阅读说明：本技术 一种基于大数据网络的监控系统及监控方法 (Monitoring system and monitoring method based on big data network ) 是由 王平 于 2021-09-10 设计创作，主要内容包括：本发明提供一种基于大数据网络的监控系统及监控方法,本发明首先建立图像拍摄设备与图像处理器的网络通信链路,然后从监控视频中选择包含有人体的人体图像,再对人体图像中的人体进行框标注,以及对标注后的图像框标记颜色和编号,再然后在确定待监控对象后,按照拍摄时间的先后顺序对所获取的人体图像进行排序,再根据待监控对象的标号和颜色来筛选出包含有待监控对象的目标图像,以及包含有目标图像的目标监控视频,最后根据目标监控视频来对待监控对象进行监控,确定待监控对象的运动信息。所以,本发明通过根据待监控对象的图像框所对应的颜色和编号来从目标监控视频中确定待监控对象的运动信息,从而可以完成对待监控对象的监控工作。(The invention provides a monitoring system and a monitoring method based on a big data network, which firstly establish a network communication link between image shooting equipment and an image processor, then select a human body image containing a human body from a monitoring video, then label the human body in the human body image with a frame, label the image frame with a color and a number, then sort the obtained human body images according to the sequence of shooting time after determining an object to be monitored, then screen out a target image containing the object to be monitored and a target monitoring video containing the target image according to the label and the color of the object to be monitored, and finally monitor the object to be monitored according to the target monitoring video and determine the motion information of the object to be monitored. Therefore, the invention determines the motion information of the object to be monitored from the target monitoring video according to the color and the number corresponding to the image frame of the object to be monitored, thereby completing the monitoring work of the object to be monitored.)

1. A monitoring method based on a big data network is characterized by comprising the following steps:

acquiring monitoring videos shot by a plurality of image shooting devices;

transmitting the monitoring video positioned at the image shooting equipment to an image processor according to a network communication link which is pre-established by the image shooting equipment and the image processor;

framing the monitoring video by using the image processor, and acquiring a human body image containing a human body from the framed image;

performing frame marking on the human body in each frame of human body image to obtain one or more corresponding human body image frames;

marking the same reference number on one or more human body image frames belonging to the same human body, and assigning the same color to one or more human body image frames belonging to the same human body;

sequencing the acquired human body images according to the sequence of the shooting time;

determining an object to be monitored, selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image;

acquiring a monitoring video containing the target image, and recording the monitoring video as a target monitoring video;

and monitoring the object to be monitored according to the target monitoring video, and determining the motion information of the object to be monitored.

2. The big data network-based monitoring method according to claim 1, wherein the network communication link pre-established between the image capturing device and the image processor, when transmitting the monitoring video at the image capturing device to the image processor, comprises:

before the monitoring video is transmitted, generating a network test data packet in the image shooting equipment, and recording the network test data packet as a first network test data packet, wherein the first network test packet comprises network test data for establishing data communication;

acquiring the sending time of the first network test data packet when entering the network communication link from the image shooting equipment;

recording a network test data packet received by the image processor from the network communication link as a second network test data packet, and acquiring the receiving time of the second network test data packet received by the image processor from the network communication link; the second network test packet comprises network test data for establishing data communication;

determining whether network delay exists in a network communication link between the image shooting equipment and the image processor or not and the existing network delay time according to the receiving time and the transmitting time;

comparing a first network test data packet generated in the image shooting equipment with a second network test data packet received by the image processor, comparing network test data contained in the first network test data packet with network test data contained in the second network test data packet, and calculating a data loss rate and a data loss amount of the network communication link;

and generating a network test analysis report of the network communication link according to the network delay result, the data loss rate and the data loss amount.

3. The big data network-based monitoring method according to claim 2, further comprising:

detecting whether a network communication link between the image shooting equipment and the image processor is abnormal or not by using the network test analysis report;

if the network communication link of the image shooting equipment is normal, determining that the access network of the image processor is abnormal;

if the network communication link of the image shooting equipment is abnormal, determining that the access network of the image processor is normal, or adjusting the network configuration of the image shooting equipment, and detecting the network communication link of the image shooting equipment again until the network communication link of the image shooting equipment is normal, detecting whether the network quality between the image shooting equipment and the image processor meets the communication transmission requirement of the monitoring video, and determining that the access network of the image processor is normal.

4. The big data network-based monitoring method according to claim 1, wherein the monitoring of the object to be monitored according to the target monitoring video, and the determining of the motion information of the object to be monitored comprises:

determining the moving distance of the object to be monitored according to each frame of image in the target monitoring video, wherein the moving distance comprises the following steps:

in the formula, S_i，jRepresenting the moving distance of a target image frame corresponding to the object to be monitored from a j-1 frame image to a j frame image in the target monitoring video;

(x_i，j，y_i，j) Representing the coordinates of the ith characteristic coordinate point of a target image frame corresponding to the object to be monitored in the jth frame image in the target monitoring video;

(x_i-1，j，y_i-1，j) Representing the ith characteristic coordinate point of the target image frame corresponding to the object to be monitored in the jth-1 frame image in the target monitoring videoCoordinates;

wherein i and j are natural numbers.

5. The big data network-based monitoring method according to claim 4, wherein the monitoring of the object to be monitored according to the target monitoring video, and the determining of the motion information of the object to be monitored further comprises:

determining a moving speed identification value of the object to be monitored according to each frame of image in the target monitoring video, wherein the moving speed identification value comprises the following steps:

in the formula, V represents a moving speed identification value of the object to be monitored;

f represents the image frame frequency in the target monitoring video;

m represents the total number of characteristic coordinate points in each frame of image in the target monitoring video;

n represents the number of image frames contained in the target surveillance video.

6. The big data network-based monitoring method according to claim 5, wherein the monitoring of the object to be monitored according to the target monitoring video, and the determining of the motion information of the object to be monitored further comprises:

wherein eta is_kA determination value representing a geographical range in which the object to be monitored may appear;

L_i，krepresenting the distance between the ith characteristic coordinate point and the kth edge coordinate point of the object to be monitored in a certain frame of image in the target monitoring video;

t represents a preset time period;

u () represents a step function, and the function value is 1 when the value in the parentheses is 0 or more and 0 when the value in the parentheses is less than 0;

when eta_kWhen the value is 0, the object to be monitored does not appear near the k-th edge coordinate point;

when eta_kWhen 1, it indicates that the object to be monitored may appear in the vicinity of the k-th edge coordinate point.

7. The big data network-based monitoring method according to claim 1, wherein the process of performing frame labeling on the human body in each frame of human body image and acquiring one or more corresponding human body image frames comprises:

pre-labeling the human body, the human head and the human face in each frame of image to obtain a pre-labeled image frame;

taking a human body frame as a parent-level image frame and a human head frame as a child-level image frame, and establishing an affiliation relationship between the human body frame and the human head frame; taking the human head frame as a parent-level image frame and the human face frame as a child-level image frame, and establishing the membership between the human head frame and the human face frame;

determining the image frame membership according to the pre-labeled image frame, and identifying whether the pre-labeled image frame has wrong labeling according to the image frame membership; wherein the error label comprises at least one of: marking the sub-level image frame which belongs to a certain parent-level image frame as not belonging to the parent-level image frame; and/or, if a child image frame not belonging to a parent image frame is marked as belonging to the parent image frame; the image frame after pre-labeling has error labeling;

and modifying the human body image frame with the wrong label, and taking the modified human body image frame as a final human body image frame.

8. The big data network-based monitoring method according to claim 1, further comprising:

identifying whether human body image frames of the object to be monitored in the current frame image and the rest frame images in the target monitoring video are in the same color and the same label;

if the label or color of a certain human body image frame in the current frame image is detected again after one or more frames of images are separated, and/or the label of the same human body image frame in the current frame image is different from the label of the adjacent one or more frames of images, and/or the color of the same human body image frame in the current frame image is different from the color of the adjacent one or more frames of images; the human body image frame of the object to be monitored of the target monitoring video is wrongly marked;

and modifying the human body image frame with the error according to the error labeling type.

9. The big data network-based monitoring method according to claim 1, wherein the process of the image capturing device establishing a network communication link with the image processor comprises:

acquiring monitoring request information input by a user on the image shooting equipment; the monitoring request information includes: monitoring time, an image shooting device number, an account name and an account password of the image shooting device;

generating a communication access request of the image shooting equipment according to the monitoring request information;

acquiring authentication information stored in an image processor by the image shooting equipment based on the communication access request; the authentication information includes at least: the system comprises an image shooting device number directory, an image shooting device account name directory and an image shooting device account password directory;

verifying the monitoring request information by using the verification information, and determining whether the image shooting equipment number, the image shooting equipment account name and the account password in the monitoring request information all exist in a directory corresponding to the verification information; if the network communication link exists completely, a network communication link between the image shooting equipment and the image processor is established; and if the image processor does not exist completely, a network communication link between the image shooting equipment and the image processor is not established.

10. A monitoring system based on big data network is characterized in that the monitoring system comprises:

the video acquisition module is used for acquiring monitoring videos shot by a plurality of image shooting devices;

the communication module is used for transmitting the monitoring video positioned at the image shooting equipment to the image processor according to a network communication link which is pre-established between the image shooting equipment and the image processor;

the image framing module is used for framing the monitoring video by using the image processor and acquiring a human body image containing a human body from the framed image;

the image labeling module is used for performing frame labeling on the human body in each frame of human body image to obtain one or more corresponding human body image frames;

the image frame marking module is used for marking one or more human body image frames belonging to the same human body with the same label and distributing the same color to the one or more human body image frames belonging to the same human body;

the image sorting module is used for sorting the acquired human body images according to the sequence of the shooting time;

the image screening module is used for determining an object to be monitored, selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image;

the monitoring video screening module is used for acquiring a monitoring video containing the target image and recording the monitoring video as a target monitoring video;

and the object monitoring module is used for monitoring the object to be monitored according to the target monitoring video and determining the motion information of the object to be monitored.

Technical Field

The invention relates to the technical field of monitoring, in particular to a monitoring system and a monitoring method based on a big data network.

Background

The webcam is a new generation of webcam produced by combining traditional webcams with web technology, and can transmit video to the other end of the earth through the web, and a remote browser can monitor the video without any specialized software, as long as the standard web browser (such as Microsoft IE or Netscape) is used. The network camera is generally composed of a lens, an image sensor, a sound sensor, an A/D converter, an image sensor, a sound sensor, a controller, a network image shooting device, an external alarm, a control interface and the like.

At present, a user can utilize a network camera to perform real-time video monitoring on a certain object, such as a certain person, a certain vehicle and a certain animal, but because the object to be monitored is moving, when the object to be monitored moves to different areas, different network cameras are needed to perform video monitoring on the object to be monitored, therefore, when the object to be monitored appears in the monitoring videos of a plurality of network cameras, if the user wants to check the videos containing the object to be monitored, the user needs to check more videos as much as possible, and the current video checking is basically performed by manual searching, so that the searching efficiency is low, the searching is difficult, and the method is very inconvenient. In addition, when a network camera is used for real-time video monitoring, due to poor network stability, the video may have network fluctuation and blocked signal transmission, which may cause network quality degradation, the transmitted monitoring video may be jammed, and the tone quality and image quality may not be clear.

Disclosure of Invention

In view of the above drawbacks of the prior art, an object of the present invention is to provide a monitoring system and a monitoring method based on a big data network, which are used to solve the problems of difficult searching, blocked transmission of monitoring video, and unclear sound quality and image quality when video monitoring is performed on an object to be monitored in the prior art.

In order to achieve the above objects and other related objects, the present invention provides a monitoring method based on big data network, comprising the following steps:

acquiring monitoring videos shot by a plurality of image shooting devices;

transmitting the monitoring video positioned at the image shooting equipment to an image processor according to a network communication link which is pre-established by the image shooting equipment and the image processor;

framing the monitoring video by using the image processor, and acquiring a human body image containing a human body from the framed image;

performing frame marking on the human body in each frame of human body image to obtain one or more corresponding human body image frames;

marking the same reference number on one or more human body image frames belonging to the same human body, and assigning the same color to one or more human body image frames belonging to the same human body;

sequencing the acquired human body images according to the sequence of the shooting time;

determining an object to be monitored, selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image;

acquiring a monitoring video containing the target image, and recording the monitoring video as a target monitoring video;

and monitoring the object to be monitored according to the target monitoring video, and determining the motion information of the object to be monitored.

Optionally, when the network communication link pre-established between the image capturing device and the image processor transmits the monitoring video located at the image capturing device to the image processor, the network communication link pre-established between the image capturing device and the image processor includes:

before the monitoring video is transmitted, generating a network test data packet in the image shooting equipment, and recording the network test data packet as a first network test data packet, wherein the first network test packet comprises network test data for establishing data communication;

acquiring the sending time of the first network test data packet when entering the network communication link from the image shooting equipment;

recording a network test data packet received by the image processor from the network communication link as a second network test data packet, and acquiring the receiving time of the second network test data packet received by the image processor from the network communication link; the second network test packet comprises network test data for establishing data communication;

determining whether network delay exists in a network communication link between the image shooting equipment and the image processor or not and the existing network delay time according to the receiving time and the transmitting time;

comparing a first network test data packet generated in the image shooting equipment with a second network test data packet received by the image processor, comparing network test data contained in the first network test data packet with network test data contained in the second network test data packet, and calculating a data loss rate and a data loss amount of the network communication link;

and generating a network test analysis report of the network communication link according to the network delay result, the data loss rate and the data loss amount.

Optionally, the method further comprises:

detecting whether a network communication link between the image shooting equipment and the image processor is abnormal or not by using the network test analysis report;

if the network communication link of the image shooting equipment is normal, determining that the access network of the image processor is abnormal;

if the network communication link of the image shooting equipment is abnormal, determining that the access network of the image processor is normal, or adjusting the network configuration of the image shooting equipment, and detecting the network communication link of the image shooting equipment again until the network communication link of the image shooting equipment is normal, detecting whether the network quality between the image shooting equipment and the image processor meets the communication transmission requirement of the monitoring video, and determining that the access network of the image processor is normal.

Optionally, monitoring the object to be monitored according to the target monitoring video, and when determining the motion information of the object to be monitored, the method includes:

determining the moving distance of the object to be monitored according to each frame of image in the target monitoring video, wherein the moving distance comprises the following steps:

in the formula, S_i，jRepresenting the moving distance of a target image frame corresponding to the object to be monitored from a j-1 frame image to a j frame image in the target monitoring video;

(x_i，j，y_i，j) Representing the coordinates of the ith characteristic coordinate point of a target image frame corresponding to the object to be monitored in the jth frame image in the target monitoring video;

(x_i-1，j，y_i-1，j) Representing the coordinates of the ith characteristic coordinate point of a target image frame corresponding to the object to be monitored in the jth-1 frame image in the target monitoring video;

wherein i and j are natural numbers.

Optionally, monitoring the object to be monitored according to the target monitoring video, and when determining the motion information of the object to be monitored, the method further includes:

determining a moving speed identification value of the object to be monitored according to each frame of image in the target monitoring video, wherein the moving speed identification value comprises the following steps:

in the formula, V represents a moving speed identification value of the object to be monitored;

f represents the image frame frequency in the target monitoring video;

m represents the total number of characteristic coordinate points in each frame of image in the target monitoring video;

n represents the number of image frames contained in the target surveillance video.

Optionally, monitoring the object to be monitored according to the target monitoring video, and when determining the motion information of the object to be monitored, the method further includes:

wherein eta is_kA determination value representing a geographical range in which the object to be monitored may appear;

L_i，krepresenting the distance between the ith characteristic coordinate point and the kth edge coordinate point of the object to be monitored in a certain frame of image in the target monitoring video;

t represents a preset time period;

u () represents a step function, and the function value is 1 when the value in the parentheses is 0 or more and 0 when the value in the parentheses is less than 0;

when eta_kWhen the value is 0, the object to be monitored does not appear near the k-th edge coordinate point;

when eta_kWhen 1, it indicates that the object to be monitored may appear in the vicinity of the k-th edge coordinate point.

Optionally, the frame labeling is performed on the human body in each frame of the human body image, and the process of obtaining one or more corresponding human body image frames includes:

pre-labeling the human body, the human head and the human face in each frame of image to obtain a pre-labeled image frame;

taking a human body frame as a parent-level image frame and a human head frame as a child-level image frame, and establishing an affiliation relationship between the human body frame and the human head frame; taking the human head frame as a parent-level image frame and the human face frame as a child-level image frame, and establishing the membership between the human head frame and the human face frame;

determining the image frame membership according to the pre-labeled image frame, and identifying whether the pre-labeled image frame has wrong labeling according to the image frame membership; wherein the error label comprises at least one of: marking the sub-level image frame which belongs to a certain parent-level image frame as not belonging to the parent-level image frame; and/or, if a child image frame not belonging to a parent image frame is marked as belonging to the parent image frame; the image frame after pre-labeling has error labeling;

and modifying the human body image frame with the wrong label, and taking the modified human body image frame as a final human body image frame.

Optionally, the method further comprises:

identifying whether human body image frames of the object to be monitored in the current frame image and the rest frame images in the target monitoring video are in the same color and the same label;

if the label or color of a certain human body image frame in the current frame image is detected again after one or more frames of images are separated, and/or the label of the same human body image frame in the current frame image is different from the label of the adjacent one or more frames of images, and/or the color of the same human body image frame in the current frame image is different from the color of the adjacent one or more frames of images; the human body image frame of the object to be monitored of the target monitoring video is wrongly marked;

and modifying the human body image frame with the error according to the error labeling type.

Optionally, the process of establishing a network communication link between the image capturing device and the image processor includes:

acquiring monitoring request information input by a user on the image shooting equipment; the monitoring request information includes: monitoring time, an image shooting device number, an account name and an account password of the image shooting device;

generating a communication access request of the image shooting equipment according to the monitoring request information;

acquiring authentication information stored in an image processor by the image shooting equipment based on the communication access request; the authentication information includes at least: the system comprises an image shooting device number directory, an image shooting device account name directory and an image shooting device account password directory;

verifying the monitoring request information by using the verification information, and determining whether the image shooting equipment number, the image shooting equipment account name and the account password in the monitoring request information all exist in a directory corresponding to the verification information; if the network communication link exists completely, a network communication link between the image shooting equipment and the image processor is established; and if the image processor does not exist completely, a network communication link between the image shooting equipment and the image processor is not established.

The invention also provides a monitoring system based on the big data network, which comprises:

the video acquisition module is used for acquiring monitoring videos shot by a plurality of image shooting devices;

the communication module is used for transmitting the monitoring video positioned at the image shooting equipment to the image processor according to a network communication link which is pre-established between the image shooting equipment and the image processor;

the image framing module is used for framing the monitoring video by using the image processor and acquiring a human body image containing a human body from the framed image;

the image labeling module is used for performing frame labeling on the human body in each frame of human body image to obtain one or more corresponding human body image frames;

the image frame marking module is used for marking one or more human body image frames belonging to the same human body with the same label and distributing the same color to the one or more human body image frames belonging to the same human body;

the image sorting module is used for sorting the acquired human body images according to the sequence of the shooting time;

the image screening module is used for determining an object to be monitored, selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image;

the monitoring video screening module is used for acquiring a monitoring video containing the target image and recording the monitoring video as a target monitoring video;

and the object monitoring module is used for monitoring the object to be monitored according to the target monitoring video and determining the motion information of the object to be monitored.

As described above, the present invention provides a monitoring system and a monitoring method based on a big data network, which have the following beneficial effects: firstly, acquiring monitoring videos shot by a plurality of image shooting devices, and transmitting the monitoring videos at the image shooting devices to an image processor according to a network communication link which is pre-established between the image shooting devices and the image processor; then, framing the monitoring video by using an image processor, and acquiring a human body image containing a human body from the framed image; performing frame marking on the human body in each frame of human body image to obtain one or more corresponding human body image frames; marking the same label on one or more human body image frames belonging to the same human body, and distributing the same color on the one or more human body image frames belonging to the same human body; sequencing the acquired human body images according to the sequence of the shooting time; determining an object to be monitored, selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image; acquiring a monitoring video containing a target image, and recording the monitoring video as a target monitoring video; and monitoring the object to be monitored according to the target monitoring video, and determining the motion information of the object to be monitored. Therefore, the network communication link between the image shooting equipment and the image processor is established firstly, then the human body image containing the human body is selected from the monitoring video, the human body in the human body image is subjected to frame marking, the marked image frame is marked with colors and numbers, the obtained human body images are sequenced according to the sequence of the shooting time after the object to be monitored is determined, the target image containing the object to be monitored and the target monitoring video containing the target image are screened out according to the mark number and the color of the object to be monitored, and finally the object to be monitored is monitored according to the target monitoring video, and the motion information of the object to be monitored is determined. Therefore, the invention determines the motion information of the object to be monitored from the target monitoring video according to the color and the number corresponding to the image frame of the object to be monitored, thereby completing the monitoring work of the object to be monitored. In addition, according to the scheme recorded by the invention, the user does not need to search the monitoring videos one by one, only the monitoring videos with the objects to be monitored need to be searched, and all the monitoring videos are equivalently primarily screened, so that the manual searching efficiency is improved, and the manual searching difficulty is reduced.

Drawings

Fig. 1 is a schematic flowchart of a monitoring method based on a big data network according to an embodiment;

fig. 2 is a schematic hardware structure diagram of a monitoring system based on a big data network according to an embodiment.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.

It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.

Referring to fig. 1, the present invention provides a monitoring method based on a big data network, including the following steps:

s10, acquiring monitoring videos shot by a plurality of image shooting devices;

s20, transmitting the monitoring video at the image shooting equipment to the image processor according to the network communication link which is pre-established by the image shooting equipment and the image processor;

s30, framing the monitoring video by using the image processor, and acquiring a human body image containing a human body from the framed image;

s40, labeling the human body in each frame of human body image to obtain one or more corresponding human body image frames;

s50, marking one or more human body image frames belonging to the same human body with the same label, and distributing the same color to one or more human body image frames belonging to the same human body;

s60, sequencing the acquired human body images according to the sequence of the shooting time;

s70, determining an object to be monitored, and selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image;

s80, acquiring a monitoring video containing the target image and recording as a target monitoring video;

s90, monitoring the object to be monitored according to the target monitoring video, and determining the motion information of the object to be monitored.

Therefore, the method comprises the steps of firstly establishing a network communication link between image shooting equipment and an image processor, then selecting a human body image containing a human body from monitoring videos, carrying out frame marking on the human body in the human body image, marking colors and numbers on an image frame after marking, then sequencing the obtained human body images according to the sequence of shooting time after determining an object to be monitored, screening out a target image containing the object to be monitored and a target monitoring video containing the target image according to the label and the color of the object to be monitored, and finally monitoring the object to be monitored according to the target monitoring video and determining the motion information of the object to be monitored. Therefore, the method determines the motion information of the object to be monitored from the target monitoring video according to the color and the number corresponding to the image frame of the object to be monitored, so that the monitoring work of the object to be monitored can be completed. In addition, according to the scheme recorded by the method, the user does not need to search the monitoring videos one by one, only the monitoring videos with the objects to be monitored need to be searched, and all the monitoring videos are equivalently primarily screened, so that the manual searching efficiency is improved, and the manual searching difficulty is reduced.

In an exemplary embodiment, the network communication link pre-established by the image capturing device and the image processor, when transmitting the monitoring video located at the image capturing device to the image processor, includes:

before the monitoring video is transmitted, generating a network test data packet in the image shooting equipment, and recording the network test data packet as a first network test data packet, wherein the first network test packet comprises network test data for establishing data communication;

acquiring the sending time of the first network test data packet when entering the network communication link from the image shooting equipment;

recording a network test data packet received by the image processor from the network communication link as a second network test data packet, and acquiring the receiving time of the second network test data packet received by the image processor from the network communication link; the second network test packet comprises network test data for establishing data communication;

determining whether network delay exists in a network communication link between the image shooting equipment and the image processor or not and the existing network delay time according to the receiving time and the transmitting time;

comparing a first network test data packet generated in the image shooting equipment with a second network test data packet received by the image processor, comparing network test data contained in the first network test data packet with network test data contained in the second network test data packet, and calculating a data loss rate and a data loss amount of the network communication link;

and generating a network test analysis report of the network communication link according to the network delay result, the data loss rate and the data loss amount.

In the embodiment, after the network communication link between the image shooting device and the image processor is subjected to network test, whether the problems of network fluctuation, blocked signal transmission, blockage, unclear tone quality and image quality and the like occur when the monitoring video is transmitted from the image shooting device to the image processor can be judged, so that the flow bandwidth of the network communication link is increased according to the corresponding problems, or the fluctuation of the network communication link is reduced, the monitoring video can be stably transmitted, and the method can stably monitor the object to be monitored in real time through the monitoring video.

According to the above, the method further comprises: detecting whether a network communication link between the image shooting equipment and the image processor is abnormal or not by using the network test analysis report; if the network communication link of the image shooting equipment is normal, determining that the access network of the image processor is abnormal; if the network communication link of the image shooting equipment is abnormal, determining that the access network of the image processor is normal, or adjusting the network configuration of the image shooting equipment, and detecting the network communication link of the image shooting equipment again until the network communication link of the image shooting equipment is normal, detecting whether the network quality between the image shooting equipment and the image processor meets the communication transmission requirement of the monitoring video, and determining that the access network of the image processor is normal. According to the embodiment, whether the communication connection between the image shooting device and the image processor is normal or not is judged, so that the problem that the monitoring of the object to be monitored fails due to the fact that real-time monitoring videos are not transmitted when the object to be monitored is solved.

In an exemplary embodiment, monitoring the object to be monitored according to the target monitoring video, and when determining the motion information of the object to be monitored, the method includes: determining the moving distance of the object to be monitored according to each frame of image in the target monitoring video, wherein the moving distance comprises the following steps:

in the formula, S_i，jRepresenting the moving distance of a target image frame corresponding to the object to be monitored from a j-1 frame image to a j frame image in the target monitoring video;

(x_i，j，y_i，j) Representing the coordinates of the ith characteristic coordinate point of a target image frame corresponding to the object to be monitored in the jth frame image in the target monitoring video;

(x_i-1，j，y_i-1，j) Representing the coordinates of the ith characteristic coordinate point of a target image frame corresponding to the object to be monitored in the jth-1 frame image in the target monitoring video; wherein i and j are natural numbers.

After the moving distance of the object to be monitored is determined, determining a moving speed identification value of the object to be monitored according to each frame of image in the target monitoring video, wherein the moving speed identification value comprises the following steps:

in the formula, V represents a moving speed identification value of the object to be monitored;

f represents the image frame frequency in the target monitoring video;

m represents the total number of characteristic coordinate points in each frame of image in the target monitoring video;

n represents the number of image frames contained in the target surveillance video.

According to the above description, after determining the moving speed identification value of the object to be monitored, the method further includes determining a geographical range in which the object to be monitored may appear, including:

wherein eta is_kA determination value representing a geographical range in which the object to be monitored may appear;

L_i，krepresenting the distance between the ith characteristic coordinate point and the kth edge coordinate point of the object to be monitored in a certain frame of image in the target monitoring video;

t represents a preset time period;

u () represents a step function, and the function value is 1 when the value in the parentheses is 0 or more and 0 when the value in the parentheses is less than 0;

when eta_kWhen the value is 0, the object to be monitored does not appear near the k-th edge coordinate point;

when eta_kWhen 1, it indicates that the object to be monitored may appear in the vicinity of the k-th edge coordinate point.

Therefore, the method can determine the moving distance of each frame of the object to be monitored in the target monitoring video according to the target monitoring video, and after the moving distance of the object to be monitored is determined, the moving speed identification value of the object to be monitored can be determined according to each frame of image in the target monitoring video, and the geographical range in which the object to be monitored may appear can be determined. The method and the device can determine the motion information of the object to be monitored according to the target monitoring video, determine the motion time and the motion track route of the object to be monitored, and facilitate the tracking and monitoring of the object to be monitored according to the corresponding motion time and the motion track route by a user.

In an exemplary embodiment, the frame labeling is performed on the human body in each frame of human body image, and the process of obtaining one or more corresponding human body image frames includes: pre-labeling the human body, the human head and the human face in each frame of image to obtain a pre-labeled image frame; taking a human body frame as a parent-level image frame and a human head frame as a child-level image frame, and establishing an affiliation relationship between the human body frame and the human head frame; taking the human head frame as a parent-level image frame and the human face frame as a child-level image frame, and establishing the membership between the human head frame and the human face frame; determining the image frame membership according to the pre-labeled image frame, and identifying whether the pre-labeled image frame has wrong labeling according to the image frame membership; wherein the error label comprises at least one of: marking the sub-level image frame which belongs to a certain parent-level image frame as not belonging to the parent-level image frame; and/or, if a child image frame not belonging to a parent image frame is marked as belonging to the parent image frame; the image frame after pre-labeling has error labeling; and modifying the human body image frame with the wrong label, and taking the modified human body image frame as a final human body image frame. According to the invention, the dependency relationship of the human body image frames is modified, so that the head of the person A can be prevented from being associated with the human body of the person B when the human body image frames are labeled.

In an exemplary embodiment, the method further comprises: identifying whether human body image frames of the object to be monitored in the current frame image and the rest frame images in the target monitoring video are in the same color and the same label; if the label or color of a certain human body image frame in the current frame image is detected again after one or more frames of images are separated, and/or the label of the same human body image frame in the current frame image is different from the label of the adjacent one or more frames of images, and/or the color of the same human body image frame in the current frame image is different from the color of the adjacent one or more frames of images; the human body image frame of the object to be monitored of the target monitoring video is wrongly marked; and modifying the human body image frame with the error according to the error labeling type. According to the invention, through modifying the human body image frame, the situation that an error monitoring video is found when the monitoring video is found by using the color and the label corresponding to the human body image frame can be avoided, so that the monitoring error rate is increased, and the monitoring efficiency is reduced.

In an exemplary embodiment, the process of the image capturing device establishing a network communication link with the image processor includes: acquiring monitoring request information input by a user on the image shooting equipment; the monitoring request information includes: monitoring time, an image shooting device number, an account name and an account password of the image shooting device; generating a communication access request of the image shooting equipment according to the monitoring request information; acquiring authentication information stored in an image processor by the image shooting equipment based on the communication access request; the authentication information includes at least: the system comprises an image shooting device number directory, an image shooting device account name directory and an image shooting device account password directory; verifying the monitoring request information by using the verification information, and determining whether the image shooting equipment number, the image shooting equipment account name and the account password in the monitoring request information all exist in a directory corresponding to the verification information; if the network communication link exists completely, a network communication link between the image shooting equipment and the image processor is established; and if the image processor does not exist completely, a network communication link between the image shooting equipment and the image processor is not established.

In summary, the present invention provides a monitoring method based on a big data network, which includes first obtaining monitoring videos captured by a plurality of image capturing devices, and transmitting the monitoring videos located at the image capturing devices to an image processor according to a network communication link pre-established between the image capturing devices and the image processor; then, framing the monitoring video by using an image processor, and acquiring a human body image containing a human body from the framed image; performing frame marking on the human body in each frame of human body image to obtain one or more corresponding human body image frames; marking the same label on one or more human body image frames belonging to the same human body, and distributing the same color on the one or more human body image frames belonging to the same human body; sequencing the acquired human body images according to the sequence of the shooting time; determining an object to be monitored, selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image; acquiring a monitoring video containing a target image, and recording the monitoring video as a target monitoring video; and monitoring the object to be monitored according to the target monitoring video, and determining the motion information of the object to be monitored. Therefore, the network communication link between the image shooting equipment and the image processor is established firstly, then the human body image containing the human body is selected from the monitoring video, the human body in the human body image is subjected to frame marking, the marked image frame is marked with colors and numbers, the obtained human body images are sequenced according to the sequence of the shooting time after the object to be monitored is determined, the target image containing the object to be monitored and the target monitoring video containing the target image are screened out according to the mark number and the color of the object to be monitored, and finally the object to be monitored is monitored according to the target monitoring video, and the motion information of the object to be monitored is determined. Therefore, the invention determines the motion information of the object to be monitored from the target monitoring video according to the color and the number corresponding to the image frame of the object to be monitored, thereby completing the monitoring work of the object to be monitored. In addition, according to the scheme recorded by the invention, the user does not need to search the monitoring videos one by one, only the monitoring videos with the objects to be monitored need to be searched, and all the monitoring videos are equivalently primarily screened, so that the manual searching efficiency is improved, and the manual searching difficulty is reduced. In addition, after the network communication link between the image shooting device and the image processor is subjected to network test, the method can judge whether the problems of network fluctuation, blocked signal transmission, blockage, unclear tone quality and image quality and the like occur when the monitoring video is transmitted from the image shooting device to the image processor, so that the flow bandwidth of the network communication link is increased according to the corresponding problems, or the fluctuation of the network communication link is reduced, the monitoring video can be stably transmitted, and the method can stably monitor the object to be monitored in real time through the monitoring video.

As shown in fig. 2, the present invention further provides a monitoring system based on big data network, which includes:

the video acquisition module M10 is used for acquiring monitoring videos shot by a plurality of image shooting devices;

the communication module M20 is used for transmitting the monitoring video at the image shooting equipment to the image processor according to the network communication link which is established in advance by the image shooting equipment and the image processor;

an image framing module M30, configured to frame the surveillance video by using the image processor, and obtain a human body image including a human body from the framed image;

the image labeling module M40 is used for performing frame labeling on the human body in each frame of human body image to obtain one or more corresponding human body image frames;

an image frame labeling module M50, for labeling the same label for one or more human body image frames belonging to the same human body, and assigning the same color for one or more human body image frames belonging to the same human body;

the image sorting module M60 is used for sorting the acquired human body images according to the sequence of the shooting time;

the image screening module M70 is used for determining an object to be monitored, and selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image;

the monitoring video screening module M80 is configured to acquire a monitoring video including the target image, and record the monitoring video as a target monitoring video;

and the object monitoring module M90 is configured to monitor the object to be monitored according to the target monitoring video, and determine motion information of the object to be monitored.

Therefore, the system firstly establishes a network communication link between the image shooting equipment and the image processor, then selects a human body image containing a human body from the monitoring video, carries out frame marking on the human body in the human body image, marks the color and the number of the image frame after marking, then sorts the obtained human body images according to the sequence of shooting time after determining the object to be monitored, then screens out a target image containing the object to be monitored and a target monitoring video containing the target image according to the label and the color of the object to be monitored, and finally monitors the object to be monitored according to the target monitoring video and determines the motion information of the object to be monitored. Therefore, the system determines the motion information of the object to be monitored from the target monitoring video according to the color and the number corresponding to the image frame of the object to be monitored, so that the monitoring work of the object to be monitored can be completed. Moreover, according to the scheme recorded by the system, the system does not need a user to search the monitoring videos one by one, only needs to search the monitoring videos with the objects to be monitored, and equivalently, all the monitoring videos are preliminarily screened, so that the manual searching efficiency is improved, and the manual searching difficulty is reduced.

In an exemplary embodiment, the network communication link pre-established by the image capturing device and the image processor, when transmitting the monitoring video located at the image capturing device to the image processor, includes:

before the monitoring video is transmitted, generating a network test data packet in the image shooting equipment, and recording the network test data packet as a first network test data packet, wherein the first network test packet comprises network test data for establishing data communication;

acquiring the sending time of the first network test data packet when entering the network communication link from the image shooting equipment;

recording a network test data packet received by the image processor from the network communication link as a second network test data packet, and acquiring the receiving time of the second network test data packet received by the image processor from the network communication link; the second network test packet comprises network test data for establishing data communication;

determining whether network delay exists in a network communication link between the image shooting equipment and the image processor or not and the existing network delay time according to the receiving time and the transmitting time;

comparing a first network test data packet generated in the image shooting equipment with a second network test data packet received by the image processor, comparing network test data contained in the first network test data packet with network test data contained in the second network test data packet, and calculating a data loss rate and a data loss amount of the network communication link;

and generating a network test analysis report of the network communication link according to the network delay result, the data loss rate and the data loss amount.

In the embodiment, after the network communication link between the image shooting device and the image processor is subjected to network test, whether the problems of network fluctuation, blocked signal transmission, blockage, unclear tone quality and picture quality and the like can occur when the monitoring video is transmitted from the image shooting device to the image processor can be judged, so that the flow bandwidth of the network communication link is increased according to the corresponding problems, or the fluctuation of the network communication link is reduced, the monitoring video can be stably transmitted, and the system can stably monitor the object to be monitored in real time through the monitoring video.

According to the above, the method further comprises: detecting whether a network communication link between the image shooting equipment and the image processor is abnormal or not by using the network test analysis report; if the network communication link of the image shooting equipment is normal, determining that the access network of the image processor is abnormal; if the network communication link of the image shooting equipment is abnormal, determining that the access network of the image processor is normal, or adjusting the network configuration of the image shooting equipment, and detecting the network communication link of the image shooting equipment again until the network communication link of the image shooting equipment is normal, detecting whether the network quality between the image shooting equipment and the image processor meets the communication transmission requirement of the monitoring video, and determining that the access network of the image processor is normal. According to the embodiment, whether the communication connection between the image shooting device and the image processor is normal or not is judged, so that the problem that the monitoring of the object to be monitored fails due to the fact that real-time monitoring videos are not transmitted when the object to be monitored is solved.

In an exemplary embodiment, monitoring the object to be monitored according to the target monitoring video, and when determining the motion information of the object to be monitored, the method includes: determining the moving distance of the object to be monitored according to each frame of image in the target monitoring video, wherein the moving distance comprises the following steps:

in the formula, S_i，jRepresenting the moving distance of a target image frame corresponding to the object to be monitored from a j-1 frame image to a j frame image in the target monitoring video;

(x_i，j，y_i，j) Representing the coordinates of the ith characteristic coordinate point of a target image frame corresponding to the object to be monitored in the jth frame image in the target monitoring video;

(x_i-1，j，y_i-1，j) Representing the coordinates of the ith characteristic coordinate point of a target image frame corresponding to the object to be monitored in the jth-1 frame image in the target monitoring video; wherein i and j are natural numbers.

After the moving distance of the object to be monitored is determined, determining a moving speed identification value of the object to be monitored according to each frame of image in the target monitoring video, wherein the moving speed identification value comprises the following steps:

in the formula, V represents a moving speed identification value of the object to be monitored;

f represents the image frame frequency in the target monitoring video;

m represents the total number of characteristic coordinate points in each frame of image in the target monitoring video;

n represents the number of image frames contained in the target surveillance video.

According to the above description, after determining the moving speed identification value of the object to be monitored, the method further includes determining a geographical range in which the object to be monitored may appear, including:

wherein eta is_kA determination value representing a geographical range in which the object to be monitored may appear;

L_i，krepresenting the distance between the ith characteristic coordinate point and the kth edge coordinate point of the object to be monitored in a certain frame of image in the target monitoring video;

i represents a preset time period;

u () represents a step function, and the function value is 1 when the value in the parentheses is 0 or more and 0 when the value in the parentheses is less than 0;

when eta_kWhen the value is 0, the object to be monitored does not appear near the k-th edge coordinate point;

when eta_kWhen 1, it indicates that the object to be monitored may appear in the vicinity of the k-th edge coordinate point.

Therefore, the system can determine the moving distance of each frame of the object to be monitored in the target monitoring video according to the target monitoring video, and after the moving distance of the object to be monitored is determined, the moving speed identification value of the object to be monitored can be determined according to each frame of image in the target monitoring video, and the geographical range in which the object to be monitored may appear can be determined. The method and the device can determine the motion information of the object to be monitored according to the target monitoring video, determine the motion time and the motion track route of the object to be monitored, and facilitate the tracking and monitoring of the object to be monitored according to the corresponding motion time and the motion track route by a user.

In an exemplary embodiment, the frame labeling is performed on the human body in each frame of human body image, and the process of obtaining one or more corresponding human body image frames includes: pre-labeling the human body, the human head and the human face in each frame of image to obtain a pre-labeled image frame; taking a human body frame as a parent-level image frame and a human head frame as a child-level image frame, and establishing an affiliation relationship between the human body frame and the human head frame; taking the human head frame as a parent-level image frame and the human face frame as a child-level image frame, and establishing the membership between the human head frame and the human face frame; determining the image frame membership according to the pre-labeled image frame, and identifying whether the pre-labeled image frame has wrong labeling according to the image frame membership; wherein the error label comprises at least one of: marking the sub-level image frame which belongs to a certain parent-level image frame as not belonging to the parent-level image frame; and/or, if a child image frame not belonging to a parent image frame is marked as belonging to the parent image frame; the image frame after pre-labeling has error labeling; and modifying the human body image frame with the wrong label, and taking the modified human body image frame as a final human body image frame. According to the invention, the dependency relationship of the human body image frames is modified, so that the head of the person A can be prevented from being associated with the human body of the person B when the human body image frames are labeled.

In an exemplary embodiment, the method further comprises: identifying whether human body image frames of the object to be monitored in the current frame image and the rest frame images in the target monitoring video are in the same color and the same label; if the label or color of a certain human body image frame in the current frame image is detected again after one or more frames of images are separated, and/or the label of the same human body image frame in the current frame image is different from the label of the adjacent one or more frames of images, and/or the color of the same human body image frame in the current frame image is different from the color of the adjacent one or more frames of images; the human body image frame of the object to be monitored of the target monitoring video is wrongly marked; and modifying the human body image frame with the error according to the error labeling type. According to the invention, through modifying the human body image frame, the situation that an error monitoring video is found when the monitoring video is found by using the color and the label corresponding to the human body image frame can be avoided, so that the monitoring error rate is increased, and the monitoring efficiency is reduced.

In an exemplary embodiment, the process of the image capturing device establishing a network communication link with the image processor includes: acquiring monitoring request information input by a user on the image shooting equipment; the monitoring request information includes: monitoring time, an image shooting device number, an account name and an account password of the image shooting device; generating a communication access request of the image shooting equipment according to the monitoring request information; acquiring authentication information stored in an image processor by the image shooting equipment based on the communication access request; the authentication information includes at least: the system comprises an image shooting device number directory, an image shooting device account name directory and an image shooting device account password directory; verifying the monitoring request information by using the verification information, and determining whether the image shooting equipment number, the image shooting equipment account name and the account password in the monitoring request information all exist in a directory corresponding to the verification information; if the network communication link exists completely, a network communication link between the image shooting equipment and the image processor is established; and if the image processor does not exist completely, a network communication link between the image shooting equipment and the image processor is not established.

In summary, the present invention provides a monitoring method based on a big data network, which includes first obtaining monitoring videos captured by a plurality of image capturing devices, and transmitting the monitoring videos located at the image capturing devices to an image processor according to a network communication link pre-established between the image capturing devices and the image processor; then, framing the monitoring video by using an image processor, and acquiring a human body image containing a human body from the framed image; performing frame marking on the human body in each frame of human body image to obtain one or more corresponding human body image frames; marking the same label on one or more human body image frames belonging to the same human body, and distributing the same color on the one or more human body image frames belonging to the same human body; sequencing the acquired human body images according to the sequence of the shooting time; determining an object to be monitored, selecting an image containing the object to be monitored from the sorted human body images according to the label and the color corresponding to the object to be monitored, and recording the image as a target image; acquiring a monitoring video containing a target image, and recording the monitoring video as a target monitoring video; and monitoring the object to be monitored according to the target monitoring video, and determining the motion information of the object to be monitored. Therefore, the network communication link between the image shooting equipment and the image processor is established firstly, then the human body image containing the human body is selected from the monitoring video, the human body in the human body image is subjected to frame marking, the marked image frame is marked with colors and numbers, the obtained human body images are sequenced according to the sequence of the shooting time after the object to be monitored is determined, the target image containing the object to be monitored and the target monitoring video containing the target image are screened out according to the mark number and the color of the object to be monitored, and finally the object to be monitored is monitored according to the target monitoring video, and the motion information of the object to be monitored is determined. Therefore, the invention determines the motion information of the object to be monitored from the target monitoring video according to the color and the number corresponding to the image frame of the object to be monitored, thereby completing the monitoring work of the object to be monitored. In addition, according to the scheme recorded by the invention, the user does not need to search the monitoring videos one by one, only the monitoring videos with the objects to be monitored need to be searched, and all the monitoring videos are equivalently primarily screened, so that the manual searching efficiency is improved, and the manual searching difficulty is reduced. In addition, after the system carries out network test on the network communication link between the image shooting device and the image processor, the system can judge whether the problems of network fluctuation, blocked signal transmission, blockage, unclear tone quality and image quality and the like can occur when the monitoring video is transmitted from the image shooting device to the image processor, thereby increasing the flow bandwidth of the network communication link according to the corresponding problems, or reducing the fluctuation of the network communication link, keeping the stable transmission of the monitoring video, and leading the system to stably monitor the object to be monitored in real time through the monitoring video.

The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.