CN111627049A

CN111627049A - High-altitude parabola determination method and device, storage medium and processor

Info

Publication number: CN111627049A
Application number: CN202010478469.1A
Authority: CN
Inventors: 汪浩; 郑凯; 刘畅; 周宇华; 王晴; 赵金岩; 胡金龙
Original assignee: Beijing Sylincom Technology Co ltd
Current assignee: Beijing Sylincom Technology Co ltd
Priority date: 2020-05-29
Filing date: 2020-05-29
Publication date: 2020-09-04
Anticipated expiration: 2040-05-29
Also published as: CN111627049B

Abstract

The application discloses a high altitude parabola determination method, a high altitude parabola determination device, a storage medium and a processor. The method comprises the following steps: acquiring target images corresponding to the target side of a building at different moments to obtain multiple frames of target images; determining a plurality of moving objects corresponding to the side surfaces of the targets based on the multi-frame target images; determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image; and tracking the target object, and judging whether the target object is a high-altitude parabola according to the speed and the acceleration of the target object in the tracking process. By the aid of the method and the device, the problem that high-altitude parabolas are difficult to accurately detect when monitoring objects with complex pictures and natural motion in the related technology is solved.

Description

High-altitude parabola determination method and device, storage medium and processor

Technical Field

The application relates to the technical field of image recognition, in particular to a method and a device for determining a high altitude parabola, a storage medium and a processor.

Background

In residential areas, some residents throw household garbage down from windows, so that the phenomenon of high-altitude object throwing is frequent, the environmental sanitation is influenced, the life safety of pedestrians is possibly threatened, and news reports are frequently reported to have the event of high-altitude falling objects to cause casualties. In order to reduce the occurrence of high-altitude parabolic phenomenon as much as possible and better guarantee the life and property safety of residents, property owners are reminded by posting warning slogans at prominent positions in a community, but the method only can play a warning role.

In order to facilitate evidence obtaining of workers and staff, a parabolic source needs to be detected, a scheme that the parabolic source is searched through monitoring videos shot by a single-path camera appears in the related technology, the workers need to search for the parabolic source by observing a large number of monitoring videos, and the method is large in workload and low in accuracy. In order to solve the problem that the difficulty of manually searching for a parabola is high, a high-altitude parabola detection method based on computer vision appears in the related technology, specifically, a superposition algorithm of a moving target detection algorithm and a target tracking algorithm is adopted to detect the high-altitude parabola, and a least square method is used to analyze the parabola, wherein the moving target detection algorithm generally uses a multi-target tracking algorithm, and the multi-target tracking algorithm has a poor effect. In addition, in an actual situation, the problems that the movement of the whole building cannot be monitored by a single-path camera, the natural object is influenced by wind power, climate and air resistance in a complicated and variable way, the non-parabolic natural object is continuously interfered and the like exist, the false detection rate and the missing detection rate are high, namely, when the monitoring picture is complicated and the natural object continuously moves back and forth, the high altitude parabolic detection algorithm in the related technology cannot achieve an ideal effect.

Aiming at the problem that the high-altitude object is difficult to accurately detect when the monitoring picture is complex and a natural moving object exists in the related technology, an effective solution is not provided at present.

Disclosure of Invention

The application provides a method and a device for determining a high-altitude parabola, a storage medium and a processor, which are used for solving the problem that the high-altitude parabola is difficult to accurately detect when monitoring objects with complex pictures and natural motion in the related art.

According to one aspect of the present application, a method of determining a high altitude parabola is provided. The method comprises the following steps: acquiring target images corresponding to the target side of a building at different moments to obtain multiple frames of target images; determining a plurality of moving objects corresponding to the side surfaces of the targets based on the multi-frame target images; determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image; and tracking the target object, and judging whether the target object is a high-altitude parabola according to the speed and the acceleration of the target object in the tracking process.

Optionally, the obtaining target images corresponding to the target side of the building at different times, and the obtaining multiple frames of target images includes: and respectively shooting the target side of the building by adopting a plurality of cameras at the target moment to obtain a plurality of sub-images, and splicing the sub-images to obtain a frame of target image corresponding to the target moment.

Optionally, before determining the target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image, the method further comprises: filtering moving objects with the size smaller than a preset size from a plurality of moving objects in each frame of target image; and/or filtering moving objects, which are smaller than a preset distance from the edge of the target image, in the plurality of moving objects in each frame of target image.

Optionally, determining the target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image comprises: respectively calculating the intersection ratio between each moving object in a frame of target image and each noise object of the frame of target image to obtain a plurality of intersection ratios of each moving object; determining the maximum intersection ratio of each moving object as the target intersection ratio of the moving object; and taking the moving object with the smallest target intersection ratio in the plurality of moving objects as the target object.

Optionally, tracking the target object comprises: taking the target object as a tracking target of a tracker, and acquiring a tracking result, wherein the tracking result comprises the tracking target and the tracking position of the tracker to the tracking target at each moment; calculating the intersection ratio between the tracking result at the current moment and the target object in the target image at the current moment to obtain a first intersection ratio; and adjusting the tracking object of the tracker based on the relation between the first intersection ratio and a preset threshold value.

Optionally, adjusting the tracked object of the tracker based on the relationship between the first intersection ratio and the preset threshold comprises: judging whether the first intersection ratio is smaller than a first preset threshold value or not; determining a tracker tracking error under the condition that the first intersection ratio is smaller than a first preset threshold value; under the condition that the first intersection ratio is greater than a first preset threshold value, judging whether the first intersection ratio is smaller than a second preset threshold value; determining that the tracking target deforms under the condition that the first intersection ratio is larger than a first preset threshold value and the first intersection ratio is smaller than a second preset threshold value; under the condition that the tracking target deforms, calculating the intersection ratio between the tracking result at the current moment and a plurality of moving objects in the target image at the current moment to obtain a plurality of intersection ratios, and updating the moving object with the largest intersection ratio as the tracking object.

Optionally, before determining whether the target object is a high altitude parabola according to the velocity and the acceleration of the target object during the tracking process, the method further includes: determining the central position of the target object in the vertical direction at each moment after the target object is tracked for a target time period, wherein the target time period is determined by the frame rate of the image processed by the image processing system; and calculating the speed and the acceleration of the target object according to the central position of the target object in the vertical direction at each moment.

Optionally, the determining whether the target object is a high altitude parabola according to the speed and the acceleration of the target object in the tracking process includes: determining the times that the speed of the target object is greater than the target speed from the target images of the preset number of frames to obtain a first time; determining the times that the acceleration of the target object is greater than the target acceleration in the target images of the preset number of frames to obtain a second time; judging whether the first time number is greater than a first target time number or not and whether the second time number is greater than a second target time number or not; and under the condition that the first times are greater than the first target times and the second times are greater than the second target times, determining that the target object is a high altitude parabola.

Optionally, after determining whether the first number is greater than the first target number of times and the second number is greater than the second target number of times, the method further includes: and determining the target object as a noise object of the target image under the condition that the first frequency is less than the first target frequency or the second frequency is less than the second target frequency.

Optionally, in a case that the first number of times is greater than the first target number of times and the second number of times is greater than the second target number of times, after determining that the target object is a high altitude parabola, the method further includes: storing a video of a first preset time period before the target time and storing a video of a second preset time period after the target time; and after a second preset time period after the target time, re-executing the step of determining the target object, wherein the target time is the time when the target object is determined to be a high altitude parabola.

According to another aspect of the present application, a high altitude parabola determination apparatus is provided. The device includes: the system comprises an acquisition unit, a display unit and a display unit, wherein the acquisition unit is used for acquiring target images corresponding to the target side of a building at different moments to obtain multiple frames of target images; a first determination unit configured to determine a plurality of moving objects corresponding to the target sides based on the plurality of frames of target images; a second determination unit configured to determine a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image; and the tracking unit is used for tracking the target object and judging whether the target object is a high-altitude parabola or not according to the speed and the acceleration of the target object in the tracking process.

In order to achieve the above object, according to another aspect of the present application, there is provided a storage medium including a stored program, wherein the program performs any one of the above-described high altitude parabola determination methods.

In order to achieve the above object, according to another aspect of the present application, there is provided a processor for executing a program, wherein the program executes any one of the above-mentioned high altitude parabola determination methods.

Through the application, the following steps are adopted: acquiring target images corresponding to the target side of a building at different moments to obtain multiple frames of target images; determining a plurality of moving objects corresponding to the side surfaces of the targets based on the multi-frame target images; determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image; the target object is tracked, whether the target object is a high-altitude object is judged according to the speed and the acceleration of the target object in the tracking process, and the problem that the high-altitude object is difficult to accurately detect when the monitoring picture is complex and a naturally moving object exists in the related technology is solved. The target object is determined according to the relevance between the moving object and the noise, and whether the object is a parabola or not is judged according to the speed and the acceleration of the target object, so that the effect of improving the detection accuracy of the high-altitude parabola is achieved when the monitoring picture is complex and a natural moving object exists.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application. In the drawings:

fig. 1 is a flowchart of a method for determining a high altitude parabola according to an embodiment of the present application;

fig. 2 is a flowchart of another high altitude parabola determination method provided in accordance with an embodiment of the present application; and

fig. 3 is a schematic diagram of a high altitude parabola determination apparatus provided according to an embodiment of the present application.

Detailed Description

It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.

In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It should be understood that the data so used may be interchanged under appropriate circumstances such that embodiments of the application described herein may be used. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

According to an embodiment of the application, a method for determining a high altitude parabola is provided.

Fig. 1 is a flow chart of a method of determining a high altitude parabola according to an embodiment of the application. As shown in fig. 1, the method comprises the steps of:

and S101, acquiring target images corresponding to the target side of the building at different moments to obtain multiple frames of target images.

Specifically, the building can be a residential building, the side of a window can be set for the residential building on the side of a target of the building, and target images corresponding to different moments can be acquired in a camera shooting mode to obtain multi-frame target images.

Optionally, in the method for determining a high altitude parabola provided in the embodiment of the present application, obtaining target images corresponding to target sides of a building at different times, and obtaining multiple frames of target images includes: and respectively shooting the target side of the building by adopting a plurality of cameras at the target moment to obtain a plurality of sub-images, and splicing the sub-images to obtain a frame of target image corresponding to the target moment.

It should be noted that, in an actual scene, a single-path camera often cannot cover a building with a high height and a wide width, and the application adopts a plurality of cameras to shoot the target side of the building, specifically, a group of M × N matrix cameras can be used to monitor the building from bottom to top, and each path of camera is responsible for monitoring different areas of the building, and meanwhile, the privacy of residents is also protected by the monitoring method from bottom to top.

For example, for a residential building with a height of 100 meters and a width of 50 meters, a matrix camera with a height of 3 × 2 can be adopted, 6 threads are correspondingly started, the collected frame images are pressed into a collection queue in a mode of controlling the starting of the threads by using signal quantity, and if the resolutions of 6 paths of videos are all W × H, an image stitching algorithm is adopted as f₂Splicing the 6 sub-images to obtain a spliced target image P: p ═ f₂(P₁₁，P₁₂，P₂₁，P₂₂，P₃₁，P₃₂) Wherein P is_ijRepresenting the camera images of the ith row and the jth column, and the resolution of the spliced image P is W_p*H_p。

Through this application embodiment, to the problem that single camera can't cover whole building body, adopt the mode of many cameras joint monitoring to obtain a plurality of subimages to adopt image mosaic technique to fuse a plurality of subimages into a picture, obtain the whole control picture at every moment, realize the control to whole scene, thereby reduce the calculated amount during tracking, reduce tracking module's calculation consumption.

Step S102, a plurality of moving objects corresponding to the side surfaces of the targets are determined based on the multi-frame target images.

Specifically, all moving objects in the image P are extracted by a moving object detection algorithm to obtain a plurality of moving objects, which form a moving object set, and any one of the moving object detection algorithms may be used, such as a Vibe algorithm, a GMM algorithm (gaussian mixture model), a background subtraction method, a frame difference method, and the like.

In order to filter moving objects in the target image that probably do not belong to a parabola, optionally, in the method for determining a high altitude parabola provided in the embodiment of the present application, before determining the target object in the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image, the method further includes: filtering moving objects with the size smaller than a preset size from a plurality of moving objects in each frame of target image; and/or filtering moving objects, which are smaller than a preset distance from the edge of the target image, in the plurality of moving objects in each frame of target image.

Specifically, let a set of moving objects consisting of a plurality of moving objects be Object, Object_iRepresents the i-th moving object,

x coordinate representing the upper left corner of the moving object,

The upper left y coordinate representing the moving object,

Indicates the width of the moving object,

Representing the height of the moving object. Filtering out the edge Object and the tiny Object in the Object according to the following formula, and enabling S_iRepresents Object_iWhether filtered or not, 1 means filtered, 0 means not filtered:

wherein the content of the first and second substances,

indicating that the distance of the moving object from the left edge of the target image is less than a first threshold,

indicating that the distance of the moving object from the right edge of the target image is less than a first threshold,

indicating that the distance of the moving object from the upper edge of the target image is less than a first threshold,

indicating that the distance of the moving object from the lower edge of the target image is less than a first threshold,

indicating that the width of the moving object is less than the second threshold value,

Indicating that the width of the moving object is less than the second threshold.

In summary, the set of moving objects obtained after filtering is T arg et, T arg et_iRepresents the i-th moving object,

and

the height of the motion object is respectively represented by the x coordinate at the upper left corner, the y coordinate at the upper left corner, the height and the width of the motion object.

By the embodiment of the application, the moving object and the micro moving object at the edge position in the target image are filtered, so that the calculation amount of parabolic analysis is reduced.

Step S103 determines a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image.

It should be noted that the high-altitude parabolic algorithm in the related art is generally a multi-target tracking algorithm, but the multi-target tracking algorithm has a poor effect, and in order to improve the tracking effect, the embodiment of the application tracks the moving object by using a single-target tracking algorithm. When a single-target tracking algorithm is used, when there are many moving targets in a monitoring picture, a suitable target object needs to be selected for tracking.

Optionally, in the method for determining a high altitude parabola provided in the embodiment of the present application, determining a target object in a plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of a target image includes: respectively calculating the intersection ratio between each moving object in a frame of target image and each noise object of the frame of target image to obtain a plurality of intersection ratios of each moving object; determining the maximum intersection ratio of each moving object as the target intersection ratio of the moving object; and taking the moving object with the smallest target intersection ratio in the plurality of moving objects as the target object.

Specifically, a set of moving objects of a frame of target image is set to T arg et, and a set of Noise of the frame of target image is Noise, Noise_jRepresenting j noise, and determining a k-th moving object in the moving object set as a target object according to the following formula:

wherein, pro_ijThe correlation between the ith moving object and the jth noise can be represented by Tarjet_kAnd Noise_jCross-over ratio of (IOU) max pro_ijRepresenting the maximum intersection of each moving object, namely the target intersection ratio, representing the maximum probability that each moving object is possible to be noise, and the moving object T arg et with the minimum target intersection ratio among a plurality of moving objects_kIs determined to be the target object, i.e. the object most likely to be a parabola.

Through the embodiment, IOUs of all moving objects and all noises are calculated, and the moving object with the minimum maximum IOU is selected as the tracking target, so that the resource consumption of a system is reduced, and the accuracy of parabolic detection is improved.

And step S104, tracking the target object, and judging whether the target object is a high-altitude parabola or not according to the speed and the acceleration of the target object in the tracking process.

It should be noted that the system can acquire the historical motion position of the target object at each moment, so as to analyze whether the object is parabolic, because in practice, the motion of the object is complex and is influenced by the view angle of the camera, and if the least square method is used to continue the judgment of the object, the detection omission is easily caused.

According to the method for determining the high-altitude parabola, multiple frames of target images are obtained by obtaining the target images corresponding to the target side of the building at different moments; determining a plurality of moving objects corresponding to the side surfaces of the targets based on the multi-frame target images; determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image; the target object is tracked, whether the target object is a high-altitude object is judged according to the speed and the acceleration of the target object in the tracking process, and the problem that the high-altitude object is difficult to accurately detect when the monitoring picture is complex and a naturally moving object exists in the related technology is solved. The target object is determined according to the relevance between the moving object and the noise, and whether the object is a parabola or not is judged according to the speed and the acceleration of the target object, so that the effect of improving the detection accuracy of the high-altitude parabola is achieved when the monitoring picture is complex and a natural moving object exists.

In order to improve the tracking accuracy, optionally, in the determination method of the high altitude parabola provided by the embodiment of the application, the tracking the target object includes: taking the target object as a tracking target of a tracker, and acquiring a tracking result, wherein the tracking result comprises the tracking target and the tracking position of the tracker to the tracking target at each moment; calculating the intersection ratio between the tracking result at the current moment and the target object in the target image at the current moment to obtain a first intersection ratio; and adjusting the tracking object of the tracker based on the relation between the first intersection ratio and a preset threshold value.

After the target object is obtained, the tracker is initialized by using the target object as a tracking target, and tracking is performed after initialization. In the tracking process of the tracker, the tracking result is inaccurate, so that the tracking result is required to be continuously repositioned and the tracker is required to be reinitialized, specifically, the intersection and parallel ratio between the target object at the tracking position and the target object in the target image at the current moment is calculated, whether tracking abnormality occurs or not is determined according to the intersection and parallel ratio, and after the tracking abnormality occurs, the tracker is reinitialized by using the moving object which is most matched with the tracking result.

Optionally, in the method for determining a high altitude parabola, the adjusting a tracked object of the tracker based on a relationship between the first intersection ratio and a preset threshold includes: judging whether the first intersection ratio is smaller than a first preset threshold value or not; determining a tracker tracking error under the condition that the first intersection ratio is smaller than a first preset threshold value; under the condition that the first intersection ratio is greater than a first preset threshold value, judging whether the first intersection ratio is smaller than a second preset threshold value; determining that the tracking target deforms under the condition that the first intersection ratio is larger than a first preset threshold value and the first intersection ratio is smaller than a second preset threshold value; under the condition that the tracking target deforms, calculating the intersection ratio between the tracking result at the current moment and a plurality of moving objects in the target image at the current moment to obtain a plurality of intersection ratios, and updating the moving object with the largest intersection ratio as the tracking object.

It should be noted that, if the tracking result of the target object tracked by the tracker is Pos, the correlation between the tracking result and the target object may be Pos and T arg et_iCross over ratio between PT_iIs shown and according to PT_iAnd judging the tracking condition of the tracker.

In particular, at PT_iWhen the tracking error is smaller than the first preset threshold, determining that the tracker has a tracking error, and directly judging that the tracking is finished; at PT_iAnd when the target deformation is larger than the first preset threshold and smaller than the second preset threshold, determining that the tracking target deforms, and the tracker is about to fail, so that the tracking target needs to be determined again. Specifically, intersection ratios between the tracking result at the current moment and a plurality of moving objects in the target image at the current moment are respectively calculated to obtain a plurality of intersection ratios, and the moving object with the largest intersection ratio is determined to be T arg et_lAt this time, T arg et is used_lThe tracker is reinitialized and tracking is performed after initialization is complete.

According to the embodiment of the application, the intersection and parallel ratio of all moving objects and the current tracking result is calculated, and the current tracking result is repositioned or interrupted to track according to the calculated result, so that the tracking accuracy is improved.

Optionally, in the method for determining a high altitude parabola, before determining whether the target object is a high altitude parabola according to the speed and the acceleration of the target object in the tracking process, the method further includes: determining the central position of the target object in the vertical direction at each moment after the target object is tracked for a target time period, wherein the target time period is determined by the frame rate of the image processed by the image processing system; and calculating the speed and the acceleration of the target object according to the central position of the target object in the vertical direction at each moment.

In the case of performing the trajectory parabolic analysis, the parabolic analysis may be started not only after the tracking is finished, but also during the tracking, and specifically, when one target is tracked for T frames, the parabolic analysis may be started to obtain a better analysis result, where T is ma x (T/2,15), where T represents the frame rate of the system processing.

Starting parabolic analysis after a target object is tracked in a target time period, specifically, calculating the central point of the target object in the vertical direction at each moment, eliminating the influence of the volume of the object, and setting historical motion positions as Tra ck and Tra ck_iIndicates the position of the ith time instant,

and

respectively representing the coordinate of the upper left corner X, the coordinate of the upper left corner y, the width and the height of the target object at the ith moment, and calculating the central position of the target object in the vertical direction at the ith moment as follows:

further, after the central point of the target object at each moment in the vertical direction is obtained, the target pair at each moment is calculatedVertical velocity and vertical acceleration of the image, wherein the vertical velocity of the target object at the ith time is v_i＝Center_i-Center_i-1The vertical acceleration of the target object at the ith moment is A_i＝v_i-v_i-1。

By the embodiment of the application, the parabolic analysis is performed in an interrupted mode in the tracking process, the historical motion position of the object is obtained, whether the object is parabolic or not is judged in real time, the tracking end does not need to be waited, invalid tracking is reduced, and resource consumption is reduced.

In addition, the speed and the acceleration of the object on the pixel point at each moment are calculated, whether the object is a parabola or not is judged according to the speed and the acceleration, adaptability to complex motion in a natural environment is improved, resistance to noise is improved, and accuracy of judging the parabola is improved.

Optionally, in the method for determining a high-altitude parabola provided in the embodiment of the present application, determining whether the target object is a high-altitude parabola according to the speed and the acceleration of the target object in the tracking process includes: determining the times that the speed of the target object is greater than the target speed from the target images of the preset number of frames to obtain a first time; determining the times that the acceleration of the target object is greater than the target acceleration in the target images of the preset number of frames to obtain a second time; judging whether the first time number is greater than a first target time number or not and whether the second time number is greater than a second target time number or not; and under the condition that the first times are greater than the first target times and the second times are greater than the second target times, determining that the target object is a high altitude parabola.

It should be noted that if the vertical velocity and the vertical acceleration of the object at most of the time are downward, the object can be directly determined to be in a parabolic motion, specifically, the object is made to move in a parabolic manner_FobjectWhether the target object is a parabola or not is represented by 1, which represents a parabola, and 0, which represents not a parabola, and the determination is made as follows:

wherein count (V > 3) represents a preset numberThe velocity of the target object in the target image of the frame is greater than the number of times of 3 pixels per second, which is greater than the product of the third preset value eps3 and the number of frames, indicating that the target object has a downward velocity, wherein eps3 may be 0.75; count (a > 2) represents the number of times the acceleration of the object is greater than 2, and in the case where the number of times is greater than the product of the fourth preset value eps4 and the number of frames, it represents that the target object has a downward speed, where eps4 may be 0.5, T_track> 5 indicates that the number of frames of the target image is greater than 5.

Optionally, in the method for determining a high altitude parabola, after determining whether the first number of times is greater than the first target number of times and the second number of times is greater than the second target number of times, the method further includes: and determining the target object as a noise object of the target image under the condition that the first frequency is less than the first target frequency or the second frequency is less than the second target frequency.

Specifically, in a case where it is determined that the target object is not a parabola, the target object is added as Noise to a Noise set Noise, where the position of the Noise is the initial position of the target object.

It should be noted that the noise set has a maximum number limit, for example, 30, and if the number is exceeded, the noise with the weakest weight is removed from the set, and in the embodiment of the present application, the weight may be set to be related to the time of adding to the set, specifically, the noise added earlier to the noise set is set to be smaller in weight, and the noise added later is set to be larger in weight.

Optionally, in the method for determining a high altitude parabola, when the first number of times is greater than the first target number of times and the second number of times is greater than the second target number of times, after determining that the target object is a high altitude parabola, the method further includes: storing a video of a first preset time period before the target time and storing a video of a second preset time period after the target time; and after a second preset time period after the target time, re-executing the step of determining the target object, wherein the target time is the time when the target object is determined to be a high altitude parabola.

For example, after the object is judged at a certain moment and is parabolic, the tracker is silent for 5s, videos 5s before and after the moment are stored, a complete video of the motion of the object can be obtained, and a new target object is determined after 5 seconds.

It should be noted that, when the object is determined to be a parabolic object at the target time, the video of the first preset time period before the target time and the video of the second preset time period after the target time are stored, so that the integrity of the stored parabolic video segment can be ensured, and specifically, the first preset time period and the second preset time period may both be the same

Video in seconds, Height represents the building Height, and g represents the acceleration of gravity. In addition, the step of determining a new target object after a second preset period of time after the target time can reduce system resource consumption.

Fig. 2 is a flow chart of another high altitude parabola determination method according to an embodiment of the application. As shown in fig. 2, the method mainly includes image acquisition, moving object extraction, parabolic analysis and determination, and video storage.

Specifically, the system configuration of the visual detection and tracking system is read firstly, then image acquisition is carried out, all cameras are opened in the image acquisition process, the side face of a building is shot, multiple paths of videos are read in the shooting process, multiple video images at each moment are spliced to obtain a complete monitoring picture at each moment, moving target extraction is carried out in the spliced images, and meanwhile, the original videos are stored.

Further, after the moving target is extracted, whether the tracker is currently in a tracking state is judged, whether a suspicious target exists in the monitoring picture is judged under the condition that the tracker is not currently in the tracking state, the suspicious target is selected under the condition that the suspicious target exists, and tracking target initialization is carried out according to the suspicious target. Under the condition that the tracker is in a tracking state currently, updating of a tracking target and correction of a tracking result are continuously carried out in the tracking process, whether tracking is finished or not is judged in the correction process of the tracking result, whether the tracking target is a parabola or not is judged under the condition that the tracking is not finished, a parabola video is stored under the condition that the tracking target is the parabola, the tracking target is counted as noise under the condition that the tracking target is not the parabola, and the tracking is finished

According to the method and the device, high-quality detection of the high-altitude parabolic object is completed through modes of multi-camera video synchronous acquisition, moving object extraction, moving target selection, tracker initialization, tracking result correction, parabolic analysis, video storage and the like, and the problems that in a vision-based high-altitude parabolic object detection algorithm in the related technology, a monitoring area is small, a tracking result is inaccurate, parabolic analysis robustness is poor, accuracy is low, omission ratio is high, intercepted videos are incomplete and the like are solved.

It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than presented herein.

The embodiments of the present application further provide a device for determining a high altitude parabola, and it should be noted that the device for determining a high altitude parabola according to the embodiments of the present application may be used to execute the method for determining a high altitude parabola provided by the embodiments of the present application. The following describes a device for determining a high altitude parabola according to an embodiment of the present application.

Fig. 3 is a schematic diagram of a high altitude parabola determination apparatus according to an embodiment of the present application. As shown in fig. 3, the apparatus includes: an acquisition unit 10, a first determination unit 20, a second determination unit 30 and a tracking unit 40.

Specifically, the acquiring unit 10 is configured to acquire target images corresponding to target sides of a building at different times to obtain multiple frames of target images;

a first determination unit 20 configured to determine a plurality of moving objects corresponding to the target sides based on the plurality of frames of target images;

a second determination unit 30 for determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image;

and the tracking unit 40 is used for tracking the target object and judging whether the target object is a high-altitude parabola or not according to the speed and the acceleration of the target object in the tracking process.

According to the device for determining the high-altitude parabola, the target images of the target side face of the building corresponding to different moments are obtained through the obtaining unit 10, and multiple frames of target images are obtained; the first determination unit 20 determines a plurality of moving objects corresponding to the target sides based on the plurality of frames of target images; the second determination unit 30 determines a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image; the tracking unit 40 tracks the target object and judges whether the target object is a high-altitude parabola according to the speed and the acceleration of the target object in the tracking process, so that the problem that the high-altitude parabola is difficult to accurately detect when the monitoring picture is complex and a natural moving object exists in the related art is solved, the target object is determined according to the relevance between the moving object and noise, and whether the target object is the parabola is judged according to the speed and the acceleration of the target object, so that the effect of improving the detection accuracy of the high-altitude parabola is achieved when the monitoring picture is complex and the natural moving object exists.

Optionally, in the apparatus for determining a high altitude parabola provided in the embodiment of the present application, the obtaining unit 10 includes: the device comprises a shooting module and a splicing module, wherein the shooting module is used for respectively shooting the side surface of a target of a building by adopting a plurality of cameras at the target moment to obtain a plurality of sub-images, and the splicing module is used for splicing the sub-images to obtain a frame of target image corresponding to the target.

Optionally, in the device for determining a high altitude parabola provided in the embodiment of the present application, the device further includes: the first filtering unit is used for filtering moving objects with the size smaller than the preset size in each frame of target images before determining the target objects in the moving objects based on the relationship between the moving objects and the noise objects of the target images; and/or the second filtering unit is used for filtering moving objects, of which the edges to the target image are smaller than a preset distance, in each frame of target image before the target object is determined in the moving objects based on the relationship between the moving objects and the noise object of the target image.

Optionally, in the high altitude parabola determination apparatus provided in the embodiment of the present application, the second determination unit 20 includes: the calculation module is used for respectively calculating the intersection ratio between each moving object in a frame of target image and each noise object of the frame of target image to obtain a plurality of intersection ratios of each moving object; the first determination module is used for determining the maximum intersection ratio of each moving object as the target intersection ratio of the moving object; and the second determining module is used for taking the moving object with the minimum target intersection ratio in the plurality of moving objects as the target object.

Optionally, in the determination apparatus for a high altitude parabola provided in the embodiment of the present application, the tracking unit 40 includes: the acquisition module is used for taking the target object as a tracking target of the tracker and acquiring a tracking result, wherein the tracking result comprises the tracking target and the tracking position of the tracker to the tracking target at each moment; the calculation module is used for calculating the intersection ratio between the tracking result at the current moment and the target object in the target image at the current moment to obtain a first intersection ratio; and the adjusting module is used for adjusting the tracking object of the tracker based on the relation between the first intersection ratio and a preset threshold value.

Optionally, in the determination apparatus for a high altitude parabola provided in this embodiment of the present application, the adjustment module includes: the first judgment submodule is used for judging whether the first intersection ratio is smaller than a first preset threshold value or not; the first determining submodule is used for determining the tracking error of the tracker under the condition that the first cross-over ratio is smaller than a first preset threshold value; the second judgment submodule is used for judging whether the first cross-over ratio is smaller than a second preset threshold value or not under the condition that the first cross-over ratio is larger than the first preset threshold value; the second determining submodule is used for determining that the tracking target deforms under the condition that the first intersection ratio is larger than a first preset threshold value and the first intersection ratio is smaller than a second preset threshold value; and the calculation submodule is used for calculating the intersection ratio between the tracking result at the current moment and the plurality of moving objects in the target image at the current moment under the condition that the tracking target deforms to obtain a plurality of intersection ratios, and updating the moving object with the largest intersection ratio as the tracking object.

Optionally, in the device for determining a high altitude parabola provided in the embodiment of the present application, the device further includes: a third determining unit, configured to determine a central position of the target object in a vertical direction at each time after a target time period after the target object is tracked before determining whether the target object is a high altitude parabola according to a velocity and an acceleration of the target object during the tracking process, where the target time period is determined by a frame rate at which the image processing system processes the image; and the calculating unit is used for calculating the speed and the acceleration of the target object according to the central position of the target object in the vertical direction at each moment.

Optionally, in the device for determining a high altitude parabola provided in the embodiment of the present application, the tracking unit 40 further includes: the third determining module is used for determining the times that the speed of the target object is greater than the target speed from the target images of the preset number of frames to obtain a first time; the fourth determining module is used for determining the times that the acceleration of the target object is greater than the target acceleration in the target images of the preset number of frames to obtain a second time; the judging module is used for judging whether the first time is greater than the first target time and whether the second time is greater than the second target time; and the fifth determining module is used for determining that the target object is a high altitude parabola under the condition that the first time number is greater than the first target time number and the second time number is greater than the second target time number.

Optionally, in the device for determining a high altitude parabola provided in the embodiment of the present application, the device further includes: and a fourth determining unit, configured to determine the target object as a noise object of the target image if the first count is less than the first target count or the second count is less than the second target count after determining whether the first count is greater than the first target count and the second count is greater than the second target count.

Optionally, in the device for determining a high altitude parabola provided in the embodiment of the present application, the device further includes: the storage unit is used for storing a video of a first preset time period before the target moment and a video of a second preset time period after the target object is determined to be a high altitude parabola under the condition that the first time number is greater than the first target time number and the second time number is greater than the second target time number; and the execution unit is used for re-executing the step of determining the target object after a second preset time period after the target moment, wherein the target moment is the moment when the target object is determined to be a high altitude parabola.

The device for determining the high altitude parabola comprises a processor and a memory, wherein the acquiring unit 10, the first determining unit 20, the second determining unit 30, the tracking unit 40 and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.

The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more than one, and the problem that the high-altitude object throwing is difficult to accurately detect when the monitoring picture is complex and the object with natural motion exists in the related technology is solved by adjusting the kernel parameters.

The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.

An embodiment of the present invention provides a storage medium on which a program is stored, which when executed by a processor implements the method for determining a high altitude parabola.

The embodiment of the invention provides a processor, which is used for running a program, wherein the program executes the high altitude parabola determination method during running.

The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps: acquiring target images corresponding to the target side of a building at different moments to obtain multiple frames of target images; determining a plurality of moving objects corresponding to the side surfaces of the targets based on the multi-frame target images; determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image; and tracking the target object, and judging whether the target object is a high-altitude parabola according to the speed and the acceleration of the target object in the tracking process.

Acquiring target images corresponding to the target side of the building at different moments, wherein the acquiring of the multi-frame target images comprises the following steps: and respectively shooting the target side of the building by adopting a plurality of cameras at the target moment to obtain a plurality of sub-images, and splicing the sub-images to obtain a frame of target image corresponding to the target moment.

Before determining the target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image, the method further includes: filtering moving objects with the size smaller than a preset size from a plurality of moving objects in each frame of target image; and/or filtering moving objects, which are smaller than a preset distance from the edge of the target image, in the plurality of moving objects in each frame of target image.

Determining the target object among the plurality of moving objects based on a relationship between the plurality of moving objects and the noise object of the target image includes: respectively calculating the intersection ratio between each moving object in a frame of target image and each noise object of the frame of target image to obtain a plurality of intersection ratios of each moving object; determining the maximum intersection ratio of each moving object as the target intersection ratio of the moving object; and taking the moving object with the smallest target intersection ratio in the plurality of moving objects as the target object.

Tracking the target object includes: taking the target object as a tracking target of a tracker, and acquiring a tracking result, wherein the tracking result comprises the tracking target and the tracking position of the tracker to the tracking target at each moment; calculating the intersection ratio between the tracking result at the current moment and the target object in the target image at the current moment to obtain a first intersection ratio; and adjusting the tracking object of the tracker based on the relation between the first intersection ratio and a preset threshold value.

Based on the relationship between the first intersection ratio and the preset threshold, adjusting the tracked object of the tracker comprises: judging whether the first intersection ratio is smaller than a first preset threshold value or not; determining a tracker tracking error under the condition that the first intersection ratio is smaller than a first preset threshold value; under the condition that the first intersection ratio is greater than a first preset threshold value, judging whether the first intersection ratio is smaller than a second preset threshold value; determining that the tracking target deforms under the condition that the first intersection ratio is larger than a first preset threshold value and the first intersection ratio is smaller than a second preset threshold value; under the condition that the tracking target deforms, calculating the intersection ratio between the tracking result at the current moment and a plurality of moving objects in the target image at the current moment to obtain a plurality of intersection ratios, and updating the moving object with the largest intersection ratio as the tracking object.

Before judging whether the target object is a high altitude parabola according to the speed and the acceleration of the target object in the tracking process, the method further comprises the following steps: determining the central position of the target object in the vertical direction at each moment after the target object is tracked for a target time period, wherein the target time period is determined by the frame rate of the image processed by the image processing system; and calculating the speed and the acceleration of the target object according to the central position of the target object in the vertical direction at each moment.

Judging whether the target object is a high-altitude parabola according to the speed and the acceleration of the target object in the tracking process comprises the following steps: determining the times that the speed of the target object is greater than the target speed from the target images of the preset number of frames to obtain a first time; determining the times that the acceleration of the target object is greater than the target acceleration in the target images of the preset number of frames to obtain a second time; judging whether the first time number is greater than a first target time number or not and whether the second time number is greater than a second target time number or not; and under the condition that the first times are greater than the first target times and the second times are greater than the second target times, determining that the target object is a high altitude parabola.

After determining whether the first number is greater than the first target number of times and the second number is greater than the second target number of times, the method further includes: and determining the target object as a noise object of the target image under the condition that the first frequency is less than the first target frequency or the second frequency is less than the second target frequency.

In the case that the first number of times is greater than the first target number of times and the second number of times is greater than the second target number of times, after determining that the target object is a high altitude parabola, the method further includes: storing a video of a first preset time period before the target time and storing a video of a second preset time period after the target time; and after a second preset time period after the target time, re-executing the step of determining the target object, wherein the target time is the time when the target object is determined to be a high altitude parabola. The device herein may be a server, a PC, a PAD, a mobile phone, etc.

The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device: acquiring target images corresponding to the target side of a building at different moments to obtain multiple frames of target images; determining a plurality of moving objects corresponding to the side surfaces of the targets based on the multi-frame target images; determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image; and tracking the target object, and judging whether the target object is a high-altitude parabola according to the speed and the acceleration of the target object in the tracking process.

In the case that the first number of times is greater than the first target number of times and the second number of times is greater than the second target number of times, after determining that the target object is a high altitude parabola, the method further includes: storing a video of a first preset time period before the target time and storing a video of a second preset time period after the target time; and after a second preset time period after the target time, re-executing the step of determining the target object, wherein the target time is the time when the target object is determined to be a high altitude parabola.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.

The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for determining a high altitude parabola, comprising:

acquiring target images corresponding to the target side of a building at different moments to obtain multiple frames of target images;

determining a plurality of moving objects corresponding to the side surfaces of the targets based on the multi-frame target images;

determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image;

and tracking the target object, and judging whether the target object is a high-altitude parabola according to the speed and the acceleration of the target object in the tracking process.

2. The method of claim 1, wherein obtaining target images corresponding to the target side of the building at different times to obtain multiple frames of target images comprises:

a plurality of cameras are adopted to respectively shoot the target side of the building at the target moment to obtain a plurality of sub-images,

and splicing the plurality of sub-images to obtain a frame of target image corresponding to the target.

3. The method of claim 1, wherein prior to determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noisy object of the target image, the method further comprises:

filtering moving objects with the size smaller than a preset size from the multiple moving objects in each frame of target image; and/or the presence of a gas in the gas,

and filtering moving objects, of which the edges with the target image are smaller than a preset distance, in each frame of target image.

4. The method of claim 1, wherein determining a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noisy object of the target image comprises:

respectively calculating the intersection ratio between each moving object in a frame of target image and each noise object of the frame of target image to obtain a plurality of intersection ratios of each moving object;

determining the maximum intersection ratio of each moving object as the target intersection ratio of the moving object;

and taking the moving object with the smallest target intersection ratio in the plurality of moving objects as the target object.

5. The method of claim 4, wherein tracking the target object comprises:

taking the target object as a tracking target of a tracker, and acquiring a tracking result, wherein the tracking result comprises the tracking target and the tracking position of the tracker to the tracking target at each moment;

calculating the intersection ratio between the tracking result at the current moment and the target object in the target image at the current moment to obtain a first intersection ratio;

and adjusting the tracked object of the tracker based on the relation between the first intersection ratio and a preset threshold value.

6. The method of claim 5, wherein adjusting the tracked object of the tracker based on the relationship between the first intersection ratio and a preset threshold comprises:

judging whether the first intersection ratio is smaller than a first preset threshold value or not;

determining that the tracker tracks an error if the first cross-over ratio is less than the first preset threshold;

under the condition that the first intersection ratio is greater than the first preset threshold, judging whether the first intersection ratio is smaller than a second preset threshold;

determining that the tracking target deforms under the condition that the first intersection ratio is larger than a first preset threshold value and the first intersection ratio is smaller than a second preset threshold value;

and under the condition that the tracking target deforms, calculating the intersection ratio between the tracking result at the current moment and a plurality of moving objects in the target image at the current moment to obtain a plurality of intersection ratios, and updating the moving object with the largest intersection ratio as the tracking object.

7. The method of claim 1, wherein before determining whether the target object is a high altitude parabola based on the velocity and acceleration of the target object during tracking, the method further comprises:

determining a central position of the target object in a vertical direction at each moment after the target object is tracked for a target period, wherein the target period is determined by a frame rate of an image processed by an image processing system;

and calculating the speed and the acceleration of the target object according to the central position of the target object in the vertical direction at each moment.

8. The method of claim 1, wherein determining whether the target object is a high altitude parabola based on the velocity and acceleration of the target object during tracking comprises:

determining the times that the speed of the target object is greater than the target speed from the target images of a preset number of frames to obtain a first time;

determining the times that the acceleration of the target object is greater than the target acceleration in the target images of the preset number of frames to obtain a second time;

judging whether the first time number is greater than a first target time number or not, and whether the second time number is greater than a second target time number or not;

and under the condition that the first times are greater than a first target times and the second times are greater than a second target times, determining that the target object is a high altitude parabola.

9. The method of claim 8, wherein after determining whether the first number of times is greater than a first target number of times and the second number of times is greater than a second target number of times, the method further comprises:

and determining the target object as a noise object of the target image under the condition that the first times are less than the first target times or the second times are less than the second target times.

10. The method of claim 8, wherein after determining that the target object is a high altitude parabola if the first number of times is greater than a first target number of times and the second number of times is greater than a second target number of times, the method further comprises:

saving a video of a first preset time period before a target time, and saving a video of a second preset time period after the target time;

after the second preset time period after the target time, re-executing the step of determining the target object, wherein the target time is a time when the target object is determined to be a high altitude parabola.

11. An apparatus for determining a high altitude parabola, comprising:

the system comprises an acquisition unit, a display unit and a display unit, wherein the acquisition unit is used for acquiring target images corresponding to the target side of a building at different moments to obtain multiple frames of target images;

a first determining unit, configured to determine, based on the multiple frames of target images, multiple moving objects corresponding to the target sides;

a second determination unit configured to determine a target object among the plurality of moving objects based on a relationship between the plurality of moving objects and a noise object of the target image;

and the tracking unit is used for tracking the target object and judging whether the target object is a high-altitude parabola object or not according to the speed and the acceleration of the target object in the tracking process.

12. A storage medium, characterized in that it comprises a stored program, wherein said program performs the method of determining a high altitude parabola according to any one of claims 1 to 10.

13. A processor, characterized in that the processor is configured to run a program, wherein the program when running performs the method of determining a high altitude parabola of any one of claims 1 to 10.