Indoor monitoring method and device based on PON home gateway, medium and equipment

Abstract

The embodiment of the invention discloses an indoor monitoring method and device based on a PON home gateway, a medium and equipment, and relates to the technical field of indoor monitoring, monitoring research and judgment are carried out by extracting a frame picture of a monitoring video, firstly, threshold segmentation is carried out on a monitoring image, and no matter how the position relation between an invading target and a curtain serving as a shielding object is, the invading target and the curtain serving as the shielding object are subjected to threshold segmentation; under multi-scale segmentation, all covered areas can be marked, then contour extraction is carried out on the areas, and most contours have high consistency from the aspect of contours, so that the contours of the curtains serving as shelters can be screened out through the similarity, the interference of sheltering is avoided, and the accuracy of screening is improved. Therefore, study and judgment focus on contours except for the contours, misjudgment is avoided, and indoor monitoring quality is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of indoor monitoring, and particularly to an indoor monitoring method, device, medium and equipment based on a PON home gateway. Background Art

[0002] A PON home gateway (Passive Optical Network Home Gateway) is a user-side access device based on passive optical network (PON) technology and is the core terminal device in the fiber to the room (FTTR) scenario. It connects the operator's fiber network with the home internal network, and at the same time integrates multiple functions such as routing, switching, and wireless access. It is the network hub of smart home and indoor monitoring systems. The indoor monitoring solution based on the PON home gateway combines the high bandwidth and stability of the fiber network with the flexibility of smart home devices.

[0003] For example, indoor monitoring is realized through a PON home gateway to achieve monitoring and alarm when there are suspicious personnel intruding. This solution collects information through monitoring devices, uses the home gateway to connect the monitoring devices and the PON network, and processes the monitoring data through a server for intrusion judgment. When an intrusion is confirmed, an alarm is sent to the user terminal. However, in the above solution, there will be a problem of false alarms in intelligent analysis, such as misidentification in edge computing. The most common situation of misidentification in edge computing is that the swinging of the curtain is misidentified as an intrusion. The prior art has an improved method of adding multi-source sensors for fusion recognition, such as adding a millimeter-wave radar or a pyroelectric sensor. These solutions will increase costs, and for the means of distinguishing static objects (curtains) and real moving targets such as millimeter-wave radars, the curtain may also move under non-intrusion conditions, so this method is not completely reliable. Summary of the Invention

[0004] The main purpose of the present invention is to provide an indoor monitoring method, device, medium and equipment based on a PON home gateway, aiming to solve the problem of low quality of indoor monitoring based on the PON home gateway in the prior art.

[0005] To achieve the above purpose, the technical solutions adopted in the embodiments of the present invention are as follows: In a first aspect, an embodiment of the present invention provides an indoor monitoring method based on a PON home gateway, including the following steps: Extract frame pictures from the indoor monitoring video to obtain monitoring images; Perform multi-scale threshold segmentation on the monitoring images and mark the segmentation regions at each scale; Extract the edge contours of the segmentation regions to obtain segmentation contours; Obtain occlusion contours according to the contours in the segmentation contours whose similarity meets the similarity threshold; Obtain the monitoring result according to the contour in the segmentation contour except the occluder contour.

[0006] In a possible implementation manner of the first aspect, obtaining the monitoring result according to the contour in the segmentation contour except the occluder contour includes: Obtain the occluded object contour according to the contour in the segmentation contour except the occluder contour; Confirm whether an intrusion occurs and obtain the monitoring result according to the similarity between the human body features and the occluded object contour.

[0007] In a possible implementation manner of the first aspect, after obtaining the occluded object contour according to the contour in the segmentation contour except the occluder contour, the method further includes: Perform a linear representation on the occluded object contour to obtain the linear features of the occluded object; Extend the linear features of the occluded object to obtain the target linear features; Confirm whether an intrusion occurs and obtain the monitoring result according to the similarity between the human body features and the occluded object contour, including: Confirm whether an intrusion occurs and obtain the monitoring result according to the similarity between the human body features and the target linear features.

[0008] In a possible implementation manner of the first aspect, performing a linear representation on the occluded object contour to obtain the linear features of the occluded object includes: Draw the minimum bounding rectangle for the occluded object contour to obtain a number of rectangular marking frames; Perform different linear representations on the rectangular marking frames according to the aspect ratio of the rectangular marking frames to obtain the linear features of the occluded object.

[0009] In a possible implementation manner of the first aspect, performing different linear representations on the rectangular marking frames according to the aspect ratio of the rectangular marking frames to obtain the linear features of the occluded object includes: Perform a single linear representation on the rectangular marking frames that exceed the ratio threshold and a multi-linear representation on the rectangular marking frames that do not exceed the ratio threshold according to the comparison between the aspect ratio of the rectangular marking frames and the ratio threshold to obtain the linear features of the occluded object.

[0010] In a possible implementation manner of the first aspect, perform multi-scale threshold segmentation on the monitoring image and mark the segmentation regions at each scale, including: Perform threshold segmentation on the monitoring image at the target scale and mark the segmentation regions at the target scale; Adjust the target scale in a predetermined direction, and return to the step of performing threshold segmentation on the monitoring image at the target scale and marking the segmentation regions at the target scale until the preset scale range is traversed.

[0011] In a possible implementation manner of the first aspect, obtaining an occluder contour according to the contours in the segmentation contour whose similarity meets the similarity threshold includes: Obtaining an occluded area according to the contours in the segmentation contour whose similarity meets the similarity threshold; Merging the occluded areas and extracting the contour of the merged area to obtain the occluder contour.

[0012] In a second aspect, an embodiment of the present invention provides an indoor monitoring device based on a PON home gateway, including: An extraction module, configured to extract frame images from the indoor monitoring video to obtain monitoring images; A marking module, configured to perform multi-scale threshold segmentation on the monitoring images and mark the segmented areas at each scale; An extraction module, configured to extract edge contours of the segmented areas to obtain a segmentation contour; An obtaining module, configured to obtain an occluder contour according to the contours in the segmentation contour whose similarity meets the similarity threshold; A monitoring module, configured to obtain a monitoring result according to the contours in the segmentation contour other than the occluder contour.

[0013] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when loaded and executed by a processor, implements the indoor monitoring method based on a PON home gateway provided in any one of the above first aspects.

[0014] In a fourth aspect, an embodiment of the present invention provides an electronic device including a processor and a memory, wherein, The memory is used to store a computer program; The processor is used to load and execute the computer program so that the electronic device executes the indoor monitoring method based on a PON home gateway provided in any one of the above first aspects.

[0015] Compared with the prior art, the beneficial effects of the present invention are: An indoor monitoring method, device, medium and equipment based on a PON home gateway proposed by an embodiment of the present invention. The method includes: extracting frame images from indoor monitoring videos to obtain monitoring images; performing multi-scale threshold segmentation on the monitoring images and marking the segmented regions at each scale; extracting edge contours of the segmented regions to obtain segmented contours; obtaining occluder contours according to the contours in the segmented contours whose similarity meets the similarity threshold; and obtaining monitoring results according to the contours in the segmented contours other than the occluder contours. By extracting frame images of the monitoring video for monitoring and judgment, the present invention first performs threshold segmentation on the monitoring images. Regardless of the positional relationship between the invading target and the curtain serving as an occluder, all covered regions can be marked under multi-scale segmentation. Then, contour extraction is performed on these regions. Different from smooth and flat plate-like targets that do not deform, the curtain may be wrinkled even in a static state, which not only affects its own threshold segmentation but also affects the segmentation of the invading target. Therefore, the extracted segmented contours may only be a scattered expression of a certain target. Considering the actual shape of the curtain in reality, whether it is static, swinging naturally, or swinging due to an invading target, most of the contours will have strong consistency in terms of the contour. Therefore, the contours of the curtain serving as an occluder can be screened out through similarity, thus avoiding the interference of occlusion and enabling the judgment to focus on the contours other than this, thereby avoiding misjudgment and improving the quality of indoor monitoring. Description of the Drawings

[0016] Figure 1 Schematic structural diagram of an electronic device for the hardware operating environment involved in an embodiment of the present invention; Figure 2 Schematic flowchart of an indoor monitoring method based on a PON home gateway provided by an embodiment of the present invention; Figure 3 Schematic module diagram of an indoor monitoring device based on a PON home gateway provided by an embodiment of the present invention; Reference numerals in the figure: 101 - processor, 102 - communication bus, 103 - network interface, 104 - user interface, 105 - memory. Detailed Embodiments

[0017] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0018] Refer to the appended Figure 1 drawing, the appended Figure 1Schematic diagram of the electronic device structure for the hardware operating environment involved in the embodiment of the present invention. The electronic device may include: a processor 101, such as a Central Processing Unit (CPU), a communication bus 102, a user interface 104, a network interface 103, and a memory 105. Among them, the communication bus 102 is used to realize the connection and communication between these components. The user interface 104 may include a display screen (Display) and an input unit such as a keyboard (Keyboard). Optionally, the user interface 104 may further include a standard wired interface and a wireless interface. The network interface 103 may optionally include a standard wired interface and a wireless interface (such as a Wireless-Fidelity (WI-FI) interface). The memory 105 may optionally be a storage device independent of the aforementioned processor 101. The memory 105 may be a high-speed Random Access Memory (RAM) or a stable Non-Volatile Memory (NVM), such as at least one disk memory. The processor 101 may be a general-purpose processor, including a central processor, a network processor, etc., or may also be a digital signal processor, an application-specific integrated circuit, a field-programmable gate array, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0019] Those skilled in the art can understand that the structure shown in the appendix Figure 1 does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0020] As shown in the appendix Figure 1 The memory 105, as a storage medium, may include an operating system, a network communication module, a user interface module, and an indoor monitoring device based on a PON home gateway.

[0021] In the electronic device shown in the appendix Figure 1 the network interface 103 is mainly used for data communication with a network server; the user interface 104 is mainly used for data interaction with users; the processor 101 and the memory 105 in the present invention may be arranged in the electronic device. The electronic device calls the indoor monitoring device based on the PON home gateway stored in the memory 105 through the processor 101 and executes the indoor monitoring method based on the PON home gateway provided by the embodiment of the present invention.

[0022] Referring to the appendix Figure 2 , based on the hardware device of the foregoing embodiment, an embodiment of the present invention provides an indoor monitoring method based on a PON home gateway, including the following steps: S10: Extract frame images from the indoor surveillance video to obtain surveillance images.

[0023] In the specific implementation process, use the surveillance cameras installed indoors to shoot videos, and then extract frame images from the videos to obtain frame images, that is, surveillance images, and use these images for image data processing for surveillance analysis.

[0024] S20: Perform multi-scale threshold segmentation on the surveillance images and mark the segmented regions at each scale.

[0025] In the specific implementation process, since threshold segmentation is to perform regional segmentation based on a set threshold, and the depth of the curtain as an occluder and the intrusion target in the image is obviously different, and the target itself will also have different depths due to different shapes and illumination levels, so in order to completely segment and mark these regions, multi-scale threshold segmentation is required, that is, multiple segmentation markings are performed based on multiple different threshold settings. It should be noted that in order to achieve faster segmentation, the surveillance images can be grayscaled to reduce the image dimension for easy and quick processing.

[0026] Specifically, performing multi-scale threshold segmentation on the surveillance images and marking the segmented regions at each scale includes: Perform threshold segmentation on the surveillance images at the target scale and mark the segmented regions at the target scale; Adjust the target scale in a predetermined direction, and return to the step of performing threshold segmentation on the surveillance images at the target scale and marking the segmented regions at the target scale until the preset scale range is traversed.

[0027] In the specific implementation process, in order to extract as many target regions as possible, multi-scale segmentation is performed on the surveillance images. First, threshold segmentation is performed at the target scale, that is, an initial threshold is set for segmentation. After segmentation, the scale is adjusted in a predetermined direction, that is, the set initial threshold is increased or decreased, specifically determined according to the size of the set initial threshold. If the set threshold is large, then the adjustment in the predetermined direction is to gradually decrease the threshold, and vice versa, increase the threshold. According to the threshold segmentation situation in the actual situation, a scale range is preset, that is, a threshold range is set. After traversing the thresholds within this range, it can be ensured that as many target ranges as possible are segmented, avoiding misjudgment caused by poor image quality.

[0028] S30: Extract the edge contours of the segmented regions to obtain the segmentation contours.

[0029] In the specific implementation process, after marking the segmented regions, extract the edge contours of these regions. An edge contour extraction algorithm can be used to find the boundaries of each segmented region to more clearly represent the segmented regions.

[0030] S40: Obtain the occluder contour according to the contours in the segmentation contour whose similarity meets the similarity threshold.

[0031] In the specific implementation process, the segmented contours are displayed on the image together as the contours representing the intrusion target and the curtain occluder. Now, it is necessary to further identify them to distinguish which contours belong to the intrusion target and which belong to the curtain. Considering the characteristics of the morphological distribution of the curtain in actual situations, it will not present a messy state. Whether it is hanging naturally, swinging, etc., the curtain itself will have strong consistency. Even if folds are formed, the extending directions of these folds are highly consistent. For example, when hanging naturally, they all extend in the vertical direction, and when swinging, they also have the same central point direction. Based on the above characteristics, the contours segmented from the curtain in the image should be highly similar.

[0032] In one embodiment, obtaining the occluder contour according to the contours in the segmentation contour whose similarity meets the similarity threshold includes: Obtain the occluded area according to the contours in the segmentation contour whose similarity meets the similarity threshold; Merge the occluded areas, and extract the contour of the merged area to obtain the occluder contour.

[0033] In the specific implementation process, determine whether it belongs to the occluded area represented by the curtain by judging the similarity between two segmentation contours. Specifically, one of them can be used as the basis, and the remaining contours are judged for similarity with it to obtain a similarity value. If this value meets the similarity threshold, it means that the two belong to the occluded area. The calculation of similarity can use existing algorithms, such as the Euclidean distance. The determination of similarity can also be based on the position of the contour distribution, and calculate the similarity of adjacent segmentation contours for determination.

[0034] For example, for a curtain in a naturally hanging state, even if there are folds, the segmented contours are mostly long and narrow rectangular-like areas and extend in the vertical direction, and these areas are highly similar. Another example is that the curtain in a swinging state is in an inclined state, and each segmented contour is also mostly long and narrow rectangular-like areas, but one end will converge towards the swinging direction. However, from the perspective of regional similarity, these contour areas are also highly similar. Since the segmented areas may be discontinuous and non-overlapping, that is, there are gaps between the segmentation contours, so these occluded areas can be merged to fill the gaps between the occluded areas and form a more complete representation of the area where the curtain is located.

[0035] S50: Obtain the monitoring result according to the contours in the segmentation contour other than the occluder contour.

[0036] In the specific implementation process, after extracting the contour of the occluder representing the curtain area, the influence of curtain swing on intrusion monitoring and judgment is avoided. Since the extraction of the segmentation contour includes the intrusion object, after screening out the contour of the occluder, the contour representing the intrusion target must exist in the remaining contours. By performing recognition and judgment based on this part of the contour, the monitoring of indoor intrusion can be completed.

[0037] In one embodiment, according to the contours in the segmentation contour other than the occluder contour, a monitoring result is obtained, including: According to the contours in the segmentation contour other than the occluder contour, an occluded object contour is obtained; Based on the similarity between the human body features and the occluded object contour, it is confirmed whether an intrusion occurs and a monitoring result is obtained.

[0038] In the specific implementation process, the embodiments of the present invention mainly conduct judgment on the situation with overlapping influence of the curtain. In actual situations, if an intrusion occurs and the recognition is not affected by the curtain, by tracking multiple frames of images, the moving target can be confirmed to identify the occurrence of an intrusion. For the situation with occlusion influence, the influence of the occluder contour is avoided, focusing on the occluded object contour, and it is determined whether an intrusion occurs. By performing similarity recognition between these contours and the human body features, that is, identifying whether the occluded object contour is a human target, it can be monitored whether an intrusion occurs.

[0039] In one embodiment, after obtaining the occluded object contour according to the contours in the segmentation contour other than the occluder contour, the method further includes: Performing a linear expression on the occluded object contour to obtain the linear feature of the occluded object; Extending the linear feature of the occluded object to obtain the target linear feature.

[0040] In the specific implementation process, human body features tend to be linearly expressed. For example, it can be simplified as a combination of the torso and limbs. To make the judgment more accurate, the occluded object contours are all transformed into linear expressions to match the linear human body feature expressions. As mentioned in the foregoing embodiments, the segmentation contours may not be continuous. Therefore, for more accurate recognition, the linear features of the occluded objects are extended and complemented. Since this part of the features represents the intrusion target and the intrusion target should also conform to the human body features, it is extended, that is, the target is restored as much as possible along the distribution law of the human body features.

[0041] In one embodiment, performing a linear expression on the occluded object contour to obtain the linear feature of the occluded object includes: Drawing a minimum bounding rectangle for the occluded object contour to obtain a number of rectangular marking frames; According to the aspect ratio of the rectangular marking frames, different linear expressions are performed on the rectangular marking frames to obtain the linear feature of the occluded object.

[0042] In the specific implementation process, intrusion monitoring depends on timeliness. The shorter the response time, the more secure it is. Improving the speed of identification and judgment is a strategy to reduce the response time. By drawing the minimum circumscribed moment of the contour, the contours with large differences in detail morphology are converted into more regular rectangular marking frames for processing, which can effectively improve the recognition efficiency. Each contour is represented by a rectangular marking frame, and its linear expression becomes a simpler centerline mark. Considering that the rectangular frame has two center lines, and the linear expression of human body features is more inclined to a single linear mark, different linear expressions are required according to the shape of the rectangular marking frame.

[0043] Based on the above steps, the similarity between the human body features and the outline of the obstructed object is used to confirm whether an intrusion has occurred and obtain monitoring results, including: Based on the similarity between human features and target linear features, confirm whether intrusion occurs and obtain monitoring results.

[0044] Specifically: according to the aspect ratio of the rectangular marking box, different linear expressions are performed on the rectangular marking box to obtain the linear features of the occluded object, including: According to the aspect ratio of the rectangular marking box and the ratio threshold, the rectangular marking box exceeding the ratio threshold is expressed unilinearly, and the rectangular marking box not exceeding the ratio threshold is expressed multilinearly to obtain the linear features of the occluded object.

[0045] In the specific implementation process, the shape of the rectangular marking box depends on the aspect ratio. The larger the aspect ratio, the narrower the rectangular marking box is, and the more it conforms to the expression of limbs in human body features. The smaller the aspect ratio, the more it tends to express human body features such as limbs and head. Therefore, a ratio threshold can be set. If the aspect ratio exceeds the threshold, only a single linear expression is performed, that is, the rectangular box is expressed with the center line along the length direction; the rest that do not exceed the threshold are multi-linearly expressed, and the rectangular box is expressed with two center lines in the length and width directions. In this way, the final intrusion identification becomes a comparison between the linear features of the obstructed object and the human body features. Both have linear expression characteristics, which can be compared and identified more quickly and accurately.

[0046] In this embodiment, by extracting the frame images of the surveillance video for surveillance analysis, first, threshold segmentation is performed on the surveillance image. Regardless of the positional relationship between the invading target and the curtain acting as an occluder, under multi-scale segmentation, all covered areas can be marked. Then, contour extraction is performed on these areas. Different from the smooth and non-deformable sheet-like targets, the curtain may be wrinkled even in a static state, which not only affects its own threshold segmentation but also affects the segmentation of the invading target. Therefore, the extracted segmentation contours may only be a scattered expression of a certain target. Considering the actual shape of the curtain in reality, whether it is static, swinging naturally, or swinging due to the invading target, most of the contours will have strong consistency in terms of the contour. So, the contour of the curtain acting as an occluder can be screened out through similarity, thereby avoiding the interference of occlusion, enabling the analysis to focus on the contours other than this, and thus avoiding misjudgment, thereby improving the quality of indoor surveillance.

[0047] Refer to the appendix Figure 3 , based on the same inventive concept as in the foregoing embodiment, the embodiment of the present invention further provides an indoor surveillance device based on a PON home gateway, including: An extraction module, configured to extract frame images from the indoor surveillance video to obtain surveillance images; A marking module, configured to perform multi-scale threshold segmentation on the surveillance images and mark the segmented areas at each scale; An extraction module, configured to perform edge contour extraction on the segmented areas to obtain segmentation contours; An obtaining module, configured to obtain the occluder contour according to the contours in the segmentation contours whose similarity meets the similarity threshold; A surveillance module, configured to obtain the surveillance result according to the contours in the segmentation contours other than the occluder contour.

[0048] Those skilled in the art should understand that the division of each module in the embodiment is only a logical function division. In actual application, it can be fully or partially integrated into one or more actual carriers, and these modules can all be implemented in the form of software called by a processing unit, or all be implemented in the form of hardware, or be implemented in the form of a combination of software and hardware. It should be noted that each module in the indoor surveillance device based on a PON home gateway in this embodiment corresponds one by one to each step in the indoor surveillance method based on a PON home gateway in the foregoing embodiment. Therefore, the specific implementation manner of this embodiment can refer to the implementation manner of the foregoing indoor surveillance method based on a PON home gateway, which will not be elaborated here.

[0049] Based on the same inventive concept as in the foregoing embodiments, an embodiment of the present invention further provides a computer-readable storage medium storing a computer program, which, when loaded and executed by a processor, implements the indoor monitoring method based on a PON home gateway provided by the embodiment of the present invention.

[0050] Based on the same inventive concept as in the foregoing embodiments, an embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein, the memory is used to store a computer program; the processor is used to load and execute the computer program so that the electronic device executes the indoor monitoring method based on a PON home gateway provided by the embodiment of the present invention.

[0051] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disc, or CD-ROM; or may be various devices including one or any combination of the above memories. The computer may be various computing devices including smart terminals and servers.

[0052] In some embodiments, the executable instructions may be in the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including being deployed as an independent program or being deployed as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0053] As an example, the executable instructions may or may not correspond to a file in the file system, and may be stored as part of a file that stores other programs or data. For example, they may be stored in one or more scripts in a Hyper Text Markup Language (HTML) document, stored in a single file dedicated to the program in question, or stored in multiple cooperating files (for example, files that store one or more modules, subroutines, or code portions).

[0054] As an example, the executable instructions may be deployed to be executed on one computing device, or on multiple computing devices located at one location, or on multiple computing devices distributed at multiple locations and interconnected by a communication network.

[0055] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0056] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented through hardware, but in many cases, the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as a read-only memory / random access memory, magnetic disk, optical disk), and includes several instructions to enable a multimedia terminal device (which can be a mobile phone, computer, television receiver, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0057] In summary, an indoor monitoring method, device, medium, and equipment based on a PON home gateway provided by an embodiment of the present invention. The method includes: extracting frame images of indoor monitoring videos to obtain monitoring images; performing multi-scale threshold segmentation on the monitoring images to mark the segmented regions at each scale; extracting the edge contours of the segmented regions to obtain segmented contours; obtaining occluder contours according to the contours in the segmented contours whose similarity meets the similarity threshold; and obtaining monitoring results according to the contours in the segmented contours other than the occluder contours. The present invention monitors and judges by extracting the frame images of the monitoring video. First, threshold segmentation is performed on the monitoring images. Regardless of the positional relationship between the invading target and the curtain acting as an occluder, under multi-scale segmentation, all covered regions can be marked. Then, contour extraction is performed on these regions. Different from smooth, flat, and non-deformable sheet-like targets, the curtain may be wrinkled even in a stationary state, which not only affects its own threshold segmentation but also affects the segmentation of the invading target. Therefore, the extracted segmented contours may only be a scattered expression of a certain target. Considering the actual shape of the curtain in reality, whether it is stationary, swinging naturally, or swinging due to the invading target, most of the contours will have strong consistency in terms of the contour. Therefore, the contours of the curtain acting as an occluder can be screened out through similarity, thereby avoiding the interference of occlusion and enabling the judgment to focus on the contours other than this, thus avoiding misjudgment and improving the quality of indoor monitoring.

[0058] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An indoor monitoring method based on a PON home gateway, characterized in that: The following steps are involved: Extract frames from indoor surveillance videos to obtain surveillance images; Performing multi-scale threshold segmentation on the monitoring image, and marking the segmented area at each scale; Extracting edge contours of the segmented area to obtain a segmented contour; Obtaining an occluder contour according to contours whose similarity satisfies a similarity threshold in the segmented contours; A monitoring result is obtained according to contours other than the contour of the obstruction in the segmented contours.

2. The indoor monitoring method based on PON home gateway according to claim 1, characterized in that: The obtaining of monitoring results according to contours other than the contour of the obstruction in the segmented contours includes: Obtaining the outline of the obstructed object according to the outlines other than the outline of the obstructing object in the segmented outline; According to the similarity between the human body features and the outline of the obstructed object, it is confirmed whether an intrusion occurs and the monitoring results are obtained.

3. The indoor monitoring method based on PON home gateway according to claim 2 is characterized in that: After obtaining the outline of the obstructed object according to the outlines other than the outline of the obstructing object in the segmented outline, the method further includes: Expressing the outline of the obstructed object linearly to obtain linear features of the obstructed object; Extending the linear features of the obstructed object to obtain target linear features; The determining whether an intrusion occurs and obtaining a monitoring result based on the similarity between the human body features and the outline of the obstructed object includes: According to the similarity between the human body features and the target linear features, whether an intrusion occurs is confirmed and the monitoring results are obtained.

4. The indoor monitoring method based on PON home gateway according to claim 3 is characterized in that: The linear expression of the outline of the obstructed object to obtain the linear features of the obstructed object includes: Performing minimum circumscribed moment drawing on the outline of the obstructed object to obtain a plurality of rectangular marking frames; According to the aspect ratio of the rectangular marking frame, different linear expressions are performed on the rectangular marking frame to obtain the linear features of the obstructed object.

5. The indoor monitoring method based on PON home gateway according to claim 4, characterized in that: According to the aspect ratio of the rectangular mark frame, different linear expressions are performed on the rectangular mark frame to obtain the linear features of the obstructed object, including: According to the aspect ratio of the rectangular marking box and the ratio threshold, the rectangular marking box exceeding the ratio threshold is expressed unilinearly, and the rectangular marking box not exceeding the ratio threshold is expressed multilinearly to obtain the linear features of the occluded object.

6. The indoor monitoring method based on PON home gateway according to claim 1, characterized in that: The multi-scale threshold segmentation is performed on the monitoring image, and the segmentation area at each scale is marked, including: Performing threshold segmentation of a target scale on the monitoring image, and marking the segmented area at the target scale; The target scale is adjusted in a predetermined direction, and the process returns to the step of performing threshold segmentation of the target scale on the monitoring image and marking the segmented area under the target scale until the preset scale range is traversed.

7. The indoor monitoring method based on PON home gateway according to claim 1, characterized in that: The step of obtaining the occluder contour according to the contours whose similarity in the segmented contours satisfies the similarity threshold comprises: Obtaining an occluded area according to contours whose similarity in the segmented contours satisfies a similarity threshold; The occluded areas are merged, and the contour of the merged area is extracted to obtain the occluder contour.

8. An indoor monitoring device based on a PON home gateway, characterized in that: include: An extraction module is used to extract frames from indoor surveillance videos to obtain surveillance images; A marking module, used to perform multi-scale threshold segmentation on the monitoring image and mark the segmented area at each scale; An extraction module, used for extracting edge contours of the segmented area to obtain a segmented contour; An obtaining module, used for obtaining the outline of the occluder according to the outlines whose similarity in the segmented outlines satisfies the similarity threshold; The monitoring module is used to obtain monitoring results according to the contours in the segmented contours except the contour of the obstruction.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is loaded and executed by the processor, the indoor monitoring method based on the PON home gateway as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: comprising a processor and a memory, wherein: The memory is used to store computer programs; The processor is used to load and execute the computer program so that the electronic device executes the indoor monitoring method based on the PON home gateway as described in any one of claims 1-7.

Images