CN114550060A - Perimeter intrusion identification method and system and electronic equipment - Google Patents
Perimeter intrusion identification method and system and electronic equipment Download PDFInfo
- Publication number
- CN114550060A CN114550060A CN202210178171.8A CN202210178171A CN114550060A CN 114550060 A CN114550060 A CN 114550060A CN 202210178171 A CN202210178171 A CN 202210178171A CN 114550060 A CN114550060 A CN 114550060A
- Authority
- CN
- China
- Prior art keywords
- image
- perimeter
- cluster
- pedestrian
- images
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000001514 detection method Methods 0.000 claims abstract description 22
- 238000013528 artificial neural network Methods 0.000 claims abstract description 14
- 230000002093 peripheral effect Effects 0.000 claims description 40
- 230000015654 memory Effects 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 230000008034 disappearance Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 5
- 238000013135 deep learning Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 6
- 238000013527 convolutional neural network Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000009545 invasion Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 238000011176 pooling Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000005672 electromagnetic field Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a perimeter intrusion identification method, a system and electronic equipment, wherein the method comprises the following steps: acquiring an image according to a real-time video stream, and setting a perimeter region on the image; automatically segmenting the image into first image clusters based on the perimeter region; zooming the images in the first image cluster to obtain a second image cluster; and inputting the second image cluster into a neural network, detecting whether continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region, and outputting a detection and identification result. The invention improves the accuracy of detecting the perimeter intrusion at the far sight, reduces the false alarm caused by the perimeter environment interference, achieves the technical effect of reducing the perimeter alarm misjudgment rate, and solves the problems that the small target cannot be identified and the perimeter alarm misjudgment rate is high in the existing perimeter intrusion scheme.
Description
Technical Field
The invention relates to the field of video analysis and identification, in particular to a perimeter intrusion identification method, a perimeter intrusion identification system and electronic equipment.
Background
When a traditional perimeter intrusion scheme detects and identifies a human body, a cable leakage mode or an infrared laser mode is often adopted, and the two modes have the same problem and are easily interfered by the environment; for example, the leaky cable system is easily interfered by metal objects or surrounding electromagnetic fields, and the infrared laser system is easily interfered by environments such as branches, wind, snow, and the like.
When the video-based perimeter intrusion scheme is adopted to detect and identify the human body, the early warning reminding is carried out by judging whether the human body enters the virtual perimeter area. A common perimeter intrusion scheme based on videos mainly identifies a human body through a single picture, cannot identify a small target, easily causes human body identification errors, is high in alarm misjudgment rate, and causes troubles to users.
Aiming at the problems that small targets cannot be identified and the perimeter alarm misjudgment rate is high in the existing perimeter intrusion scheme, an effective solution is not provided at present.
Disclosure of Invention
The invention mainly aims to provide a perimeter intrusion identification method and a system, which are used for solving the problems that small targets cannot be identified and the perimeter alarm misjudgment rate is high in the existing perimeter intrusion scheme.
In order to achieve the above object, a first aspect of the present invention provides a perimeter intrusion identification method, including:
acquiring an image according to a real-time video stream, and setting a perimeter region on the image;
automatically segmenting the image into first image clusters based on the perimeter region;
zooming the images in the first image cluster to obtain a second image cluster;
and inputting the second image cluster into a neural network, detecting whether continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region, and outputting a detection and identification result.
Optionally, the automatically segmenting the image into the first image cluster based on the perimeter region comprises:
configuring a gaze disappearance direction within the perimeter region;
and automatically segmenting the image into a plurality of segments of pixel clusters according to the visual line disappearance direction to obtain a first image cluster.
Optionally, the scaling the image in the first image cluster to obtain a second image cluster includes:
determining a scaling according to the detection scene and the requirement;
and reducing the image of the relatively near end of the visual line in the first image cluster according to the scaling, and amplifying the image of the relatively far end to obtain a second image cluster.
Optionally, the inputting the second image cluster into a neural network, detecting whether a continuous multi-frame image of the second image cluster contains a pedestrian invading the perimeter region, and outputting a detection recognition result includes:
detecting whether the image of the second image cluster contains a pedestrian or not by using a deep learning algorithm;
if the image is detected to contain the pedestrian, background difference is carried out on the area where the pedestrian is detected by using a vibe algorithm, and the intra-area mobility judgment is carried out by judging a difference image.
Further, the method further comprises:
and if no pedestrian is detected in the image, automatically segmenting the image based on the peripheral area again.
Further, the performing a background difference on the detected pedestrian region by using the vibe algorithm, and performing a region mobility determination by determining a difference image includes:
judging whether the pedestrian is in the peripheral area;
if the pedestrian is in the peripheral area, judging whether the pedestrian is in the peripheral area of the continuous multi-frame images and whether the images in the peripheral area have differential changes;
if the pedestrian is in the peripheral area of the continuous multi-frame images and the images in the peripheral area have difference changes, alarm snapshot is carried out; otherwise, whether the pedestrian is in the peripheral area is judged again.
Further, the method further comprises:
and if the pedestrian is not in the peripheral area, automatically segmenting the image based on the peripheral area again.
A second aspect of the present invention provides a perimeter intrusion identification system, comprising:
the setting unit is used for acquiring an image according to a real-time video stream and setting a perimeter region on the image;
a segmentation unit for automatically segmenting the image into first image clusters based on the perimeter region;
the zooming unit is used for zooming the images in the first image cluster to obtain a second image cluster;
and the detection unit is used for inputting the second image cluster into a neural network, detecting whether the continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region or not, and outputting a detection identification result.
A third aspect of the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the perimeter intrusion identification method provided in any one of the first aspects.
A fourth aspect of the present invention provides an electronic apparatus, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to cause the at least one processor to perform the perimeter intrusion identification method provided in any one of the first aspect.
In the perimeter intrusion identification method provided by the embodiment of the invention, the images in the first image cluster are zoomed, so that the proportion difference of pedestrians in different areas is reduced, small targets are accurately identified, and the accuracy rate of detecting perimeter intrusion at a far sight line is improved;
whether the continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region or not is detected by inputting the second image cluster into a neural network, false alarm caused by perimeter environment interference is reduced, the technical effect of reducing perimeter alarm misjudgment rate is achieved, and the problems that small targets cannot be identified and the perimeter alarm misjudgment rate is high in the existing perimeter invasion scheme are solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic flow chart of a perimeter intrusion identification method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a perimeter region and a direction of disappearance provided by an embodiment of the invention;
FIG. 3 is a flowchart of a pedestrian detection method based on Faster R-CNN according to an embodiment of the present invention;
FIG. 4 is a flowchart of a perimeter intrusion identification method according to an embodiment of the present invention;
FIG. 5 is a block diagram of a perimeter intrusion identification system according to an embodiment of the present invention;
fig. 6 is a block diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the 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 invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention 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 is to be understood that the data so used may be interchanged under appropriate circumstances in order to facilitate the description of the embodiments of the invention herein. 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.
When the video-based perimeter intrusion scheme is adopted to detect and identify the human body, the early warning reminding is carried out by judging whether the human body enters the virtual perimeter area. The common perimeter intrusion scheme based on videos mainly identifies a human body through a single picture, cannot identify a small target, easily causes human body identification errors, has high alarm misjudgment rate and causes troubles to users. The perimeter intrusion identification method provided by the invention is suitable for scenes including an agricultural and pasture safety management scene, such as a pasture biological safety management scene, and solves the problems that the pasture perimeter alarm misjudgment rate is high and small targets cannot be identified.
The perimeter intrusion identification method provided by the embodiment of the invention is applied to edge computing equipment, and as shown in fig. 1, the method comprises the following steps S101 to S104:
step S101: acquiring an image according to a real-time video stream, and setting a perimeter region on the image;
the method comprises the steps of presetting a camera, acquiring a real-time video stream acquired by the camera through a Web page, then acquiring an image according to the real-time video stream, and dividing a perimeter area on the image according to requirements.
Step S102: automatically segmenting the image into a first image cluster based on the perimeter region; in order to accurately detect the pedestrian invasion condition on the perimeter area, the image is automatically segmented to generate a plurality of sub-images, and the detection rate is improved by detecting each sub-image one by one.
Specifically, the step S102 includes:
configuring a gaze disappearance direction within the perimeter region; a schematic diagram of the perimeter region and the direction of line of sight disappearance provided by an embodiment of the invention is shown in fig. 2.
And automatically segmenting the image into a plurality of segments of pixel clusters according to the visual line disappearance direction to obtain a first image cluster.
Step S103: zooming the images in the first image cluster to obtain a second image cluster;
specifically, the step S103 includes:
determining a scaling according to the detection scene and the requirement;
the scale of zoom-in/zoom-out is determined according to different detection scenarios and requirements: under the condition that a camera lens and a CCD (Charge Coupled Device) are the same, if the distance needing to be detected is longer, the number of images needing to be segmented is larger, and the scaling ratio is larger; when the distance and the CCD are the same, if the focal length of the lens is shorter, the number of the images needing to be sliced is larger, and the scaling is larger.
And according to the principle of near-far and near-far, reducing the image of the visual line relative to the near end in the first image cluster according to the scaling, and amplifying the image relative to the far end to obtain a second image cluster.
According to the method, the images in the first image cluster are zoomed, the near-end image is reduced, the far-end image is enlarged, the proportion difference of pedestrians in different areas is reduced, small targets are accurately identified, and the accuracy rate of detecting perimeter intrusion at the far sight is improved.
Step S104: and inputting the second image cluster into a neural network, detecting whether continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region, and outputting a detection and identification result.
Whether the continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region or not is detected by inputting the second image cluster into a neural network, false alarm caused by perimeter environment interference is reduced, the technical effect of reducing perimeter alarm misjudgment rate is achieved, and the problems that small targets cannot be identified and the perimeter alarm misjudgment rate is high in the existing perimeter invasion scheme are solved.
Specifically, the step S104 includes:
detecting whether the image of the second image cluster contains a pedestrian or not by using a deep learning algorithm; the deep learning algorithm can be a target detection algorithm, and comprises any one of R-CNN, Fast R-CNN, Faster R-CNN, SSD, YOLO v1, YOLO v2, YOLO v4 and YOLO v 5.
A flow chart of the pedestrian detection method based on the Faster R-CNN adopted in the embodiment of the present invention is shown in fig. 3, wherein image is an image, conv layers are convolutional layers, feature maps are feature maps, and a Region pro-social Network (RPN) is a Region generation Network or a Region candidate Network for extracting a candidate frame; the ROI posing is Region-of-Interest pooling, is used for a neural network layer of a target detection task, and performs pooling operation on the ROI (Region of Interest), wherein the operation aims to obtain output feature maps with fixed sizes by using a pooling method for ROIs with different sizes in the input feature maps; classifier is a classifier;
structurally, feature extraction, forward extraction, bounding box regression (rectangle regression) and classification of feature extraction have been integrated into one network by fast R-CNN, so that the comprehensive performance is greatly improved, and the detection speed is particularly obvious.
And if no pedestrian is detected in the image, automatically segmenting the image based on the peripheral area again. That is, if no pedestrian is included in the image, the step of automatically segmenting the image based on the peripheral region is returned to.
If the image is detected to contain the pedestrian, background difference is carried out on the area where the pedestrian is detected by using a vibe algorithm, and the intra-area mobility judgment is carried out by judging a difference image. Background difference operation is carried out by using a vibe algorithm, samples of pixels needing to be replaced are randomly selected, and neighborhood pixels are randomly selected for updating; and when the model of the pixel change cannot be determined, the strategy is randomly updated, and the uncertainty of the pixel change is simulated.
The principle of the method is that a sample set of pixel points is established by extracting pixel values around the pixel points (x, y) and previous pixel values, then the pixel values at another frame (x, y) are compared with the pixel values in the sample set, if the distance between the pixel values in the sample set and the pixel values in another frame (x, y) is larger than a preset threshold value, the pixel points are regarded as foreground pixel points, and if not, the pixel points are background pixel points.
The method for carrying out background difference on the areas where the pedestrians are detected by using the vibe algorithm and carrying out intra-area mobility judgment by judging the difference images comprises the following steps:
judging whether the pedestrian is in the peripheral area;
and if the pedestrian is not in the peripheral area, automatically segmenting the image based on the peripheral area again. That is, if the pedestrian is not within the perimeter region, return is made to the step of automatically segmenting the image based on the perimeter region.
If the pedestrian is in the peripheral area, judging whether the pedestrian is in the peripheral area of the continuous multi-frame images and whether the images in the peripheral area have differential changes;
if the pedestrian is in the peripheral area of the continuous multi-frame images and the images in the peripheral area have difference changes, alarm snapshot is carried out;
otherwise, whether the pedestrian is in the peripheral area is judged again. That is, if the pedestrian does not appear in the peripheral area of the consecutive multi-frame images, or there is no differential change in the images within the peripheral area, the process returns to the step of determining whether the pedestrian is within the peripheral area.
The method has the advantages that background difference is carried out on the detected pedestrian area by using the vibe algorithm, if the pedestrian is in the peripheral area of the continuous multi-frame images and the difference change exists in the images in the peripheral area, alarm snapshot is carried out, false alarm caused by peripheral environment interference is reduced, the technical effect of reducing the peripheral alarm misjudgment rate is achieved, and the problems that small targets cannot be identified and the peripheral alarm misjudgment rate is high in the existing peripheral intrusion scheme are solved.
A flowchart of the perimeter intrusion identification method provided by the embodiment of the present invention is shown in fig. 4, wherein the method sequentially includes the following steps:
setting a perimeter area;
according to the set perimeter region and the principle of the near, the far and the near, the image is automatically divided into regions and is cut into picture clusters;
the near-end image is reduced, the far-end image is enlarged, and the proportion difference of pedestrians in different areas is reduced;
inputting the processed picture cluster into a neural network for calculation;
judging whether a pedestrian is detected and identified; if not, the automatic segmentation area is cut into the picture clusters again;
if yes, judging whether the pedestrian is in the perimeter area; if not, the automatic segmentation area is cut into the picture clusters again;
if yes, further judging whether multiple frames are continuously detected in the area and whether the image in the target area has difference change;
if yes, carrying out alarm snapshot; if not, whether the pedestrian is in the perimeter area is judged again.
From the above description, it can be seen that the present invention achieves the following technical effects:
according to the method, the images in the first image cluster are zoomed, the near-end images are reduced, the far-end images are enlarged, the proportion difference of pedestrians in different areas is reduced, small targets are accurately identified, and the accuracy of detecting perimeter intrusion of far sight is improved;
whether the continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region or not is detected by inputting the second image cluster into the neural network, false alarm caused by perimeter environment interference is reduced, the technical effect of reducing perimeter alarm misjudgment rate is achieved, and the problems that small targets cannot be identified and the perimeter alarm misjudgment rate is high in the existing perimeter invasion scheme 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 here.
An embodiment of the present invention further provides a perimeter intrusion identification system for implementing the perimeter intrusion identification method, as shown in fig. 5, the system includes:
a setting unit 51, configured to obtain an image according to a real-time video stream, and set a perimeter region on the image;
a segmentation unit 52 for automatically segmenting the image into first image clusters based on the perimeter region;
a scaling unit 53, configured to scale an image in the first image cluster to obtain a second image cluster;
and the detection unit 54 is configured to input the second image cluster into a neural network, detect whether a continuous multi-frame image of the second image cluster contains a pedestrian invading the perimeter region, and output a detection identification result.
An embodiment of the present invention further provides an electronic device, as shown in fig. 6, the electronic device includes one or more processors 61 and a memory 62, where one processor 61 is taken as an example in fig. 6.
The controller may further include: an input device 63 and an output device 64.
The processor 61, the memory 62, the input device 63 and the output device 64 may be connected by a bus or other means, as exemplified by the bus connection in fig. 6.
The Processor 61 may be a Central Processing Unit (CPU), the Processor 61 may also be other general-purpose processors, Digital Signal Processors (DSP), Application Specific Integrated Circuits (ASIC), Field Programmable Gate Arrays (FPGA), other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or any combination thereof, and the general-purpose Processor may be a microprocessor or any conventional Processor.
The memory 62, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the control methods in the embodiments of the present invention. The processor 61 executes various functional applications of the server and data processing by running non-transitory software programs, instructions and modules stored in the memory 62, namely, implements the perimeter intrusion identification method of the above-described method embodiment.
The memory 62 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of a processing device operated by the server, and the like. Further, the memory 62 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 62 may optionally include memory located remotely from the processor 61, which may be connected to a network connection device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 63 may receive input numeric or character information and generate key signal inputs related to user settings and function control of the processing device of the server. The output device 64 may include a display device such as a display screen.
One or more modules are stored in the memory 62, which when executed by the one or more processors 61, perform the method as shown in fig. 1.
Those skilled in the art will appreciate that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and the processes of the embodiments of the motor control methods described above can be included when the computer program is executed. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (FM), a Hard Disk (Hard Disk Drive, HDD), or a Solid-State Drive (SSD); the storage medium may also comprise a combination of memories of the kind described above.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims (10)
1. A perimeter intrusion identification method, comprising:
acquiring an image according to a real-time video stream, and setting a perimeter region on the image;
automatically segmenting the image into first image clusters based on the perimeter region;
zooming the images in the first image cluster to obtain a second image cluster;
and inputting the second image cluster into a neural network, detecting whether continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region, and outputting a detection and identification result.
2. The method of claim 1, wherein automatically segmenting the image into the first cluster of images based on the perimeter region comprises:
configuring a gaze disappearance direction within the perimeter region;
and automatically segmenting the image into a plurality of segments of pixel clusters according to the visual line disappearance direction to obtain a first image cluster.
3. The method of claim 1, wherein scaling the images in the first image cluster to obtain a second image cluster comprises:
determining a scaling according to the detection scene and the requirement;
and reducing the image of the relatively near end of the visual line in the first image cluster according to the scaling, and amplifying the image of the relatively far end to obtain a second image cluster.
4. The method according to claim 1, wherein the inputting the second image cluster into a neural network, detecting whether the continuous multiframe images of the second image cluster contain pedestrians invading the perimeter area, and outputting the detection recognition result comprises:
detecting whether the image of the second image cluster contains a pedestrian or not by using a deep learning algorithm;
if the image is detected to contain the pedestrian, background difference is carried out on the area where the pedestrian is detected by using a vibe algorithm, and the intra-area mobility judgment is carried out by judging a difference image.
5. The method of claim 4, further comprising:
and if no pedestrian is detected in the image, automatically segmenting the image based on the peripheral area again.
6. The method according to claim 4, wherein the background difference of the pedestrian-detected region by using the vibe algorithm and the intra-region mobility judgment by judging the difference image comprise:
judging whether the pedestrian is in the peripheral area;
if the pedestrian is in the peripheral area, judging whether the pedestrian is in the peripheral area of the continuous multi-frame images and whether the images in the peripheral area have differential changes;
if the pedestrian is in the peripheral area of the continuous multi-frame images and the images in the peripheral area have difference changes, alarm snapshot is carried out; otherwise, whether the pedestrian is in the peripheral area is judged again.
7. The method of claim 6, further comprising:
and if the pedestrian is not in the peripheral area, automatically segmenting the image based on the peripheral area again.
8. A perimeter intrusion identification system, comprising:
the setting unit is used for acquiring an image according to a real-time video stream and setting a perimeter region on the image;
a segmentation unit for automatically segmenting the image into first image clusters based on the perimeter region;
the zooming unit is used for zooming the images in the first image cluster to obtain a second image cluster;
and the detection unit is used for inputting the second image cluster into a neural network, detecting whether the continuous multi-frame images of the second image cluster contain pedestrians invading the perimeter region or not, and outputting a detection identification result.
9. A computer-readable storage medium storing computer instructions for causing a computer to perform the perimeter intrusion identification method according to any one of claims 1 to 7.
10. An electronic device, characterized in that the electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to cause the at least one processor to perform the perimeter intrusion identification method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210178171.8A CN114550060A (en) | 2022-02-25 | 2022-02-25 | Perimeter intrusion identification method and system and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210178171.8A CN114550060A (en) | 2022-02-25 | 2022-02-25 | Perimeter intrusion identification method and system and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114550060A true CN114550060A (en) | 2022-05-27 |
Family
ID=81679693
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210178171.8A Pending CN114550060A (en) | 2022-02-25 | 2022-02-25 | Perimeter intrusion identification method and system and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114550060A (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008123159A (en) * | 2006-11-10 | 2008-05-29 | Meidensha Corp | Intruder detecting apparatus |
| JP2009064287A (en) * | 2007-09-07 | 2009-03-26 | Meidensha Corp | Intruder detector |
| CN107316024A (en) * | 2017-06-28 | 2017-11-03 | 北京博睿视科技有限责任公司 | perimeter alarm algorithm based on deep learning |
| CN108009506A (en) * | 2017-12-07 | 2018-05-08 | 平安科技（深圳）有限公司 | Intrusion detection method, application server and computer-readable recording medium |
| CN109886245A (en) * | 2019-03-02 | 2019-06-14 | 山东大学 | A kind of pedestrian detection recognition methods based on deep learning cascade neural network |
| CN111564015A (en) * | 2020-05-20 | 2020-08-21 | 中铁二院工程集团有限责任公司 | Method and device for monitoring perimeter intrusion of rail transit |
| CN111582060A (en) * | 2020-04-20 | 2020-08-25 | 浙江大华技术股份有限公司 | Automatic line drawing perimeter alarm method, computer equipment and storage device |
| CN111813997A (en) * | 2020-09-08 | 2020-10-23 | 平安国际智慧城市科技股份有限公司 | Intrusion analysis method, device, equipment and storage medium |
| CN112085010A (en) * | 2020-10-28 | 2020-12-15 | 成都信息工程大学 | Mask detection and deployment system and method based on image recognition |
| CN113593161A (en) * | 2021-06-28 | 2021-11-02 | 华迪计算机集团有限公司 | Perimeter intrusion detection method |
-
2022
- 2022-02-25 CN CN202210178171.8A patent/CN114550060A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008123159A (en) * | 2006-11-10 | 2008-05-29 | Meidensha Corp | Intruder detecting apparatus |
| JP2009064287A (en) * | 2007-09-07 | 2009-03-26 | Meidensha Corp | Intruder detector |
| CN107316024A (en) * | 2017-06-28 | 2017-11-03 | 北京博睿视科技有限责任公司 | perimeter alarm algorithm based on deep learning |
| CN108009506A (en) * | 2017-12-07 | 2018-05-08 | 平安科技（深圳）有限公司 | Intrusion detection method, application server and computer-readable recording medium |
| CN109886245A (en) * | 2019-03-02 | 2019-06-14 | 山东大学 | A kind of pedestrian detection recognition methods based on deep learning cascade neural network |
| CN111582060A (en) * | 2020-04-20 | 2020-08-25 | 浙江大华技术股份有限公司 | Automatic line drawing perimeter alarm method, computer equipment and storage device |
| CN111564015A (en) * | 2020-05-20 | 2020-08-21 | 中铁二院工程集团有限责任公司 | Method and device for monitoring perimeter intrusion of rail transit |
| CN111813997A (en) * | 2020-09-08 | 2020-10-23 | 平安国际智慧城市科技股份有限公司 | Intrusion analysis method, device, equipment and storage medium |
| CN112085010A (en) * | 2020-10-28 | 2020-12-15 | 成都信息工程大学 | Mask detection and deployment system and method based on image recognition |
| CN113593161A (en) * | 2021-06-28 | 2021-11-02 | 华迪计算机集团有限公司 | Perimeter intrusion detection method |
Non-Patent Citations (2)
| Title |
|---|
| 杜海龙;: "分布式光纤传感周界安防系统的入侵信号处理研究", 枣庄学院学报, no. 05, 1 September 2020 (2020-09-01) * |
| 闫春江;王闯;方华林;王毅轩;杜觉晓;项学智;郭鑫立;: "基于深度学习的输电线路工程车辆入侵检测", 信息技术, no. 07, 24 July 2018 (2018-07-24) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| GB2526658A (en) | An efficient method of offline training a special-type parked vehicle detector for video-based on-street parking occupancy detection systems | |
| CN103729858A (en) | Method for detecting article left over in video monitoring system | |
| KR102159954B1 (en) | Method for establishing region of interest in intelligent video analytics and video analysis apparatus using the same | |
| CN113396423A (en) | Method of processing information from event-based sensors | |
| CN110647818A (en) | Identification method and device for shielding target object | |
| CN111582060A (en) | Automatic line drawing perimeter alarm method, computer equipment and storage device | |
| JP6413318B2 (en) | Vehicle detection device, system, and program | |
| CN111582166A (en) | Remnant detection method based on Gaussian modeling and YoLo V3 target detection | |
| US20200394802A1 (en) | Real-time object detection method for multiple camera images using frame segmentation and intelligent detection pool | |
| Sharma et al. | Automatic vehicle detection using spatial time frame and object based classification | |
| KR102171384B1 (en) | Object recognition system and method using image correction filter | |
| JP6413319B2 (en) | Vehicle detection device, system, and program | |
| Chandrasekhar et al. | A survey of techniques for background subtraction and traffic analysis on surveillance video | |
| CN114550060A (en) | Perimeter intrusion identification method and system and electronic equipment | |
| JP2024516642A (en) | Behavior detection method, electronic device and computer-readable storage medium | |
| JP2023036054A (en) | Target counting method, apparatus, device, and storage medium | |
| JP2019067377A (en) | Image processing device and method, as well as monitoring system | |
| CN114821441A (en) | Deep learning-based airport scene moving target identification method combined with ADS-B information | |
| CN113673362A (en) | Method and device for determining motion state of object, computer equipment and storage medium | |
| CN112967321A (en) | Moving object detection method and device, terminal equipment and storage medium | |
| CN112465835A (en) | Method for jadeite image segmentation and model training method | |
| CN112085002A (en) | Portrait segmentation method, portrait segmentation device, storage medium and electronic equipment | |
| CN111091526A (en) | Video blur detection method and system | |
| CN110766644A (en) | Image down-sampling method and device | |
| Zaihidee et al. | Comparison of human segmentation using thermal and color image in outdoor environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |