Distributed heterogeneous device interconnection method and system in narrowband weak connection environment

Abstract

This invention provides a method and system for interconnecting distributed heterogeneous devices in a narrowband, weakly connected environment. The method includes: identifying all uploading devices with interconnection relationships with a receiving device, and determining the amount of data transmitted by each uploading device corresponding to different transmitted images at the same data transmission time; performing redundancy allocation for each uploading device based on the amount of transmitted data, and determining the redundancy ratio of the transmitted data amount corresponding to the same uploading device based on the obtained redundant data amount; determining the data extraction target for the corresponding data transmission time, and performing image redundancy on the transmitted image based on the data extraction target according to the corresponding redundancy ratio, obtaining redundant data corresponding to the transmitted image; and controlling the uploading devices to uniformly send the transmitted image and the obtained redundant data of the corresponding transmitted image to the receiving device. This invention at least improves transmission reliability.

Description

Technical Field

[0001] This invention relates to data processing technology, and more particularly to a method and system for interconnecting distributed heterogeneous devices in a narrowband, weakly connected environment. Background Technology

[0002] As IoT technology extends to remote areas and complex industrial scenarios, the demand for interconnection of distributed heterogeneous devices in narrowband weak connection environments (such as mountain communication and underground mine data transmission) is growing. In such environments, the data transmission bandwidth between devices is limited and the signal stability is poor. Transmitted images, as key data carriers (such as remote monitoring images and industrial inspection images), directly affect subsequent data processing and decision-making.

[0003] Currently, narrowband environments with weak connectivity are prone to data packet loss. If image data packets are lost during transmission, the image information may be incomplete or even completely lost. Therefore, there is an urgent need to provide a method and system for interconnecting distributed heterogeneous devices in a narrowband weak connection environment that ensures transmission reliability through image redundancy. Summary of the Invention

[0004] In view of the above problems, the present invention is proposed to provide a method and system for interconnecting distributed heterogeneous devices in a narrowband weak connection environment to overcome or at least partially solve the above problems.

[0005] According to one aspect of the present invention, a method for interconnecting distributed heterogeneous devices in a narrowband weak connection environment is provided, comprising the following steps: Identify all uploading devices that have device interconnection relationships with the receiving device, and determine the amount of data transmitted by each uploading device for different transmitted images at the same data transmission time; Redundancy allocation is performed for each upload device based on the amount of transmitted data, and the redundancy ratio of the transmitted data to the corresponding upload device is determined based on the obtained redundant data amount. Determine the data extraction target for the corresponding data transmission time, and perform image redundancy with a corresponding redundancy ratio on the transmitted image based on the data extraction target to obtain redundant data corresponding to the transmitted image; The control upload device sends the transmitted image and the corresponding redundant data of the transmitted image to the receiving device.

[0006] Optionally, in the method according to the present invention, determining all uploading devices that have a device interconnection relationship with the receiving device, and determining the amount of data transmitted by each uploading device corresponding to different transmitted images at the same data transmission time, includes: The system receives a receiving instruction carrying a device code sent by the management terminal, determines the receiving device based on the device code, and sorts the device code and the instruction sending time of the corresponding receiving instruction in a corresponding random order. The obtained information sequence is converted based on a hash algorithm, and a distribution identification code corresponding to the receiving device is generated based on the obtained hash string. Control the receiving device to distribute the distribution identification code based on a distributed heterogeneous network, and establish a timed task corresponding to the data distribution; In response to any other device in the corresponding distributed heterogeneous network, the distribution identification code is scanned and identified within the device response period of the corresponding timed task to obtain the device type corresponding to the other device. In response to the fact that the device type is within the preset upload range corresponding to the receiving device, the other device is identified as an upload device that has a device interconnection relationship with the receiving device; Establish a data transmission channel between the receiving device and each uploading device, and determine the amount of data transmitted by each uploading device based on the data transmission channel for different transmitted images at the same data transmission time.

[0007] Optionally, in the method according to the invention, the method further includes: If the number of first devices corresponding to all upload devices with device interconnection relationship with the receiving device, as determined by the timed task, is less than the preset number of devices, the first location information corresponding to each upload device is obtained. Based on the first location information, a device communication circle is established with each uploading device as the center. Based on the distributed heterogeneous network, the second location information corresponding to each other device is determined. Based on the second location information, other devices located in the device communication circle are determined as terminal supplementary groups. If the number of second devices corresponding to other devices in all terminal supplementary groups is less than or equal to the number of preset devices corresponding to the number of first devices, the other devices in all terminal supplementary groups are identified as newly added upload devices. The device type and device deviation distance between each other device in the terminal supplement group and the upload device are obtained by responding to the device difference between the second number of other devices in all terminal supplement groups and the first number of devices corresponding to a preset number of devices; Based on the device deviation distance, determine the distance deviation value of the upload device corresponding to each other device, and based on the device type, determine the type deviation value of the preset upload range corresponding to each other device; The distance deviation value and type deviation value corresponding to the same other device are summed. Based on the obtained comprehensive deviation value, all other devices are sorted from smallest to largest terminal. Based on the obtained terminal sequence, other devices with corresponding device differences are identified as newly added upload devices.

[0008] Optionally, in the method according to the present invention, redundancy allocation is performed on each uploading device based on the amount of transmitted data, and the redundancy ratio of the transmitted data amount corresponding to the same uploading device is determined based on the obtained redundant data amount, including: The amount of data transmitted by each uploading device is summed, and the difference between the preset allocated data amount and the obtained total transmission amount is calculated to obtain the redundancy allocation amount. Determine the transmission ratio of each transmitted data volume to the total transmission volume, and calculate the redundant data volume corresponding to each upload device based on the product of the redundancy allocation and each transmission ratio. The redundancy ratio is obtained by calculating the ratio of the amount of redundant data corresponding to the same upload device to the amount of transmitted data.

[0009] Optionally, in the method according to the present invention, determining the data extraction target corresponding to the data transmission time, and performing image redundancy with a corresponding redundancy ratio on the transmitted image based on the data extraction target to obtain redundant data corresponding to the transmitted image, includes: In response to receiving a target extraction signal carrying any warning target at any target reception time, the data extraction targets of all data transmission times corresponding to a preset extraction duration after the target reception time are determined as the warning targets, and the redundancy ratio is numerically compared with the preset redundancy number. In response to the redundancy ratio being greater than or equal to a preset number of redundancy counts, the transmitted image is subjected to data redundancy based on complete redundancy and corresponding to the preset number of redundancy counts to obtain redundant data corresponding to the transmitted image; In response to the redundancy ratio being less than a preset number of redundancies, the ratio of the redundancy ratio to the corresponding preset number of redundancies is calculated to obtain the unit ratio; Based on the data extraction target, the corresponding single-unit proportion of the transmitted image is extracted, and the transmitted image is subjected to data redundancy based on partial redundancy and corresponding preset redundancy number according to the extracted target content, so as to obtain the redundant data corresponding to the transmitted image.

[0010] Optionally, in the method according to the present invention, the transmitted image is subjected to content extraction based on the data extraction target, corresponding to the proportion of individual components, and the transmitted image is subjected to data redundancy based on partial redundancy and corresponding to a preset number of redundancy times according to the extracted target content, to obtain redundant data corresponding to the transmitted image, including: Image recognition is performed on the transmitted image, and each recognition element located in the transmitted image is determined based on the recognition result; If any identified element is the same as any preset element in the retrieved element priority group, the preset priority corresponding to the preset element will be determined as the element priority of the corresponding identified element. In response to the existence of a corresponding data extraction target identification element in the transmitted image, the element priority corresponding to the identification element is adjusted to the highest element priority, and the element priority corresponding to each remaining identification element is updated; The transmitted image is processed into arrays to obtain array blocks with the same partitioning specifications, and all array blocks that make up the same identification element are identified as a block group of the same element. All element tile groups are sorted from largest to smallest priority based on element priority to obtain a priority sequence, and the total number of groups of all element image tiles included in each element tile group is determined. The total number of arrays corresponding to all array images is multiplied by the percentage of individual units to obtain the total number of selections; Based on the priority sequence, the total number of corresponding groups of element tile groups is superimposed one by one, and based on the comparison between the number of group superpositions corresponding to each superposition number and the selected total number, the transmitted image is subjected to data redundancy based on partial redundancy and corresponding preset redundancy number.

[0011] Optionally, in the method according to the present invention, the transmitted image is subjected to data redundancy based on partial redundancy and corresponding to a preset number of redundancy times, based on the comparison result between the number of group superpositions corresponding to each superposition number and the total number of selections, including: If the number of group stacks corresponding to the maximum stack count is less than or equal to the total number of selections, a redundant template with the same image specification as the transmitted image is created. The redundant template includes a mapping slot that has a redundant mapping relationship with each array tile. All element image blocks located in each element tile group are filled into the mapping slots located in the redundant template that have a redundant mapping relationship with each element image block, and the resulting redundant image is copied for a corresponding preset number of redundant times.

[0012] Optionally, in the method according to the present invention, the transmitted image is subjected to data redundancy based on partial redundancy and corresponding to a preset number of redundancy times, based on the comparison result between the number of group superpositions corresponding to each superposition number and the total number of selections, including: If the number of group stacks corresponding to any stacking number is greater than the total number of selections, all element tile groups corresponding to stacking numbers before that stacking number are determined as the first group set, and all remaining element tile groups are determined as the second group set based on the priority sequence; Determine the selection difference between the number of group superpositions corresponding to the first group set and the total number of selections, and determine the element tile group that is first in the second group set as the secondary selection group based on the priority sequence; All array tiles selected based on the selection difference of the secondary selection group are aggregated into the new tile group, and data redundancy is performed based on the first group set and the new tile group, with a corresponding preset number of redundancy times.

[0013] Optionally, in the method according to the invention, summarizing all array tiles corresponding to the selection difference selected based on the secondary selection group into a new tile group includes: The array of images arranged radially around the center point of the corresponding transmitted image is divided into regions to obtain regions corresponding to different hierarchical orders from the inside to the outside. Obtain the number of all array tiles in the secondary selection group included in each divided region, and calculate the number of regions in each divided region from the inside to the outside to obtain the number of regions superimposed for different superimposed times. If the number of regions corresponding to any number of superpositions is greater than the selection difference, the region difference between the number of regions superpositions and the selection difference is determined, and the region with the largest corresponding hierarchical order for the number of superpositions is determined as the secondary selection region. All array tiles corresponding to the number of superpositions, except those located in the secondary selection area, are aggregated into a new tile group, and all array tiles located in the secondary selection area corresponding to the area difference are aggregated into a new tile group.

[0014] According to another aspect of the present invention, a distributed heterogeneous device interconnection system in a narrowband weak connection environment is provided, comprising: a transmission determination module configured to determine all uploading devices that have device interconnection relationships with a receiving device, and to determine the amount of data transmitted by each uploading device corresponding to different transmitted images at the same data transmission time; The redundancy allocation module is configured to perform redundancy allocation for each upload device based on the amount of transmitted data, and determine the redundancy ratio of the transmitted data amount to the corresponding upload device based on the obtained redundancy data amount. The image redundancy module is configured to determine the data extraction target at the corresponding data transmission time, and perform image redundancy on the transmitted image with a corresponding redundancy ratio based on the data extraction target to obtain redundant data corresponding to the transmitted image; The unified sending module is configured to control the uploading device to uniformly send the transmitted image and the corresponding redundant data of the transmitted image to the receiving device.

[0015] According to the present invention, by constructing an image redundancy mechanism, this invention effectively solves the problem of complete data loss due to packet loss in image transmission under narrowband weak connection environments, significantly improving the transmission reliability of distributed heterogeneous device interconnection. Specifically, regarding the accuracy of redundancy allocation, this invention first determines the amount of data transmitted by each uploading device for different images at the same time, then performs differentiated redundancy allocation based on the data volume and calculates the redundancy ratio. This breaks the limitation of the uniform redundancy ratio in existing technologies. Higher redundancy can be allocated to images with larger data volumes (avoiding complete data loss due to large data volume and high packet loss probability), while the redundancy ratio is reasonably controlled for images with smaller data volumes (avoiding bandwidth waste). This achieves precise matching of redundancy resources and transmission requirements, maximizing packet loss resistance under the constraint of limited narrowband bandwidth. Regarding the targeted nature of the redundancy strategy... This invention designs an image redundancy scheme by combining data extraction targets. The redundancy priority can be adjusted according to actual needs, ensuring that core image information receives priority redundancy protection. This avoids the impact of packet loss in ordinary areas on the integrity of critical data, further enhancing the reliability and practicality of image transmission and meeting the data processing and decision-making needs in different scenarios. In terms of transmission coordination, this invention controls the uploading device to send the transmitted image and corresponding redundant data in a unified manner. Compared to existing technologies that separate redundant data from the original data or rely on retransmission requests, this invention can complete the delivery of critical data in a single step in a narrowband, weakly connected environment. Even if some packets of the original image are lost, the receiving end can complete the image through synchronously transmitted redundant data, completely avoiding the problem of complete image failure due to retransmission request timeouts or loss. It also simplifies the transmission process and reduces the communication overhead of device interconnection. Attached Figure Description

[0016] Figure 1 A flowchart of a method for interconnecting distributed heterogeneous devices in a narrowband weak connection environment according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of redundant images in this embodiment is shown; Figure 3 A structural block diagram of a distributed heterogeneous device interconnection system in a narrowband weak connection environment according to another embodiment of the present invention is shown. Detailed Implementation

[0017] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0018] To address the problems existing in the prior art, the inventors proposed the solution of this invention. One embodiment of this invention provides a method for interconnecting distributed heterogeneous devices in a narrowband, weakly connected environment. This method can be executed in a computing device, which can be understood as a terminal with data processing capabilities, such as a mobile phone or a computer.

[0019] Figure 1 A flowchart of a method for interconnecting distributed heterogeneous devices in a narrowband weak connection environment according to an embodiment of the present invention is shown, such as... Figure 1 As shown, the method begins with step S1, which includes the following: Identify all uploading devices that have interconnection relationships with the receiving device, and determine the amount of data transmitted by each uploading device for different transmitted images at the same data transmission time.

[0020] For example, in a narrowband, weakly connected environment, during the data transmission of images, to prevent the complete loss of the transmitted images due to packet loss, this embodiment performs image redundancy based on the transmitted images. The pre-process includes determining the amount of data to be transmitted for the uploading device, the receiving device, and the corresponding transmitted images. It can be explained that the receiving device can be understood as the terminal that ultimately receives the transmitted images and redundant data, while the uploading device is the source device that sends the transmitted images to the receiving device. The interconnection between these two devices is a prerequisite for data transmission and image redundancy. In a narrowband, weakly connected environment, only by clearly identifying which devices are uploading devices interconnected with the receiving device can we avoid the receiving device receiving data from irrelevant devices, thus preventing bandwidth consumption. Simultaneously, it ensures that subsequent redundant data can be accurately sent from the uploading device to the receiving device, preventing data transmission link interruptions due to chaotic device interconnection, and providing device-level guarantees for the stable transmission of transmitted images and redundant data. Furthermore, "the same data transmission time" refers to all uploading devices... The transmission data volume refers to the specific time point at which the uploading device synchronously transmits data. It represents the data size of each transmitted image when an uploading device sends one or more different transmitted images to a receiving device at that moment (e.g., if an uploading device sends image A at time t1, the transmission data volume is 2MB; if it sends image B, the transmission data volume is 1.5MB). Defining this data volume provides a precise quantitative reference for subsequent "redundancy allocation based on transmission data volume." For example, transmitted images with larger transmission data volumes require more redundant data to cope with packet loss risks; transmitted images with smaller transmission data volumes can have their redundant data volume appropriately reduced to save limited bandwidth in narrowband environments. Redundancy allocation based on actual transmission data volume avoids bandwidth waste or insufficient redundancy caused by blindly allocating redundant data. This ensures that in narrowband, weakly connected environments, the redundant data for each transmitted image is compatible with its own data volume, maximizing the compensation effect of redundant data for packet loss and providing data support for the rationality and effectiveness of subsequent image redundancy operations.

[0021] Furthermore, in this embodiment, the aforementioned "determining all uploading devices that have a device interconnection relationship with the receiving device, and determining the amount of data transmitted by each uploading device corresponding to different transmitted images at the same data transmission time" may further include the following steps: The system receives a receiving instruction carrying a device code sent by the management terminal, determines the receiving device based on the device code, and sorts the device code and the instruction sending time of the corresponding receiving instruction in a corresponding random order. The obtained information sequence is converted based on a hash algorithm, and a distribution identification code corresponding to the receiving device is generated based on the obtained hash string. Control the receiving device to distribute the distribution identification code based on a distributed heterogeneous network, and establish a timed task corresponding to the data distribution; In response to any other device in the corresponding distributed heterogeneous network, the distribution identification code is scanned and identified within the device response period of the corresponding timed task to obtain the device type corresponding to the other device. In response to the fact that the device type is within the preset upload range corresponding to the receiving device, the other device is identified as an upload device that has a device interconnection relationship with the receiving device; Establish a data transmission channel between the receiving device and each uploading device, and determine the amount of data transmitted by each uploading device based on the data transmission channel for different transmitted images at the same data transmission time.

[0022] For example, in this embodiment, the determination of the uploading device, the receiving device, and the amount of data transmitted for the corresponding transmitted image may specifically include the following technical contents: First, the server responds to the receiving instruction carrying the device code sent by the management terminal. In this embodiment, the server determines the receiving device based on the device code and sorts the device code and the instruction sending time of the corresponding receiving instruction in a random order. It can be explained that the management terminal is the terminal controlling data transmission. It can control the receiving device to receive corresponding transmitted images based on its needs. For example, in a power grid scenario, the management terminal can periodically monitor the safety of corresponding power grid equipment. The management terminal can control the receiving device corresponding to the power grid equipment to receive transmitted images captured by various uploading devices (e.g., cameras) that collect images of the power grid equipment. The device code is the unique identifier of the receiving device. Determining the receiving device based on this code avoids confusion between different receiving devices in a distributed heterogeneous network, ensuring that subsequent data transmission and redundant operations accurately target the receiving device. Sorting the device code and instruction sending time in a random order prevents malicious cracking of the fixed sorting pattern, thus avoiding the leakage of the receiving device's identity and improving the security of device identification. This lays the foundation for the stability of subsequent data transmission. The corresponding management terminal... Next, after determining the receiving device, the server can further perform data transformation on the obtained information sequence based on a hash algorithm, and generate a distribution identification code corresponding to the receiving device based on the obtained hash string. Here, the hash algorithm can convert the sorted information sequence into a fixed-length hash string. This string is unique and irreversible. The distribution identification code generated based on it can accurately represent the receiving device and is not easily tampered with. In a narrowband weak connection environment, a short and unique distribution identification code facilitates quick identification and transmission by the device, avoids transmission delays or packet loss due to excessively long identification codes, and ensures that other devices can efficiently locate the target receiving device. The corresponding distribution identification code can be a QR code or barcode, etc. Then, the server can control the receiving device to distribute the distribution identification code based on a distributed heterogeneous network, and establish a timed task corresponding to the data distribution. It can be explained that distributing the distribution identification code in a distributed heterogeneous network can make other devices in the network aware of the existence of the target receiving device; establishing a timed task can limit the time range of device response (device response cycle), avoid other devices from continuously responding during unnecessary periods, reduce invalid bandwidth occupation in narrowband environments, and ensure that the device response is timely, preventing the failure to establish device interconnection relationships due to response delays. Subsequently, any other device in the corresponding distributed heterogeneous network scans and identifies the distribution identification code within the device response period of the corresponding timed task. The server can obtain the device type corresponding to the other device. That is, only the other devices that scan and identify the distribution identification code within the device response period are processed. This can further filter out devices with real-time connection capabilities and exclude devices that cannot respond in time due to weak network connections. This ensures that subsequent interconnected devices can transmit data stably. Obtaining the device type is to determine whether the device has the ability to upload the corresponding transmitted image, avoiding including devices without the function of uploading the transmitted image in the interconnection range and improving the accuracy of device screening. Subsequently, in response to the device type being within the preset upload range corresponding to the receiving device, the server can identify the other device as an upload device that has a device interconnection relationship with the receiving device. It can be explained that the preset upload range is a set of device types with qualified upload capabilities set by the receiving device according to its own needs (such as cameras that can simultaneously capture images of the same type of power grid equipment). Identifying other devices with device types within this range as upload devices can ensure that the upload devices have the hardware and functional foundation for stable image transmission, avoid increased packet loss rate due to insufficient device capabilities, and provide a reliable device source for subsequent image redundancy and data transmission. Finally, the server can establish a data transmission channel for each uploading device corresponding to the receiving device, and determine the amount of data transmitted by each uploading device for different transmitted images based on the data transmission channel at the same data transmission time. Here, the dedicated data transmission channel can reduce data transmission interference between different uploading devices and receiving devices, improve the stability of data transmission in a narrowband weak connection environment, and reduce the risk of packet loss caused by channel sharing. Determining the amount of data transmitted by each uploading device for different transmitted images at the same data transmission time can provide a precise quantitative basis for the redundancy allocation based on the amount of transmitted data mentioned later. For example, more redundant data can be allocated to transmitted images with a larger amount of transmitted data to ensure that the redundant data can effectively compensate for packet loss, while the redundant data can be appropriately reduced for transmitted images with a smaller amount of transmitted data to save bandwidth. Ultimately, a balance is achieved between the image redundancy anti-packet loss effect and the bandwidth resource utilization efficiency in a narrowband weak connection environment.

[0023] Furthermore, if the number of first devices with interconnection relationships with the receiving device, determined based on the timed task, is less than the preset number of devices, it will result in an insufficient number of uploading devices. This will not only reduce the transmission efficiency of transmitted images and redundant data (e.g., a single device has to bear an excessive transmission task, which can easily lead to transmission delays due to narrowband bandwidth limitations), but may also increase the risk of packet loss due to the small number of devices and the single transmission link. That is, once the connection of an uploading device is interrupted, the receiving device may not be able to obtain the corresponding transmitted images and redundant data, ultimately affecting the reliability of image redundancy and packet loss prevention. Therefore, further, this embodiment may also include the following steps: If the number of first devices corresponding to all upload devices with device interconnection relationship with the receiving device, as determined by the timed task, is less than the preset number of devices, the first location information corresponding to each upload device is obtained. Based on the first location information, a device communication circle is established with each uploading device as the center. Based on the distributed heterogeneous network, the second location information corresponding to each other device is determined. Based on the second location information, other devices located in the device communication circle are determined as terminal supplementary groups. If the number of second devices corresponding to other devices in all terminal supplementary groups is less than or equal to the number of preset devices corresponding to the number of first devices, the other devices in all terminal supplementary groups are identified as newly added upload devices. The device type and device deviation distance between each other device in the terminal supplement group and the upload device are obtained by responding to the device difference between the second number of other devices in all terminal supplement groups and the first number of devices corresponding to a preset number of devices; Based on the device deviation distance, determine the distance deviation value of the upload device corresponding to each other device, and based on the device type, determine the type deviation value of the preset upload range corresponding to each other device; The distance deviation value and type deviation value corresponding to the same other device are summed. Based on the obtained comprehensive deviation value, all other devices are sorted from smallest to largest terminal. Based on the obtained terminal sequence, other devices with corresponding device differences are identified as newly added upload devices.

[0024] For example, in this embodiment, in order to improve the reliability of image redundancy and packet loss prevention, this embodiment can be implemented based on the following technical content: First, if the number of first devices corresponding to all upload devices with device interconnection relationships with the receiving device, as determined by the timed task, is less than the preset number of devices, the server can obtain the first location information for each upload device. Here, the first location information is the specific location data of the determined upload device in the distributed heterogeneous network. When the number of first devices is insufficient, obtaining this information can provide a location benchmark for subsequent screening of other devices that can be supplemented, avoiding blindly searching for supplementary devices in the entire network, reducing invalid information interaction in narrowband environments, and improving the targeting of supplementary device screening. Next, the server can establish a device communication circle centered on each uploading device based on the first positioning information, determine the second positioning information corresponding to each other device based on the distributed heterogeneous network, and determine the other devices located within the device communication circle as terminal supplementary groups based on the second positioning information. It can be noted that the device communication circle is a range that can stably communicate, centered on the existing uploading device (considering the communication distance limitations in a narrowband weak connection environment, the range delineation needs to ensure the connection stability between the devices within the circle and the uploading device); the second positioning information is the location data of other devices in the network that have not been determined as uploading devices. In this embodiment, by determining the other devices located within the device communication circle as terminal supplementary groups, it can ensure that these supplementary devices have basic communication conditions with the existing uploading devices, reduce the risk of connection interruption or packet loss due to excessive distance, and lay a location foundation for stable data transmission of newly added uploading devices in the future; Then, in response to the device difference between the number of second devices in all terminal supplementary groups and the number of preset devices corresponding to the number of first devices, the server can identify the other devices in all terminal supplementary groups as newly added upload devices. It can be noted that the device difference is the difference between the preset number of devices and the number of first devices (for example, when the preset number of devices is 10 and the number of first devices is 6, the corresponding device difference is 4). When the number of second devices (for example, 3) is less than or equal to the device difference, the server can directly identify the other devices in all terminal supplementary groups as newly added upload devices. This can quickly fill the device quantity gap, avoid the problem of insufficient upload device quantity due to excessively long screening process, ensure that the transmission tasks of transmitting images and redundant data can be allocated to sufficient devices in a timely manner, improve transmission efficiency, enrich transmission links, and reduce the risk of data loss due to packet loss in a single link. Alternatively, if the number of second devices corresponding to other devices in all terminal supplement groups is greater than the number of preset devices corresponding to the number of first devices, the server can obtain the device type of each other device in the terminal supplement group and the device deviation distance between it and the uploading device. That is, when the number of second devices (e.g., 6) is greater than the device difference (e.g., 4), it is necessary to further screen suitable devices for uploading supplementation. Obtaining the device type is to determine whether other devices can perform image acquisition on the same type of power grid equipment (i.e., it needs to fit the preset upload range of the receiving device). Obtaining the device deviation distance is to evaluate the connection stability between the device and the existing uploading device (the closer the distance, the lower the communication packet loss rate in a narrowband weak connection environment is usually). The combination of the two can ensure that the newly added uploading devices selected have both functional adaptability and connection reliability. Subsequently, the server can determine the distance deviation value of the upload device corresponding to each other device based on the device deviation distance, and determine the type deviation value of the preset upload range corresponding to each other device based on the device type. It can be noted that, in this embodiment, the distance deviation value is a quantitative assessment of the device deviation distance (for example, the farther the distance, the larger the distance deviation value, which means the worse the connection stability); while the type deviation value is a quantitative assessment of the matching degree between the device type and the preset upload range (for example, if the device type completely matches the preset upload range, the type deviation value is 0; the larger the deviation, the larger the type deviation value). By quantifying the deviation value, the device's location and functional adaptability can be transformed into comparable data, providing an objective basis for subsequent screening and avoiding screening errors caused by subjective judgment. Finally, the server calculates the sum of the distance deviation value and the type deviation value corresponding to other devices. Based on the obtained comprehensive deviation value, all other devices are sorted from smallest to largest. Based on the obtained terminal sequence, other devices with corresponding device differences are identified as new upload devices. Here, based on the above, it is known that the smaller the comprehensive deviation value, the better the connection stability and the higher the functional adaptability of the device. Prioritizing such devices as new upload devices (for example, selecting the first 4 devices with the smallest comprehensive deviation value from the terminal sequence) can ensure that the new upload devices have optimal transmission capabilities in narrowband weak connection environments. That is, it reduces the risk of packet loss due to connection instability and can efficiently complete the upload tasks of transmitting images and redundant data. Ultimately, it achieves the goal of meeting the preset requirements for the number of upload devices and ensuring transmission quality, providing sufficient and reliable device support for subsequent image redundancy operations.

[0025] Step S2 includes the following: Redundancy allocation is performed on each upload device based on the amount of data transmitted, and the redundancy ratio of the transmitted data to the corresponding upload device is determined based on the obtained redundant data amount.

[0026] For example, in this embodiment, by allocating redundant data to each uploading device based on the amount of transmitted data and calculating the redundancy ratio, it can be ensured that the amount of redundant data is accurately matched with the transmission needs of the uploading device. This allows for a balance between the effectiveness of redundancy in preventing packet loss and the efficiency of bandwidth resource utilization in narrowband, weakly connected environments. Specifically, the amount of transmitted data refers to the size of the data transmitted by each uploading device for different images at the same data transmission moment. It directly reflects the data scale of the images transmitted by that uploading device. A larger amount of transmitted data means that the images transmitted by that device contain richer information, and in the event of packet loss, more redundant data is needed to fill the information gap. Conversely, a smaller amount of transmitted data requires relatively less redundant data. Redundancy allocation based on this data avoids a "one-size-fits-all" redundancy allocation approach, ensuring that uploading devices with large amounts of transmitted data are allocated sufficient redundant data to cope with possible packet loss and reduce the risk of complete data loss. At the same time, an appropriate amount of redundant data is allocated to uploading devices with small amounts of transmitted data to avoid... To avoid excessive consumption of narrowband bandwidth, ensure smooth transmission links for other uploading devices, and improve the overall stability of data transmission; the redundancy ratio is the ratio of the amount of redundant data to the amount of transmitted data for the same uploading device (for example, if an uploading device transmits 5MB of data and has 1MB of redundant data at a certain data transmission moment, then the corresponding redundancy ratio is 20%). By calculating this ratio, the relationship between the amount of redundant data and the amount of transmitted data can be transformed into an intuitive proportional indicator, providing a clear quantitative basis for the subsequent discussion of image redundancy based on the data extraction target and corresponding redundancy ratio of transmitted images. For example, if the redundancy ratio of an uploading device is 30%, then when performing image redundancy, the redundant data of the transmitted image of that device can be accurately generated according to the 30% ratio, ensuring that the scale of redundant data matches the scale of the original data of the transmitted image. This ensures that there is neither too little redundant data due to a redundancy ratio, which would be unable to compensate for packet loss, nor too much redundant data due to a redundancy ratio, which would waste narrowband bandwidth.

[0027] Furthermore, in this embodiment, the aforementioned "performing redundancy allocation for each uploading device based on the amount of transmitted data, and determining the redundancy ratio of the transmitted data amount corresponding to the same uploading device based on the obtained redundant data amount" may further include the following steps: The amount of data transmitted by each uploading device is summed, and the difference between the preset allocated data amount and the obtained total transmission amount is calculated to obtain the redundancy allocation amount. Determine the transmission ratio of each transmitted data volume to the total transmission volume, and calculate the redundant data volume corresponding to each upload device based on the product of the redundancy allocation and each transmission ratio. The redundancy ratio is obtained by calculating the ratio of the amount of redundant data corresponding to the same upload device to the amount of transmitted data.

[0028] For example, in this embodiment, determining the corresponding redundancy ratio based on redundancy allocation can be specifically implemented based on the following technical aspects: First, the server can sum the amount of data transmitted by each uploading device at the same data transmission moment, and then calculate the difference between the preset allocated data amount and the obtained total transmission amount to obtain the redundancy allocation amount. It can be explained that the total transmission amount is the sum of the data transmitted by all uploading devices at the same data transmission moment (for example, if uploading device A transmits 3MB of data and uploading device B transmits 2MB of data, then the corresponding total transmission amount is 5MB). The preset allocated data amount, on the other hand, is the total amount of data that can be used to transmit images and redundant data in a narrowband, weak connection environment, considering both bandwidth capacity and packet loss prevention requirements (e.g., the preset allocated data amount is 7MB). The redundancy allocation amount (7MB-5MB=2MB) calculated by the difference between the two clearly defines the total amount of redundant resources available for allocation to all uploading devices, preventing the allocation of redundant resources from exceeding the narrowband bandwidth capacity, ensuring that redundant data transmission does not lead to bandwidth congestion, and providing a total constraint for the subsequent redundancy allocation of each device, balancing redundancy requirements and bandwidth resource limitations. Next, the server can determine the corresponding transmission proportion of the total transmission volume based on each transmitted data volume, and calculate the redundant data volume for each upload device based on the product of the redundancy allocation and each transmission proportion. Here, the transmission proportion is the ratio of the transmitted data volume of a single upload device to the total transmission volume (for example, the transmission proportion of upload device A is 3MB ÷ 5MB = 60%, and the transmission proportion of upload device B is 2MB ÷ 5MB = 40%). Based on the product of this proportion and the redundancy allocation (2MB), the redundant data volume is calculated (upload device A: 2MB × 60% = 1.2MB; upload device B: 2MB × 40% = 0.8MB). This ensures that the allocation of redundant data volume matches the scale of the transmitted data volume of the upload device. That is, the larger the transmitted data volume of the device, the more redundant data volume it obtains, which can better cope with the high packet loss risk caused by the large data volume; the smaller the transmitted data volume of the device, the less redundant data volume it obtains, avoiding waste of redundant resources. This proportional allocation method can maximize the use of limited redundant resources in narrowband weak connection environments and improve the overall packet loss prevention effect. Finally, the server can calculate the ratio of redundant data volume to transmitted data volume corresponding to the same upload device to obtain the redundancy ratio. It can be explained that the redundancy ratio is the ratio of redundant data volume of a single upload device to its own transmitted data volume (for example, the redundancy ratio of upload device A is 1.2MB ÷ 3MB = 40%, and the redundancy ratio of upload device B is 0.8MB ÷ 2MB = 40%). This calculation transforms the relationship between redundant data volume and transmitted data volume into an intuitive proportional indicator, providing a precise quantitative basis for subsequent "image redundancy based on the data extraction target and corresponding redundancy ratio of the transmitted image." For example, when performing image redundancy on the transmitted image of upload device A, redundant data can be directly generated according to a 40% redundancy ratio, ensuring that the scale of redundant data is highly compatible with the original data scale of the transmitted image. This avoids insufficient redundant data due to a low redundancy ratio, which would fail to compensate for packet loss, and excessive redundant data due to a high redundancy ratio, which would waste narrowband bandwidth. Ultimately, this achieves the accuracy and effectiveness of image redundancy in narrowband weak connection environments.

[0029] Step S3 includes the following steps: The data extraction target corresponding to the data transmission time is determined, and the transmitted image is redundant according to the corresponding redundancy ratio based on the data extraction target to obtain the redundant data corresponding to the transmitted image.

[0030] For example, in this embodiment, by defining the data extraction target to clarify the redundancy focus and controlling the redundancy scale based on the redundancy ratio, it ensures that image redundancy in a narrowband weak connection environment accurately focuses on key content while making reasonable use of bandwidth and effectively addressing packet loss risks. Specifically, the data extraction target is the information content that the receiving device or management end needs to extract from the transmitted image at a specific data transmission moment (e.g., in a security monitoring scenario, the key information to extract is "personal facial information" at a certain data transmission moment; in an industrial inspection scenario, the key information to extract is "equipment fault location"). In a narrowband weak connection environment, due to limited transmission bandwidth, clearly defining the data extraction target allows image redundancy operations to focus on key content, avoiding bandwidth waste caused by redundant processing of irrelevant content. For example, when the data extraction target is "equipment fault location," subsequent image redundancy can prioritize the redundant data for that location, ensuring that even if packet loss occurs, the key fault location information can be recovered from the redundant data. To reduce the risk of complete data failure due to the loss of critical information, image redundancy is guided by clear content. The redundancy ratio, previously determined based on the amount of data transmitted by the uploading device, is the ratio of redundant data to transmitted data (e.g., a redundancy ratio of 30% for an uploading device). It quantifies the upper limit of the scale of redundant data, ensuring that redundant data does not exceed the carrying capacity of narrowband bandwidth. In this embodiment, image redundancy based on the data extraction target and corresponding redundancy ratio means that within the scale limited by the redundancy ratio, priority is given to redundancy processing of content related to the data extraction target in the transmitted image. For example, if the redundancy ratio is 30%, the server can extract content related to the data extraction target from the overall data of the transmitted image and generate redundant data with a ratio of 30%, or ensure that the content related to the target accounts for no less than 30% of the overall redundant data. Thus, the total scale of redundant data can be controlled by the redundancy ratio to avoid bandwidth waste, and the data extraction target can be used to ensure that the redundant data focuses on critical content.

[0031] Furthermore, in this embodiment, the aforementioned "determining the data extraction target corresponding to the data transmission time, and performing image redundancy on the transmitted image based on the data extraction target with a corresponding redundancy ratio to obtain redundant data corresponding to the transmitted image" may further include the following steps: In response to receiving a target extraction signal carrying any warning target at any target reception time, the data extraction targets of all data transmission times corresponding to a preset extraction duration after the target reception time are determined as the warning targets, and the redundancy ratio is numerically compared with the preset redundancy number. In response to the redundancy ratio being greater than or equal to a preset number of redundancy counts, the transmitted image is subjected to data redundancy based on complete redundancy and corresponding to the preset number of redundancy counts to obtain redundant data corresponding to the transmitted image; In response to the redundancy ratio being less than a preset number of redundancies, the ratio of the redundancy ratio to the corresponding preset number of redundancies is calculated to obtain the unit ratio; Based on the data extraction target, the corresponding single-unit proportion of the transmitted image is extracted, and the transmitted image is subjected to data redundancy based on partial redundancy and corresponding preset redundancy number according to the extracted target content, so as to obtain the redundant data corresponding to the transmitted image.

[0032] For example, in this embodiment, the image redundancy of the transmitted image according to the corresponding redundancy ratio can be specifically implemented based on the following technical content: First, in response to receiving a target extraction signal carrying any warning target at any target reception time, the server can determine the data extraction targets of all data transmission times within a corresponding preset extraction duration after the target reception time as the warning targets. The server then compares the redundancy ratio with a preset redundancy count. In this embodiment, the target reception time is a specific point in time when the target extraction signal is received, the warning target is a critical content that needs to be monitored or protected (e.g., "abnormal moving targets" in security scenarios, "abnormal temperature areas of equipment" in power grid environments), and the target extraction signal is the instruction that triggers key redundancy. By determining the data extraction targets of all data transmission times within a preset extraction duration (e.g., 5 minutes) after the target reception time as the warning targets, the server can clearly define the key redundancy period, ensuring that during the transmission of warning-related images, redundancy operations always focus on the warning target, avoiding wasting bandwidth during non-critical periods. Simultaneously, comparing the redundancy ratio with a preset redundancy count (the number of redundancies to be completed based on narrowband bandwidth and packet loss prevention requirements, e.g., 3 times) provides a basis for subsequent selection of full or partial redundancy, ensuring that the redundancy method is compatible with the scale of redundancy resources. Secondly, in response to the redundancy ratio being greater than or equal to the preset redundancy number, the server can perform data redundancy on the transmitted image based on complete redundancy and corresponding to the preset redundancy number, to obtain the redundant data corresponding to the transmitted image. That is, when the redundancy ratio is sufficient to support the preset redundancy number (for example, the redundancy ratio is 350%, the preset redundancy number is 3 times, and the redundancy requirement is 350%≥3 times), performing complete redundancy means generating redundant data for all the content of the transmitted image according to the preset redundancy number (such as copying the entire transmitted image 3 times as redundant data). This ensures that each part of the transmitted image has sufficient redundancy protection. Even if multiple packet losses occur in a narrowband weak connection environment, all image information can be recovered through complete redundant data, minimizing the risk of complete data loss. At the same time, sufficient redundancy ratio is used to meet multiple redundancy requirements, avoiding the reduction in anti-packet loss effect caused by insufficient redundancy. In addition, if the redundancy percentage is less than the preset number of redundancies, the server needs to calculate the ratio of the redundancy percentage to the preset number of redundancies to obtain the unit percentage. That is, when the redundancy percentage cannot support the preset number of redundancies (for example, the redundancy percentage is 150%, the preset number of redundancies is 3, and 150% < the redundancy requirement corresponding to 3), the unit percentage obtained by the ratio calculation (for example, 150% ÷ 3 times ≈ 50%) can evenly distribute the limited redundancy percentage to each redundancy operation, ensuring that each redundancy has a clear redundancy scale constraint. This avoids the situation where subsequent redundancy operations have no resources available due to a single redundancy percentage being too high, or a single redundancy percentage being too low to play a role in preventing packet loss. It provides an accurate basis for the single redundancy ratio for subsequent partial redundancy. Finally, after determining the percentage of individual units, the server can extract content corresponding to the percentage of individual units from the transmitted image based on the data extraction target. Then, based on the extracted target content, the server performs data redundancy on the transmitted image with partial redundancy and a preset number of redundancy counts to obtain redundant data for the transmitted image. It can be explained that the data extraction target (i.e., the warning target) is the key content in the transmitted image. Extracting this part of the content based on the percentage of individual units (e.g., if the percentage of individual units is 50%, then extracting 50% of the content related to the warning target in the transmitted image) and then performing redundancy on the target content according to a preset number of redundancy counts (e.g., copying the extracted target content 3 times as redundant data) prioritizes multiple redundancies of key content within a limited redundancy percentage in a narrowband, weakly connected environment. Even if non-critical parts of the transmitted image are lost due to packet loss, core information can be retained through the redundant target content, preventing complete data loss. Simultaneously, controlling the scale of each redundancy count by the percentage of individual units ensures that the preset number of redundancy counts is completed, balancing redundancy counts with key content protection, achieving optimal packet loss prevention under limited bandwidth resources.

[0033] Furthermore, in this embodiment, the aforementioned "extracting content from the transmitted image based on the target data extraction, corresponding to the proportion of each individual element, and performing data redundancy on the transmitted image based on partial redundancy and a preset number of redundancy times according to the extracted target content, to obtain redundant data corresponding to the transmitted image" may further include the following steps: Image recognition is performed on the transmitted image, and each recognition element located in the transmitted image is determined based on the recognition result; If any identified element is the same as any preset element in the retrieved element priority group, the preset priority corresponding to the preset element will be determined as the element priority of the corresponding identified element. In response to the existence of a corresponding data extraction target identification element in the transmitted image, the element priority corresponding to the identification element is adjusted to the highest element priority, and the element priority corresponding to each remaining identification element is updated; The transmitted image is processed into arrays to obtain array blocks with the same partitioning specifications, and all array blocks that make up the same identification element are identified as a block group of the same element. All element tile groups are sorted from largest to smallest priority based on element priority to obtain a priority sequence, and the total number of groups of all element image tiles included in each element tile group is determined. The total number of arrays corresponding to all array images is multiplied by the percentage of individual units to obtain the total number of selections; Based on the priority sequence, the total number of corresponding groups of element tile groups is superimposed one by one, and based on the comparison between the number of group superpositions corresponding to each superposition number and the selected total number, the transmitted image is subjected to data redundancy based on partial redundancy and corresponding preset redundancy number.

[0034] For example, in this embodiment, the data redundancy of the transmitted image based on partial redundancy and corresponding to a preset number of redundancy times can be specifically implemented based on the following technical content: First, the server can perform image recognition on the transmitted image and determine the identification elements located in the transmitted image based on the recognition results. It can be explained that image recognition is a process of parsing the content of the transmitted image through algorithms. The identification elements are information units in the transmitted image that can be independently identified (such as "personnel", "vehicles", and "signs" in security images, and "power grid equipment", "lines", and "indicator lights" in power grid images). By determining each identification element, the abstract content of the transmitted image can be transformed into specific and distinguishable units, providing a basis for subsequent priority division and content extraction, avoiding blind extraction caused by the inability to identify image content, and ensuring that subsequent operations can accurately locate key information. Here, image recognition can be specifically based on neural network models or machine learning models to complete the corresponding recognition process. Since the construction, training, and recognition process of neural network models or machine learning models can all be implemented based on existing technologies, this embodiment will not elaborate on this in detail. Secondly, if any identified element is the same as any preset element in the retrieved element priority group, the server can determine the preset priority of the corresponding preset element as the element priority of the identified element. It can be explained that the element priority group is a set of elements containing high-value information that is preset according to the regular transmission requirements (for example, "equipment fault indicator" and "key components" are preset elements in the power grid scenario). It can be stored in the server in advance, while the preset priority is the order of importance set for each preset element (for example, "fault indicator" has a higher priority than "ordinary shell"). In this embodiment, by matching the identified element with the preset element and assigning the corresponding priority, high-value elements in the transmitted image can be initially screened out. This ensures that in the case of limited narrowband bandwidth, the regular important elements are given priority for redundant protection, improving the utilization efficiency of redundant data and avoiding non-important elements from occupying redundant resources. Then, in response to the presence of a corresponding data extraction target identification element in the transmitted image, the server can adjust the element priority of the corresponding identification element to the highest element priority, and update the element priority of each remaining identification element. Based on the foregoing, the data extraction target is the core content that needs to be protected at the current transmission moment (such as "abnormal moving target" in an early warning scenario). By adjusting the priority of its corresponding identification element to the highest, it can be ensured that the element is given priority in subsequent extraction, avoiding the omission of key targets due to conventional priority settings. At the same time, dynamically updating the priority of the remaining elements can maintain the order of importance among elements, providing a clear basis for subsequent extraction of content according to priority, ensuring that redundant data always focuses on the most critical information at present, and in the event of packet loss, the core target content can be recovered through redundant data, reducing the risk of complete data loss. Next, the server can synchronously perform array processing on the transmitted image to obtain array patches with the same division specifications, and determine all array patches that make up the same identification element as the same element patch group. Here, array processing is the operation of dividing the transmitted image into multiple array patches of uniform specifications according to a fixed size (e.g., 10×10 pixels). The element patch group is a collection of all array patches that make up the same identification element (e.g., the "person" identification element is composed of array patches corresponding to the head, torso, and limbs, and these patches together constitute the "person" element patch group). This can transform the identification element into a quantifiable and countable set of patches, providing an operable unit for subsequent calculation of the total number of patches based on the proportion of individual components and superimposed filtering of patch groups. This avoids the inability to control the extraction scale due to the abstraction of image content and ensures that the content extraction can accurately match the individual component proportion requirements. Subsequently, the server can prioritize all element tile groups from largest to smallest based on element priority, obtaining a priority sequence and determining the total number of groups for all element image tiles included in each element tile group. It can be explained that the priority sequence is ordered according to element importance (e.g., "data extraction target element" > "element priority group element" > "ordinary element"), while the total number of groups is the number of array tiles contained in a single element tile group (e.g., the "personnel" element tile group contains 50 array tiles, and the total number of groups is 50). Determining the priority sequence clarifies the order of content extraction, ensuring that high-priority elements are extracted first; while determining the total number of groups provides quantitative data for subsequent overlay calculations, avoiding over- or under-extraction due to unknown element tile group size, laying the foundation for accurate matching of individual component proportions. Next, the server can multiply the total number of arrays corresponding to all array images with the single-unit percentage to obtain the total number of selections. It can be noted that the total number of arrays is the total number of all array tiles after the transmitted image is processed by the array (for example, if the transmitted image is divided into 200 array tiles, then the corresponding total number of arrays is 200). The single-unit percentage is the previously calculated percentage of content to be extracted for a single redundancy (e.g., 50%). In this embodiment, the total number of selections obtained by multiplication (e.g., 200 × 50% = 100) can clearly define the total number of array tiles to be extracted based on the single-unit percentage, setting a clear upper limit for the scale of subsequent content extraction, avoiding excessive extraction of content that occupies narrowband bandwidth, or insufficient extraction of content that cannot meet the packet loss prevention requirements, and ensuring that the scale of partial redundancy is adapted to the narrowband environment. Finally, the server can successively calculate the total number of corresponding groups for the element tile groups based on the priority sequence, and perform data redundancy on the transmitted image based on partial redundancy and a preset number of redundancy times based on the comparison between the number of groups for each number of times of redundancy and the total number of selections. Here, the total number of groups of element tile groups is superimposed according to the priority sequence (for example, first superimpose 120 tiles of "data extraction target elements", the number of groups is 120; if it exceeds the total number of selections of 100, then stop superimposing and adjust). By comparing the number of groups superimposed with the total number of selections, tile groups that meet the single-unit ratio can be accurately selected (for example, after superimposing 100 tiles of "data extraction target elements", the total number of selections is reached). By performing data redundancy on these selected tile groups according to the preset number of redundancy times (such as copying 3 times), the obtained redundant data can focus on the key content corresponding to the data extraction target, and the redundancy scale can be controlled to adapt to narrowband bandwidth. In a narrowband weak connection environment, even if packet loss occurs, the core information of the transmitted image can be recovered through the redundant data containing key tile groups, effectively preventing complete data loss, while maximizing the use of limited bandwidth resources to improve the redundancy effect.

[0035] In one case, the aforementioned "performing data redundancy on the transmitted image based on partial redundancy and corresponding to a preset number of redundancy times, based on the comparison result between the number of group superpositions corresponding to each superposition number and the total number of selections" may further include the following steps: If the number of group stacks corresponding to the maximum stack count is less than or equal to the total number of selections, a redundant template with the same image specification as the transmitted image is created. The redundant template includes a mapping slot that has a redundant mapping relationship with each array tile. All element image blocks located in each element tile group are filled into the mapping slots located in the redundant template that have a redundant mapping relationship with each element image block, and the resulting redundant image is copied for a corresponding preset number of redundant times.

[0036] For example, in this embodiment, when the number of stacks in a group corresponding to the maximum number of stacks is less than or equal to the total number of selections, data redundancy can be implemented based on the following technical aspects: First, if the number of group overlays corresponding to the maximum overlay count is less than or equal to the total number of selections, the server can create a redundant template with the same image specification as the transmitted image. This redundant template includes mapping slots with redundant mapping relationships to each array tile. It can be noted that the maximum overlay count is the final number of times element tile groups are overlaid according to priority, and the group overlay count is the total number of array tiles in all element tile groups at that number of overlays. When the group overlay count is less than or equal to the total number of selections, it indicates that the filtered element tile groups have not exceeded the required extraction scale, and no additional filtering of other elements is needed. In this case, the server can create a redundant template with the same specification as the transmitted image. The template (e.g., if the transmitted image is 100×100 pixels, then the corresponding redundant template is also 100×100 pixels) ensures that the size of the redundant data matches the original image perfectly. The mapping slot is the position in the redundant template that corresponds one-to-one with each array patch of the original transmitted image (e.g., the array patch in the first row and first column of the original image corresponds to the mapping slot in the first row and first column of the redundant template). This one-to-one redundant mapping relationship can avoid the redundant data from being misaligned with the original image, ensuring that when the receiving device loses packets, it can accurately recover the lost portion of the original image through the redundant data in the corresponding slot of the redundant template, providing structural guarantees for the effective use of redundant data. Secondly, the service area can fill all the element image blocks located in each element tile group into the mapping slots of the redundant template that have a redundant mapping relationship with each element image block, and perform image copying on the obtained redundant image for a corresponding preset number of redundancy times. It can be explained that filling the element image blocks (array tiles) contained in the element tile group into the corresponding mapping slots of the redundant template ensures that the redundant template contains only image information of high-priority elements. Thus, in a narrowband weak connection environment, based on the filling method that focuses on high-priority content, non-critical content can be avoided from occupying redundant resources, ensuring that the limited bandwidth is used only to protect core information. At the same time, copying the filled redundant image according to the preset number of redundancy times (such as 3 times) can generate multiple identical redundant data (such as 3 redundant images containing high-priority elements). Even if 1-2 packet losses occur during narrowband transmission, there is still remaining redundant data that can be used to recover the core content of the original image, effectively reducing the risk of complete data loss due to packet loss, and balancing the redundancy effect with the limited bandwidth of narrowband.

[0037] For example, Figure 2 This diagram illustrates the redundant images in this embodiment, based on Figure 2 As can be seen from the example, the redundant image includes 64 mapping slots, of which 12 mapping slots contain redundant array tiles, and the content displayed in the redundant image corresponding to the 12 mapping slots is "personnel", which can correspond to the data extraction target.

[0038] In another scenario, the aforementioned "performing data redundancy on the transmitted image based on partial redundancy and corresponding to a preset number of redundancy counts, based on the comparison result between the number of group overlays corresponding to each overlay count and the total number of selections" may further include the following steps: If the number of group stacks corresponding to any stacking number is greater than the total number of selections, all element tile groups corresponding to stacking numbers before that stacking number are determined as the first group set, and all remaining element tile groups are determined as the second group set based on the priority sequence; Determine the selection difference between the number of group superpositions corresponding to the first group set and the total number of selections, and determine the element tile group that is first in the second group set as the secondary selection group based on the priority sequence; All array tiles selected based on the selection difference of the secondary selection group are aggregated into the new tile group, and data redundancy is performed based on the first group set and the new tile group, with a corresponding preset number of redundancy times.

[0039] For example, in this embodiment, when the number of stacks in a group corresponding to the maximum number of stacks is greater than the total number of selections, data redundancy can be specifically implemented based on the following technical aspects: First, if the number of group stacks corresponding to any stacking number is greater than the total number of selections, the server can determine all element tile groups corresponding to stacking numbers preceding that stacking number as the first group set, and determine all remaining element tile groups as the second group set based on the priority sequence. It can be noted that any stacking number is a specific number of times element tile groups are stacked sequentially according to the priority sequence, and the group stacking number is the total number of array tiles in all stacked element tile groups at that number. When the group stacking number is greater than the total number of selections (for example, if the total number of selections is 100, and the group stacking number at a certain stacking number is 120), the current stacking is stopped and the set is divided, thus revealing the first group set. The first set consists of all the higher-priority element tile groups that have been superimposed before this number of superpositions (e.g., the "data extraction target element tile group" and "high-priority preset element tile group" from the first two superpositions). The second set consists of the remaining lower-priority element tile groups in the priority sequence (e.g., the "ordinary element tile group"). This allows for the retention of high-priority element tile groups, preventing the discarding of key content due to excessive superposition. At the same time, low-priority elements are grouped into the second set, providing a source for accurate gap filling in the future. This ensures that redundant data always focuses on high-priority content, improving the targeting of packet loss prevention. It should be noted that the total number of superpositions corresponding to the first set should be less than or equal to the total number of selections. Secondly, the server can further determine the selection difference between the group stack number corresponding to the first group set and the total number of selections, and determine the first element tile group in the second group set as the secondary selection group based on the priority sequence. Here, the selection difference is the difference between the group stack number of the first group set and the total number of selections (for example, if the group stack number of the first group set is 90 and the total number of selections is 100, the selection difference is 10). This quantifies the number of array tiles that need to be supplemented. The corresponding secondary selection group is the highest priority element tile group in the second group set (such as the "second highest priority ordinary element tile group" in the second set). By calculating the selection difference, the specific scale to be supplemented can be clearly defined, avoiding over- or under-selection. Selecting the first element tile group in the second set as the secondary selection group ensures that the supplemented content is still relatively high-priority information, avoiding the consumption of bandwidth by supplementing low-value content, and maximizing the utilization efficiency of redundant data in a narrowband environment. Finally, the server can aggregate all array tiles selected based on the selection difference of the secondary selection group into a new tile group, and perform data redundancy based on partial redundancy and a preset number of redundancy times according to the first group set and the new tile group. Selecting array tiles from the secondary selection group with the same number as the selection difference (for example, if the selection difference is 10, then 10 array tiles can be selected from the secondary selection group) and aggregating them into the new tile group ensures that the total number of array tiles in the first group set and the new tile group accurately matches the total number of selections (90+10=100), avoiding exceeding or falling short of the extracted content scale, and fitting the limited carrying capacity of narrowband bandwidth. When performing partial redundancy based on these two sets, the redundant data only includes the high-priority content of the first group set and the supplementary content of the new tile group, ensuring that the core information is fully covered. At the same time, generating redundant data according to the preset number of redundancy times (for example, 3 times) can provide multiple sets of redundancy protection for key content in narrowband weak connection environments. That is, even if packet loss occurs, the receiving device can recover the core information of the transmitted image through these redundant data, effectively preventing complete data loss and balancing redundancy effect with reasonable utilization of bandwidth resources.

[0040] Furthermore, in this embodiment, the aforementioned "aggregating all array tiles selected based on the corresponding selection difference into a new tile group" may further include the following steps: The array of images arranged radially around the center point of the corresponding transmitted image is divided into regions to obtain regions corresponding to different hierarchical orders from the inside to the outside. Obtain the number of all array tiles in the secondary selection group included in each divided region, and calculate the number of regions in each divided region from the inside to the outside to obtain the number of regions superimposed for different superimposed times. If the number of regions corresponding to any number of superpositions is greater than the selection difference, the region difference between the number of regions superpositions and the selection difference is determined, and the region with the largest corresponding hierarchical order for the number of superpositions is determined as the secondary selection region. All array tiles corresponding to the number of superpositions, except those located in the secondary selection area, are aggregated into a new tile group, and all array tiles located in the secondary selection area corresponding to the area difference are aggregated into a new tile group.

[0041] For example, in this embodiment, the selection of all array tiles corresponding to the selection difference based on the secondary selection group can be specifically implemented based on the following technical content: First, the server can divide the array of tiles arranged radially around the center point of the corresponding transmitted image into regions, resulting in different hierarchical regions from the inside out. It can be explained that the center point of the transmitted image is the geometric center of the image. The array of tiles arranged radially around this point generally means that tiles closer to the center are more likely to contain core information of the image (for example, in a portrait image, the central area may correspond to the face; in an image of power grid equipment, the central area may correspond to key components). Therefore, dividing the regions into different hierarchical regions from the inside out (e.g., dividing from the center into level 1, level 2, and level 3 regions) provides a basis for prioritizing the selection of core region tiles, avoiding prioritizing edge tiles during secondary selection, ensuring that redundant data covers high-value information within the secondary selection group as much as possible, and laying the content foundation for information recovery after packet loss. Secondly, the server can obtain the number of array tiles in each division region that are located in the secondary selection group, and calculate the number of regions by superimposing them from the inside out to obtain the superposition number of each region corresponding to different superposition numbers. Here, the number of regions is the total number of array tiles belonging to the secondary selection group in a single division region (for example, a level 1 division region contains 8 array tiles of 8 secondary selection groups, so the number of regions is 8; a level 2 division region contains 6, so the number of regions is 6). By superimposing them from the inside out (for example, first superimposing the 8 level 1 tiles, the number of regions is 8; then superimposing the 6 level 2 tiles, the number of regions is 14), the total number of tiles in each level of the secondary selection group can be quantified in the order of "core region, edge region". This superposition method not only conforms to the logic of "prioritizing the protection of core information", but also allows for an intuitive understanding of the scale of selectable tiles under different superposition numbers through the number of regions superimposed, providing a clear quantitative reference for subsequent matching and selection differences and avoiding blind screening. Then, in response to a situation where the number of overlapping regions corresponding to any number of overlapping events is greater than the selection difference, the server can determine the region difference between the number of overlapping regions and the selection difference, and determine the region with the largest corresponding hierarchical order for that number of overlapping events as the secondary selection region. Here, any number of overlapping events is a specific number of times the region is overlapping from the inside out. When the number of overlapping regions under that number of events is greater than the selection difference (for example, if the selection difference is 10, the number of overlapping regions when overlapping level 1 + level 2 regions is 14 > 10), the overlapping stops and the region difference is calculated (14 - 10 = 4). This region difference is the number of tiles that exceed the selection difference under the current number of overlapping events. The region with the largest hierarchical order is the outermost region under that number of overlapping events (for example, when overlapping level 1 + level 2, the level 2 region is the region with the largest hierarchical order). Determining it as the secondary selection region can clearly identify which tiles need to be selected to accurately match the range of the selection difference, avoiding excessive tiles due to the overall selection of the outer region, while ensuring that the secondary selection region is still a relatively close region to the core, and that the selected tiles have certain information value. Finally, the server can aggregate all array tiles corresponding to the overlay count, excluding those located in the secondary selection region, into a new tile group, and aggregate all array tiles located in the secondary selection region corresponding to the region difference into the new tile group. It can be noted that the total number of regions (8) in other subdivisions besides the secondary selection region (e.g., the level 1 subdivisions under the aforementioned overlay count) does not exceed the selection difference. Aggregating all secondary selection group array tiles in these regions into the new tile group ensures that complete tiles from the core region are prioritized, maximizing the retention of high-value information; simultaneously, from the secondary... Selecting array tiles from the selected region (level 2 region division) with the same number of region differences (4) and summarizing them into the new tile group ensures that the total size of the new tile group accurately matches the selection difference (8+4=10). This avoids over-selection, which wastes narrowband bandwidth, or under-selection, which fails to fill the gap. When performing partial redundancy based on this new tile group and the first group set, the redundant data contains high-priority core content and precisely controls the scale. Even if packet loss occurs, the receiving device can recover the key information of the transmitted image through this redundant data, effectively preventing complete data loss and balancing redundancy effect with the rational use of bandwidth resources.

[0042] It can be noted that, in this embodiment, the partial redundancy based on the newly added tile group and the first group set can be achieved, for example, by adopting the specific method mentioned above: "creating a redundancy template, filling all element image blocks located in the newly added tile group and the first group set into the mapping slots located in the redundancy template that have a redundancy mapping relationship with each element image block, and performing image copying on the obtained redundant image with a corresponding preset number of redundancy."

[0043] Step S4 includes the following: The control upload device sends the transmitted image and the corresponding redundant data of the transmitted image to the receiving device.

[0044] For example, in this embodiment, by controlling the uploading device to send the transmitted image and corresponding redundant data in a unified manner, it is ensured that the two are transmitted to the receiving device synchronously and in a matching manner. This maximizes the anti-packet loss effect of the redundant data in a narrowband weak connection environment and effectively avoids the complete loss of information due to data loss. It can be explained that the redundant data corresponding to the transmitted image is backup data previously generated based on the amount of transmitted data, the redundancy ratio, and the data extraction target of the transmitted image. It is used to compensate for possible packet loss of the transmitted image (e.g., multiple identical images generated based on complete redundancy, or backups of key content generated based on partial redundancy). Clarifying the correspondence between the two ensures that the redundant data sent by the uploading device accurately matches the specific transmitted image, avoiding misalignment between redundant data and the transmitted image. This provides a prerequisite for the receiving device to recover the lost packet content through the redundant data. For example, the redundant data of transmitted image A is only used to compensate for packet loss of image A and will not cause confusion. In the recovery process of image B; simultaneously, in a narrowband weak connection environment, the uploading device may experience disordered autonomous sending behavior due to network fluctuations (such as delayed sending or missing data). Therefore, by actively controlling the sending actions of the uploading device, the timing and content of sending can be forcibly standardized, ensuring that when each transmitted image is sent, its corresponding redundant data is simultaneously transmitted, avoiding asynchronous reception caused by differences in the timing of sending. In a narrowband weak connection environment, even if some packets are lost during the transmission of the image, the receiving device can immediately call the corresponding redundant data received synchronously, and recover the lost packets through the complete information or key content in the redundant data. This avoids the complete loss of data due to the loss of a single transmitted image without matching redundant data to supplement it, achieving the core goal of preventing complete data loss. At the same time, it avoids the problem of low recovery efficiency caused by scattered data reception, improving the reliability of data transmission in a narrowband weak connection environment.

[0045] In summary, according to the solution of this embodiment, by constructing an image redundancy mechanism, this embodiment effectively solves the problem of complete data loss due to packet loss when transmitting images in a narrowband weak connection environment, significantly improving the transmission reliability of distributed heterogeneous device interconnection. Specifically, regarding the accuracy of redundancy allocation, this embodiment first determines the amount of data transmitted by each uploading device for different images at the same time, then performs differentiated redundancy allocation based on the data volume and calculates the redundancy ratio. This breaks the limitation of the uniform redundancy ratio in existing technologies. Higher redundancy can be allocated to images with larger data volumes (avoiding complete data loss due to large data volume and high packet loss probability), while the redundancy ratio is reasonably controlled for images with smaller data volumes (avoiding bandwidth waste). This achieves precise matching of redundancy resources and transmission requirements, maximizing packet loss resistance under the constraint of limited narrowband bandwidth. The redundancy strategy targets… In terms of reliability, this embodiment designs an image redundancy scheme by combining data extraction targets. The redundancy focus can be adjusted according to actual needs to ensure that core image information receives priority redundancy protection. This avoids the impact of packet loss in ordinary areas on the integrity of key data, further enhancing the reliability and practicality of image transmission and meeting the data processing and decision-making needs in different scenarios. In terms of transmission coordination, this embodiment controls the uploading device to send the transmitted image and corresponding redundant data in a unified manner. Compared with the existing technology of transmitting redundant data separately from the original data or relying on retransmission requests, key data delivery can be completed in one go in a narrowband weak connection environment. Even if some packets of the original image are lost, the receiving end can be completed by synchronously transmitting redundant data, completely avoiding the problem of complete image failure due to retransmission request timeout or loss. At the same time, it simplifies the transmission process and reduces the communication overhead of device interconnection.

[0046] Another embodiment of the present invention provides a distributed heterogeneous device interconnection system in a narrowband weak connection environment. Figure 3 Its corresponding system block diagram, such as Figure 3 As shown, the system includes: The transmission determination module is configured to determine all uploading devices that have a device interconnection relationship with the receiving device, and to determine the amount of data transmitted by each uploading device for different transmitted images at the same data transmission time. The redundancy allocation module is configured to perform redundancy allocation for each upload device based on the amount of transmitted data, and determine the redundancy ratio of the transmitted data amount to the corresponding upload device based on the obtained redundancy data amount. The image redundancy module is configured to determine the data extraction target at the corresponding data transmission time, and perform image redundancy on the transmitted image with a corresponding redundancy ratio based on the data extraction target to obtain redundant data corresponding to the transmitted image; The unified sending module is configured to control the uploading device to uniformly send the transmitted image and the corresponding redundant data of the transmitted image to the receiving device.

[0047] In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with the examples of this invention. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing preferred embodiments of the invention.

[0048] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0049] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various aspects of the invention, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof.

[0050] Those skilled in the art will understand that modules, units, or components of the devices disclosed in the examples herein can be arranged in the devices described in this embodiment, or alternatively, can be located in one or more devices different from the devices in this example. The modules in the foregoing examples can be combined into a single module or, in addition, can be divided into multiple sub-modules.

[0051] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components.

[0052] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0053] Furthermore, some of the embodiments described herein are methods or combinations of method elements that can be implemented by a processor of a computer system or by other means of performing the functions. Therefore, a processor having the necessary instructions for implementing the methods or method elements forms means for implementing the methods or method elements. Furthermore, the elements described herein in the apparatus embodiments are examples of means for implementing the functions performed by elements for the purposes of carrying out the invention.

[0054] As used herein, unless otherwise specified, the use of ordinal numbers such as “first,” “second,” “third,” etc., to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects being described must have a given order in time, space, ordering, or any other manner.

[0055] Although the invention has been described with respect to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and edibility purposes, and not for the purpose of explaining or limiting the subject matter of the invention.

Claims

1. A method for interconnecting distributed heterogeneous devices in a narrowband weak connection environment, characterized in that, Includes the following steps: Identify all uploading devices that have device interconnection relationships with the receiving device, and determine the amount of data transmitted by each uploading device for different transmitted images at the same data transmission time; Redundancy allocation is performed for each upload device based on the amount of transmitted data, and the redundancy ratio of the transmitted data to the corresponding upload device is determined based on the obtained redundant data amount. Determine the data extraction target for the corresponding data transmission time, and perform image redundancy with a corresponding redundancy ratio on the transmitted image based on the data extraction target to obtain redundant data corresponding to the transmitted image; The control upload device sends the transmitted image and the corresponding redundant data of the transmitted image to the receiving device. Specifically, determining the data extraction target for the corresponding data transmission time, and performing image redundancy on the transmitted image based on the data extraction target with a corresponding redundancy ratio to obtain redundant data corresponding to the transmitted image, includes: In response to receiving a target extraction signal carrying any warning target at any target reception time, the data extraction targets of all data transmission times corresponding to a preset extraction duration after the target reception time are determined as the warning targets, and the redundancy ratio is numerically compared with the preset redundancy number. In response to the redundancy ratio being greater than or equal to a preset number of redundancy counts, the transmitted image is subjected to data redundancy based on complete redundancy and corresponding to the preset number of redundancy counts to obtain redundant data corresponding to the transmitted image; In response to the redundancy ratio being less than a preset number of redundancies, the ratio of the redundancy ratio to the corresponding preset number of redundancies is calculated to obtain the unit ratio; Image recognition is performed on the transmitted image, and each recognition element located in the transmitted image is determined based on the recognition result; If any identified element is the same as any preset element in the retrieved element priority group, the preset priority corresponding to the preset element will be determined as the element priority of the corresponding identified element. In response to the existence of a corresponding data extraction target identification element in the transmitted image, the element priority corresponding to the identification element is adjusted to the highest element priority, and the element priority corresponding to each remaining identification element is updated; The transmitted image is processed into arrays to obtain array blocks with the same partitioning specifications, and all array blocks that make up the same identification element are identified as a block group of the same element. All element tile groups are sorted from largest to smallest priority based on element priority to obtain a priority sequence, and the total number of groups of all element image tiles included in each element tile group is determined. The total number of arrays corresponding to all array images is multiplied by the percentage of individual units to obtain the total number of selections; Based on the priority sequence, the total number of corresponding groups of element tile groups is superimposed one by one, and based on the comparison between the number of group superpositions corresponding to each superposition number and the selected total number, the transmitted image is subjected to data redundancy based on partial redundancy and corresponding preset redundancy number.

2. The method according to claim 1, characterized in that, Identify all uploading devices that have interconnection relationships with the receiving device, and determine the amount of data transmitted by each uploading device for different transmitted images at the same data transmission time, including: The system receives a receiving instruction carrying a device code sent by the management terminal, determines the receiving device based on the device code, and sorts the device code and the instruction sending time of the corresponding receiving instruction in a corresponding random order. The obtained information sequence is converted based on a hash algorithm, and a distribution identification code corresponding to the receiving device is generated based on the obtained hash string. Control the receiving device to distribute the distribution identification code based on a distributed heterogeneous network, and establish a timed task corresponding to the data distribution; In response to any other device in the corresponding distributed heterogeneous network, the distribution identification code is scanned and identified within the device response period of the corresponding timed task to obtain the device type corresponding to the other device. In response to the fact that the device type is within the preset upload range corresponding to the receiving device, the other device is identified as an upload device that has a device interconnection relationship with the receiving device; Establish a data transmission channel between the receiving device and each uploading device, and determine the amount of data transmitted by each uploading device based on the data transmission channel for different transmitted images at the same data transmission time.

3. The method according to claim 2, characterized in that, The method further includes: If the number of first devices corresponding to all upload devices with device interconnection relationship with the receiving device, as determined by the timed task, is less than the preset number of devices, the first location information corresponding to each upload device is obtained. Based on the first location information, a device communication circle is established with each uploading device as the center. Based on the distributed heterogeneous network, the second location information corresponding to each other device is determined. Based on the second location information, other devices located in the device communication circle are determined as terminal supplementary groups. If the number of second devices corresponding to other devices in all terminal supplementary groups is less than or equal to the number of preset devices corresponding to the number of first devices, the other devices in all terminal supplementary groups are identified as newly added upload devices. The device type and device deviation distance between each other device in the terminal supplement group and the upload device are obtained by responding to the device difference between the second number of other devices in all terminal supplement groups and the first number of devices corresponding to a preset number of devices; Based on the device deviation distance, determine the distance deviation value of the upload device corresponding to each other device, and based on the device type, determine the type deviation value of the preset upload range corresponding to each other device; The distance deviation value and type deviation value corresponding to the same other device are summed. Based on the obtained comprehensive deviation value, all other devices are sorted from smallest to largest terminal. Based on the obtained terminal sequence, other devices with corresponding device differences are identified as newly added upload devices.

4. The method according to claim 1, characterized in that, Redundancy allocation is performed on each upload device based on the amount of data transmitted, and the redundancy ratio of the transmitted data to the corresponding upload device is determined based on the obtained redundant data amount, including: The amount of data transmitted by each uploading device is summed, and the difference between the preset allocated data amount and the obtained total transmission amount is calculated to obtain the redundancy allocation amount. Determine the transmission ratio of each transmitted data volume to the total transmission volume, and calculate the redundant data volume corresponding to each upload device based on the product of the redundancy allocation and each transmission ratio. The redundancy ratio is obtained by calculating the ratio of the amount of redundant data corresponding to the same upload device to the amount of transmitted data.

5. The method according to claim 1, characterized in that, Based on the comparison between the number of group stacks corresponding to each stacking number and the total number of selections, the transmitted image is subjected to data redundancy based on partial redundancy and corresponding to a preset number of redundancy times, including: If the number of group stacks corresponding to the maximum stack count is less than or equal to the total number of selections, a redundant template with the same image specification as the transmitted image is created. The redundant template includes a mapping slot that has a redundant mapping relationship with each array tile. All element image blocks located in each element tile group are filled into the mapping slots located in the redundant template that have a redundant mapping relationship with each element image block, and the resulting redundant image is copied for a corresponding preset number of redundant times.

6. The method according to claim 1, characterized in that, Based on the comparison between the number of group stacks corresponding to each stacking number and the total number of selections, the transmitted image is subjected to data redundancy based on partial redundancy and corresponding to a preset number of redundancy times, including: If the number of group stacks corresponding to any stacking number is greater than the total number of selections, all element tile groups corresponding to stacking numbers before that stacking number are determined as the first group set, and all remaining element tile groups are determined as the second group set based on the priority sequence; Determine the selection difference between the number of group superpositions corresponding to the first group set and the total number of selections, and determine the element tile group that is first in the second group set as the secondary selection group based on the priority sequence; All array tiles selected based on the selection difference of the secondary selection group are aggregated into the new tile group, and data redundancy is performed based on the first group set and the new tile group, with a corresponding preset number of redundancy times.

7. The method according to claim 6, characterized in that, All array tiles selected based on the corresponding selection difference based on the secondary selection group are aggregated into a new tile group, including: The array of images arranged radially around the center point of the corresponding transmitted image is divided into regions to obtain regions corresponding to different hierarchical orders from the inside to the outside. Obtain the number of all array tiles in the secondary selection group included in each divided region, and calculate the number of regions in each divided region from the inside to the outside to obtain the number of regions superimposed for different superimposed times. If the number of regions corresponding to any number of superpositions is greater than the selection difference, the region difference between the number of regions superpositions and the selection difference is determined, and the region with the largest corresponding hierarchical order for the number of superpositions is determined as the secondary selection region. All array tiles corresponding to the number of superpositions, except those located in the secondary selection area, are aggregated into a new tile group, and all array tiles located in the secondary selection area corresponding to the area difference are aggregated into a new tile group.

8. A distributed heterogeneous device interworking system in a narrowband weak connection environment, characterized by, include: The transmission determination module is configured to determine all uploading devices that have a device interconnection relationship with the receiving device, and to determine the amount of data transmitted by each uploading device for different transmitted images at the same data transmission time. The redundancy allocation module is configured to perform redundancy allocation for each upload device based on the amount of transmitted data, and determine the redundancy ratio of the transmitted data amount to the corresponding upload device based on the obtained redundancy data amount. The image redundancy module is configured to determine the data extraction target at the corresponding data transmission time, and perform image redundancy on the transmitted image with a corresponding redundancy ratio based on the data extraction target to obtain redundant data corresponding to the transmitted image; The unified sending module is configured to control the uploading device to send the transmitted image and the corresponding redundant data of the transmitted image to the receiving device in a unified manner; Specifically, determining the data extraction target for the corresponding data transmission time, and performing image redundancy on the transmitted image based on the data extraction target with a corresponding redundancy ratio to obtain redundant data corresponding to the transmitted image, includes: In response to receiving a target extraction signal carrying any warning target at any target reception time, the data extraction targets of all data transmission times corresponding to a preset extraction duration after the target reception time are determined as the warning targets, and the redundancy ratio is numerically compared with the preset redundancy number. In response to the redundancy ratio being greater than or equal to a preset number of redundancy counts, the transmitted image is subjected to data redundancy based on complete redundancy and corresponding to the preset number of redundancy counts to obtain redundant data corresponding to the transmitted image; In response to the redundancy ratio being less than a preset number of redundancies, the ratio of the redundancy ratio to the corresponding preset number of redundancies is calculated to obtain the unit ratio; Image recognition is performed on the transmitted image, and each recognition element located in the transmitted image is determined based on the recognition result; If any identified element is the same as any preset element in the retrieved element priority group, the preset priority corresponding to the preset element will be determined as the element priority of the corresponding identified element. In response to the existence of a corresponding data extraction target identification element in the transmitted image, the element priority corresponding to the identification element is adjusted to the highest element priority, and the element priority corresponding to each remaining identification element is updated; The transmitted image is processed into arrays to obtain array blocks with the same partitioning specifications, and all array blocks that make up the same identification element are identified as a block group of the same element. All element tile groups are sorted from largest to smallest priority based on element priority to obtain a priority sequence, and the total number of groups of all element image tiles included in each element tile group is determined. The total number of arrays corresponding to all array images is multiplied by the percentage of individual units to obtain the total number of selections; Based on the priority sequence, the total number of corresponding groups of element tile groups is superimposed one by one, and based on the comparison between the number of group superpositions corresponding to each superposition number and the selected total number, the transmitted image is subjected to data redundancy based on partial redundancy and corresponding preset redundancy number.

Images