System and method for generating standardized pictures based on multi-source intelligent terminal thermal infrared image acquisition

Abstract

The application provides a system and method for generating standardized pictures based on multi-source intelligent terminals, wherein the system is provided with a task scheduling and communication module, a file receiving module, an SDK dynamic loading and adaptation module, a temperature data extraction module and a DL / T664 standard picture generation module; the method steps implemented by the system are as follows: step one, the task scheduling and communication module issues a collection task to a designated intelligent terminal; step two, the file receiving module receives an infrared image file uploaded by the terminal; step three, the SDK dynamic loading and adaptation module automatically selects and loads a corresponding SDK library according to the file source manufacturer; step four, the temperature data extraction module extracts an original temperature matrix and metadata by using the loaded SDK; step five, the DL / T664 standard picture generation module renders and generates a standard-compliant picture by using the extracted data; and step six, the final standard picture is output and stored, and is connected to a subsequent analysis and diagnosis platform.

Description

Technical Field

[0001] This invention relates to the field of power equipment condition monitoring and image processing technology, specifically to a system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals. Background Technology

[0002] In the daily inspection and condition monitoring of power systems, infrared thermal imaging technology is an indispensable detection method that can effectively detect potential faults such as equipment overheating and defects. The State Grid Corporation of China's DL / T664-2016 "Application Specification for Infrared Diagnosis of Live Equipment" puts forward clear requirements for the format of infrared thermal images, temperature data, analysis and annotation information to ensure the accuracy and comparability of diagnostic results.

[0003] Currently, the market offers a variety of smart terminals and drones, handheld thermal imagers, and infrared cameras from different manufacturers, such as DJI, FLIR, Testo, and Hikvision, all of which can be used to collect infrared data. However, these devices suffer from the following problems: 1) Inconsistent data formats: The thermal infrared image formats output by devices from different manufacturers, such as JPEG, R-JPG, and TIFF, vary, as do their internal temperature data storage methods. Some images are even composites of visible light and infrared images, lacking original, extractable temperature matrix data; 2) Heterogeneous SDK interfaces: The software development kits (SDKs) provided by different manufacturers are different, making it complex to integrate and call multiple SDK technologies in a single system. The system is difficult to maintain and expand due to several drawbacks. 3) Lack of standardized processing: Images directly acquired from terminals typically do not conform to the DL / T664 standard, requiring manual post-processing using specialized software to add temperature scales, hotspot annotations, equipment information, etc. This process is cumbersome, inefficient, and prone to human error. Therefore, based on these requirements, the applicant proposes a system and method for generating standardized images from thermal infrared images acquired from multi-source intelligent terminals. This system is compatible with various intelligent terminals, automatically acquires raw temperature data, and generates standardized DL / T664 infrared thermal images with a single click, eliminating data barriers and achieving standardized and automated processing of inspection data. This effectively improves the automation level of power inspections and enhances the accuracy of data analysis. Summary of the Invention

[0004] To address the aforementioned technical issues, this invention proposes a system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals. The system comprises a task scheduling and communication module, a file receiving module, an SDK dynamic loading and adaptation module, a temperature data extraction module, and a DL / T664 standard image generation module. The method implemented by the system involves the task scheduling and communication module issuing acquisition tasks to designated intelligent terminals; the file receiving module receiving infrared image files uploaded by the terminals; the SDK dynamic loading and adaptation module automatically selecting and loading the corresponding SDK library based on the file's source manufacturer; the temperature data extraction module extracting the original temperature matrix and metadata using the loaded SDK; and the DL / T664 standard image generation module using the extracted data to render and generate standard-compliant images, outputting and storing the final standard image, and connecting it to a subsequent analysis and diagnostic platform. This eliminates data barriers, achieves standardized and automated processing of inspection data, and effectively improves the automation level of power inspection.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals, characterized in that: the system for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals includes a task scheduling and communication module, a file receiving module, an SDK dynamic loading and adaptation module, a temperature data extraction module, and a DL / T664 standard image generation module; the task scheduling and communication module constructs standard TCP packets according to the State Grid Document No. 57 protocol and sends image acquisition task instructions to intelligent terminals from different manufacturers that have been connected to the network; the file receiving module receives the original thermal infrared image files uploaded by the intelligent terminals after inspection through the FTPS file transfer protocol, ensuring the security of data transmission; the SDK dynamic loading and adaptation module uses Java... JNA technology dynamically loads native SDK libraries from different vendors. The module includes an adapter layer that shields the API differences between different SDKs and provides a unified calling interface. The temperature data extraction module calls the unified interface of the SDK dynamic loading and adaptation module to parse and extract the most critical raw temperature matrix data and metadata from raw image files of different formats. The DL / T664 standard image generation module receives the extracted temperature data and metadata from the image. The module first performs a set of temperature feature extraction and statistical analysis algorithms on the raw two-dimensional temperature matrix data to obtain advanced diagnostic information. Then, according to the requirements of the DL / T664 specification, combined with the diagnostic information, it automatically generates a standardized infrared thermogram.

[0007] Furthermore, the temperature data extraction module of the system setting for generating standardized images based on multi-source intelligent terminal thermal infrared image acquisition executes the following processes through algorithm output: generating color bars with temperature scales, automatically marking identified abnormal overheated areas, adding device information and measurement parameter information, and outputting an image format that meets the specifications. The temperature feature extraction and statistical analysis algorithm of the temperature data extraction module includes:

[0008] 1) Global statistics calculation: Traverse the two-dimensional temperature matrix and calculate its global maximum temperature value, global minimum temperature value and global average temperature value;

[0009] 2) Region of interest division; Based on a preset device template or temperature gradient, at least one region of interest is defined in the temperature matrix;

[0010] 3) Calculation of regional statistics; Within each region of interest, calculate its highest temperature, lowest temperature, and average temperature.

[0011] 4) Calculation and diagnosis of critical temperature difference: Based on global and regional statistics, the critical temperature difference is calculated and compared with a preset safety threshold to achieve automated initial screening of overheating defects.

[0012] Furthermore, the method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals is as follows:

[0013] Step 1: The task scheduling and communication module sends the data collection task to the designated smart terminal;

[0014] Step 2: The file receiving module receives the infrared image file uploaded by the terminal;

[0015] Step 3: The SDK dynamic loading and adaptation module automatically selects and loads the corresponding SDK library based on the file source manufacturer;

[0016] Step 4: The temperature data extraction module uses the loaded SDK to extract the original temperature matrix and metadata;

[0017] Step 5: The DL / T664 standard image generation module uses the extracted data to render and generate an image that conforms to the standard.

[0018] Step 6: Output and store the final standard image, and connect it to the subsequent analysis and diagnostic platform.

[0019] The benefits of this application are:

[0020] 1. A system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals. By using JNA technology and adapter design pattern, the system perfectly solves the problem of heterogeneous integration of SDKs from multiple manufacturers. One system can process data from multiple terminals, reducing development and maintenance costs.

[0021] 2. The system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals realizes full automation from task assignment to standard image generation, without the need for manual intervention in data conversion and format processing, which greatly improves work efficiency;

[0022] 3. The system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals directly reads raw temperature data through the official SDKs of various manufacturers, ensuring the authenticity and accuracy of the data, laying the foundation for subsequent accurate diagnosis, and ensuring the fidelity and accuracy of the data.

[0023] 4. A system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals forces the output of unified DL / T664 standard images, ensuring the comparability of data collected at different times, with different devices, and by different personnel, thereby improving the reliability of power equipment condition assessment.

[0024] 5. The system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals realizes the intelligence and automation of the diagnostic process; through the built-in temperature feature extraction and statistical analysis algorithm, the system can automatically output key temperature statistics and preliminary diagnostic conclusions, which greatly reduces the workload of manual analysis and reduces the dependence on the professional experience of operators.

[0025] 6. The system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals improve the accuracy and objectivity of condition assessment; its precise calculation based on full matrix data avoids the problems of strong subjectivity and easy omission of details in human eye observation, making temperature readings, temperature rise calculations and equipment condition judgments more accurate and objective.

[0026] 7. The system and method for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals enhance the system's real-time processing capabilities; the algorithm is optimized for large-scale two-dimensional temperature arrays, has high computational efficiency, and can meet the needs of rapid and real-time analysis of a large number of devices in power inspection. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the overall system architecture of the present invention;

[0028] Figure 2 This is a schematic diagram of the steps in the system implementation method of the present invention;

[0029] Figure 3 This is a schematic diagram illustrating the workflow of the dynamic loading and adaptation module of the SDK of this invention. Detailed Implementation

[0030] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0031] like Figure 1-3 The diagram illustrates a system and method for generating standardized images from thermal infrared images acquired by multi-source intelligent terminals. The system includes a task scheduling and communication module, a file receiving module, an SDK dynamic loading and adaptation module, a temperature data extraction module, and a DL / T664 standard image generation module. The task scheduling and communication module constructs standard TCP packets according to the State Grid Document No. 57 protocol and sends image acquisition task instructions to intelligent terminals from different manufacturers that are already connected to the network. The file receiving module receives the original thermal infrared image files uploaded by the intelligent terminals after inspection using the FTPS file transfer protocol, ensuring the security of data transmission. The SDK dynamic loading and adaptation module uses Java... JNA technology dynamically loads native SDK libraries from different vendors. The module includes an adapter layer that shields the API differences between different SDKs and provides a unified calling interface. The temperature data extraction module calls the unified interface of the SDK dynamic loading and adaptation module to parse and extract the most critical raw temperature matrix data and metadata from raw image files of different formats. The DL / T664 standard image generation module receives the extracted temperature data and metadata from the image. The module first performs a set of temperature feature extraction and statistical analysis algorithms on the raw two-dimensional temperature matrix data to obtain advanced diagnostic information. Then, according to the requirements of the DL / T664 specification, combined with the diagnostic information, it automatically generates a standardized infrared thermogram.

[0032] The temperature data extraction module of the system setup shown, which generates standardized images based on thermal infrared image acquisition from multi-source intelligent terminals, outputs results through algorithms. It performs a standardization process including generating color bars with temperature scales, automatically marking identified abnormal overheated areas, adding device information and measurement parameter information, and outputting an image in a format that meets specifications. The temperature feature extraction and statistical analysis algorithms of the temperature data extraction module include:

[0033] 1) Global statistics calculation: Traverse the given two-dimensional temperature matrix and calculate its global maximum temperature value, global minimum temperature value, and global average temperature value;

[0034] 2) Region of Interest (ROI) Delineation: Based on a preset device template or temperature gradient, delineate at least one ROI within the shown temperature matrix.

[0035] 3) Calculation of regional statistics; For each shown region of interest, calculate its highest temperature, lowest temperature, and average temperature.

[0036] 4) Calculation and diagnosis of critical temperature difference: Based on global and regional statistics, the critical temperature difference is calculated and compared with the preset safety threshold to achieve automated initial screening of overheating defects.

[0037] The implementation process of the system for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals is as follows:

[0038] Step 1: The task scheduling and communication module issues a data collection task to the designated smart terminal; the system generates a TCP message conforming to the protocol of "State Grid Document No. 57" according to the inspection plan; the message content may include: task ID, terminal ID, latitude and longitude of the inspection point, and shooting parameters, such as whether visible light is superimposed; the message is sent to the designated smart terminal through a Socket connection.

[0039] Step 2: The file receiving module receives the infrared image files uploaded by the terminal; an FTPS server is set up on the server, and an independent directory and authentication credentials are assigned to each terminal; after the smart terminal completes the task, it automatically uploads the collected files, such as DJI_XXXX.jpg and FLIR_XXXX.tiff, to their corresponding FTPS directory; the system obtains the files through the file path carried by the infrared smart terminal in the patrol results report.

[0040] Step 3: The SDK dynamic loading and adaptation module automatically selects and loads the corresponding SDK library based on the file source manufacturer. This module, as the core processing unit of the system, is responsible for interacting with the native software development kits of various manufacturers. Internally, it constructs a unified adapter interface to shield the differences in calling methods, function naming, and parameter passing between SDKs from different manufacturers. When the system needs to extract temperature data from a thermal infrared image file, this module begins its work. First, the system determines the device manufacturer identifier corresponding to the file based on the access terminal source information, such as the filename, metadata, or preset configuration. Subsequently, the adapter dynamically selects and loads the corresponding SDK library based on this manufacturer identifier. The adapter loads the corresponding manufacturer's native SDK library. This process is achieved through Java native access technology, which allows Java programs to seamlessly call native functions located in the native dynamic link library. Specifically, after determining the target manufacturer, such as DJI, the adapter will call functions specifically encapsulated for that manufacturer through a pre-declared interface. These methods include, but are not limited to: parsing and returning a temperature data matrix representing the temperature value of the entire image from the specified image file path; and obtaining the physical parameters set when capturing the image, such as emissivity. After successfully obtaining the above raw data through the native SDK, the adapter will perform a data standardization process. It will encapsulate the data, which may have different formats obtained from different manufacturers' SDKs, into a unified standard data structure defined internally by the system. This structure contains complete physical quantity information such as temperature matrix, emissivity, and ambient temperature. Finally, this standardized data object containing all the core temperature information is returned to the upstream calling module of the system, thus providing an accurate and unified data foundation for the subsequent generation of infrared thermal images conforming to the DL / T664 standard.

[0041] For devices from other manufacturers, such as FLIR, the system adopts the exact same workflow and interface specifications. Only the corresponding calling logic needs to be implemented for that manufacturer inside the adapter. This fully demonstrates the good scalability and compatibility of the design. The system determines the source manufacturer's SDKAdapter based on the metadata or configuration table, uses JNA to dynamically load the corresponding manufacturer's SDK library, and calls its native functions to extract the original temperature matrix and other key parameters. The original temperature matrix is ​​a two-dimensional array of type double, with each pixel corresponding to a temperature value.

[0042] Step 4: The temperature data extraction module uses the loaded SDK to extract the original temperature matrix and metadata; the implementation details of the temperature feature extraction algorithm are as follows:

[0043] 1) In the DL / T664 data generation module, the temperature matrix is ​​set to a two-dimensional array T[m][n] consisting of M rows and N columns of floating-point numbers, where 0 ≤ m < M, 0 ≤ n < N, and each element represents the Celsius temperature value of the corresponding pixel on the image;

[0044] 2) Global statistical feature extraction:

[0045] Maximum temperature extraction: Initialize a variable T_max to the minimum possible value, iterate through each element of the entire matrix T using a double loop, compare and update it with T_max, and finally obtain the global maximum temperature value;

[0046] Minimum temperature extraction: Initialize a variable T_min to the maximum possible value, and then obtain the global minimum temperature value by iterating and comparing.

[0047] Average temperature calculation: Summate all M × N temperature values, then divide by the total number of pixels to obtain the global average temperature T_avg. In the specific implementation, to prevent accumulation overflow, a segmented or incremental averaging algorithm can be used;

[0048] 3) Region of Interest (ROI) segmentation and feature extraction; To achieve more refined analysis, the system supports defining one or more ROIs on the temperature matrix based on preset device templates or image segmentation algorithms, such as edge detection or threshold segmentation; Let a ROI_k be defined by a set of pixel coordinates, then:

[0049] Extraction of the highest / lowest temperature in a region: Perform a similar traversal comparison operation as in step one only within the pixel set corresponding to ROI_k to obtain T_max_k and T_min_k of the region;

[0050] Calculate the average temperature of the region: sum the temperature values ​​of all pixels in the ROI_k region and divide by the number of pixels in that region to get T_avg_k;

[0051] 4) Calculation and generation of critical temperature differences; the system automatically calculates a series of critical temperature differences, such as:

[0052] ΔT_critical = T_max_k - T_avg, the temperature difference between a hot spot area and the overall average;

[0053] ΔT_relative = T_max_k1 - T_avg_k2, the relative temperature difference between different regions;

[0054] The calculated ΔT_critical is compared with the safe operating thresholds stored in the database. If the threshold is exceeded, the system will automatically add a prominent diagnostic mark, such as a flashing red crosshair, to the abnormal point when generating the DL / T664 standard image. A warning record will also be generated in the diagnostic report attached to the image, such as: "The temperature of the B phase clamp is abnormal, with a temperature rise of 45K, which exceeds the standard value. It is recommended to deal with it in time."

[0055] Step 5: The DL / T664 standard image generation module uses the extracted data to render and generate a standard-compliant image. It maps the extracted temperature matrix data to pixel colors (e.g., iron red, rainbow) according to a set color band, and draws the main thermal image. A color bar corresponding to the temperature value is generated on one side of the image. The temperature matrix is ​​traversed to find the highest and lowest temperatures globally or regionally, automatically marking them on the image with crosshairs or dots, and displaying their specific temperature values ​​and temperature differences. In the reserved blank areas of the image, usually at the bottom or side, information such as the device name, number, inspection conditions (ambient temperature, humidity, distance, emissivity), inspection personnel, date, and time is filled in text form. Some of this information is obtained from metadata, while some needs to be obtained from inspection tasks or manually entered. For devices that do not support direct extraction of the temperature matrix, the system can be configured to a downgraded processing mode, that is, only adding DL / T664 standard annotation information to the uploaded ordinary infrared images, but still achieving some standardization functions.

[0056] Step 6: Output and store the final standard image, and connect it to the subsequent analysis and diagnostic platform. The image output is in standard JPEG or PNG format and stored in a database or file system for subsequent analysis system calls.

[0057] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any modifications or equivalent changes made based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A system for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals, characterized in that: The system for generating standardized images based on thermal infrared image acquisition from multi-source intelligent terminals includes a task scheduling and communication module, a file receiving module, an SDK dynamic loading and adaptation module, a temperature data extraction module, and a DL / T664 standard image generation module. The task scheduling and communication module constructs standard TCP packets according to the State Grid Document No. 57 protocol and sends image acquisition task instructions to intelligent terminals from different manufacturers that have been connected to the network. The file receiving module receives the raw thermal infrared image files uploaded by the smart terminal after inspection via the FTPS file transfer protocol, ensuring the security of data transmission. The SDK dynamic loading and adaptation module dynamically loads native SDK libraries provided by different manufacturers using Java JNA technology. The module includes an adapter layer to shield the API differences between different SDKs and provide a unified calling interface. The temperature data extraction module calls the unified interface of the SDK dynamic loading and adaptation module to parse and extract the most critical raw temperature matrix data and metadata from raw image files of different formats. The DL / T664 standard image generation module receives the extracted temperature data and metadata. The module first performs a set of temperature feature extraction and statistical analysis algorithms on the raw two-dimensional temperature matrix data to obtain advanced diagnostic information. Then, according to the requirements of the DL / T664 specification, combined with the diagnostic information, it automatically generates a standardized infrared thermal image.

2. The system for generating standardized images based on thermal infrared image acquisition using a multi-source intelligent terminal according to claim 1, characterized in that: include: 1) Global statistics calculation: Traverse the two-dimensional temperature matrix and calculate its global maximum temperature value, global minimum temperature value and global average temperature value; 2) Region of interest division; Based on a preset device template or temperature gradient, at least one region of interest is defined in the temperature matrix; 3) Calculation of regional statistics; Within each region of interest, calculate its highest temperature, lowest temperature, and average temperature. 4) Calculation and diagnosis of key temperature differences; Based on global and regional statistics, a critical temperature difference is calculated and compared with a preset safety threshold to achieve automated initial screening of overheating defects.

3. The method for generating standardized images based on thermal infrared image acquisition using a multi-source intelligent terminal according to claim 1, characterized in that: The method is implemented as follows: Step 1: The task scheduling and communication module sends the data collection task to the designated smart terminal; Step 2: The file receiving module receives the infrared image file uploaded by the terminal; Step 3: The SDK dynamic loading and adaptation module automatically selects and loads the corresponding SDK library based on the file source manufacturer; Step 4: The temperature data extraction module uses the loaded SDK to extract the original temperature matrix and metadata; Step 5: The DL / T664 standard image generation module uses the extracted data to render and generate an image that conforms to the standard. Step 6: Output and store the final standard image, and connect it to the subsequent analysis and diagnostic platform.

Images