image data is converted into 8-bit pixel data. Then, the image data for one frame (hereinafter, referred to as first image data) is written in the first image memory 1 by the first data processing unit 11.

When the moving object detecting apparatus is started up, the first input image data for one frame is written in the second image memory 2 first, and the image for the second one frame is displayed. The data is written in the first image memory 1. In the subtracter 5, the first image data read from the first image memory 1 by the first data processing means 11 and the second data processing means 12
The pixel-by-pixel subtraction is performed with the second image data read from the second image memory 2, and the absolute value processing of the subtraction result is performed. Then, the calculation result (hereinafter, referred to as third image data) is written into the third image memory 3 by the third data processing unit 13 in pixel units.

In the binarized comparison / determination device 6, the third image data read out from the third image memory 3 by the third data processing means 13 is compared and determined with the set predetermined threshold value. When the result of the determination here is larger than the threshold value for which the third image data is set (that is, the difference amount is large, and the second image data of the first image data is used as the background image). (When it is determined that the influence is large)
Is smaller than the threshold value for which the third image data is set (that is, the difference amount is small, and the influence of the first image data on the second image data as the background image is In the normal case where it is determined that there is almost no, "0" is output to the switching adder 7 and the fourth data processing means 14, respectively.

Regarding the threshold value used in the threshold value determination of the difference amount, the movement having the estimated predetermined brightness level with respect to the average brightness level of the pixel unit determined in consideration of the change with time of the second image data. It is set as the difference amount required to detect the body. In the switching adder 7 (see FIG. 2 for the detailed configuration thereof), the first image data read from the first image memory 1 by the first data processing unit 11 and the second data The processing means 12 simultaneously performs the multiplication process on each of the second image data read from the second image memory 2 and adds the respective multiplication results.

In this case, the first image data is multiplied by the multiplier a (where 0≤a≤1), and the second image data is multiplied by "1-" which is the complement of 1 to a. a ”is multiplied as a multiplier. Also, the binarization comparison / determination unit
The combination of the two sets of multipliers a and b (however, a> b) to be used is switched according to the determination result of. The multiplier a and the multiplier b are set as follows. For the multiplier a, the degree of influence of the second image data on the next-ordered first image data is predicted by taking into consideration the change over time of the second image data expected in the state where no moving body is detected. Is set as the predetermined value determined by Specifically, in a state where the moving body is not detected, the brightness level of each pixel included in the first image data and the second image data does not change extremely. However, only if there is a slight change in brightness over time,
A change in brightness level will occur.

Therefore, in such a case, it is considered that the second image data has little influence on the first image data in the next order, so that the multiplier a has no or little change over time. Is set to 0.5, and if there is any change over time, it is set to a value smaller than 0.5 and close to 0.5. That is, when the multiplier a is set to 0.5, the complement to 1 is also 0.5, so that the influence of the second image data on the next first image data is 0. Become. Further, when considering a little influence, how close the set value is to 0.5 is set according to the degree of change with time so that the smaller the degree of change with time, the closer to 0.5. To do. However, specific numerical values will be decided depending on the situation.

Regarding the multiplier b, when a moving object having an estimated predetermined brightness level is detected, the second image data is calculated as follows when the expected change with time of the second image data is taken into consideration. It is set as a predetermined value determined by predicting the degree of influence exerted on the first image data in the order. Specifically, when a moving object is detected, the influence of the second image data on the next-order first image data may increase depending on the brightness level of the moving object. Further, in such a case, in order to ensure the accuracy of detection, it is necessary to consider the change over time of the second image data.

Therefore, the multiplier b to be set in such a case is determined by the ratio to the multiplier a.
In that case, it is determined according to the threshold value set by the binarization comparison / determination unit 6. That is, when the threshold value for the difference amount determination is set low, the detection sensitivity of the moving body is high,
Since it is assumed that the difference between the estimated brightness level of the moving object and the background brightness level is small, the degree of influence of the second image data on the next first image data is low. Conceivable. Therefore, for example, when a = 0.5, the ratio is set to 1/10 and the multiplier b is set to 0.05.

On the contrary, when the threshold value for the difference amount determination is set high, the detection sensitivity of the moving object is low, and the difference between the estimated brightness level of the moving object and the background brightness level is large. Since the state is assumed, it is considered that the degree of influence of the second image data on the next first image data is large. Therefore, for example, when a = 0.5, the ratio is set to 1/5, which is lower than 1/10, and the multiplier b is set to 0.01. Note that a specific multiplier b
Similarly to the multiplier a, the numerical value of is determined according to the situation.

Subsequently, the addition result of the switching adder 7 is the second addition result.
The data processing means 12 writes the second image data in the second image memory 2 and updates the previously written second image data. Furthermore, at the end of the update, that is,
At the time when the data writing to the second image memory 2 is completed, the second data processing means 12 instructs the first data processing means 11 to sample the next video signal.

On the other hand, the fourth data processing means 14 writes the judgment result of the binarization judging device 6 in the fourth image memory 4 as the fourth image data, and further reads it out. The image is displayed on the image monitor 8. Here, general image processing software for displaying the movement status of the target pixel group on the screen by using the binarized data which is the determination result of the binarization determiner 6 is used. Here, the user can accurately grasp the movement status of the moving body such as an intruder by visually observing the video displayed on the image monitor 8.

FIG. 2 is a block diagram showing the internal structure of the switching adder 7 shown in FIG. Reference numerals 21 to 24 are amplifiers for amplifying the respective input signals. As shown in the figure, the output from the first image memory 1 is input to the amplifier 21 and the amplifier 22 at the same time, and the output from the second image memory 2 is input to the amplifier 23 and the amplifier 2 at the same time.
4 is input.

Here, 21 is an a-times amplifier (where 0 ≦ a
≦ 1), 22 is a b-fold amplifier (where 0 ≦ b ≦ 1), 23 is a (1-a) -fold amplifier, and 24 is (1-
b) It is a double amplifier. Note that the multipliers a and b are a> b, and are set as predetermined values (for example, a = 0.5 and b = 0.1), respectively, as described above.

Reference numeral 25 is a switch of 4 inputs and 2 outputs, which is used when the output signal from the binarized comparison / determination unit 6 is "0" (that is,
In the normal case where the difference amount is small and it is determined that the first image data has almost no influence on the second image data serving as the background image), the simultaneous output of the amplifier 21 and the amplifier 23 is selected. When the output signal from the binarization comparison / determination unit 6 is “1” (that is, the difference amount is large, the influence of the first image data on the second image data as the background image is large. If so, the simultaneous output of the amplifier 22 and the amplifier 24 is selected.
Further, 26 is an adder for adding and mixing the two outputs from the switch 25, and the addition result is as follows.
As described above, the data is output to the second data processing means 12.

By selecting each amplifier in this way, the multiplier a and "1-" for the first image data are selected.
The combination of "a" and the combination of the multiplier b and "1-b" for the second image data is selected. As a result, even when a moving object with a large brightness level difference is detected and further moved, the previous background image uses a small multiplier b for the first image data of the next sampled order. Since it is updated by multiplication, it is possible to suppress the adverse effect of the moving body on the background image as much as possible.

FIG. 3 is a flow chart showing the flow of processing operations in the moving body detecting apparatus shown in FIG. First, an initial process for starting up the moving body detection device is performed (S1). In this process, the first data processing means 11 converts the one frame of the image data consisting of 8 bits per pixel converted by the A / D converter 10 into the first background image to be referred to by the first data processing means 11. Image data is second
Is written in the image memory 2.

Subsequent processing from steps S2 to S12 is processing performed when the moving body is detected. First, the first data processing means 11 takes in one frame of image data and writes it in the first image memory 1 (S2). Next, the first data processing means 11 and the second data processing means 12 subtract the data of the same address in the first image memory 1 and the second image memory 2 from the subtracter 5
The data is sequentially read out, and the absolute value of the difference between the read data of the first image memory 1 (first image data) and the read data of the second image memory 2 (second image data) is calculated. Then, the calculation result is written in the third image memory 3 by the third data processing means 13 (S3).

Subsequently, the third data processing means 13 converts the data in the third image memory 3 into a binarized comparison / determination unit 6.
Are sequentially read out and binarized by comparing with a preset value t ((S4). Here, for the set value t, for example,
Assuming that t = 64, “1” is output when the data to be sequentially read exceeds t, and “0” is output when the data is t or less. What it means is
When the binarized data is "0", it is considered that there is no change between the background image and the sampling image for the pixel of interest, and when the binarized data is "1", it is changed between them. Is considered to have occurred (however, in this case, the image captured as the background is sufficiently bright and the quantized data is 64 or more).

Next, it is judged what the binarized data is (S5), and the judgment result is written in the fourth image memory 4 by the fourth data processing means 14 (S6). Then, the data content of the fourth image memory 4 is processed into a moving body detection image and displayed on the image monitor 8 (S7). Here, the user can confirm the moving status of the intruder, the position and the form thereof through the image displayed there.

If the determination result in step S5 is "0", the switching adder 7 multiplies the data in the first image memory 1 by a times (for example, 0.5 times) and the second image memory. The data of 2 is multiplied by "1-a" and the product of them is added (S8). When the result of the determination in step S5 is "1", the switching adder 7 multiplies the data in the first image memory 1 by b times (for example, 0.1 times) and the second image memory 2 Is multiplied by "1-b" and the product of them is added (S9).

The meaning of this calculation processing is as follows. That is, as described above, when the binarized data is “0”, the sampling image has almost no change compared with the background image, and therefore the ratio of the sampling image added to the background image is increased, The background image is actively adapted to changes in the sampling image. By doing so, even when the input video changes slowly with time, the background image can be made to follow the change.

However, when the binarized data is "1", there is a noticeable change in the sampling image compared with the background image, and when an intruder occurs, the target pixel in the background image is set. A situation may occur in which the quantization value 64 is far exceeded. In such a case, the influence of the background image to be referred to is further increased by reducing the ratio of the sampling image added to the background image. By doing so, even if there is an intruder having a brightness level greatly different from the brightness level of the background image, it is possible to suppress the degree of influence on the background image as much as possible. It should be noted that the actually used multiplier value should be appropriately determined according to various environmental conditions for detecting a moving object.

FIG. 4 is a block diagram showing an example of the internal configuration of a multilevel comparison comparator / determinator and a multistage switching adder in a moving body detection apparatus according to a second embodiment of the present invention. here,
As the multi-valued comparison / determination device, a four-valued comparison / determination device 40 is shown, which is installed in place of the binarization comparison / determination device 6 in FIG. As shown in the figure, the four-valued comparison / determination unit 40 is composed of four comparators CP1 to CP4. In the comparator CP1, the moving body is not detected.
Judgment is made based on the threshold value, and data “00” is output. Here, the threshold value is set as an average value of the difference amount from the first image data, which is determined in consideration of the change over time of the second image data serving as the background image.

Further, in the comparator CP2, in consideration of the second image data which changes with time, the first state in which the moving body having the estimated predetermined luminance level can be detected is determined by the threshold value, and the data " 10 "is output. Similarly, the comparator CP3 determines the second detectable state of the moving body by a threshold value, outputs data “10”, and outputs the data to the comparator CP3.
In 4, the third detectable state of the moving body is determined by the threshold value, and the data “11” is output.

Output data from the four-valued comparison / determination unit 40 (that is, any one of 00, 01, 10 or 11)
Are simultaneously input to the first selector 42 and the second selector 43 in the multistage switching adder 41. In the first selector 42 and the second selector 43, the input 4-valued comparison / determination unit 4
The output data from 0 determines the next amplifier combination to use. A is an A times amplifier, B is a B times amplifier, C is a C times amplifier, D is a D times amplifier, "1-A" is a "1-A" times amplifier, and "1. −
"B" is a "1-B" times amplifier, and "1-C" is "1-
It is a C "times amplifier, and" 1-D "is a" 1-D "times amplifier, which is a multiplier of input data.

Here, the data "10" is a combination of A and "1-A" for the data "00" and the combination B and "1-B" is for the data "01". For the data "11" of the combination of C and "1-C"
, Two amplifiers with a combination of D and "1-D" are selected. In addition, the first selector 42
The output from the first image memory input to
The output from the second image memory, which is amplified by either the C or D amplifier and is also input to the first selector 42, is "1-A", "1-B", "1-C", Or "1
It is to be amplified by any of the amplifiers of "-D". Further, the outputs from the amplifiers of each combination (that is, the product which is the multiplication result) are added by the adder 44 and output to the second data processing means 12.

Regarding the set values (multiplication values) of the amplifiers A, B, C and D, for the same reason as explained in the explanation of FIG. 1, for example, when A = 0.5, B = 0.4, C = 0.3, and D = 0.2 are set so as to sequentially decrease from the numerical value 0.5. Specifically, the set threshold value to the threshold value is set to In consideration of,
It is decided according to the situation.

In the above description of the embodiment of the present invention, the binarized or multi-valued data stored in the third image memory 3 is used as the first image data (sampling image data). Second image data (background image data)
Although the mixing ratio with and is fixed, if the mixing ratio is variable and the optimum mixing ratio is set according to the situation, more accurate moving object detection can be performed.

[0046]

As described above, according to the present invention, even when an intruding object having a brightness level greatly different from that of the background image is detected, the ratio of the sampling image to be added to the background image is updated when the background image is updated. By making the size smaller than the case where the intruder is not detected, the adverse effect of the image of the intruder on the background image is suppressed as much as possible. As a result, even if the intruder further moves, the probability of erroneously detecting the state before moving as a new intruder is low, and the intruder or moving object can be detected with high accuracy. It becomes possible to do.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration example of a moving body detection device according to a first embodiment of the present invention.

2 is a block diagram showing an internal configuration of a switching adder 7 shown in FIG.

FIG. 3 is a flowchart showing a processing operation flow of the moving body detection device shown in FIG.

FIG. 4 is a block diagram showing an example of an internal configuration of a multilevel comparison comparator / determinator and a multistage switching adder in a moving body detection apparatus according to a second embodiment of the present invention.

[Explanation of symbols]

 1 1st image memory 2 2nd image memory 3 3rd image memory 4 4th image memory 5 Subtractor 6 Binarization comparison judgment device 7 Switching adder 8 Image monitor 10 A / D converter 11 1st Data processing means 12 2nd data processing means 13 3rd data processing means 14 4th data processing means 21 a times amplifier 22 b times amplifier 23 (1-a) times amplifier 24 (1-b) times amplifier 25 4-input 2-output switcher 26 adder 40 multi-valued comparison / determination device 41 multi-stage switching adder 42 first selector 43 second selector 44 adder

Claims (7)

[Claims] 1. A difference amount calculation for calculating a difference amount between sampled first image data and second image data one frame before stored as a background image referred to for detecting a moving object. Means, first multiplication means for multiplying the first image data by a first multiplier, and second means for multiplying the second image data by a complement of 1 of the first multiplier. And the second product obtained by the second multiplication means, and the second image data is obtained by using the addition value of the first product obtained by the first multiplication means and the second product obtained by the second multiplication means. An image data updating unit for updating the image data, a threshold value for the difference amount calculated by the difference amount calculating unit, and a multivalued data conversion unit for converting the judgment result into multivalued data including binary data, Value obtained by data conversion means According multivalued data, the first in the multiplication by said first multiplication means
Of the multi-valued data obtained by the multi-valued data conversion means A moving object detection device comprising: a monitor display unit for displaying the movement status of an object on a monitor based on the above. 2. The difference amount calculation means calculates a difference amount for each corresponding pixel between the first and second image data sampled in frame units, and converts the difference amount into an absolute value. The moving body detection device according to 1. 3. The first multiplier multiplied by the first multiplying unit is converted into the first image data by the second image data under a change over time expected in a state where no moving body is detected. The second multiplier, which is a predetermined value set in consideration of the degree of influence exerted, is multiplied by the first multiplication means and has a predetermined brightness level estimated under the expected change over time. The predetermined value set in consideration of the degree of influence of the second image data on the first image data when a moving object is detected. 2. The moving body detection device according to 2. 4. The first multiplier is a positive numerical value smaller than a numerical value 0.5 set when the second image data does not change with time, and the second multiplier is the first numerical value. A positive number smaller than a multiplier of 1 and a ratio to the first multiplier is a predetermined number determined according to a threshold level used by the multi-valued data conversion unit for threshold determination of a difference amount. The moving body detection device according to claim 3, wherein the moving body detection device has a positive numerical value. 5. The image data update means further comprises image data acquisition instruction means for instructing acquisition of the next image data when the update of the second image data is completed. The moving body detection device according to claim 1 or 4. 6. A threshold value used by the multi-valued data conversion means in the threshold value determination of the difference amount is a moving object exhibiting several levels of brightness levels estimated corresponding to the change with time of the second image data. The moving body detection device according to claim 1 or 5, wherein the difference amount value is set as a difference amount value of several stages necessary for detection. 7. The multiplier switching means uses the first multiplier when the multi-valued data obtained by the multi-valued data converter is data corresponding to a case where a moving body is not detected. When the multi-valued data obtained by the multi-valued data conversion means is multi-valued data according to the brightness level of the moving body, the corresponding second multiplier is used according to the multi-valued data. 2. The method according to claim 1, wherein
Alternatively, the moving body detection device according to claim 6.