Information processing equipment, monitor system, information processing method

Abstract

To reduce processing load by performing monitoring with a necessary processing load for a monitored area.SOLUTION: It divides a captured image data obtained from an imaging unit into one or more sub-regions, and detects objects according to a detection cycle set for each small region based on the transit time required for the detection target to pass through the space corresponding to the small region.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technique for detecting an object from image data captured by an imaging device. [Background technology]

[0002] Conventionally, there are known surveillance systems that detect surrounding objects around an imaging device from captured image data, acquire position, speed, and acceleration information, and perform tasks such as determining collisions between the surveillance system and surrounding objects, determining collisions between surrounding objects, and monitoring for intrusions of surrounding objects into specific areas.

[0003] In surveillance systems, in order to expand the surveillance area and ensure detection accuracy, higher resolution imaging devices and the ability to perform detection processing on a larger amount of image data are required, and this is expected to increase the amount of detection processing. Meanwhile, for devices that operate surveillance systems, it is expected that the specifications of processing devices will be lowered to reduce installation costs. For these reasons, there is a need for a method to reduce the amount of image data that surveillance systems perform detection processing on.

[0004] For example, in Patent Document 1, image data is divided into a plurality of small regions, and whether or not a monitored object exists in each small region is determined according to a presence probability indicating the probability that a monitored object exists, which is set in advance for each partial region, and the execution order of the detection process is adjusted, and a method is disclosed in which the detection frequency in regions with a low presence probability is reduced.

[0005] Furthermore, Patent Document 2 discloses a method of dividing image data into a plurality of small regions, extracting image features from each small region, classifying the small regions as to whether a moving object is likely to exist or not using a classification method such as SVM, and executing a detection process on the small image regions classified as having a moving object, and then reducing the image data to be subjected to the detection process, thereby reducing the processing load. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent No. 6125201 [Patent Document 2] Patent Publication No. 2013-037549 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in the method disclosed in Patent Document 1, the probability of a monitored object being present in a small area where the object is stationary at the same position increases, which increases the number of times detection processing is performed for the small area. This poses a problem in that stationary objects are less important in monitoring, and unnecessary detection processing is performed.

[0008] Furthermore, the method disclosed in Patent Document 2 has a problem in that the classification model is applied to all small regions, and therefore the amount of processing required to operate the classification model increases.

[0009] Therefore, an object of the present invention is to perform monitoring with a processing load necessary for the monitoring area, thereby reducing the overall processing load. [Means for solving the problem]

[0010] The information processing device according to the present invention has the following configuration: a dividing means for dividing image data acquired from an imaging unit into one or more small regions, and a detecting means for detecting an object in accordance with a detection period set for each small region based on the transit time required for a detection target to pass through a space corresponding to the small region. [Effects of the Invention]

[0011] According to the present invention, the amount of calculation required for detection processing in a monitoring system can be appropriately reduced. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram showing a schematic diagram of a monitoring system according to the present invention; [Figure 2]1 is a diagram illustrating an example of the configuration of a monitoring system according to an embodiment of the present invention. [Figure 3] FIG. 10 is a diagram showing detection processing information according to an embodiment of the present invention. [Figure 4] FIG. 1 is a diagram showing a flow of a monitoring system 101 according to an embodiment of the present invention. [Figure 5] FIG. 2 is a diagram showing an example of imaging data in the present invention. [Figure 6] 1 is a diagram illustrating an example of the configuration of a monitoring system according to a second embodiment of the present invention. [Figure 7] FIG. 10 is a diagram showing detection processing information according to the second embodiment of the present invention. [Figure 8] FIG. 5 is a diagram showing the flow of a monitoring system 501 according to two embodiments of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0014] A schematic diagram of a monitoring system envisioned in the present invention is shown in Fig. 1. Fig. 1 shows a monitoring system that monitors vehicles traveling on roads as monitoring targets based on image information.

[0015] (First embodiment) The configuration of a monitoring system according to a first embodiment of the present invention will be described below with reference to Fig. 2. The monitoring system 101 has an imaging unit 102 equipped with one or more imaging devices, a detection processing information storage unit 103 that holds detection processing information 201 shown in Fig. 3, and a detection unit 104 that performs object detection processing on imaging data captured by the imaging unit 102 based on the detection processing information 201. An information processing device has the detection processing information storage unit 103 and the detection unit 104. The imaging device referred to here is a device for obtaining two-dimensional or three-dimensional images of the surrounding environment, such as a camera or a LiDAR (Light Detection and Ranging) sensor, or a device that combines these.

[0016] Next, the processing performed by the monitoring system will be explained using the flow diagram in Figure 4. The flowchart below is assumed to be realized by the CPU executing a control program.

[0017] In step S301, the imaging unit 102 captures an image and sends the image data 401 including the image capture time t as shown in FIG.

[0018] In step S302, the detection unit 104 divides the imaging data 401 acquired from the imaging unit 102. That is, the detection unit 104 divides the imaging data 401 into small areas 402 and 403 shown in Fig. 5 based on small area information defined from a rectangular area connecting two points on the imaging plane included in the detection processing information 201 acquired from the detection processing information storage unit 103. The detection processing information 201 is stored in the detection processing information storage unit 103 for each small area.

[0019] In step S303, the detection unit 104 uses the detection cycle S, the previous detection time T, and the image capture time t included in the detection processing information 201 corresponding to the small regions 402 and 403 to calculate the following:

[0020]

number

[0021] The method for calculating the detection period S will be described later.

[0022] In step S304, the detection unit 104 executes object detection processing for the small area of ​​the detection target, measures the position, speed, and acceleration information of the monitoring target while it is passing through the small area, and outputs the results as detection results.

[0023] In step S305, the detection unit 104 updates the previous detection time T for the small region for which the detection process has been executed with the image capture time t, and ends the process.

[0024] Next, a method for calculating the detection period S in the first embodiment will be described.

[0025] The detection period S is set to a value that satisfies the monitoring function and maximizes the effect of reducing the amount of calculation without causing any missed detections. For example, the calculation (determination) is performed using the following method.

[0026] In the monitoring system 101 for monitoring the road environment as shown in FIG. 1, the vehicle to be monitored is -Movement on the road surface (restrictions on movement direction due to freedom of movement) -Moving through spaces surrounded by structures such as curbs and guardrails (restrictions on movement direction due to surrounding structures) - Move along the center line and lane boundary lines (restrictions on movement direction due to traffic laws) There are constraints on the movement direction.

[0027] From the constraint conditions on the movement direction, it is possible to estimate the shortest route of the vehicle passing through small areas 402 and 403. In this case, the distance of the shortest route for small area 402 is set to L1, and the distance of the shortest route for small area 403 is set to L2.

[0028] In addition, the vehicle being monitored is -The maximum speed that the vehicle can output (speed limit due to power limit) -Legal speed (speed limit according to traffic laws) - Maximum speed of the monitored object that has moved in the past (speed limit based on past statistics) There are constraints on the speed.

[0029] From the constraints on the speed, the maximum speed V1 of the vehicle passing through the small area 402 and the maximum speed V2 of the vehicle passing through the small area 403 are estimated.

[0030] From the distances L1 and L2 of the shortest route and the passing speeds V1 and V2, the shortest passing time D1 required for a vehicle, which is a target object, to pass through the small area 402 and the shortest passing time D2 required to pass through the passing route 403 are calculated as follows:

[0031]

number

[0032]

number

[0033] It can be calculated as follows.

[0034] In order to satisfy the monitoring function, when the number of times that the same monitoring target needs to be detected in the small area 402 is N1 and the number of times that the same monitoring target needs to be detected in the small area 403 is N2, the detection cycle S1 held as the detection process information 201 for the small area 402 and the detection cycle S2 held as the detection process information 201 for the small area 403 are defined as follows:

[0035]

number

[0036]

number

[0037] It is stipulated that:

[0038] In step S303, the detection unit 104 excludes from the detection target small regions any small region whose passage time tT from the previous detection time T is less than the detection period S based on the detection period calculated by the above calculation method.

[0039] Thus, according to this embodiment, the monitoring system 101 reduces redundant detection processing to fulfill the monitoring function based on the detection period S calculated from the shortest route and maximum speed of the vehicle passing through each small area, and performs detection processing only for the small areas necessary for the monitoring function, thereby making it possible to reduce the detection processing while maintaining the monitoring function.

[0040] (Second embodiment) The configuration of a monitoring system according to the second embodiment of the present invention will be described below with reference to Fig. 6. The monitoring system 501 has the following configuration. The imaging unit 102 includes one or more imaging devices. The detection processing information storage unit 503 holds detection processing information 601 shown in Fig. 7. The detection unit 104 performs detection processing on the imaging data captured by the imaging unit 102 based on the detection processing information 601. The detection cycle calculation unit 502 updates the detection cycle based on the detection results obtained by the detection unit 104 and the detection processing information 601 held in the detection processing information storage unit 503.

[0041] Next, the processing performed by the monitoring system will be described with reference to the flow chart of FIG.

[0042] In step S701, the imaging unit 102 captures an image and sends the captured image data 401 including the image capture time t as shown in FIG.

[0043] In step S702, the detection unit 104 divides the imaging data 401 acquired from the imaging unit 102. That is, based on small area information defined from a rectangular area connecting two points on the imaging plane, which is included in the detection processing information 601 acquired from the detection processing information storage unit 503, the detection unit 104 divides the imaging data 401 into small areas 402 and 403 shown in FIG.

[0044] In step S703, the detection unit 104 excludes the small region satisfying formula (1) from the detection target.

[0045] In step S704, the detection unit 104 executes detection processing for the small area of ​​the detection target, measures the position, speed, and acceleration information of the monitoring target, and outputs the results of detection.

[0046] In step S705, the detection unit 104 updates the previous detection time T for the small region for which the detection process has been executed with the image capture time t.

[0047] In step S706, for the small area where detection has been performed, the detection period calculation unit 502 estimates the maximum speed V' of the monitoring target currently existing in the small area from the speed and acceleration information included in the detection result.

[0048] In steps S707 and S708, if the maximum speed V' of the monitoring target currently existing in the small area is greater than the maximum speed V included in the detection process information 503, the maximum speed V included in the detection process information 503 is updated with V'.

[0049] Although the preferred embodiment of the present invention has been described above, the object to be monitored is not limited to a vehicle, but can be applied to various moving bodies such as autonomous mobile robots, drones, aircraft, etc. The small area information is not limited to a two-dimensional rectangular area on the imaging plane, but can arbitrarily define two-dimensional or three-dimensional space within the monitoring area.

[0050] Furthermore, the present invention is not limited to these embodiments, and various modifications and changes are possible within the scope of the gist of the present invention.

[0051] The present invention can also be realized by executing the following process. That is, software (programs) that realize the functions of the above-described embodiments are supplied to a system or device via a network or various storage media, and the computer (or CPU, MPU, etc.) of the system or device reads and executes the programs. The programs may also be provided by recording them on a computer-readable recording medium.

Claims

1. a dividing means for dividing the image data acquired from the imaging unit into one or more small regions; and detecting means for detecting an object in accordance with a detection period set for each small region based on the time it takes for the detection target to pass through a space corresponding to the small region.

2. an imaging means having one or more imaging devices for imaging a monitoring target within a monitoring area; a dividing means for dividing the image data acquired from the imaging means into one or more small regions; a detection means for detecting an object in accordance with a detection period set for each small area based on the transit time it takes for the monitored object to pass through the space corresponding to the small area.

3. The transit time is The distance and speed of the route traveled by the monitored object are calculated. The information processing device according to claim 1 .

4. The distance of the passage path is Restricting movement of the monitored subject by degrees of freedom of movement; Restrictions on movement due to surrounding structures of the monitored object; Restrictions on movement due to traffic laws relating to the subject of said surveillance; characterized by being determined by one or more of the following: The information processing device according to claim 3 .

5. The passing speed is Limiting the speed due to the output limit of the monitored object; Speed ​​limits according to traffic laws relating to the subject of said monitoring; speed and acceleration information of the monitoring target that has passed through the small area in the past; speed and acceleration information of the monitored object passing through the small area; determined by one or more of The information processing device according to claim 3 .

6. The speed and acceleration information is It is updated at any interval, The detection period is updated based on the passage time determined from the updated speed and acceleration information. The information processing device according to claim 5 .

7. a dividing step of dividing the imaging data acquired from the imaging unit into one or more small regions; and a detection step of detecting an object in accordance with a detection period set for each small region based on the time it takes for the detection target to pass through the space corresponding to the small region.

8. a division step of dividing the image data acquired from the image capturing unit into one or more small regions; and a detection procedure for detecting an object in accordance with a detection period set for each small area based on the transit time required for the detection target to pass through the space corresponding to the small area.

Images