JP2013164823A - Space monitoring method and device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a space monitoring method and device for detecting whether or not there is an object of a fixed dimension or larger inside a monitoring space area in real time in the monitoring space area whose range is arbitrarily set, and generating alarm signals when determined that there is the object.SOLUTION: According to the present invention, a device is for photographing a space region including a monitoring space area whose range is arbitrarily set by a digital camera, detecting whether or not there is an object inside the monitoring space area from photographed image data in real time, and outputting image information and signals for reporting presence/absence of the object, and includes: a plurality of digital cameras for photographing the space area including the monitoring space area whose range is arbitrarily set; a control mechanism for photographing the space area by the digital cameras; and means and an arithmetic unit for obtaining a parallax for each pixel by a stereo matching method from the photographed image data, extracting a pixel set included inside the monitoring space region, discriminating the feature, and determining the presence/absence of the object of a fixed dimension or larger in the monitoring space area in real time.

Description

Detailed Description of the Invention

Field of Invention

The present invention relates to an apparatus for detecting whether or not an object having a certain size or more is present in an arbitrarily set monitoring space area, and more specifically, a plurality of space areas including an arbitrarily set monitoring space area. Taking a picture with a digital camera, obtaining the parallax from the obtained digital image, calculating the spatial coordinate position of the object reflected in the pixel, discriminating a set of adjacent pixels, and detecting the size of the object based on the number of pixels included in the set Is. In other words, when the size of an object occupying the monitoring space area is detected from the number of pixels in the position included in the monitoring space area among the pixels included in the adjacent pixel set, and the detected size is more than a certain level The present invention relates to a space monitoring method and apparatus for determining in real time when an object to be detected is present in a monitoring space area and generating an alarm signal.

As a method for determining whether or not an object such as an obstacle is present in a specific monitoring space area, a method for determining the presence or absence of an object using light or sound waves is generally used. As a method using light, an interrupter method in which a beam-shaped light beam generating device and a light receiving device are installed so that the beam optical axis intersects the surveillance space region, and an object is present when the light beam is blocked. There are many. As a method using sound waves, there are many echo radar methods in which a sound wave emitting device and a receiving device are installed close to each other and the presence is confirmed by the presence or absence of reflected sound from an object in the monitoring space region.

However, as a method for detecting the presence of an object in a monitoring space area set arbitrarily, either method has a major drawback. For example, in the interrupter method using light, in order to monitor a relatively wide space area or to detect a relatively small object, a beam is generated by a method such as preparing a large number of beam light generators or scanning and emitting beam light. The optical axis extends over the entire monitoring space area with high monitoring density, and a large number of light receiving devices need to be installed in the same manner, so that the entire monitoring device becomes large. The echo radar system using sound waves is susceptible to environmental effects such as temperature and humidity due to the propagation characteristics of sound waves, and it is difficult to achieve high spatial resolution due to the long wavelength of sound waves used. Accurate measurement of the distance to the object is difficult.

There is also a method of detecting a target object such as an obstacle or an intruder with a surveillance camera. Although the conventional monitoring method using a camera can detect a relatively small object regardless of the size of the object, it is difficult to measure the distance to the target object with high accuracy, and the object appears in the image obtained from the monitoring camera. However, it is impossible to accurately determine in real time whether it is inside or outside the set monitoring space area.

As a distance measuring method to an object, there is a method of calculating a distance to an object by a triangulation method using parallax of image data obtained by a plurality of imaging devices. This method is a good method for measuring the distance to an object in a wide spatial range, but in order to measure at high speed with high spatial resolution, it is necessary to process a large amount of data at high speed. In order to determine in real time whether there is an object in the monitored space area, it is necessary to perform high-speed processing for a large number of pixels, and it is difficult to perform measurement easily with a small device.

Problems to be solved by the invention

In the present invention, a spatial region including an arbitrarily set surveillance space region is photographed by a plurality of digital cameras arranged to generate parallax, and the parallax of all pixels is obtained from the obtained digital image by a stereo matching method. The object to be detected by calculating the spatial coordinate position of the object that appears in each pixel from the obtained parallax, and discriminating a pixel set in which the number of pixels at the adjacent spatial coordinate position is a certain number or more The number of pixels included in the discriminated pixel set and the number of pixels located in the monitoring space area is counted to check the size occupied by the object to be detected in the monitoring space area. The object is to determine in real time that there is an object to be detected in the monitoring space area when a certain dimension is exceeded, and to output the determination result as an alarm signal.

Means for solving the problem

According to the first aspect of the present invention, parallax of all pixels is obtained by stereo matching method from digital image data photographed by a plurality of digital cameras arranged to generate parallax in an arbitrarily set surveillance space region. Generate a parallax map of the entire area captured in space, calculate the spatial coordinate position of the object shown on the pixel by triangulation method from the parallax map, extract the pixels at the adjacent spatial coordinate position, and the number of extracted pixels is constant By discriminating the pixel set having the above number, it is determined whether or not there is an object having a size to be detected, and among the pixels included in the discriminated pixel set, the number of pixels whose spatial coordinate position is in the monitoring space area is determined. In-space object determination that detects the size of an object in the surveillance space area after inspection and determines that there is an object to be detected in the surveillance space area if the size of the object is greater than a certain level It is the law.

The space monitoring device according to claim 2 monitors whether or not the whole or a part of the object occupies the monitoring space region in the monitoring space region arbitrarily set by using the method of claim 1, It is a space monitoring device that generates an alarm signal when it detects the state that an object occupies the monitoring space area. It is a digital camera with multiple digital cameras that shoot the space area that includes the monitoring space area, and at intervals that generate parallax in the shooting space area. The number of pixels located in the surveillance space area by performing arithmetic processing using stereo matching processing on the camera frame where the camera is placed and fixed, the frame guide that is fixed toward the space area where the camera frame is photographed, and the image data of multiple digital cameras When the number exceeds a certain number, a spatial distance calculation device that determines that an object exists in the surveillance space area, and repeats shooting with a digital camera The digital image data is a space monitoring device which is characterized in that it comprises a main control unit for outputting the processed and alarm signal to perform spatial distance calculation is transferred to the spatial distance calculation unit to the outside.

The spatial distance calculation device according to claim 3 is a specific example of the spatial distance calculation device in the constituent elements of claim 2. That is, the spatial distance calculation device of the present invention includes a filter unit that sets the average brightness between the standard image data and the reference image data to the same level, and a parallax map generation unit that generates a parallax map between the standard image data and the reference image data. An object discriminating unit that discriminates and aggregates pixels in a parallax map and discriminates the set according to the number of pixels in the set; a distance calculation unit that calculates a spatial coordinate position of the object from the parallax map; and a spatial coordinate position of the object Output information of the determination unit that confirms the inclusion relationship of the monitoring space region and generates an alarm signal when the object is present in the monitoring space region, the filter unit, the parallax map generation unit, the object discrimination unit, the distance calculation unit, and the determination unit And a storage unit for outputting image information.

Action

According to the present invention, when monitoring whether or not an object is present in an arbitrarily set surveillance space area, a plurality of digital cameras that capture the space area including the surveillance space area are at least within the surveillance space area. The captured images of the object are arranged at an interval of the optical axis so as to generate a parallax necessary and sufficient for distance measurement, and are simultaneously photographed. The range (width) and distance resolution of the imaging space are determined by a fixed relational expression using the image sensor size of the digital camera, pixel resolution, lens focal length, and optical axis interval (baseline length) of the digital camera as variables. Therefore, the image sensor size and pixel resolution of the digital camera, the focal length of the camera lens, and the optical axis interval of the camera may be determined according to the distance and range from the digital camera to the surveillance space region and the minimum size of the object to be detected.

Since a plurality of digital cameras shoot a space area at the same time, the digital images obtained from the cameras all have similar light and dark and contour pattern characteristics. Therefore, by comparing digital image data by stereo matching method and extracting contour patterns with similar brightness and darkness, the object that appears in the pixel by detecting the shift amount of the similar pattern position (pixel interval of the camera image sensor) on the image The parallax of the object is obtained, and the parallax map of the entire shooting region is generated by obtaining the parallax for all the pixels.

The spatial coordinate position of the object shown in the pixel is obtained by triangulation using the distance between the optical axes of the captured digital camera as the base line length and the parallax amount and lens focal length as variables. Therefore, based on the parallax map, the spatial coordinate position of the object shown in each pixel in the entire shooting space is calculated, and the pixels at the adjacent spatial coordinate position are extracted in order to discriminate the object having the size to be detected. Thus, a pixel set in which the number of extracted pixels is a certain number or more is discriminated. The size of the object to be detected can be arbitrarily set by the discrimination process on the condition of the number of pixels at adjacent spatial coordinate positions.

In order to determine whether an object of a size to be detected exists in the monitoring space area, the spatial coordinate position of the pixel included in the pixel set discriminated as the detection target candidate is set to a position included in the monitoring space area. Check for the presence and count the pixels at the included positions. When the number of pixels located in the monitoring space area becomes a certain value or more, it is determined that the detection target object exists in the monitoring space area. By the determination process using the number of pixels as a condition, the size of the object to be detected in the monitoring space region can be arbitrarily set.

If it is determined that there is an object to be detected in the monitoring space area, an alarm signal is generated. In the present invention, the determination condition of the size of the object to be detected and the determination condition of the proportion of the object in the monitoring space area can be set independently, so the determination request “there is an object in the monitoring space area” On the other hand, various determination conditions can be set as shown in FIG. 5 or FIG.

FIG. 5 shows an example of the size of an object to be detected. Example 1 of the determination condition is when the size of the detection target is greater than or equal to the object a, Example 2 is when the size of the detection target is set within the range of the object a and the object b, and Example 3 is a certain specific size b This is a case where only the detection target is set.

FIG. 6 shows the ratio of the object of the detection target in the monitoring space area. In the determination condition example 4, the range of the object c that is 100% and the object d that is 50%, that is, the range of 50% to 100%, is set as the detection target. The example 5 is 50%. A case is shown in which the range of the object d contained and the object e contained 5%, that is, the range of 5% to 50% is the detection target. As described above, the size of the detection target and how much the detection target is included in the monitoring space area can be freely set as the object detection determination condition.

Since the function of determining whether or not an object is present in the monitoring space area is processed in real time, monitoring can be continuously performed by repeatedly executing the processing. Furthermore, by continuously monitoring, an alarm is generated only when the state in which the target object is in the monitoring space area lasts for a certain period of time, or the target object is in the monitoring space area and an alarm signal is generated. It is also possible to give the alarm signal hysteresis, such as not interrupting the alarm when the target object momentarily moves out of the monitoring space area and immediately enters the monitoring space area. In addition, by extracting features of time-series changes in light and darkness and contour patterns from digital images repeatedly shot using the storage function, it is possible to further improve the detection accuracy of the size of objects captured in pixels and spatial coordinate positions.

Embodiment of the Invention

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Here, an example using two digital cameras will be described.

FIG. 1 is a flowchart showing the processing content of the in-space object determination method according to claim 1. Monitoring is continuously performed by repeating the steps shown in this flowchart.

In step S101, the vertex coordinates of the three-dimensional shape of the monitoring space area are set in order to set the monitoring space area. As a method for setting the monitoring space area, it is possible to set all the coordinates in the space determined by the edge function, the surface function, and the resolution of the monitoring device.

In step S102, a standard image and a reference image are acquired and stored from the standard image camera and the reference image camera, respectively, for stereo matching processing.

In step S103, stereo matching processing is performed using the reference image in order to obtain the parallax of the pixels in the standard image. The parallax of each pixel is held as data representing the position of the pixel of the reference image and the amount of parallax as a parallax map. At this time, edge enhancement of the reference image and reference image and processing for making the average brightness between the two images the same level are performed as necessary.

In step S104, the pixels in the parallax map are aggregated according to the adjacency discrimination rule so that only objects having a certain size or larger are detected, and are discriminated based on the number of pixels in the aggregate. The adjacency discrimination rule is a method for discriminating that an object shown in each pixel is adjacent in space, and for each pixel of the parallax map, the difference in the amount of parallax from the adjacent pixel at the spatial coordinate position is examined. Are connected and assembled as pixels that photograph the same object. Then, only a set of pixels in which the number of connected pixels is equal to or greater than a certain value is stored as a processing target for confirming whether or not it exists in the monitoring space area.

In step S105, in order to confirm whether or not an object reflected in the pixel is in the monitoring space area, the spatial coordinate position indicated by each pixel in the distinguished pixel set is represented by the amount of parallax, and the baseline length of the two cameras. Calculate using the focal length of the lens.

In step S106, in order to determine whether or not the object is included in the monitoring space area, the inclusion relationship between the space coordinate position of the object reflected in each pixel calculated in step S105 and the monitoring space area is checked. If the number of pixels in the monitoring space area is equal to or greater than a certain number, it is determined that the object exists in the monitoring space area, and an alarm signal is output in step S107.

In step S108, the reference image, the reference image, the parallax map, and the spatial coordinate position of the object are output as image information separately from the alarm for real-time confirmation and post-mortem analysis of the space monitoring situation.

In step S109, it is confirmed whether there is an instruction to end monitoring, and monitoring is ended if necessary.

FIG. 2 is a functional configuration diagram of the space monitoring device according to claim 2.
It comprises a reference image camera 21, a reference image camera 22, a main control device 26, and a spatial distance calculation device 27.

The control signal 25a and the control signal 25b are signals for instructing the base image camera 21 and the reference image camera 22 to perform repeated shooting, respectively.

The control signal 25c is a signal that causes the spatial distance calculation device 27 to execute a spatial distance calculation.

The reference image data 23 is an output of the reference image camera 21, and the reference image data 24 is an output of the reference image camera 22.

The main control device 26 transfers the standard image data 23 and the reference image data 24 to the spatial distance calculation device 27 and outputs a process for executing the spatial distance calculation and an alarm signal 28 to the outside.

Image information 29 is also output from the spatial distance calculation device 27.

FIG. 3 is a structural diagram of the space monitoring device according to claim 2.
The reference image camera 21 is a camera that acquires the reference image data 23.

The reference image camera 22 is a camera that acquires reference image data 24.

The camera mount mechanism 31 arranges the reference image camera 21 and the reference image camera 22 with an interval in which the optical axes of the two cameras are parallel and the captured image of the object in the monitoring space region generates a sufficient parallax for distance measurement.・ Fix it. Further, the standard image camera 21 and the reference image camera 22 are maintained at desired elevation angles.

The main control device 26 transfers the image data obtained by the standard image camera 21 and the reference image camera 22 to the spatial distance computing device 27 and outputs a processing for executing the spatial distance computation and an alarm signal 28 to the outside.

The spatial distance calculation device 27 is described in claim 3 and will be described in detail with reference to FIG.

The gantry guide mechanism 32 fixes each component of the camera gantry mechanism 31, the main controller 26, and the spatial distance calculation device 27 as a device, and further fixes the entire device in the monitoring space direction.

1, FIG. 2, and FIG. 3, an example in which one reference image camera and one reference image camera are used is shown. However, one reference image camera and a plurality of reference image cameras are used. The same can be achieved by using a reference image camera.

FIG. 4 shows an example of the functional configuration of the spatial distance calculation apparatus according to claim 3, and constitutes the spatial distance calculation apparatus 27 as a whole. The standard image data 23 and the reference image data 24 are the image data shown in step S102 and are input to the filter unit 41. The filter unit 41 executes the process for enhancing the edge enhancement of the standard image and the reference image and the average brightness between the two images shown in step S102. The parallax map generation unit 42 to which the output signal of the filter unit 41 is input executes step S103.

An output signal of the parallax map generation unit 42 is input to the object discrimination unit 43. The object discriminating unit 43 executes Step S104.

The output signal of the object discriminating unit 43 is input to the distance calculating unit 44. The distance calculation unit 44 executes step S105.

The output signal of the distance calculation unit 44 is input to the determination unit 45. The determination unit 45 executes step S101 to set a monitoring space area, and outputs a warning signal 28 when it is determined that the object exists in the monitoring space area by executing step S106.

The storage unit 46 temporarily stores data (parallax map or the like) handled in each step in a memory.

The alarm signal 28 is an alarm signal that is processed by the determination unit 45 and output in step S107.

The image information 29 is image information processed in the storage unit 46 and output in step S108.

Effect of the invention

According to the present invention, a spatial region including an arbitrarily set surveillance space region is photographed by a plurality of digital cameras arranged so as to generate parallax, and all the pixels are obtained from the obtained digital image by a stereo matching method. The parallax is obtained, the spatial coordinate position of the object reflected in each pixel is calculated from the obtained parallax, and the detection target is discriminated by discriminating a pixel set in which the number of pixels at adjacent spatial coordinate positions is a certain number or more. The size of the object to be detected is inspected, and the size of the object to be detected in the monitoring space area is inspected by counting the pixels located in the monitoring space area among the pixels included in the discriminated pixel set. When the distance exceeds a certain dimension, it can be determined in real time that there is an object to be detected in the monitoring space area, and the determination result can be output as an alarm signal.

Because it is possible to monitor a set surveillance space area, it is ideal for monitoring an object in a narrow space area, for example, detecting an object that obstructs opening and closing of a door in a narrow area near the door of an automatic door. Monitoring conditions can be set.

As for the spatial resolution, high resolution can be realized by appropriately selecting the pixel resolution and lens focal length of the digital camera to be used, so monitoring of small objects, for example, the tip of an umbrella protruding from the vehicle by station platform door monitoring, etc. Optimal monitoring conditions can be set for detecting objects that obstruct the opening and closing of vehicles and platform doors.

In addition, since a digital camera is used, it is possible to detect an object under conditions that are difficult for human vision by adjusting the optical wavelength sensitivity of the image sensor, such as detecting an object with high reflectivity in the infrared region or ultraviolet region, Optimum conditions can be set and used as a security sensor for buildings, people's houses, warehouses, etc. combined with visual field lighting.

Processing flow diagram of space monitoring Functional configuration diagram of the space monitoring device Structure diagram of space monitoring device Functional diagram of the spatial distance calculation device Setting example of size judgment condition Setting example of ratio judgment condition

Claims (3)

For an arbitrarily set surveillance space area, the parallax of all pixels is obtained by the stereo matching method using digital image data obtained by capturing the spatial area including the surveillance space area so that the parallax is generated, and the space of the object reflected in the pixel The coordinate position is calculated from the parallax, the pixel set in which the number of pixels at the adjacent spatial coordinate position is a certain number or more is discriminated, and the pixels included in the discriminated pixel set are the pixels located in the monitoring space area. A space monitoring method in which the number is inspected and it is determined that there is an object to be detected in the monitoring space area when the number of pixels located in the monitoring space area exceeds a certain number. A digital camera that captures the reference digital image data, a digital camera that aims to capture the reference image data, and performs arithmetic processing based on the reference digital image data and the reference digital image data to be in the surveillance space area. A spatial distance calculation device that determines that an object is present in an area when the number of pixels that are located exceeds a certain number, and a main control device that has an external output function of a digital camera, control of the spatial distance calculation device, and an alarm signal And a space monitoring device. 3. The spatial monitoring device according to claim 2, wherein the spatial distance calculation device has a filter unit that sets the average brightness between the standard image data and the reference image data to the same level, and a parallax map that generates a parallax map between the standard image data and the reference image data. A generating unit; an object discriminating unit that discriminates and aggregates pixels in the parallax map and discriminates the set according to the number of pixels in the set; a distance calculation unit that calculates a spatial coordinate position of the object from the parallax map; and an object space A determination unit that confirms the inclusion relationship between the coordinate position and the monitoring space region and generates an alarm signal when the object exists in the monitoring space region, a filter unit, a parallax map generation unit, an object discrimination unit, a distance calculation unit, and a determination unit A space monitoring apparatus comprising: a storage unit that stores each output information and outputs image information.

Images