Perceptual data decoding method and system

Abstract

The invention provides a sensing data decoding method and system. A decoding process and multiple paths of sensing devices communicate with a pull flow process, and the sensing data decoding method comprises the following steps: acquiring a configuration file of the sensing devices, and entering an SIP pull flow module or an RTSP pull flow module in the pull flow process according to the configuration file; obtaining perception data based on a current pull stream module, and obtaining code stream data according to the perception data; and performing decoding based on the code stream data through the decoding process to obtain target data. The system is compatible with RTSP and SIP protocols, various different sensing devices can be accessed, and both a common RTSP camera and a novel SIP intelligent sensor can be conveniently integrated into the same system, so that more device selection space is provided for a user, the system can flexibly configure different sensing devices according to actual requirements, and the system is convenient to use. And diversified application scenes are satisfied.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of decoding, in particular to a perception data decoding method and system. BACKGROUND

[0002] Vehicle-road-cloud integration is an important development direction in the field of intelligent transportation, and its core is to realize the technical system of intelligent driving through the cooperation of vehicle, roadside and cloud. Among them, the roadside is an important part of the vehicle-road-cloud integrated system, and plays an important role in providing traffic information, assisting vehicle decision-making, improving traffic efficiency and safety. Roadside infrastructure is an indispensable part of the vehicle-road-cloud integrated system, including roadside perception devices, communication devices, edge computing devices, and traffic ancillary facilities. Cameras as roadside perception devices can provide clear and continuous picture data for edge computing platforms, realize multi-target detection and prediction, including vehicle position, speed, trajectory, traffic events, etc., and perform preliminary processing and analysis of data to provide basic information for vehicle-road cooperation and other functions.

[0003] Currently, perception devices mainly support RTSP (Real Time Streaming Protocol) and SIP (Session Initiation Protocol), and the mainstream pull stream program basically supports the RTSP protocol, but many do not support the SIP protocol, which limits the application range of perception devices using the SIP protocol. SUMMARY

[0004] To solve the above technical problems, the present application provides a perception data decoding method and system compatible with RTSP and SIP protocols.

[0005] Specifically, the present application provides a perception data decoding method, and a decoding process and a plurality of perception devices are in communication with a pull stream process. The perception data decoding method comprises the following steps: obtaining a configuration file of a perception device, and entering an SIP pull stream module or an RTSP pull stream module in the pull stream process according to the configuration file; obtaining perception data based on the current pull stream module to obtain code stream data according to the perception data; and decoding target data based on the code stream data by the decoding process.

[0006] In the above technical solution, RTSP and SIP protocols are compatible, and a variety of different perception devices can be accessed. Whether it is a common RTSP camera or a new SIP intelligent sensor, it can be easily integrated into the same system, which provides more device selection space for users, so that the system can flexibly configure different perception devices according to actual needs to meet diversified application scenarios.

[0007] Further, the stream pulling process further comprises a reading module, a first parsing module, a second parsing module and a first interface; the configuration file of the sensing device is obtained by reading the configuration file through the reading module and storing the configuration file in the cache.

[0008] In the above technical solution, the configuration file is stored in the cache, which facilitates the quick access of subsequent programs.

[0009] Further, in the SIP stream pulling module, the sensing data is obtained by receiving the SIP signaling sent by the sensing device and feeding back the first reply information to the sensing device based on the SIP signaling; wherein, when the SIP signaling is a registration signaling, further comprising: based on the registration signaling and the corresponding sensing device to establish a connection to receive the sensing data sent by the sensing device.

[0010] In the above technical solution, the signaling is replied to confirm the receipt of the signaling and the corresponding processing.

[0011] Further, the configuration file comprises sensing device information; in the RTSP stream pulling module, the sensing data is obtained by obtaining the sensing device information, establishing a connection with the corresponding sensing device based on the sensing device information, and receiving the sensing data sent by the sensing device; wherein, further comprising: sending the setting protocol data to the sensing device to receive the second reply information fed back by the sensing device based on the setting protocol data.

[0012] In the above technical solution, the sensing device information in the configuration file is read, the information is used to connect the sensing device, and the corresponding protocol interaction is carried out, the reply information returned by the sensing device is processed, and the reply information is processed accordingly.

[0013] Further, the code stream data is obtained according to the sensing data, which comprises: the sensing data is parsed by the first parsing module and the second parsing module to obtain the code stream data, and the code stream data is sent to the decoding thread through the first interface.

[0014] In the above technical solution, the first parsing module is used to parse the received RTP / RTCP data, and the second parsing module is used to parse the H264 / H265 encoded audio and video data.

[0015] Furthermore, the decoding process includes at least a second interface and a decoding module; the step of decoding the target data based on the bitstream data includes: obtaining hardware configuration information through the decoding module to initialize the second interface according to the hardware configuration information; receiving the bitstream data through the second interface and assembling the bitstream data into packets through the decoding module; obtaining a decoding handle based on the assembled bitstream data and storing the bitstream data of each frame in a buffer based on the decoding handle; and sending the bitstream data in the buffer to the decoder to obtain the decoded target data from the decoder.

[0016] In the above technical solution, the packet reassembly operation can recombine these scattered data packets into complete bitstream data, which facilitates subsequent decoding processing. This packet reassembly mechanism can improve the efficiency of data processing, reduce errors and delays in the decoding process, and ensure the accuracy and smoothness of decoding. By managing the cache through the decoding handle, data can be accessed and processed more efficiently, improving the real-time performance of decoding.

[0017] Furthermore, the acquisition of the configuration file of the sensing device also includes: real-time monitoring of the configuration file based on the cache through the reading module; wherein, when the configuration file is detected to be modified, the configuration file is compared with historical configuration information to obtain the working status of each sensing device, and when any sensing device is in a disconnected state, the corresponding link is disconnected and system resources are released at the same time; and when any sensing device is in an online state, the SIP streaming module or RTSP streaming module in the streaming process is entered according to the modified configuration file.

[0018] In the above technical solution, when the configuration file is modified, the corresponding processing logic will be triggered. By comparing the configuration file, it is possible to distinguish which sensing devices have exited the connection and which are newly connected devices. For sensing devices that need to exit the connection, the corresponding link is disconnected and the relevant resources are released, which can avoid the waste of resources and improve the performance of the system. For newly connected sensing devices, the corresponding streaming module can be re-executed according to their configuration information.

[0019] Furthermore, storing the configuration file or bitstream data in the cache includes: obtaining the tail pointer of the cache, determining whether the current distance to the tail of the queue has reached a preset length based on the tail pointer, if so, acquiring a spinlock, copying the data to be stored to the head position of the cache, updating the head pointer and tail pointer, and releasing the spinlock; otherwise, directly copying the data to be stored to the position of the tail pointer and updating the position of the tail pointer.

[0020] Further, the cache reading code stream data or configuration file comprises: performing a lock operation through a spin lock, and acquiring a head pointer and a tail pointer of the cache to calculate a readable data length; and based on the readable data length and the head pointer, acquiring to-be-read data and updating a position of the head pointer, while releasing the spin lock.

[0021] In the above technical solution, the spin lock is used simultaneously, avoiding context switching when multiple threads compete, and the lock-free cache is continuous, and the reading and writing of the cache are realized through the head pointer and the tail pointer; when the tail pointer is close to the tail, the to-be-stored data needs to be copied to the head, which is in conflict with the reading operation, and the spin lock is needed to ensure thread safety, and vice versa when the tail pointer is far away from the tail, which does not interfere with the reading operation, and direct writing can be performed; and the reading and writing operations are both completed based on the cache, reducing the number of copying.

[0022] Based on the same concept, the application further provides a decoding system of perception data, which comprises: a pull stream process configured to acquire a configuration file of a perception device, and enter a SIP pull stream module or an RTSP pull stream module according to the configuration file, so as to acquire perception data based on a current pull stream module, and further obtain code stream data according to the perception data; and a decoding process configured to decode the code stream data to obtain target data.

[0023] In the above technical solution, the RTSP and SIP protocols are compatible, and a plurality of different perception devices can be accessed, which provides more device selection space for users, so that the system can flexibly configure different perception devices according to actual needs, and meet diversified application scenarios.

[0024] Compared with the prior art, the application has the following beneficial effects: The application supports the RTSP and SIP protocols, so that various types of perception devices can seamlessly access the same system; for users, this greatly expands the range of device selection, so that users can flexibly configure different perception devices according to specific business needs and scene characteristics. This flexibility enables the system to accurately adapt to diversified application scenarios, bringing higher operation efficiency and better user experience to enterprises and users. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 A flowchart of the decoding method of perception data described in the application.

[0026] Figure 2 A timing diagram of the pull stream process described in the application.

[0027] Figure 3 A timing diagram of the decoding process described in the application.

[0028] Figure 4Architecture diagram of a decoding system of perception data described in the present application. DETAILED DESCRIPTION

[0029] The decoding method and system of perception data described in the present application are further described in detail below in combination with specific embodiments and drawings.

[0030] Referring to Figure 1 The present application provides a decoding method of perception data, a decoding process and a multi-path perception device are in communication with a pull stream process; the decoding method of perception data comprises the following steps S100-S300.

[0031] In a feasible embodiment, the Conifg obtains the configuration file of the perception device, and monitors whether the configuration file is modified. If the configuration file is modified, the modified configuration file is compared with the old configuration file, and then the corresponding link of the exit device is disconnected, and whether to start the SIP service is determined based on the latest configuration file. If yes, the SIP pull stream module is entered, otherwise, if the RTSP service is set in the configuration file, the RTSP pull stream module is entered; in the SIP pull stream module or the RTSP pull stream module, the connection with the perception device is established based on the current stream process of the pull stream module to receive the perception data sent by the perception device. After the perception data is obtained, the RTP / RTCP data and the H264 / H265 data are parsed in sequence to obtain the code stream data; the code stream data is further sent to the decoding process through the communication abstract interface of the pull stream process, the decoding process receives the code stream data through the communication abstract interface thereof, and packeting and a decoding handle are obtained to cache each frame of code stream data based on the decoding handle, so that the code stream data in the cache is sent to the decoder for decoding, and finally the decoded data, i.e. target data, is obtained from the decoder. After the data is received and parsed by the decoding process, the data is cached, and the cached data is taken out in the decoding thread for decoding.

[0032] Next, each implementation step of steps S100-S300 is described in detail.

[0033] Referring to Figure 2 Step S100: obtaining the configuration file of the perception device, and entering the SIP pull stream module (SIP Server) or the RTSP pull stream module (RTSP Client) in the pull stream process according to the configuration file.

[0034] The pull stream process further includes a reading module Conifg, a first parsing module RTP / RTCP Parse, a second parsing module H264 / H265 Parse, and a first interface communication Interface (pub).

[0035] In some embodiments, the configuration file includes various information required for system operation, such as whether the SIP service is started, the sensing device information, the pull stream mode (RTSP configuration information and SIP configuration information), etc. After reading the configuration file, these information are stored in the cache. At the same time, it is judged whether the SIP service is started according to the setting in the configuration file. If the SIP service is specified to be started in the configuration file, the SIP pull stream module is entered. If the RTSP service is specified to be started in the configuration file, the RTSP pull stream module is entered.

[0036] In the above technical solution, the configuration file is stored in the cache, which can facilitate the quick access of subsequent programs.

[0037] Step S200: obtaining sensing data based on the current pull stream module, so as to obtain code stream data according to the sensing data.

[0038] In the SIP pull stream module, the obtaining of the sensing data includes receiving the SIP signaling sent by the sensing device and feeding back first reply information to the sensing device based on the SIP signaling. When the SIP signaling is a registration signaling, it further includes establishing a connection with the corresponding sensing device based on the registration signaling, so as to receive the sensing data sent by the sensing device.

[0039] In some embodiments, referring to Figure 2 , after entering the SIP pull stream module, the listening is started, and the connection request initiated by the sensing device such as a camera is waited. The SIP service listens to a specific port to receive the SIP signaling from the sensing device, such as Register / Message / Bye signaling. These signalings are used for the registration, message transmission and session end of the sensing device. According to the SIP protocol, the received SIP signaling is replied (i.e. the first reply information). Further, according to the registration information of the sensing device, the invite request is sent to invite the sensing device to establish a media session. The invite request includes the related information of the session, such as the media type, the coding format, etc. After the sensing device receives the invite request, the connection request is initiated, and the connection with the edge computing platform is established. After the connection is established, the sensing device starts to send the audio and video data to the edge computing platform.

[0040] In the technical solution, the signaling is replied to confirm the receipt of the signaling and to perform corresponding processing.

[0041] Further, the configuration file at least includes the sensing device information; in the RTSP pull streaming module, the obtaining of the sensing data includes: acquiring the sensing device information, establishing a connection with the corresponding sensing device based on the sensing device information, and receiving the sensing data sent by the sensing device; wherein, further includes: sending the set protocol data to the sensing device to receive the second reply information fed back by the sensing device based on the set protocol data.

[0042] In some embodiments, referring to Figure 2 , the sensing device information in the configuration file is read, and the sensing device is connected using the information, and the sensing device information is, for example, the IP address, port number, and the like of the sensing device; further, the RTSP protocol data such as OPTIONS / DESCRIBE / SETUP / PLAY / TEARDOWN is sent to the sensing device, and the protocol data is used to acquire the capability information of the sensing device, describe the media stream, set the session parameter, start playing, and end the session, and the like; at the same time, the Reply data (i.e., the second reply information) returned by the sensing device is processed, and corresponding processing is performed according to the reply information. After the connection is established, the audio and video data sent by the sensing device is received.

[0043] In the technical solution, the sensing device information in the configuration file is read, the sensing device is connected using the information, and corresponding protocol interaction is performed, the reply information returned by the sensing device is processed, and corresponding processing is performed according to the reply information.

[0044] Further, the obtaining of the code stream data based on the sensing data includes: performing corresponding analysis work on the sensing data by the first analysis module and the second analysis module to obtain the code stream data, and sending the code stream data to the decoding thread through the first interface.

[0045] In some embodiments, referring to Figure 2 , the first analysis module is, for example, to analyze the received RTP (Real-Time Transport Protocol) / RTCP (Real-Time Control Protocol) data, and the second analysis module is, for example, to analyze the audio and video data encoded by H264 / H265.

[0046] Wherein, RTP is a protocol specially designed for real-time data (such as audio, video) transmission on the network, and its parsing work mainly revolves around its header and payload data, that is, the encoded audio or video data, which will be further processed according to the payload type; The role of RTCP is to provide control information related to RTP stream, and different types of RTCP packet formats are different.

[0047] H264 video stream includes a series of NAL units (Network Abstraction Layer Unit), each NAL unit has a start code, and the boundary of each NAL unit is determined by searching the start code to divide the video stream into independent NAL units, the first byte of each NAL unit includes NAL unit type information, and the type of NAL unit can be obtained by bit operation on the byte, different types of NAL units have different roles in video decoding process, such as key frame, non-key frame, sequence parameter set, etc. H265 is also composed of NAL units, but the unit type information is contained in the first two bytes of NAL unit, and the type of NAL unit is obtained by bit operation on the first byte.

[0048] Step S300: obtaining target data by decoding based on the code stream data through the decoding process.

[0049] Wherein, please refer to Figure 3 , the decoding process at least includes second interface communication Interface (sub) and decoding module Decode Interface; the decoding based on the code stream data to obtain target data includes: obtaining hardware configuration information through the decoding module, initializing the second interface according to the hardware configuration information; receiving the code stream data through the second interface, and packeting the code stream data through the decoding module; obtaining decoding handle according to the packeted code stream data, and storing each frame of code stream data into cache based on the decoding handle; sending the code stream data in cache to decoder to obtain decoded target data from the decoder.

[0050] In some embodiments, the second interface is initialized, such as setting the transmission rate, data format and other parameters of the interface, when the initialization is completed, the code stream data can be prepared to be received, and packeting can be carried out after receiving the code stream data.

[0051] Wherein, the decoding handle such as includes the source device, timestamp and other information of data.

[0052] It should be noted that the above communication interface abstraction can be replaced by those skilled in the art based on different project requirements, such as replacing it with ROS2, AP AUTOSAR, etc.

[0053] In the above technical solution, the packet reassembly operation can recombine these scattered data packets into complete bitstream data, which facilitates subsequent decoding processing. This packet reassembly mechanism can improve the efficiency of data processing, reduce errors and delays in the decoding process, and ensure the accuracy and smoothness of decoding. By managing the cache through the decoding handle, data can be accessed and processed more efficiently, improving the real-time performance of decoding.

[0054] In addition, please see Figure 2 The decoding method for perceived data described in this application supports dynamic configuration; specifically: The acquisition of the configuration file of the sensing device further includes: real-time monitoring of the configuration file based on the cache through the reading module; wherein, when the configuration file is detected to be modified, the configuration file is compared with historical configuration information to obtain the working status of each sensing device, and when any sensing device is in a disconnected state, the corresponding link is disconnected and system resources are released at the same time; and when any sensing device is in an online state, the SIP streaming module or RTSP streaming module in the streaming process is entered according to the modified configuration file.

[0055] In some embodiments, changes to the configuration file are continuously monitored. If the configuration information is modified, the corresponding processing logic is triggered to reload the updated configuration file and compare it with the previously cached old configuration file. By comparing, it can be determined which sensing devices have disconnected and which are newly connected devices. For sensing devices that need to disconnect, the corresponding link is disconnected and related resources, such as network connection and memory, are released. This can avoid resource waste and improve system performance. For newly connected sensing devices, the corresponding streaming module can be re-executed according to their configuration information.

[0056] Furthermore, storing the configuration file or bitstream data in the cache includes: obtaining the tail pointer of the cache, determining whether the current distance to the tail of the queue has reached a preset length based on the tail pointer, if so, acquiring a spinlock, copying the data to be stored to the head position of the cache, updating the head pointer and tail pointer, and releasing the spinlock; otherwise, directly copying the data to be stored to the position of the tail pointer and updating the position of the tail pointer.

[0057] Wherein, the preset length is set by the person skilled in the art to determine whether the current data length is close to or far from the tail, if close to the tail, Tail+length>bufferSize (buffer size); Tail represents the tail pointer, and length represents the preset fixed length.

[0058] Further, based on the cache reading code stream data or configuration file, including: locking operation through spin lock, and obtaining the head pointer and tail pointer of the cache to calculate the readable data length; based on the readable data length and the head pointer, obtaining the to-be-read data, and updating the position of the head pointer, while releasing the spin lock.

[0059] Wherein, based on the readable data length and the head pointer, the to-be-read data is obtained using a callback function, and the length of each reading is fixed, generally greater than the size of a packet, so as to reduce the number of system calls, if the to-be-read data includes a packet of completed data, the size of the packet data is returned, otherwise 0 is returned.

[0060] In the above technical solution, the spin lock is used at the same time, avoiding the context switching when multiple threads compete, the lock-free cache is continuous, and the reading and writing of the cache are realized through the head pointer and the tail pointer; when the tail pointer is close to the tail, the to-be-stored data needs to be copied to the head, at this time, the reading operation is in conflict, and the spin lock is needed to ensure thread safety, otherwise when the tail pointer is far away from the tail, the reading operation does not interfere with each other, and direct writing can be realized; and the reading and writing operations are completed based on the cache, reducing the number of copying.

[0061] In summary, the decoding method of the perception data is compatible with the RTSP and SIP protocols, can access a variety of different perception devices, whether it is a common RTSP camera or a new SIP intelligent sensor, can be easily integrated into the same system, which provides more device selection space for users, so that the system can flexibly configure different perception devices according to actual needs, and meet diversified application scenarios.

[0062] Based on the same concept, please refer to Figure 4 The application further provides a decoding system of perception data, the system comprising: a streaming process for obtaining a configuration file of a perception device, and entering an SIP streaming module or an RTSP streaming module according to the configuration file, so as to obtain perception data based on the current streaming module, and then obtain code stream data according to the perception data; and a decoding process for decoding the target data based on the code stream data.

[0063] In some embodiments, the stream-pulling process includes Config, RTSPClient, SIPServer, RTP / RTCP Parse, H264 / H265 Parse and communication Interface (pub); and the decoding process includes communication Interface (sub) and Decode Interface.

[0064] The Config is responsible for reading the configuration file information of the camera, and simultaneously monitors the configuration file in real time. When the file is modified, the configuration file is reloaded, and the new and old configurations are compared to determine which sensing device exits the connection and which sensing device accesses the edge computing device.

[0065] RTSPClient: mainly implements RTSP protocol stream pulling, supports OPTIONS / DESCRIBE / SETUP / PLAY / TEARDOWN and other protocols.

[0066] SIPServer: mainly implements SIP protocol stream pulling, supports REGISTER / MESSAGE / BYE / INVITE and other protocols.

[0067] RTP / RTCP Parse: mainly responsible for parsing the RTP / RTCP protocol.

[0068] H264 / H265 Parse: mainly responsible for parsing the H264 / H265 protocol and adding Startcode.

[0069] communication Interface (pub): an abstract interface of IPC communication, used for sending code stream data.

[0070] communication Interface (sub): an abstract interface of IPC communication, used for receiving code stream data.

[0071] Decode Interface: an abstract interface of hardware decoding, suitable for multiple hardware, and realizing hardware decoding of code stream data.

[0072] It should be noted that the specific working process of the stream-pulling process and the decoding process can be referred to Figures 2-3 , and each embodiment in the decoding method of the sensing data, which will not be described herein.

[0073] In the technical solution, RTSP and SIP protocols are compatible, and various sensing devices can be accessed, which provides more device selection space for users, and enables the system to flexibly configure different sensing devices according to actual needs, and meet diversified application scenarios.

[0074] Although the example embodiments have been described herein with reference to the accompanying drawings, it is to be understood that the example embodiments are only exemplary and are not intended to limit the scope of the present application. Those of ordinary skill in the art can make various changes and modifications without departing from the scope and spirit of the present application. All such changes and modifications are intended to be included within the scope of the present application as claimed in the appended claims.

[0075] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in connection with the embodiments disclosed herein can be realized in electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those of ordinary skill in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0076] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0077] The various component embodiments of the present application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some modules according to the embodiments of the present application. The present application can also be implemented as a device program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such program implementing the present application can be stored on a computer readable medium, or can have one or more signals in the form. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0078] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting; it is not intended to exclude myriad other embodiments of the present application that other inventors can develop based on the description provided herein. It must be stressed, however, that any combination of the components or features taught according to any aspect of the present application can be important sufficient to confer "patentable" subject matter status on a single, stand-alone claim. It is therefore intended to embrace all such alternatives, modifications and variations that fall within the spirit and broad scope of the appended claims. One skilled in the art will appreciate that the application can be practiced with embodiments other than those described. The disclosed embodiments are illustrative only and not restrictive.

[0079] While the application has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended to cover and embrace all adaptations or modifications of the application.

Claims

1. A method of decoding perception data, the method comprising: The decoding process and the multi-sensing device are in communication with the streaming process; the decoding method of the sensing data comprises the following steps: Obtain the configuration file of the sensing device, and enter the SIP streaming module or the RTSP streaming module in the streaming process according to the configuration file; Obtain the sensing data based on the current streaming module to obtain the code stream data according to the sensing data; And obtain the target data by decoding based on the code stream data through the decoding process.

2. The method of claim 1, wherein, The streaming process further comprises a reading module, a first parsing module, a second parsing module and a first interface; the configuration file of the sensing device comprises: Read the configuration file through the reading module and store the configuration file in the cache.

3. The method of claim 2, wherein, In the SIP streaming module, the sensing data comprises: Receive the SIP signaling sent by the sensing device, and feed back the first reply information to the sensing device based on the SIP signaling; Wherein, when the SIP signaling is a registration signaling, it further comprises: Establish a connection with the corresponding sensing device based on the registration signaling to receive the sensing data sent by the sensing device.

4. The method of claim 3, wherein, The configuration file at least comprises sensing device information; in the RTSP streaming module, the sensing data comprises: Obtain the sensing device information to establish a connection with the corresponding sensing device based on the sensing device information and receive the sensing data sent by the sensing device; Wherein, it further comprises: sending the set protocol data to the sensing device to receive the second reply information fed back by the sensing device based on the set protocol data.

5. The method of claim 4, wherein, The code stream data obtained according to the sensing data comprises: The sensing data is parsed through the first parsing module and the second parsing module to obtain the code stream data, and the code stream data is sent to the decoding thread through the first interface.

6. The method of claim 5, wherein, The decoding process at least comprises a second interface and a decoding module; the decoding based on the code stream data to obtain the target data comprises: Obtain the hardware configuration information through the decoding module to initialize the second interface according to the hardware configuration information; Receive the code stream data through the second interface and pack the code stream data through the decoding module; Obtain the decoding handle according to the packed code stream data, and store each frame of code stream data in the cache based on the decoding handle; Send the code stream data in the cache to the decoder to obtain the decoded target data from the decoder.

7. The method of claim 2, wherein, The configuration file of the sensing device further comprises: Real-time monitoring of the configuration file based on the cache through the reading module; Wherein, when the configuration file is modified, compare the configuration file with the historical configuration information to obtain the working state of each sensing device, and when any sensing device is in the disconnected state, disconnect the corresponding link and release the system resources; And when any sensing device is in the online state, enter the SIP streaming module or the RTSP streaming module in the streaming process according to the modified configuration file.

8. The method of claim 6, wherein, Storing the configuration file or the code stream data in the cache comprises: An end pointer of the cache is acquired, whether the current distance from the tail end reaches a preset length is determined based on the end pointer, if yes, a spin lock is acquired, the data to be stored is copied to a head position of the cache, and a head pointer and the end pointer are updated, and the spin lock is released; otherwise, the data to be stored is directly copied to a position of the end pointer, and a position of the end pointer is updated.

9. The method of claim 7, wherein, The code stream data or the configuration file is read based on the cache, including: A lock operation is performed through the spin lock, and a head pointer and an end pointer of the cache are acquired to calculate a readable data length; The data to be read is acquired based on the readable data length and the head pointer, and a position of the head pointer is updated, and the spin lock is released.

10. A system employing a decoding method of perceptual data as claimed in any one of claims 1-9, characterized in that, The system includes: A pull stream process for acquiring a configuration file of a sensing device, and entering a SIP pull stream module or an RTSP pull stream module according to the configuration file, so as to acquire sensing data based on a current pull stream module, and further acquire code stream data according to the sensing data; And a decoding process for decoding to acquire target data based on the code stream data.

Images