JP7328778B2 - Image processing device and image processing program - Google Patents

Description

The present invention relates to an image processing apparatus and an image processing program for identifying detection targets from captured images.

Conventionally, an image of a predetermined monitoring area (referred to as a predetermined area) is photographed with a camera, and the photographed image is compared with a reference image that does not include a detection target to obtain a changed area, and the size of the changed area is obtained. There is known an image processing apparatus that determines the presence or absence of a detection target such as an intruder based on the following. In this image processing device, when an insect that exists in the vicinity of the camera is photographed, the size of the image becomes about the same as that of a person, and there is a possibility that the insect is erroneously detected as a person.

Therefore, in Patent Document 1, a lit image with proximity lighting that illuminates only the immediate vicinity of the camera is compared with an unlit image that does not light up the proximity lighting, and the difference in brightness between the lit image and the unlit image. Insects and humans are discriminated by judging insects when is large. In other words, an insect near the photographing device is illuminated by the proximity illumination, and thus appears bright in the lit image and dark in the unlit image, resulting in a large difference in brightness.

JP-A-2000-348266

However, when the predetermined area is bright due to sunlight or indoor lighting, the difference in brightness of the insect between the lit image and the unlit image becomes small. For this reason, the accuracy of discriminating between insects and humans may be low.

The present invention provides an image processing apparatus capable of improving the accuracy of identifying a detection target (eg, a person) and a non-detection target (eg, an insect) even when a predetermined area is relatively bright.

An image processing apparatus according to the present invention includes a photographing unit that obtains photographed images obtained by sequentially photographing a predetermined region, and a photographed image to be processed among the images, which is compared with a reference image photographed in the past from the photographed image, and an image processing unit that extracts a changed area representing a moving object from a captured image, and determines a detection target that exists at a position away from the imaging unit and a non-detection target that exists near the imaging unit based on the changed area; wherein the image processing unit obtains a dark area made up of pixels with a low luminance value and a bright area made up of pixels with a higher luminance value than the dark area in the changed area of the captured image, a feature amount calculating means for obtaining, as a feature amount, a ratio of the bright area to an image area in the vicinity of the dark area; and determination means for determining the changed region as the non-detection target.

Preferably, the feature amount calculation means sets a threshold value based on the luminance value of the captured image from which the changed region is extracted, and obtains the dark region and the bright region from the threshold value.

It is preferable that the feature amount calculation means regards an image area formed of pixels adjacent to the dark area among the image areas located in the vicinity of the dark area as the image area in the vicinity of the dark area.

Further, the image processing apparatus according to the present invention includes a photographing unit that obtains photographed images obtained by sequentially photographing a predetermined region, the photographed image to be processed among the photographed images, and a reference image photographed before the photographed image. A changed area representing a moving object is extracted from the captured image by comparison, and based on the extracted changed area, a detection target existing at a position away from the imaging unit and a non-detection target existing near the imaging unit are determined. and an illumination means for illuminating the vicinity of the photographing unit, wherein the image processing unit processes a dark area made up of pixels having a low luminance value in the changing area of the photographed image, and a dark area more than the dark area. A bright area composed of pixels having a high luminance value is obtained, and a ratio of the bright area to an image area in the vicinity of the dark area is obtained as a first feature value of the photographed image, and the photograph is taken with the lighting means turned on. feature quantity calculating means for obtaining, as a second feature quantity, a difference in luminance value of the changing region extracted based on the lighted image and the non-lighting image photographed without lighting the lighting means; determination means for determining whether the changed area is a non-detection target using the first feature amount and the second feature amount, and the determination means determines that the changed area is the non-detection target as the ratio is higher. Alternatively, the greater the difference between the luminance values, the easier it is to determine the changed region as the non-detection target.

The determining means determines that the higher the ratio, the smaller the difference in luminance value necessary for determining the non-detection target, or the greater the difference in luminance value, the greater the non-detection target determination. The required ratio should be low.

The determining means sets the ratio necessary for determining the non-detection target to a higher value as the predetermined area is brighter, or sets the luminance value necessary for determining the non-detection target as the predetermined area is darker. The difference should be a large value.

Further, an image processing program according to the present invention is an image processing program that causes an image processing device to distinguish between a detection target that exists at a position distant from an imaging unit and a non-detection target that exists near the imaging unit, causing an image processing device to compare a photographic image to be processed among photographic images obtained by sequentially photographing a predetermined region with a reference image that was previously photographed from the photographic image, and to extract a changed region representing a moving object from the photographic image ; A dark area composed of pixels having a low luminance value and a bright area composed of pixels having a luminance value higher than that of the dark area are obtained for the changed area of the photographed image, and occupying an image area in the vicinity of the dark area of the photographed image. A ratio of the bright area is obtained as a feature amount, and if the ratio obtained for the changed area using the feature amount is equal to or greater than a predetermined value, the changed area is determined as the non-detection target. do.

Further, an image processing program according to the present invention is an image processing program that causes an image processing device to distinguish between a detection target that exists at a position distant from an imaging unit and a non-detection target that exists near the imaging unit, causing an image processing device to compare a photographic image to be processed among photographic images obtained by sequentially photographing a predetermined region with a reference image that was previously photographed from the photographic image, and to extract a changed region representing a moving object from the photographic image; A dark area composed of pixels having a low luminance value and a bright area composed of pixels having a luminance value higher than that of the dark area are obtained for the changed area of the photographed image, and occupying an image area in the vicinity of the dark area of the photographed image. Obtaining the ratio of the bright region as a first feature quantity, and extracting based on a lit image captured with lighting means for illuminating the vicinity of the imaging unit and a non-lighting image captured without lighting the lighting means A difference in luminance value of the changed region obtained as a second feature amount is obtained, and using the first feature value and the second feature value obtained for the changed region, the higher the ratio, the more the changed region is the It is characterized by making it easier to determine that the changed region is a non-detected target, or making it easier to determine that the changed region is a non-detected target as the difference between the luminance values increases, thereby determining whether the changed region is a non-detected target. do.

According to the present invention, the non-detection target is determined by determining whether the change region is a non-detection target such as an insect using the ratio of the bright region occupying the periphery of the dark region in the image region corresponding to the change region. can be accurately identified.

1 is a schematic configuration diagram of a security system according to this embodiment; FIG. 1 is a schematic configuration diagram of an image processing apparatus; FIG. It is a figure which shows the example of a reference|standard image and a picked-up image, (a) shows a reference|standard image, (b) shows a picked-up image. FIG. 10 is a diagram showing an example of a binarized image showing changed areas of a captured image; FIG. 10 is an explanatory diagram showing a case where the ratio calculation method is applied to insects (non-detection targets); FIG. 10 is an explanatory diagram showing a case where the ratio calculation method is applied to a person (detection target); 10 is a flowchart of processing (abnormality detection processing) for identifying and detecting humans (detection targets) and insects (non-detection targets). 10 is a flowchart for determining whether or not an object is an insect based on two feature amounts;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described based on the drawings. It should be noted that the invention is not limited to the embodiments described herein.

"System configuration"
FIG. 1 is a schematic diagram of the configuration of a security system 10 according to this embodiment. The security system 10 includes a security device 14 installed in each property 12 to be monitored such as a store, an office, an apartment building, a warehouse, a house, etc., and a security center connected to each security device 14 via a communication network 16 such as a public telephone line. It comprises a device 18 and a user device 20 . Furthermore, the security system 10 detects a detection target (in this embodiment, a person: an intruder) from the monitored area within the monitored object 12 based on the photographed image of the monitored area (also referred to as a predetermined area) of the monitored object 12. It includes one or more image processing devices 22 for detection and a recording device 24 for recording captured images captured by the image processing devices 22 . The image processing device 22 and the recording device 24 are communicably connected to the security device 14 .

When the security device 14 receives an alarm signal from the image processing device 22 connected to itself via a local LAN or the like, the alarm signal and the identification signal of the security device 14 itself, or the property 12 to be monitored or an abnormality is detected. The identification signal of the image processing device 22 is transmitted to the security center device 18 . Therefore, the security device 14 has a communication interface for communicating with the image processing device 22, a communication interface for communicating with the security center device 18 and the user device 20, and a control unit for controlling them.

The security center device 18 is composed of a so-called computer, and has a communication interface for communicating with the security device 14 via the communication network 16, a display device such as a liquid crystal display, and a notification unit composed of a buzzer, an LED, and the like. . When the security center device 18 receives an alarm signal from the security device 14 via the communication network 16, the security center device 18 notifies the property 12 to be monitored where the security device 14 that sent the alarm signal and the content of the detected abnormality are sent to the reporting unit and Notify the observer through the display device.

The user device 20 is also composed of a so-called computer, and includes a communication interface for communicating with the security device 14 via the communication network 16, a display device such as a liquid crystal display, and a keyboard, mouse, etc. for remotely operating the security device 14. A user interface for inputting operation commands for When the user device 20 is operated to observe the object 12 to be monitored which has been registered in advance via the user interface, the user device 20 is currently photographing the security device 14 installed at the object 12 to be monitored which has been registered. Various image request signals are transmitted to request the user device 20 to transmit the captured image inside or the captured image recorded in the recording device 24 . Then, when the photographed image is received from the security device 14, the user device 20 displays the requested photographed image on the display unit.

The recording device 24 includes recording media such as a magnetic disk device such as an HDD, a magnetic tape such as a DAT, and an optical recording medium such as a DVD-RAM. It consists of a device that reads and writes The recording device 24 receives the photographed image photographed by the image processing device 22 from the security device 14 and records it in association with the photographing time.

"Configuration of image processing device"
FIG. 2 is a schematic diagram of the configuration of the image processing device 22. As shown in FIG. The image processing device 22 includes a computer and executes an image processing program to identify a person (detection target) or an insect (non-detection target) from the captured image.

The communication unit 30 is an input/output interface for transmitting and receiving various setting signals and control signals between the image processing device 22 and the security device 14 via a communication network such as a local LAN. It consists of various communication interface circuits and driver software for driving them. Specifically, when the image processing unit 44 (to be described later) detects a person who has intruded into the monitored area, the communication unit 30 sends an intrusion alarm signal to the security device 14 to indicate that the intruder has been detected. Output.

The photographing unit 32 photographs a monitoring area in the property 12 to be monitored and sequentially obtains photographed images. The imaging unit 32 includes a two-dimensional detector composed of a photoelectric converter having sensitivity to visible light and near-infrared light, such as a CCD or CMOS, and an image of a monitoring area formed on the two-dimensional detector. It consists of an image optical system.

Further, the photographing unit 32 is capable of photographing the monitoring area at the same angle of view, and acquires photographed images by photographing at regular time intervals (for example, 1/5 second). Then, the acquired photographed image is stored in the storage unit 38 .

Here, the image processing device 22 has an illumination section 36, and the illumination section 36 includes, for example, a near-infrared LED. The illumination unit 36 is a proximity illumination means, and brightens only the vicinity area in front of the photographing unit 32 . That is, the illumination is limited to an area in which the image of a small flying object such as an insect obtained by the photographing unit 32 has the same size as a person who has entered the monitoring area.

Then, the photographing unit 32 turns on the lighting unit 36 to photograph a lit image photographed by illuminating only the immediate vicinity of the photographing unit 32 and a light-off image photographed by turning off the lighting unit 36 . The turned-on image and the turned-off image are taken at the same time. For example, it is preferable to alternately capture a lit image and an unlit image. Alternatively, the lit image and the unlit image may be shot alternately at regular time intervals (for example, 1/5 second). Another image may be captured between the lit image and the unlit image.

The illuminance sensor 34 detects the illuminance of the monitored area (brightness of the monitored area) and outputs an illuminance signal indicating the detected illuminance to the image processing unit 44 described later. The illuminance sensor 34 includes, for example, a photodiode, and various known illuminance sensors can be used.

The storage unit 38 is composed of a recording medium such as a nonvolatile semiconductor memory, a magnetic disk, or an optical disk. The storage unit 38 stores an image processing program for operating each unit of the image processing device 22 . Further, as shown in FIG. 2 , the storage unit 38 stores a captured image 40 and a reference image 42 acquired by the imaging unit 32 . The captured image 40 includes a lighting image 40a and a lighting image 40b, and the reference image 42 includes a lighting reference image 42a and a lighting reference image 42b.

The photographed image 40 is an image acquired by the photographing unit 32 sequentially photographing the monitoring area. 40 is an extinguished image 40b. The storage unit 38 stores captured images 40 for several tens of frames (images). The latest frame is stored after deleting the old frame. That is, the storage unit 38 accumulates and stores captured images 40 (lighted image 40a and unlit image 40b) for a predetermined period.

The reference image 42 is a background image that serves as a reference when there is no detection target such as a person or non-detection target such as an insect within the monitoring area. The captured image 40 captured when the image processing device 22 was initially installed may be used, or it may be generated based on a past image. For example, the image processing unit 44 obtains an inter-frame difference between two captured images adjacent in time series, the difference between the frames is equal to or less than a predetermined value, and the difference between the inter-frame differenced captured image and the reference image. is equal to or less than a predetermined value, the photographed image obtained by obtaining the difference is stored in the storage unit 38 as the reference image 42 . Note that the reference image 42 is preferably updated at regular intervals (for example, every 10 minutes) in order to cope with variations in the brightness of the monitoring area due to variations in lighting conditions, diurnal variations in the sun, and the like. Also, the stored reference image 42 may be a captured image that was captured earlier than the past image stored in the storage unit 38 . Also, the reference image 42 may be created from a plurality of captured images 40 .

Here, the reference image 42 includes two of a lighting reference image 42a that is a reference image of the lighting image 40a and a lighting reference image 42b that is a reference image of the lighting image 40b.

The image processing unit 44 is a data processing execution unit of a computer, has a built-in microprocessor unit, memories such as ROM and RAM, and peripheral circuits thereof, and executes image processing programs. Further, as shown in FIG. 2, the image processing section 44 includes an extraction means 46, a feature amount calculation means 48, and a determination means 50. FIG.

"extraction means"
The extracting means 46 compares the unlit image 40b and the unlit reference image 42b and extracts a different region between them as a changed region. For example, the luminance difference of each pixel is obtained, and one or more pixels with a difference of a predetermined amount or more are extracted as the changed area. Alternatively, the changed area may be extracted by comparing the lit image 40a and the lit reference image 42a, or the areas that differ between the lit image 40b and the lit reference image 42b and the areas that differ between the lit image 40a and the lit reference image 42a. may be extracted as a changed region, or an overlapping region among different regions between the two may be extracted as a changed region.

Note that the extracting means 46 compares the captured image 40 with the reference image 42, which is a background image, and extracts a change area from the captured image 40. A past image may be used as the reference image 42, and the reference image 42 and the photographed image 40 may be compared to extract the changed area. Note that the comparison of the captured images 40 is a comparison between the lit images 40a or the unlit images 40b.

FIG. 3(a) is an example of an extinguished reference image 42b, and FIG. 3(b) is an example of an extinguished image 40b. In this example, a person 2 is shown in the lower left, and an insect 3 is shown in the upper right.

"Feature value calculation means"
The feature amount calculating means 48 analyzes the changed regions extracted by the extracting means 46, and determines whether the extracted changed regions are insects (non-detection targets) or humans (detection targets). A feature amount for determining likelihood is calculated.

Here, the feature amount calculation means 48 calculates two types of feature amounts. i.e.
(1) difference in brightness in each change area between the lit image 40a and the unlit image 40b;
(2) the ratio of bright regions (bright regions with high brightness values) existing around dark regions (regions with low brightness values) in the change region;
is obtained as a feature quantity.

<Brightness difference in change area>
When obtaining the difference in brightness (difference in luminance) for the changed area, the following processing is performed on the captured image 40 and the reference image 42 in the changed area.

First, the light-off image 40b and the light-on image 40a of the photographed image 40 are compared (the difference in luminance is obtained) to create an off-light difference image. Also, the non-lighting reference image 42b and the lighting reference image 42a of the reference image 42 are compared (obtaining the difference in luminance) to create a non-lighting difference reference image.

The created non-lighting difference image and the non-lighting difference reference image are compared (the difference is obtained), and the non-lighting difference image corresponding to the image area of the change region extracted by the extraction means 46 and the difference image of the non-lighting difference reference image. The brightness difference (luminance difference) of the changed area is obtained from the brightness value of the image area of . This difference in brightness may be the difference in the average luminance value of the pixels in the changed area, or the difference in the total luminance value of the changed area that is not divided by the number of pixels. Alternatively, other known methods may be used to obtain the difference in brightness for the changed region.

The difference between the brightness of the change area obtained in this way between when the light is on and when the light is off is calculated as one of the feature amounts. That is, the lighting unit 36 illuminates the vicinity of the photographing unit 32, and brightly illuminates insects (non-detection targets) flying near the photographing unit 32, but does not affect people (detection targets) far from the photographing unit 32. is less. Therefore, if the difference in brightness of the changed area is large, it can be determined as an insect.

<ratio>
When the shooting environment is bright, the difference in brightness between the lit image 40a and the unlit image 40b is small. Furthermore, a person (detection target) at a position away from the photographing unit 32 and an insect (non-detection target) flying near the camera are similar in size on the image. Therefore, there is a demand for improving the accuracy of identifying humans and insects when the shooting environment is bright.

For such a determination, it is preferable to use the ratio of the bright area to the image area positioned near the periphery of the dark area for the changed area. That is, the ratio between the number of pixels in the expanded area obtained by expanding the dark area and the number of pixels in the bright area within the expanded area is calculated, and humans and insects are identified. This ratio is obtained by the following ratio calculation method.

First, the average brightness value within the ratio calculation area, which is the image area of the unlit image 40b corresponding to the image area of the change area extracted by the extracting means, is calculated. Since the photographing environment is bright, the influence of lighting of the lighting unit 36 is small, and either the lighted image 40a or the lighted image 40b of the photographed image 40 may be used. It is preferable to use the unlit image 40b. Then, a predetermined offset value is added to this average luminance value, and this value is used as a threshold value. This offset value may be fixed or may be a value such as 10% of the average value. Alternatively, the offset value may be set to 0, and the average brightness value may be used as the threshold as it is.

When the threshold is obtained in this way, the luminance of each pixel in the ratio calculation area of the unlit image 40b is compared with the threshold, and the bright area that is equal to or greater than the threshold and the dark area that is less than the threshold are obtained. Create a valued image.

FIG. 4 shows an example of a binarized image showing the change area of the photographed image 40. Regarding the ratio calculation area 5, the ratio calculation area 5 for the person 2 is entirely a dark area, and the insect 3 is a peripheral area. part is a bright region. The binarized image of FIG. 4 is obtained by binarizing the image area (ratio calculation area) of the captured image 40 corresponding to the changed area of the captured image 40 and the reference image 42 of FIGS. It is an image.

In the case of an insect, since it is in the vicinity of the photographing unit 32, the influence of the diffracted light appears strongly on the image. This is a phenomenon similar to the case of photographing an object in backlight. Therefore, in the case of images of insects, the contours of dark regions are often adjacent to bright regions. For this reason, the ratio of pixels in the bright area increases around the dark area (expansion area). On the other hand, since the person is far from the photographing unit 32, the effect of diffracted light on the image is less likely to appear. This is because when comparing an insect, which is the same size as a person, and a person, the actual size of the person in one pixel is sufficiently larger when the person is farther away. Therefore, the outline of the dark area is less likely to be adjacent to the bright area, and more often coincides with the outline of the ratio calculation area (change area). Therefore, the ratio of pixels of the background image, which is a non-change area, increases around the dark area (expansion area). That is, the number of pixels in the bright area within the change area is reduced. From the above, as shown in FIG. 3(b), the insect is strongly affected by the diffracted light, so a relatively bright area is formed around the insect 3, but the person in the distance is less affected by the diffracted light. Therefore, the peripheral portion of the person 2 is not captured brightly.

Next, an image area consisting of pixels adjacent to the outline of the dark area is determined. For example, the outline of the dark region in the obtained binarized image is expanded toward the periphery. Each peripheral pixel in the dark region may be expanded once in four neighborhoods (up, down, left, and right), or may be expanded once in eight neighborhoods (up, down, left, right, and obliquely). Further, it may be inflated multiple times rather than once. An expanded region (only the expanded region) is obtained by removing the dark region from the expanded dark region.

When the expansion area is obtained in this way, a target area formed by pixels that are bright areas in the expansion area is extracted. That is, a target area (and area between a bright area and an expanded area) that is both a bright area and an expanded area is extracted.

A ratio is calculated by comparing the number of pixels in the obtained target area and the number of pixels in the expansion area. For example, the ratio is obtained from (the number of pixels in the target area)/(the number of pixels in the expansion area)×100.

FIG. 5 shows a case where the ratio calculation method described above is applied to the insect 3 in FIG. First, the inside of the ratio calculation area 5 is divided into a bright area 6 and a dark area 7 . Next, the dark area 7 is taken out and expanded to the periphery. By removing the dark area 7 an expanded area 8 is obtained. The dilated area 8 and the bright area 6 are overlapped to obtain a target area 9 where both overlap. In this case, the region of interest 9 coincides with the expansion region 8 .

FIG. 6 shows a case where the ratio calculation method described above is applied to the person 2 in FIG. First, the inside of the ratio calculation area 5 is divided into a bright area 6 and a dark area 7 . Next, the dark area 7 is taken out and expanded to the periphery. By removing the dark area 7 an expanded area 8 is obtained. The expanded region 8 overlaps the bright region 6. In this example, the ratio calculation region 5 is all the dark region 7, so the expanded region 8 overlaps the unchanged background region. no.

Thus, the person 2 and the insect 3 have different distributions of the bright area 6 and the dark area 7 when the ratio calculation area 5 is binarized. Therefore, even when the shooting environment is bright, the person 2 and the insect 3 can be distinguished from each other by the ratio. It should be noted that even the person 2 has a white portion around it, which may be included in the expansion region 8 . However, even in this case, compared to the insect 3, the bright area 6 included in the expansion area 8 is not so large, and the person 2 and the insect 3 can be distinguished from each other by the ratio. Note that the identification method described above is particularly effective for a person who appears black on the captured image 40 (for example, a person wearing black clothes).

<About the threshold>
Here, the threshold for binarizing the ratio calculation area may be other than the average luminance value in the ratio calculation area. For example, a preset target luminance value used in exposure control may be set as a threshold. This target brightness value is used, for example, to perform exposure control so that the brightness (average brightness value) of the entire image becomes the target brightness value. If the shooting environment is bright, the insect will block the light from the shooting environment, and will appear darker than the target luminance value. Therefore, the target luminance value can be used as the threshold instead of the average luminance value within the ratio calculation area. Alternatively, the average brightness value of the entire image may be used as the threshold. Also, there may be a plurality of thresholds. For example, the average brightness value in the ratio calculation area is set as the first threshold, and the value obtained by adding the offset value to the average brightness value is set as the second threshold. The area may be determined to be a dark area, and an image area composed of pixels having a second threshold value or more within the ratio calculation area may be determined to be a bright area.

Furthermore, a desired offset value may be added (subtracted) to the target luminance value. This offset value may be fixed or may be changed according to the exposure of the captured image. For example, if the exposure amount is small, the insects tend to appear dark because the environment is bright, and the effect of diffracted light tends to be strong, so the offset value is increased.

"Judgment means"
The determining means 50 obtains the size (width and height) of the changed area extracted by the extracting means 46, and determines whether or not the obtained size is equivalent to the pre-stored size representing humanness. Then, if the size of the changed region is equivalent to the size of a person, the feature quantity calculated by the feature quantity calculating means 48 is used to determine whether or not the changed region is an insect (non-detection target). Note that the determination means 50 also determines whether or not the change area is a person (detection target).

<Judgment by Brightness Difference>
The determining means 50 determines whether or not the changed area extracted by the extracting means 46 is an insect based on the feature amount calculated by the feature amount calculating means 48 .

First, when the brightness difference calculated by the feature amount calculation means 48 is equal to or greater than a preset threshold value (set brightness value), it is determined to be an insect, and the obtained brightness difference is less than the preset threshold value. If so, it is determined not to be an insect. This is because, as described above, if the changing area is an insect in the vicinity of the photographing unit 32, it is greatly affected by the illumination by the illumination unit 36, resulting in a large difference in brightness.

<Judgment by ratio>
If the calculated ratio is equal to or greater than the preset ratio, it is determined to be an insect, and if the calculated ratio is less than the preset ratio, it is determined not to be an insect.

The size of the change area varies depending on the position of the person or insect. For example, the larger the change area, the larger the number of pixels in the expansion area. Therefore, by using the ratio of the bright areas instead of the number of pixels in the bright areas in the expansion area, highly accurate determination can be performed.

<Procedure of detection processing>
FIG. 7 shows a flowchart of processing (abnormality detection processing) for identifying and detecting humans (detection targets) and insects (non-detection targets). This processing is performed by the extracting means 46 , feature quantity calculating means 48 and determining means 50 of the image processing section 44 .

First, the extracting means 46 compares the unlit image 40b of the captured image 40 with the unlit reference image 42b, and extracts a region having a significant difference in pixel data (for example, brightness) as a changed region (S10). Then, it is determined whether or not there is a change area in the unlit image 40b (S11).

If the determination in S11 is NO, there is neither a detection target nor a non-detection target, so the process ends. On the other hand, if the determination in S11 is YES, the determining means 50 determines whether or not the extracted change area is human size (S12).

If it is determined in S12 that the changed area does not correspond to the size of a person, it is determined that the changed area is not a person (S13), it is determined to be normal (S14), and the process ends.

If the determination in S12 is YES, the feature amount calculation means 48 obtains a feature amount for determining whether the extracted changed region is a person or an insect (S15). As described above, this feature amount includes (i) the difference in brightness between the lit image and the unlit image 40b in the extracted changing region, (ii) dividing the extracted changing region into a dark region and a bright region, and the ratio of bright areas around dark areas.

Then, the determining means 50 determines whether the difference in brightness, which is one of the feature amounts, is equal to or greater than a predetermined value (S16). If the determination in S16 is YES, it is determined to be an insect (non-detection object) (S17), it is determined to be normal (S18), and the process ends.

If the determination in S16 is NO, it is determined whether the ratio, which is another characteristic quantity, is equal to or greater than a predetermined value (S19). Then, if the determination in S19 is YES, the process proceeds to S17, determines that the changed area is an insect, determines normal in S18, and terminates the process.

If the determination in S19 is NO and the ratio is less than the predetermined value, it is determined whether or not it is a person (S20). Here, since it is determined whether or not the object is a person based on size in S12, determination other than size is performed here. For example, between a plurality of time-series captured images 40, after performing tracking processing of the changed regions, the moving speed of the changed regions is calculated from the position of the changed region in each captured image 40 and the acquisition time of each captured image 40. , the moving speed of the changed area is further compared with a pre-stored human-like speed, and it is determined whether or not the changed area is a human area. Also, whether or not the object is a person may be determined from the number of edges in the change area. For the number of edges, for example, the number of edges of pixels in which the change in luminance value is a predetermined number or more in the changed area is obtained. In addition, you may determine whether it is a person by applying various well-known methods.

If the determination in S20 is NO, it is determined to be normal (S21), and the process ends. On the other hand, if the determination in S20 is YES, it means that a person has been detected and is determined to be abnormal (S22). In this case, for example, an intrusion alarm signal is transmitted from the communication unit 30 to the security device 14 . It is also preferable to sound an alarm provided in the monitored area.

Only the determination of S19 may be performed without performing the determination of S16, or only the determination of S16 may be performed without performing the determination of S19.

Thus, according to this embodiment, it is possible to effectively exclude insects (non-detection targets) and detect humans.

"Variation"
<Combination of feature values>
In the above-described flowchart, the brightness difference and ratio are determined independently, but two feature amounts may be used to determine whether or not the object is an insect. Also, the difference in brightness and the ratio may be determined in multiple stages.

When the difference in brightness is small, insects are detected with priority given to the ratio compared to when the difference in brightness is large. judgment is made. For example, the brightness difference threshold and the ratio threshold are set in three steps, respectively, in the order of the third threshold > second threshold > first threshold, and whether or not the insect is an insect is determined based on the determination result of the brightness difference and ratio. do. Note that a specific threshold value differs depending on the brightness difference and the ratio. FIG. 8 is a flowchart for determining whether or not an object is a non-detection target (insect) from two feature amounts.

First, it is determined whether the difference in brightness is equal to or greater than the third threshold (S31). If the determination in S31 is YES, it is determined whether the ratio is equal to or greater than the first threshold (S32), and if YES, it is determined as an insect.

If the determination in S31 is NO, it is determined whether the difference in brightness is equal to or greater than the second threshold (S33). If the determination in S33 is YES, it is determined whether the ratio is equal to or greater than the second threshold (S34), and if YES, it is determined as an insect.

If the determination in S33 is NO, it is determined whether the difference in brightness is greater than or equal to the first threshold (S35). If the determination in S35 is YES, it is determined whether the ratio is equal to or greater than the third threshold (S36), and if YES, it is determined as an insect.

If the determination in S35 or S36 is NO, it is determined that the object is not an insect, so the process proceeds to determine whether the object is a person.

If the difference in brightness is greater than or equal to the third threshold value, or if the ratio is greater than or equal to the third threshold value, it may be determined that the object is an insect based solely on that determination.

Also, if the integrated value obtained by integrating the two feature amounts is greater than or equal to a preset integrated setting value, it may be determined as an insect. For example, an integrated value is obtained by summing the insect-likeness value obtained from the brightness difference and the insect-likeness value obtained from the ratio.

<Consideration of brightness>
If the shooting environment is bright, the difference in brightness becomes small. For this reason, when the brightness of the shooting environment is bright, the determination criterion of the ratio is emphasized rather than the difference in brightness to determine whether it is an insect or not. Whether or not an object is an insect may be determined by emphasizing the criterion of the difference in size. For example, in the flow of FIG. 7, if the brightness of the shooting environment exceeds a predetermined value, the determination in S16 is not performed, and the determination in S19 is performed. may be skipped, the determination of S16 may be performed, and the determination of S17 or S20 may be performed. In the flow of FIG. 8, when the brightness of the shooting environment exceeds a predetermined value, the ratio is equal to or greater than the third threshold, and when the difference in brightness is equal to or greater than the first threshold, it is determined to be an insect. However, if the brightness of the imaging environment is less than a predetermined value, the difference in brightness is greater than or equal to the third threshold, and if the ratio is greater than or equal to the first threshold, it may be determined to be an insect. In addition to the brightness of the imaging environment, the difference in brightness and the size of the ratio of brightness may also be considered to determine whether or not the object is an insect. For example, when the brightness of the shooting environment is bright, the smaller the difference in brightness, the more important the judgment criterion for the ratio is. Whether it is an insect or not is determined with emphasis. A dark shooting environment means that the indoor lighting is on or the sunlight is shining, but the shooting environment is not so bright because the indoor lighting or sunlight is weak. point to

When calculating the feature amount of the brightness difference and ratio, the brightness of the shooting environment is determined using a setting value set lower than the predetermined value for determining the brightness of the shooting environment described above, and the brightness of the shooting environment is determined. If the brightness is less than the set value (no sunlight is shining or the indoor lighting is turned off), only the difference in brightness is calculated and the shooting environment is greater than the set value (sunlight is turned off). or the indoor lighting is turned on), the brightness difference and ratio may be calculated. That is, if the shooting environment is bright, it is determined whether the changed area is an insect based on the brightness difference and the ratio. It may be determined whether

In this embodiment, even if the imaging environment is bright, humans and insects can be accurately distinguished and determined.

<Determination of brightness of shooting environment>
Here, the brightness of the shooting environment can be determined from the output of the illuminance sensor 34, but may be determined from the captured image 40 without providing the illuminance sensor 34. FIG. Here, the photographing unit 32 has an exposure control function for controlling the amount of exposure according to the brightness of the photographing environment. That is, when the shooting environment is dark, the exposure is increased, and when the shooting environment is bright, the exposure is decreased. Therefore, it is preferable to determine that the captured image 40 is dark when the exposure amount is equal to or greater than the threshold, and to be bright when the exposure amount is less than the threshold. It should be noted that the threshold value is not limited to one, and the threshold values for the bright determination and the dark determination may be different. In this case, in the intermediate region, as described above, it is possible to combine the two determinations of the difference in brightness and the ratio, so that only the determination of the ratio is performed when the region is bright, and only the determination of the difference in brightness is performed when the region is dark. .

<Consideration of Edge Strength>
In addition to the feature amount of the ratio, the edge strength in the expanded region may be obtained as the feature amount. Strong edge pixels whose change in luminance value is greater than or equal to a predetermined value are extracted in the expanded region, and if the number of strong edge pixels is less than a predetermined edge numerical value, the likelihood of an insect is increased. For example, when the number of edges is small, the threshold value for the difference in brightness or the ratio is made small to facilitate determination as an insect. On the other hand, when the number of strong edge pixels is equal to or greater than the predetermined edge value, the threshold value is increased to make it difficult to determine insects. Note that the determination result may be changed without changing the threshold.

Insects have a smooth luminance change because the diffracted light and the luminance value of the insect itself are mixed. Therefore, by judging the edge strength in the expanded region, it is possible to judge the insects more accurately.

It should be noted that the edge intensity may be determined not only for the expansion area but also for the entire change area. If it is an insect, it will appear uniformly dark, and the expanded area will also have a smooth luminance change as described above. As a result, the difference in luminance between pixels of the insect body becomes smaller, and the proportion of strong edges becomes smaller than that of a human.

Note that the image processing device 22 is not limited to processing images stored in the storage unit 38 within the device itself. For example, the configuration may be such that an image read from another external device is subjected to image processing, or an image received from an external camera is subjected to image processing. Alternatively, a plurality of cameras may be provided, and the turned-on image and the turned-off image may be captured by each camera. In that case, the lit image and the unlit image may be captured at the same time.

2 people, 3 insects, 5 changing area, 6 bright area, 7 dark area, 8 expansion area, 9 target area, 10 security system, 12 objects to be monitored, 12 objects to be monitored, 14 security device, 16 communication network, 18 Security center device, 20 user device, 22 image processing, 24 recording device, 30 communication unit, 32 photographing unit, 34 illuminance sensor, 36 lighting unit, 38 storage unit, 40 photographed image, 40a lighting image, 40b lighting image, 42 Reference image 42a lighting reference image 42b lighting reference image 44 image processing section 46 extraction means 48 feature quantity calculation means 50 determination means.

Claims (8)

a photographing unit that obtains a photographed image obtained by sequentially photographing a predetermined area;
A photographed image to be processed among the images is compared with a reference image photographed in the past from the photographed image, a changed area representing a moving object is extracted from the photographed image, and the moving object is separated from the photographing unit based on the changed area. an image processing unit that determines a detection target that exists at a position close to the imaging unit and a non-detection target that exists near the imaging unit;
including
The image processing unit
A dark area composed of pixels having a low luminance value and a bright area composed of pixels having a luminance value higher than that of the dark area are obtained for the changed area of the photographed image, and occupying an image area in the vicinity of the dark area of the photographed image. a feature amount calculating means for obtaining the ratio of the bright area as a feature amount;
determination means for determining the changed region as the non-detection target when the ratio obtained for the changed region using the feature amount is equal to or greater than a predetermined value;
An image processing device comprising: The image processing device according to claim 1,
The image processing apparatus, wherein the feature quantity calculating means sets a threshold value based on the brightness value of the photographed image from which the changed region is extracted, and obtains the dark region and the bright region from the threshold value. The image processing device according to claim 1 or 2,
The image characterized in that the feature quantity calculating means defines an image area formed of pixels adjacent to the dark area among the image areas positioned in the vicinity of the dark area as the image area in the vicinity of the dark area. processing equipment. a photographing unit that obtains a photographed image obtained by sequentially photographing a predetermined area;
comparing the photographed image to be processed among the photographed images with a reference image photographed in the past from the photographed image , extracting a changed region representing a moving object from the photographed image ; an image processing unit that determines a detection target that exists at a position away from an imaging unit and a non-detection target that exists near the imaging unit;
lighting means for illuminating the vicinity of the photographing unit;
including
The image processing unit
A dark area composed of pixels having a low luminance value and a bright area composed of pixels having a luminance value higher than that of the dark area are obtained for the changed area of the photographed image, and occupying an image area in the vicinity of the dark area of the photographed image. The ratio of the bright area is obtained as a first feature amount, and the luminance value of the changed area is extracted based on the lit image captured with the lighting means turned on and the non-lighted image captured without lighting the lighting means. A feature amount calculation means for obtaining the difference as a second feature amount;
determination means for determining whether the changed region is a non-detection target using the first feature amount and the second feature value obtained for the changed region;
with
The determining means makes it easier to determine the changed region as the non-detection target as the ratio is higher, or to make it easier to determine the changed region as the non-detection target as the difference in luminance value is larger. An image processing device characterized by: The image processing device according to claim 4,
The determination means is
The higher the ratio, the smaller the difference in luminance value required to determine the non-detection target; An image processing device characterized in that a low value is set. The image processing device according to claim 4 or 5,
The determining means sets the ratio necessary for determining the non-detection target to a higher value as the predetermined area is brighter, or sets the luminance value necessary for determining the non-detection target as the predetermined area is darker. An image processing apparatus characterized in that the difference is set to a large value. An image processing program that causes an image processing device to distinguish between a detection target that exists at a position distant from an imaging unit and a non-detection target that exists near the imaging unit,
image processing equipment,
comparing a photographic image to be processed among photographic images obtained by sequentially photographing a predetermined area with a reference image obtained before the photographic image, and extracting a changed area representing a moving object from the photographic image ;
A dark area composed of pixels having a low luminance value and a bright area composed of pixels having a luminance value higher than that of the dark area are obtained for the changed area of the photographed image, and occupying an image area in the vicinity of the dark area of the photographed image. Obtaining the ratio of the bright area as a feature amount,
If the ratio obtained for the changed region using the feature amount is equal to or greater than a predetermined value, determining the changed region as the non-detection target;
Image processing program. An image processing program that causes an image processing device to distinguish between a detection target that exists at a position distant from an imaging unit and a non-detection target that exists near the imaging unit,
image processing equipment,
comparing a photographic image to be processed among photographic images obtained by sequentially photographing a predetermined area with a reference image obtained before the photographic image, and extracting a changed area representing a moving object from the photographic image;
A dark area composed of pixels having a low luminance value and a bright area composed of pixels having a luminance value higher than that of the dark area are obtained for the changed area of the photographed image, and occupying an image area in the vicinity of the dark area of the photographed image. Obtaining the ratio of the bright region as a first feature quantity, and extracting based on a lit image captured with lighting means for illuminating the vicinity of the imaging unit and a non-lighting image captured without lighting the lighting means obtaining a difference in brightness value of the changed region as a second feature quantity;
Using the first feature amount and the second feature amount obtained for the changing area, the higher the ratio, the easier it is to determine the changing area as the non-detection target, or the difference in luminance value is large. the more easily it is determined that the changed region is the non-detected target, and it is determined whether the changed region is the non-detected target.
Image processing program.

Images