Intrusion detection method, device, equipment, medium and product

Abstract

The invention provides an intrusion detection method and device, equipment, a medium and a product. The method comprises the following steps: receiving video image data sent by terminal equipment; performing first preprocessing on the video image data to obtain a to-be-detected image; performing second preprocessing on the to-be-detected image to determine whether a moving object exists in the to-be-detected image; if the moving object exists in the to-be-detected image, a trained detection model is adopted to identify the to-be-detected image so as to determine whether an invader exists in the to-be-detected image, and the trained detection model is obtained through training of a training data set; the training data set comprises a plurality of training image samples and invader labels corresponding to the training image samples; and if the invader exists in the to-be-detected image, emitting an intrusion alarm. The method improves the accuracy of intrusion detection.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an intrusion detection method, device, equipment, medium and product. Background Art

[0002] Video surveillance technology can be applied to many fields such as residential communities, public security and traffic control to monitor target areas and provide feedback on whether the monitored target has entered the target area. Traditional manual monitoring methods cannot achieve real-time monitoring throughout the day.

[0003] In the prior art, deep learning-based intrusion monitoring technology is used to identify monitoring targets. However, due to the interference of the actual monitoring environment, the monitoring results of the intrusion monitoring technology in the prior art are not accurate enough and therefore cannot meet user needs.

[0004] Based on this, there is a problem of insufficient accuracy of intrusion monitoring in the prior art. Summary of the invention

[0005] The present application provides intrusion detection methods, devices, equipment, media and products to improve the accuracy of intrusion monitoring technology.

[0006] In a first aspect, the present application provides an intrusion detection method, which is applied to a cloud server, and the method comprises:

[0007] Receiving video image data sent by a terminal device;

[0008] Performing a first preprocessing on the video image data to obtain an image to be detected;

[0009] Performing a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected;

[0010] If there is a moving object in the image to be detected, the trained detection model is used to identify the image to be detected to determine whether there is an intruder in the image to be detected, wherein the trained detection model is obtained by training a training data set, and the training data set includes multiple training image samples and intruder labels corresponding to the training image samples;

[0011] If there is an intruder in the image to be detected, an intrusion alarm will be issued.

[0012] In a possible embodiment, a trained detection model is used to identify the image to be detected to determine whether there is an intruder in the image to be detected, including:

[0013] The trained detection model is used to identify the image to be detected, so as to obtain the intruder label corresponding to the image to be detected through the trained detection model;

[0014] Identify the scene corresponding to the image to be detected;

[0015] Determine the intruder identification threshold based on the scenario;

[0016] According to the intruder label corresponding to the image to be detected and the intruder identification threshold, it is determined whether there is an intruder in the image to be detected.

[0017] In a possible embodiment, identifying a scene corresponding to the image to be detected includes:

[0018] Acquire environmental data of the image to be detected, wherein the environmental data includes at least one of time, weather and lighting conditions;

[0019] According to the environmental data, the scene corresponding to the image to be detected is determined.

[0020] In a possible embodiment, before using the trained detection model to identify the image to be detected so as to obtain the intruder label corresponding to the image to be detected through the trained detection model, the method further includes:

[0021] Get the training dataset;

[0022] The initial detection model is trained using the training data set to obtain a trained detection model.

[0023] In a possible embodiment, the initial detection model is trained using a training data set to obtain a trained detection model, including:

[0024] The initial detection model is trained using the training data set to obtain a trained detection model;

[0025] Get the test dataset;

[0026] The trained detection model is trained using the test data set to obtain a trained detection model.

[0027] In a possible embodiment, the detection model is a target detection model that integrates a dilated convolution layer and a separable convolution that combines depthwise convolution with point-by-point convolution.

[0028] In a possible embodiment, performing a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected includes:

[0029] The difference operation of adjacent frames is performed on the image to be detected to determine whether there is a moving object in the image to be detected.

[0030] In a possible embodiment, performing a first preprocessing on the video image data to obtain an image to be detected includes:

[0031] The video image data is first preprocessed by using a plurality of video stream encoding, contrast enhancement, shot segmentation, brightness adjustment, video stream decoding and storage to obtain an image to be detected.

[0032] In a second aspect, the present application provides an intrusion detection device, including: a memory, a processor;

[0033] Memory stores computer-executable instructions;

[0034] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0035] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the above first aspect and / or various possible implementations of the first aspect.

[0036] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementations of the first aspect.

[0037] The intrusion detection method, device, equipment, medium and product provided by the present application perform a first preprocessing on the video image data sent by the receiving terminal device to obtain the image to be detected; so as to improve the image quality and transmission efficiency of the video image to be detected; perform a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected; so as to reduce the number of training images to be detected, thereby improving the detection efficiency; if there is a moving object in the image to be detected, use the trained detection model to identify the image to be detected to determine whether there is an intruder in the image to be detected; compared with the deep learning-based detection model in the prior art, the present application trains the detection model through multiple training image samples in the training data set and the intruder labels corresponding to the training image samples, thereby improving the accuracy of intrusion detection; if there is an intruder in the image to be detected, an intrusion alarm is issued to improve the security of intrusion detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0039] Figure 1 A schematic diagram of an intrusion detection system architecture provided for this application;

[0040] Figure 2 A schematic diagram of an intrusion detection method provided in this application Figure 1 ;

[0041] Figure 3 A schematic diagram of the process of detecting intrusion provided in this application Figure 2 ;

[0042] Figure 4 A schematic diagram of the process of detecting intrusion provided in this application Figure 3 ;

[0043] Figure 5 A schematic diagram of the structure of an intrusion detection device provided in this application;

[0044] Figure 6 A schematic diagram of the structure of an intrusion detection device provided in this application.

[0045] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0046] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0047] It should be noted that the information (including but not limited to device information, user information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0048] Video surveillance technology is a technology that uses video cameras to monitor and record target areas in real time. It is widely used in the fields of security and management to prevent suspicious persons from invading prohibited areas.

[0049] The intrusion detection algorithms in the existing technology rely on deep learning or artificial intelligence technology, which requires a lot of computing resources and storage space, has low detection efficiency and cannot respond in a timely manner. In addition, in actual detection scenarios, if the detection target is small or obscured, or there are lighting problems and noise interference, the resolution of the image to be detected is low and the target to be detected cannot be accurately identified.

[0050] Based on this, in the prior art, there is a problem of low accuracy of intrusion detection.

[0051] In order to solve the above problems, the core concept of the present application is to: perform a first preprocessing on the received video image data to improve the quality of the image to be detected and further improve the accuracy of intrusion detection; perform a second preprocessing on the image to be detected, and use a trained detection model to identify the image to be detected with moving objects, so as to determine whether there is an intruder in the image to be detected, thereby reducing the number of invalid images to be detected and improving the efficiency of intrusion detection; if there is an intruder in the image to be detected, an intrusion alarm is issued to ensure the security of intrusion detection.

[0052] Figure 1 A schematic diagram of an intrusion detection system architecture provided for this application. Figure 1 In the above framework, the above framework includes at least one of a data acquisition device 101, a processing device 102, and a display device 103.

[0053] It is understandable that the structure illustrated in the embodiment of the present application does not constitute a specific limitation on the intrusion detection system architecture. In other feasible implementations of the present application, the above architecture may include more or fewer components than shown in the figure, or combine certain components, or split certain components, or arrange the components differently, which can be determined according to the actual application scenario and is not limited here. Figure 1 The components shown may be implemented in hardware, software, or a combination of software and hardware.

[0054] In a specific implementation process, the data acquisition device 101 may include an input / output interface and may also include a communication interface. The data acquisition device 101 may be connected to the processing device via the input / output interface or the communication interface.

[0055] The processing device 102 can be used to perform a first preprocessing on the video image data to obtain the image to be detected; perform a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected; if there is a moving object in the image to be detected, use the trained detection model to identify the image to be detected to determine whether there is an intruder in the image to be detected; if there is an intruder in the image to be detected, issue an intrusion alarm.

[0056] The display device 103 may also be a touch display screen or a screen of a terminal device, which is used to receive user instructions while displaying the above content to achieve interaction with the user.

[0057] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0058] Figure 2 A schematic diagram of an intrusion detection method provided in this application Figure 1 ,like Figure 2 As shown, the method is applied to a cloud server and includes:

[0059] S201. Receive video image data sent by a terminal device.

[0060] In this embodiment, the terminal device may be a surveillance camera; the video image data may be a video image of the target area to be detected collected by the surveillance camera; wherein the target area to be detected includes but is not limited to traffic roads, construction sites, scenic spots or supermarket warehouses.

[0061] S202: Perform a first preprocessing on the video image data to obtain an image to be detected.

[0062] Optionally, performing a first preprocessing on the video image data to obtain an image to be detected includes:

[0063] The video image data is first preprocessed by using a plurality of video stream encoding, contrast enhancement, shot segmentation, brightness adjustment, video stream decoding and storage to obtain an image to be detected.

[0064] In this embodiment, video stream encoding refers to reducing the size of video data through compression technology to save storage space and improve transmission efficiency; video stream decoding refers to decoding the video stream when the video is needed so as to perform real-time processing and analysis of the video; by encoding and decoding the video image data, the storage and transmission of the image to be detected are optimized.

[0065] Shot segmentation refers to dividing the video stream into different shots or scenes to facilitate analysis of specific scenes; by performing shot segmentation on video image data, the accuracy of the image to be detected is improved.

[0066] Contrast enhancement refers to an image processing technology that increases the difference between different brightness areas in an image to make image details clearer and easier to identify; brightness adjustment refers to an image processing technology that improves image visibility by changing the overall brightness level of the image; by performing contrast enhancement and brightness adjustment on video image data, the detection accuracy of the image to be detected under different lighting conditions is improved.

[0067] S203: Perform a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected.

[0068] Optionally, performing a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected includes:

[0069] The difference operation of adjacent frames is performed on the image to be detected to determine whether there is a moving object in the image to be detected.

[0070] In this embodiment, the difference operation of adjacent frames is performed on the image to be detected by using the inter-frame difference method to reduce the calculation complexity and reduce the calculation resources, so as to determine whether there is a moving object in the image to be detected.

[0071] Optionally, the inter-frame difference method in this embodiment may also be replaced by a single Gaussian model, a mixed Gaussian model or a visual background extraction algorithm to determine whether there is a moving object in the image to be detected.

[0072] S204. If there is a moving object in the image to be detected, a trained detection model is used to identify the image to be detected to determine whether there is an intruder in the image to be detected, wherein the trained detection model is obtained by training a training data set, and the training data set includes multiple training image samples and intruder labels corresponding to the training image samples.

[0073] In this embodiment, the detection model is a target detection model that integrates a dilated convolution layer and a separable convolution that combines depthwise convolution with point-by-point convolution.

[0074] Deep convolution refers to the convolution operation that performs convolution on each input channel separately, which reduces the amount of calculation and the number of parameters; the hole convolution layer refers to the convolution operation that inserts a hole into the standard convolution layer to increase the receptive field of the convolution kernel without increasing the number of parameters; point-by-point convolution refers to the use of 1 1 The convolution kernel performs a convolution operation and linearly combines the channels to change the number of channels of the feature map; separable convolution refers to a depth-wise separable convolution that combines depth-wise convolution and point-by-point convolution, which reduces the amount of computation and the number of parameters while maintaining the performance of the detection model.

[0075] In this embodiment, a trained detection model is used to identify the image to be detected to determine whether there is an intruder in the image to be detected; the computational complexity is reduced, the receptive field of the convolution kernel is increased, the feature details are retained, and the detection accuracy of the intruder is improved; and the detection model includes an attention mechanism, which can adjust the weights of different channels to highlight the key features of the target area and optimize the detection model's processing of noise, occlusion, and low-resolution scenes.

[0076] S205: If there is an intruder in the image to be detected, an intrusion alarm is issued.

[0077] In this embodiment, if the trained detection model detects that there is an intruder in the image to be detected, an intrusion alarm is issued to warn the intruder;

[0078] For example, the method proposed in the embodiment of the present application also tracks the movement trajectory of the intruder and records the intruder's stay time to improve the safety and reliability of security.

[0079] The intrusion detection method provided in the embodiment of the present application obtains the image to be detected by performing a first preprocessing on the video image data sent by the terminal device in the cloud server, thereby optimizing the storage, transmission efficiency, and accuracy of the image to be detected; performing a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected, thereby reducing the computational complexity and reducing the computational resources; if there is a moving object in the image to be detected, using a trained detection model to identify the image to be detected to determine whether there is an intruder in the image to be detected; compared with the traditional YOLOV5 (target detection algorithm) network, the trained detection model of the present application effectively reduces the computational complexity and improves the detection accuracy of intruders in scenarios with interference; if there is an intruder in the image to be detected, an intrusion alarm is issued, thereby improving security and reliability.

[0080] Figure 3 A schematic diagram of the process of detecting intrusion provided in this application Figure 2 ,like Figure 3 As shown, in this embodiment Figure 2 Based on the embodiment, the trained detection model is used in the above step S204 to identify the image to be detected to determine whether there is an intruder in the image to be detected, including:

[0081] S301: Use the trained detection model to identify the image to be detected, so as to obtain the intruder label corresponding to the image to be detected through the trained detection model.

[0082] In this embodiment, the detection model uses an ASESD (Atrous-Separable-Excitation-Space-Depth) block to replace the strided convolution and pooling layers in the original YOLOV5 network.

[0083] The ASESD block includes a first channel and a second channel, wherein the first channel downsamples the input feature data with a scale of 2 to obtain four sub-feature data; the second channel is divided into a first branch and a second branch, and the second channel is an ASE (Attention with Spatial Encoding) block, which downsamples the input feature data through a dilated convolution to obtain the first feature data; based on the depthwise separable convolution, the first feature data input to the first branch is firstly subjected to depthwise convolution and pointwise convolution, and the first feature data input to the second branch is directly transferred, and the sub-feature data output in the first branch and the second branch are subjected to feature fusion through multiplication operation to obtain the intruder label corresponding to the image to be detected.

[0084] S302: Identify the scene corresponding to the image to be detected.

[0085] Optionally, identifying a scene corresponding to the image to be detected includes:

[0086] Acquire environmental data of the image to be detected, wherein the environmental data includes at least one of time, weather and lighting conditions;

[0087] According to the environmental data, the scene corresponding to the image to be detected is determined.

[0088] In this embodiment, for example, the time, weather and lighting conditions of the image to be detected are detected through a preset scenario perception system to determine the environmental data of the image to be detected, wherein the preset scenario perception system includes sensors and a network information integration unit, the sensors include temperature sensors, humidity sensors, light intensity sensors, and wind speed sensors, and each sensor is used to monitor the environmental changes of the area to be detected in real time; the network information integration unit obtains environmental data by integrating weather forecast information on the Internet, or monitoring the environmental changes of the area to be detected based on each sensor.

[0089] S303: Determine an intruder identification threshold according to the scenario.

[0090] In this embodiment, the intruder recognition threshold is dynamically adjusted according to the environmental data of the image to be detected to improve the monitoring accuracy.

[0091] For example, the initial value of the intruder recognition threshold is not a default threshold. If the environmental data is a rainy or snowy environment, the intruder recognition threshold can be lowered; if the environmental data is at night or the light intensity is lower than the preset threshold, the intruder recognition threshold is increased.

[0092] Optionally, the intruder identification threshold may be adjusted according to time. For example, different intruder identification thresholds may be adjusted according to working hours and holidays.

[0093] S304: Determine whether there is an intruder in the image to be detected according to the intruder label corresponding to the image to be detected and the intruder identification threshold.

[0094] In this embodiment, the detection model compares the intruder label corresponding to the image to be detected and the intruder recognition threshold. If the confidence score corresponding to the intruder label corresponding to the image to be detected is greater than or equal to the intruder recognition threshold, there is an intruder in the image to be detected; if the confidence score corresponding to the intruder label corresponding to the image to be detected is lower than the intruder recognition threshold, there is no intruder in the image to be detected.

[0095] The intrusion detection method provided in the embodiment of the present application adopts a trained detection model to identify the image to be detected and obtain an intruder label corresponding to the image to be detected; adjusts the intruder recognition threshold according to different scenarios of the image to be detected; and determines whether there is an intruder in the image to be detected based on the intruder label and the intruder recognition threshold corresponding to the image to be detected, thereby improving the accuracy of intruder detection in different scenarios.

[0096] Figure 4 A schematic diagram of the process of detecting intrusion provided in this application Figure 3 ,like Figure 4 As shown, in this embodiment Figure 2 Based on the embodiment, before using the trained detection model in the above step S301 to identify the image to be detected, so as to obtain the intruder label corresponding to the image to be detected through the trained detection model, the method further includes:

[0097] S401: Obtain a training data set.

[0098] In this embodiment, the training data set is a public data set that has undergone data enhancement and data supplementation; wherein the data enhancement includes image inversion, image rotation, image cropping, image translation, image affine transformation, and hue transformation; and the data supplementation includes a preset number of small-sized personnel training data sets, and a training data set that induces false detection.

[0099] The initial detection model is trained using the training data set to obtain a trained detection model.

[0100] Optionally, the initial detection model is trained using a training data set to obtain a trained detection model, including:

[0101] S402: Use the training data set to train the initial detection model to obtain a trained detection model.

[0102] In this embodiment, the initial detection model is trained by using a training data set with data enhancement and data supplementation, thereby preventing overfitting and improving the accuracy of the detection model.

[0103] S403: Obtain a test data set.

[0104] In this embodiment, data enhancement is performed on the public data set to obtain a test data set.

[0105] S404: Use the test data set to train the trained detection model to obtain a trained detection model.

[0106] In this embodiment, the trained detection model is trained by using a test data set to obtain a trained detection model, which effectively improves the generalization ability of the detection model and verifies the accuracy of the detection model.

[0107] Optionally, the initial detection model can be trained through transfer learning to obtain a trained detection model, including:

[0108] Training the initial detection model using simulated data greater than a first quantity threshold to obtain initial network parameters;

[0109] The initial network parameters are adjusted through experimental data below the second quantity threshold to obtain the network parameters of the trained detection model.

[0110] In this embodiment, transfer learning is used to save time and computing resources for training the detection model, and improve the applicability and performance of the detection model in different tasks.

[0111] Figure 5 A schematic diagram of the structure of an intrusion detection device provided in this application, such as Figure 5 As shown, the intrusion detection device provided in this embodiment includes:

[0112] The receiving module 501 is used to receive video image data sent by a terminal device.

[0113] The first preprocessing module 502 is used to perform a first preprocessing on the video image data to obtain an image to be detected.

[0114] The second preprocessing module 503 is used to perform second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected.

[0115] The determination module 504 is used to use the trained detection model to identify the image to be detected if there is a moving object in the image to be detected, so as to determine whether there is an intruder in the image to be detected, wherein the trained detection model is obtained by training a training data set, and the training data set contains multiple training image samples and intruder labels corresponding to the training image samples.

[0116] Optionally, the detection model is a target detection model that integrates a dilated convolution layer and a separable convolution that combines depthwise convolution with point-by-point convolution.

[0117] The alarm module 505 is used to issue an intrusion alarm if there is an intruder in the image to be detected.

[0118] In a possible implementation, the first preprocessing module 502 may also be specifically used for:

[0119] The video image data is first preprocessed by using a plurality of video stream encoding, contrast enhancement, shot segmentation, brightness adjustment, video stream decoding and storage to obtain an image to be detected.

[0120] In a possible implementation manner, the second preprocessing module 503 may also be specifically used for:

[0121] The difference operation of adjacent frames is performed on the image to be detected to determine whether there is a moving object in the image to be detected.

[0122] In a possible implementation, the determination module 504 may also be specifically configured to:

[0123] The trained detection model is used to identify the image to be detected, so as to obtain the intruder label corresponding to the image to be detected through the trained detection model;

[0124] Identify the scene corresponding to the image to be detected;

[0125] Determine the intruder identification threshold based on the scenario;

[0126] According to the intruder label corresponding to the image to be detected and the intruder identification threshold, it is determined whether there is an intruder in the image to be detected.

[0127] In a possible implementation, the determination module 504 may also be specifically configured to:

[0128] Acquire environmental data of the image to be detected, wherein the environmental data includes at least one of time, weather and lighting conditions;

[0129] According to the environmental data, the scene corresponding to the image to be detected is determined.

[0130] In a possible implementation, the determination module 504 may also be specifically configured to:

[0131] Get the training dataset;

[0132] The initial detection model is trained using the training data set to obtain a trained detection model.

[0133] In a possible implementation, the determination module 504 may also be specifically configured to:

[0134] The initial detection model is trained using the training data set to obtain a trained detection model;

[0135] Get the test dataset;

[0136] The trained detection model is trained using the test data set to obtain a trained detection model.

[0137] The intrusion detection device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be described in detail here.

[0138] Figure 6 This is a schematic diagram of the structure of an intrusion detection device provided by this application. Figure 6 As shown, the electronic device 60 provided in this embodiment includes: at least one processor 601 and a memory 602. Optionally, the device 60 also includes a communication component 603. The processor 601, the memory 602 and the communication component 603 are connected via a bus 604.

[0139] In a specific implementation process, at least one processor 601 executes the computer execution instructions stored in the memory 602, so that at least one processor 601 executes the above method.

[0140] The specific implementation process of the processor 601 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0141] In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the invention can be directly implemented as a hardware processor, or can be implemented by a combination of hardware and software modules in the processor.

[0142] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.

[0143] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0144] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0145] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0146] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0147] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0148] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0149] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0150] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0151] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0152] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0153] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. An intrusion detection method, characterized in that: Applied to a cloud server, the method comprises: Receiving video image data sent by a terminal device; Performing a first preprocessing on the video image data to obtain an image to be detected; Performing a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected; If there is a moving object in the image to be detected, the trained detection model is used to identify the image to be detected to determine whether there is an intruder in the image to be detected, wherein the trained detection model is obtained by training a training data set, and the training data set includes multiple training image samples and intruder labels corresponding to the training image samples; If there is an intruder in the image to be detected, an intrusion alarm is issued.

2. The method according to claim 1, characterized in that The method of using the trained detection model to identify the image to be detected to determine whether there is an intruder in the image to be detected includes: Using the trained detection model to identify the image to be detected, so as to obtain an intruder label corresponding to the image to be detected through the trained detection model; Identifying a scene corresponding to the image to be detected; According to the described scenario, determine the intruder identification threshold; It is determined whether there is an intruder in the image to be detected according to the intruder label corresponding to the image to be detected and the intruder identification threshold.

3. The method according to claim 2, characterized in that The identifying the scene corresponding to the image to be detected includes: Acquire environmental data of the image to be detected, wherein the environmental data includes at least one of time, weather and lighting conditions; The scene corresponding to the image to be detected is determined according to the environmental data.

4. The method according to claim 2, characterized in that: Before the trained detection model is used to identify the image to be detected, so as to obtain the intruder label corresponding to the image to be detected through the trained detection model, the method further includes: Get the training dataset; The training data set is used to train the initial detection model to obtain a trained detection model.

5. The method according to claim 4, characterized in that The using the training data set to train the initial detection model to obtain a trained detection model includes: Using the training data set to train the initial detection model to obtain a trained detection model; Get the test dataset; The trained detection model is trained using the test data set to obtain a trained detection model.

6. The method according to any one of claims 2 to 4, characterized in that: The detection model is a target detection model that integrates a dilated convolution layer and a separable convolution that combines depthwise convolution with point-by-point convolution.

7. The method according to any one of claims 1 to 5, characterized in that: The performing a second preprocessing on the image to be detected to determine whether there is a moving object in the image to be detected includes: A differential operation of adjacent frames is performed on the image to be detected to determine whether there is a moving object in the image to be detected.

8. The method according to any one of claims 1 to 5, characterized in that: The first preprocessing of the video image data to obtain the image to be detected includes: The video image data is first preprocessed by using a plurality of video stream encoding, contrast enhancement, shot segmentation, brightness adjustment, video stream decoding and storage to obtain an image to be detected.

9. An intrusion detection device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the intrusion detection method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to implement the intrusion detection method according to any one of claims 1 to 8 when executed by a processor.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the intrusion detection method according to any one of claims 1 to 8 is implemented.

Images