Fan fire early warning method, device and equipment based on multi-source video data fusion and storage medium

Abstract

This application discloses a method, apparatus, device, and storage medium for wind turbine fire early warning based on multi-source video data fusion, relating to the field of image data fusion technology. The application includes: when the video data of the wind turbine exhibits fire early warning characteristics, determining the fire early warning location based on the video data; determining multiple target monitoring devices based on the fire early warning location and extracting video data from the multiple target monitoring devices; fusing the video data with the video data from the multiple target monitoring devices, using the fire early warning location as the center point, to obtain fused video data; performing fire early warning detection on the fused video data to obtain the fire detection result; and when the fire detection result is a preset fire detection result, outputting the fire location based on the fire detection result, thereby achieving fire early warning detection of the wind turbine from multiple angles and ensuring the accuracy of the fire early warning detection.

Description

Technical Field

[0001] This application relates to the field of image data fusion technology, and in particular to a method, apparatus, equipment and storage medium for early warning of wind turbine fires based on multi-source video data fusion. Background Technology

[0002] Wind turbine fires are a prominent issue in wind power accidents, with large wind turbines being particularly prone to fires. Lightning strikes and high-temperature operation are key factors that can cause fires in electrical components and wiring. Currently, smoke detectors are used in wind power systems to detect smoke and determine if a fire has occurred within the wind turbine. However, this method is limited by the smoke detector's ability to judge smoke concentration, and by the time smoke is produced, an ignition source has already been identified. By then, the warning has already caused damage to the wind turbine equipment, severely impacting its normal operation. Therefore, current fire warning methods cannot meet the production needs of wind turbines and cannot quickly and accurately locate potential fire sources.

[0003] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this application is to provide a method, device, equipment, and storage medium for early warning of wind turbine fires based on multi-source video data fusion, aiming to solve the technical problem that existing technologies cannot intervene in advance when a fire occurs.

[0005] To achieve the above objectives, this application proposes a wind turbine fire early warning method based on multi-source video data fusion, wherein the multi-source video data fusion wind turbine fire early warning method includes:

[0006] When the video data of the wind turbine shows fire warning characteristics, the location of the fire warning is determined based on the video data;

[0007] Multiple target monitoring devices are determined based on the fire warning location, and video data of the multiple target monitoring devices is extracted, wherein the video data of the target monitoring devices includes the fire warning location;

[0008] Using the fire warning location as the center point, the video data is fused with the video data from multiple target monitoring devices to obtain fused video data;

[0009] Fire early warning detection is performed on the fused video data to obtain the fire detection result. When the fire detection result is a preset fire detection result, the fire location is output according to the fire detection result.

[0010] In one embodiment, the step of determining the location of a fire warning based on the video data when the video data of the wind turbine unit contains fire warning characteristics includes:

[0011] Feature recognition is performed on the video data of the wind turbine to determine the temperature feature map of the wind turbine in the video data;

[0012] Determine the temperature gradient of the temperature characteristic map of the wind turbine generator, and when the temperature gradient is greater than a preset temperature gradient, determine the temperature change vector based on the temperature gradient;

[0013] The temperature change vectors are aggregated to obtain the vector center, which is then used as the location for fire early warning.

[0014] In one embodiment, the step of determining multiple target monitoring devices based on the fire warning location and extracting video data from the multiple target monitoring devices, wherein the video data of the target monitoring devices includes the fire warning location, includes:

[0015] Determine the spatial coordinates of the fire warning location based on the fire warning location;

[0016] Obtain a map of monitoring locations of the wind turbine monitoring equipment and the monitoring area of ​​each of the wind turbine monitoring equipment;

[0017] The candidate wind turbine monitoring equipment is determined based on the spatial location coordinates and the monitoring point map of the wind turbine monitoring equipment.

[0018] Multiple target wind turbine monitoring devices are determined based on the monitoring area of ​​the candidate wind turbine monitoring devices and the spatial location coordinates.

[0019] Video data from multiple target wind turbine monitoring devices are extracted.

[0020] In one embodiment, the step of fusing the video data with the video data of multiple target monitoring devices, using the fire warning location as the center point, to obtain fused video data includes:

[0021] Determine the fire warning location and set the fire warning location as the center point;

[0022] The fire warning locations in the video data of multiple target monitoring devices are determined respectively, and the fire warning locations in the video data of the target monitoring devices are determined as fusion reference points;

[0023] The edge features of the center point are determined, and the edge features of the center point are used as the fusion base.

[0024] Determine the edge features of the fusion reference point and align the fusion reference point with the center point;

[0025] Based on the fusion substrate, the edge features of the fusion point are fused with the edge features of the center point to obtain fused video data.

[0026] In one embodiment, the step of fusing the edge features of the fusion point with the edge features of the center point to obtain fused video data includes:

[0027] Determine similar features between the edge features of the fusion point and the edge features of the center point, wherein the similar features exist simultaneously in both the edge features of the fusion point and the edge features of the center point;

[0028] The fusion reference point and the center point are fused in an orthogonal position based on the video frame to obtain an orthogonally fused video.

[0029] The feature offset is determined by comparing the edge features of the fusion point with the edge features of the center point in the positive-position fused video. The feature offset is the feature offset data of the edge features of the fusion point with respect to the edge features of the center point.

[0030] The positive-position fused video is offset-calibrated based on the characteristic offset to obtain fused video data.

[0031] In one embodiment, the step of offset calibration of the orthogonal fused video based on the feature offset to obtain fused video data includes:

[0032] Determine the offset distance and offset angle of the feature offset;

[0033] Using the offset angle as the first calibration parameter, the edge features of the fusion point are rotated and offset.

[0034] After completing the rotation offset, the offset direction and the offset distance can be updated based on the edge features of the fusion point after the rotation offset and the edge features of the center point, and the offset vector can be obtained according to the offset direction and the offset distance;

[0035] The edge features of the fusion point are adjusted using the offset vector so that the edge features of the fusion point coincide with the edge features of the center point, thereby obtaining fused video data.

[0036] In one embodiment, the step of performing fire early warning detection on the fused video data to obtain the fire detection result, and outputting the fire location based on the fire detection result when the fire detection result is a preset fire detection result, includes:

[0037] Fire early warning detection is performed on the fused video data to determine the infrared detection data in the fused video data and to determine the infrared imaging image corresponding to the fused video data.

[0038] Abnormal high temperature points are determined based on the temperature thresholds of each component of the wind turbine and the infrared imaging image.

[0039] Fire detection results are output based on the abnormal high temperature points;

[0040] When the fire detection result is a preset fire detection result, the fire location is output according to the fire detection result.

[0041] Furthermore, to achieve the above objectives, this application also proposes a wind turbine fire early warning device based on multi-source video data fusion, wherein the multi-source video data fusion wind turbine fire early warning device includes:

[0042] The location early warning module is used to determine the fire early warning location based on the video data when the video data of the wind turbine unit shows fire early warning characteristics;

[0043] The video extraction module is used to determine multiple target monitoring devices based on the fire warning location and extract video data from the multiple target monitoring devices, wherein the video data of the target monitoring devices includes the fire warning location;

[0044] The video fusion module is used to fuse the video data with the video data of multiple target monitoring devices, with the fire warning location as the center point, to obtain fused video data;

[0045] The fire early warning module is used to perform fire early warning detection on the fused video data, obtain the fire detection result, and output the fire location according to the fire detection result when the fire detection result is a preset fire detection result.

[0046] Furthermore, to achieve the above objectives, this application also proposes a wind turbine fire early warning device based on multi-source video data fusion. The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The computer program is configured to implement the steps of the wind turbine fire early warning method based on multi-source video data fusion as described above.

[0047] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the wind turbine fire early warning method based on multi-source video data fusion as described above.

[0048] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the wind turbine fire early warning method based on multi-source video data fusion as described above.

[0049] One or more technical solutions proposed in this application have at least the following technical effects: when the video data of the wind turbine has fire warning characteristics, the fire warning location is determined based on the video data; multiple target monitoring devices are determined based on the fire warning location; video data of the multiple target monitoring devices is extracted; the video data is fused with the video data of the multiple target monitoring devices with the fire warning location as the center point to obtain fused video data; fire warning detection is performed on the fused video data to obtain the fire detection result; when the fire detection result is a preset fire detection result, the fire location is output based on the fire detection result, thereby realizing fire warning detection of the wind turbine from multiple angles and ensuring the accuracy of fire warning detection. Attached Figure Description

[0050] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0051] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart illustrating an embodiment of the wind turbine fire early warning method based on multi-source video data fusion in this application.

[0053] Figure 2 This is a schematic diagram of infrared imaging of a wind turbine generator provided in an embodiment of the wind turbine fire early warning method based on multi-source video data fusion of this application.

[0054] Figure 3 This is a schematic diagram of the center point location provided in an embodiment of the wind turbine fire early warning method based on multi-source video data fusion of this application.

[0055] Figure 4 This is a schematic diagram illustrating the determination of offset distance and offset angle in an embodiment of the wind turbine fire early warning method based on multi-source video data fusion of this application.

[0056] Figure 5 This is a schematic diagram of the module structure of the wind turbine fire early warning device based on multi-source video data fusion according to an embodiment of this application;

[0057] Figure 6 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the wind turbine fire early warning method based on multi-source video data fusion in the embodiments of this application.

[0058] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0059] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0060] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific embodiments.

[0061] The main solution of this application embodiment is as follows: when the video data of the wind turbine has fire warning characteristics, the fire warning location is determined based on the video data; multiple target monitoring devices are determined based on the fire warning location, and video data of the multiple target monitoring devices is extracted, wherein the video data of the target monitoring devices includes the fire warning location; with the fire warning location as the center point, the video data is fused with the video data of the multiple target monitoring devices to obtain fused video data; fire warning detection is performed on the fused video data to obtain the fire detection result; when the fire detection result is a preset fire detection result, the fire location is output based on the fire detection result.

[0062] In this embodiment, for ease of description, the following description will focus on a wind turbine fire early warning device that identifies multi-source video data fusion.

[0063] Current fire early warning technology typically involves installing smoke detectors in fire-prone locations. These detectors are triggered when smoke is generated during a fire. However, this approach, which focuses on the fire-to-warning process, involves intervention after the fire has already started and combustion has begun, thus preventing further escalation. In other words, current fire early warning systems only issue warnings after the fire has already occurred. Furthermore, to avoid false alarms, smoke detectors set thresholds for smoke concentration, only triggering an alarm when the concentration exceeds these thresholds. Consequently, situations can arise where a fire occurs but no warning is issued. Therefore, it is clear that current fire early warning systems cannot intervene before a fire occurs.

[0064] This application provides a solution that can identify and output potentially dangerous locations that may be subject to fire before a fire actually occurs, and implement relevant fire early warning measures to reduce the probability of fire and improve the operational safety of wind turbine units.

[0065] As can be seen from the above embodiments, this application determines the fire warning location based on the video data when the video data of the wind turbine has fire warning characteristics, identifies multiple target monitoring devices based on the fire warning location, extracts the video data of the multiple target monitoring devices, and performs video fusion with the video data of the multiple target monitoring devices using the fire warning location as the center point to obtain fused video data. Fire warning detection is then performed on the fused video data to obtain the fire detection result. When the fire detection result is a preset fire detection result, the fire location is output based on the fire detection result. This achieves fire warning detection of the wind turbine from multiple angles, ensuring the accuracy of fire warning detection.

[0066] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions, such as a wind turbine fire early warning device with multi-source video data fusion. The following description uses a wind turbine fire early warning device with multi-source video data fusion as an example to illustrate this embodiment and the subsequent embodiments.

[0067] Based on this, embodiments of this application provide a wind turbine fire early warning method based on multi-source video data fusion, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the wind turbine fire early warning method based on multi-source video data fusion according to this application.

[0068] In this embodiment, the wind turbine fire early warning method based on multi-source video data fusion includes steps S10 to S40:

[0069] Step S10: When the video data of the wind turbine unit shows fire warning characteristics, determine the fire warning location based on the video data.

[0070] It should be noted that a wind turbine is a device that converts wind energy into mechanical energy, and then into electrical energy through a generator. During the operation of a wind turbine, various factors may lead to a fire, especially mechanical or electrical faults. According to the principles of fire occurrence, the temperature at the ignition point will rise abnormally before a fire occurs. Therefore, local temperature anomalies can be used as a fire warning feature, and the point of local temperature anomaly is the fire warning location, indicating that a fire may occur at the current location.

[0071] In practical implementation, for fire early warning of wind turbine units, to cope with the occurrence of fires, multiple-angle video monitoring is conducted on various areas of the wind turbine unit. To ensure data accuracy, at least two operational monitoring devices are present at any location on the wind turbine unit for video surveillance. The video monitoring includes temperature information from various locations within the wind turbine unit, outputting the temperature characteristics of the wind turbine unit in the form of infrared imaging. The fire early warning feature is a localized high-temperature point. When the temperature characteristic detection in the wind turbine unit's video data identifies a localized high-temperature point, this identified localized high-temperature point can be designated as a fire early warning location with fire risk, and the current fire early warning location can be output.

[0072] In one feasible implementation, the step of determining the fire warning location based on the video data when the video data of the wind turbine unit contains fire warning characteristics includes:

[0073] Feature recognition is performed on the video data of the wind turbine to determine the temperature feature map of the wind turbine in the video data;

[0074] Determine the temperature gradient of the temperature characteristic map of the wind turbine generator, and when the temperature gradient is greater than a preset temperature gradient, determine the temperature change vector based on the temperature gradient;

[0075] The temperature change vectors are aggregated to obtain the vector center, which is then used as the location for fire early warning.

[0076] It should be noted that temperature feature maps refer to images captured by infrared thermal imaging technology, reflecting the temperature distribution on an object's surface. In these images, different temperatures represent different temperatures; typically, blue represents low temperature and red represents high temperature. Temperature gradient refers to the rate of temperature change on an object's surface. A larger temperature gradient indicates a faster rate of temperature change and reflects a significant difference between the current temperature and adjacent temperatures, suggesting a localized high-temperature point. Accordingly, to describe the degree of temperature change, the temperature gradient is transformed into a temperature change vector. This vector describes both the quantity and direction of temperature change. Furthermore, the combined effect of multiple temperature change vectors allows for the determination of the vector center, which is generally located in the same area as the fire warning location, exhibiting consistency between the vector center and the fire warning location.

[0077] In practical implementation, before performing feature recognition on the video data of wind turbine generators, preprocessing operations are required, including noise reduction and contrast enhancement, to improve the accuracy of subsequent feature recognition. After preprocessing, infrared imaging data can be extracted from the video data to output a temperature feature map. Simultaneously, edge detection and region growing are performed on the video data to separate the wind turbine generator or its components from the background, extracting the individual wind turbine generator or its components. This extracted components are then fused with the temperature feature map to obtain the wind turbine generator's temperature feature map. When determining the temperature feature map of the wind turbine generator, such as... Figure 2 As shown, Figure 2 This is a schematic diagram of infrared imaging of a wind turbine. When obtaining the temperature feature map, different color zones can be determined based on the color partitions of the infrared imaging corresponding to the temperature conditions. Each color zone is defined as a temperature range. Temperature differences exist between different temperature ranges, and these temperature differences determine the vector direction. In this embodiment, the vector direction is from the high-temperature range to the low-temperature range. Therefore, the temperature gradient is determined based on the temperature feature map. The temperature gradient represents the degree of temperature change, and the degree of temperature change is positively correlated with the temperature distribution. When there is a sudden change in temperature, the temperature gradient is larger. Therefore, when detecting the temperature gradient, if the temperature gradient is greater than a preset temperature gradient, the temperature change vector can be determined based on the temperature gradient. The temperature vectors are aggregated, and an aggregation region is formed based on the tail of the temperature vector. This aggregation region is the vector center. The area corresponding to this vector center is a high-temperature area, forming a local high-temperature point compared to adjacent areas. Therefore, this vector center can be determined as a fire warning location.

[0078] Step S20: Determine multiple target monitoring devices based on the fire warning location, and extract video data from the multiple target monitoring devices, wherein the video data of the target monitoring devices includes the fire warning location.

[0079] It should be noted that the monitoring equipment is used for video monitoring of wind turbine units, and can acquire monitoring videos of wind turbine units, while the target monitoring equipment is a monitoring device among many monitoring devices that can monitor the location of fire warnings.

[0080] In specific implementation, when a fire warning location is determined, the system can identify target monitoring devices that can detect the current fire warning location and extract the monitoring video recorded by the corresponding target monitoring devices. When identifying target monitoring devices, the system can determine the spatial coordinates of the fire warning location based on the fire warning location; obtain the monitoring location map of the wind turbine monitoring devices and the monitoring area of ​​each wind turbine monitoring device; determine candidate wind turbine monitoring devices based on the spatial coordinates and the monitoring location map; determine multiple target wind turbine monitoring devices based on the monitoring area of ​​the candidate wind turbine monitoring devices and the spatial coordinates; and extract video data from the multiple target wind turbine monitoring devices.

[0081] When deploying monitoring equipment, a 3D model of the wind turbine's environment and the turbine itself can be created, establishing a 3D spatial coordinate system. The spatial coordinates within this system point to a unique location in space. The monitoring range and location of each device are recorded, creating a monitoring point map of the wind turbine monitoring equipment. Therefore, upon detecting a fire warning location, the spatial coordinates of the fire can be determined. Simultaneously, the monitoring point map and monitoring areas of each wind turbine monitoring device are acquired. By aligning these monitoring areas with the fire warning location, the area containing the fire warning location can be identified as the target monitoring area. The corresponding target monitoring equipment can then be locked, and its video data can be extracted.

[0082] Step S30: Using the fire warning location as the center point, the video data is fused with the video data of multiple target monitoring devices to obtain fused video data.

[0083] It should be noted that the center point refers to the reference point in the video fusion process. It is used to indicate that the fire warning location in the image where the fire warning location is determined is the main reference point and is set as the center point. In other words, the subsequent video fusion process needs to use the center point as a reference for video fusion, and the data after fusing multiple video data is called fused video data.

[0084] In practical implementation, the fire warning location is used as the center point for video fusion, and this center point serves as the reference point for video fusion. Firstly, during the video fusion process, the timestamp data of multiple target monitoring devices is used to perform time calibration, unifying the timelines of each video data point. This ensures that videos from the same moment are fused, avoiding monitoring errors caused by video data asynchrony due to video latency. After unifying the time information, a suitable video fusion algorithm can be selected. For example, video frames from different cameras can be stitched together to form a panoramic view, or image blending technology can be used to smoothly transition between different video frames. The fused video is then optimized, including eliminating stitching seams, adjusting color consistency, and enhancing details to improve the naturalness and usability of the fused video, resulting in fused video data.

[0085] In one feasible implementation, the step of fusing the video data with the video data of multiple target monitoring devices, using the fire warning location as the center point, to obtain fused video data includes:

[0086] Determine the fire warning location and set the fire warning location as the center point;

[0087] The fire warning locations in the video data of multiple target monitoring devices are determined respectively, and the fire warning locations in the video data of the target monitoring devices are determined as fusion reference points;

[0088] The edge features of the center point are determined, and the edge features of the center point are used as the fusion base.

[0089] Determine the edge features of the fusion reference point and align the fusion reference point with the center point;

[0090] Based on the fusion substrate, the edge features of the fusion point are fused with the edge features of the center point to obtain fused video data.

[0091] It should be noted that the fusion reference point is the fire warning location in the target monitoring equipment, the fusion base refers to the reference point in the video fusion process, the center point edge features and the fusion point edge features are the object features near the center point and the object features near the fusion reference point, respectively.

[0092] In the specific implementation, the first step is to determine the image at the location where the fire warning is detected, identify the fire warning location, and then define the position of the fire warning location in the image as the center point, referring to... Figure 3 , Figure 3This is a schematic diagram showing the location of the center point. It's important to note that this center point is not the actual center of the monitored video, but rather the fusion center during the subsequent fusion process. Therefore, as shown... Figure 3 As shown, this can represent the edge region of the image. Simultaneously, it's possible to determine the fire warning location in the video data of other target monitoring devices. Due to the angle of the monitoring equipment, there is an angular deviation between the video data of the target monitoring device and the video data showing the fire warning location. Since the fire warning location is the same in all video data, it is used as a reference point to fuse multiple video data sets.

[0093] During the fusion process, the edge features of the center point can be determined. The edge strength of the center point can be determined according to system presets, either by extending from fixed pixels or by determining the contour of the identified object; this embodiment does not impose any limitations. When the edge features of the center point are determined, the edge features of the intermediate point can be used as reference features to guide the fusion of the edge features of the fusion point. At this point, the fusion reference point and the center point can be overlapped first, and then the edge features of the fusion point and the edge features of the center point can be fused. This may involve video rotation and scaling to obtain fused video data.

[0094] In one feasible implementation, the step of fusing the edge features of the fusion point with the edge features of the center point to obtain fused video data includes:

[0095] Determine similar features between the edge features of the fusion point and the edge features of the center point, wherein the similar features exist simultaneously in both the edge features of the fusion point and the edge features of the center point;

[0096] The fusion reference point and the center point are fused in an orthogonal position based on the video frame to obtain an orthogonally fused video.

[0097] The feature offset is determined by comparing the edge features of the fusion point with the edge features of the center point in the positive-position fused video. The feature offset is the feature offset data of the edge features of the fusion point with respect to the edge features of the center point.

[0098] The positive-position fused video is offset-calibrated based on the characteristic offset to obtain fused video data.

[0099] In the specific implementation, a similarity analysis is first performed on the edge features of the fusion point and the center point to identify the edge features they share. Edge detection is then performed on the edge features of the fusion point and the center point using the Canny algorithm, yielding the edge features of the fusion point and the center point respectively. These edge features are then vectorized to facilitate similarity calculation. When determining the edge features of the fusion point and the center point, the formula can be used:

[0100]

[0101] Where, μ x and μ y These are the average brightness of the edge features at the fusion point and the edge features at the center point, respectively. It is the variance, σ xy C1 and C2 are covariances, and C1 and C2 are constants used to maintain stability.

[0102] Subsequently, using the video frame as a reference, the fusion reference point and the center point are precisely fused to generate an orthogonal fused video. Next, the feature offsets of the fusion point edge features and the center point edge features in the orthogonal fused video are calculated to identify feature points that simultaneously exist in both the fusion point edge features and the center point edge features, and their feature offsets are calculated using the following formula:

[0103]

[0104] Where P1 = (x1, y1) and P2 = (x2, y2) are the coordinates of the feature points.

[0105] This offset reflects the displacement of the edge features at the fusion point relative to the edge features at the center point. Based on the calculated feature offset, offset calibration is performed on the orthogonal fused video to obtain the calibrated fused video data.

[0106] In one feasible implementation, the step of offset calibration of the orthogonal fused video based on the feature offset to obtain fused video data includes:

[0107] Determine the offset distance and offset angle of the feature offset;

[0108] Using the offset angle as the first calibration parameter, the edge features of the fusion point are rotated and offset.

[0109] After completing the rotation offset, the offset direction and the offset distance can be updated based on the edge features of the fusion point after the rotation offset and the edge features of the center point, and the offset vector can be obtained according to the offset direction and the offset distance;

[0110] The edge features of the fusion point are adjusted using the offset vector so that the edge features of the fusion point coincide with the edge features of the center point, thereby obtaining fused video data.

[0111] In practical implementation, when determining the feature offsets of feature points in the center point edge features and the fusion point edge features, the offset distance and offset angle within the feature offsets can be determined. The offset distance refers to the distance deviation and angle of the feature point between the center point edge features and the fusion point edge features. Since the center point and the fusion point coincide during fusion, it is possible to radiate outwards from the center point. Therefore, when determining the offset distance and offset angle, reference is made to... Figure 4 , Figure 4 A schematic diagram is provided to determine the offset distance and offset angle. When determining the offset angle, clockwise and counter-clockwise directions can be used as positive angles. For ease of description, the offset angle α can be set to (-180°, 180°), where L1 to L4 represent the distances of the feature points from the center point. The circle and rhombus represent two pairs of feature points. Taking the circular feature point as an example, first, the edge feature points of the center point and the edge feature points of the fusion point are determined. Starting from the edge feature point of the center point, the position of the corresponding edge feature point of the fusion point is determined, and its relative position to the edge feature point of the center point is determined. In this diagram, the edge feature point of the fusion point is located counter-clockwise from the edge feature point of the center point; therefore, the offset angle is negative. Simultaneously, the distance between the edge feature point of the center point and the center point can be determined as L1, and the distance between the edge feature point of the fusion point and the center point as L2. When determining the offset distance, the starting point is also the edge feature point of the center point, with the direction closer to the center point being negative and the direction farther from the center point being positive. Therefore, when determining the offset distance L, it can be calculated using L = L1 - L2. Using the offset angle as the first calibration parameter, and -α as the calibration parameter, the edge features of the fusion point are rotated and offset. After the offset is completed, the offset vector is obtained based on the offset distance and offset angle, and the offset vector is used to adjust the fusion edge features so that they coincide with the center point edge features, thus obtaining the fused video data.

[0112] Step S40: Perform fire early warning detection on the fused video data to obtain the fire detection result. When the fire detection result is a preset fire detection result, output the fire location according to the fire detection result.

[0113] In the specific implementation, fire early warning detection is performed on the fused video data to identify infrared detection data and corresponding infrared imaging images. Based on these infrared imaging images, the temperature thresholds of each component in the wind turbine and abnormal high-temperature points are determined. It's important to note that a high temperature in a certain area, but not an extreme one, does not necessarily mean the current temperature is normal. To determine if a point is an abnormal high-temperature point, the temperature threshold of the component at its current location must be determined based on its location and characteristics. The current temperature is then compared with this threshold to identify abnormal high-temperature points in the image. After identifying these abnormal high-temperature points, a fire detection result is output. When evaluating the fire detection result, comparing it with preset fire detection results indicates the location of a potential fire. Therefore, the fire location can be determined based on the fire detection results, allowing personnel to respond proactively and perform shutdown, maintenance, and other operations.

[0114] This embodiment provides a wind turbine fire early warning method based on multi-source video data fusion. When fire early warning characteristics are present in the video data of the wind turbine, the fire early warning location is determined based on the video data. Multiple target monitoring devices are identified based on the fire early warning location, and video data from these devices is extracted. Using the fire early warning location as the center point, the video data is fused with the video data from the multiple target monitoring devices to obtain fused video data. Fire early warning detection is performed on the fused video data to obtain the fire detection result. When the fire detection result matches a preset fire detection result, the fire location is output based on the fire detection result. This method enables fire early warning detection of the wind turbine from multiple angles, ensuring the accuracy of the fire early warning detection.

[0115] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the wind turbine fire early warning method based on multi-source video data fusion in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0116] This application also provides a wind turbine fire early warning device based on multi-source video data fusion; please refer to... Figure 5 The multi-source video data fusion wind turbine fire early warning device includes:

[0117] The location early warning module 10 is used to determine the fire early warning location based on the video data when the video data of the wind turbine unit contains fire early warning characteristics.

[0118] The video extraction module 20 is used to determine multiple target monitoring devices based on the fire warning location and extract video data from the multiple target monitoring devices, wherein the video data of the target monitoring devices includes the fire warning location;

[0119] The video fusion module 30 is used to fuse the video data with the video data of multiple target monitoring devices, with the fire warning location as the center point, to obtain fused video data;

[0120] The fire early warning module 40 is used to perform fire early warning detection on the fused video data, obtain the fire detection result, and output the fire location according to the fire detection result when the fire detection result is a preset fire detection result.

[0121] In one embodiment, the location early warning module 10 is further configured to perform feature recognition on the video data of the wind turbine, determine the temperature feature map of the wind turbine in the video data; determine the temperature gradient of the temperature feature map of the wind turbine, and when the temperature gradient is greater than a preset temperature gradient, determine the temperature change vector based on the temperature gradient; aggregate the temperature change vector to obtain the vector center, and determine the vector center as the fire early warning location.

[0122] In one embodiment, the video extraction module 20 is further configured to: determine the spatial coordinates of the fire warning location based on the fire warning location; acquire a monitoring location map of the wind turbine monitoring equipment and the monitoring area of ​​each wind turbine monitoring equipment; determine candidate wind turbine monitoring equipment based on the spatial coordinates and the monitoring location map of the wind turbine monitoring equipment; determine multiple target wind turbine monitoring equipment based on the monitoring area of ​​the candidate wind turbine monitoring equipment and the spatial coordinates; and extract video data from the multiple target wind turbine monitoring equipment.

[0123] In one embodiment, the video fusion module 30 is further configured to: determine the fire warning location and define the fire warning location as a center point; determine the fire warning location in the video data of multiple target monitoring devices and define the fire warning location in the video data of the target monitoring devices as a fusion reference point; determine the edge features of the center point and define the edge features of the center point as a fusion base; determine the fusion point edge features of the fusion reference point and make the fusion reference point coincide with the center point; and fuse the fusion point edge features with the center point edge features based on the fusion base to obtain fused video data.

[0124] In one embodiment, the video fusion module 30 is further configured to: determine similar features between the edge features of the fusion point and the edge features of the center point, wherein the similar features exist simultaneously in both the edge features of the fusion point and the edge features of the center point; perform orthogonal fusion of the fusion reference point and the center point based on the video frame to obtain an orthogonal fused video; determine the feature offset between the edge features of the fusion point and the edge features of the center point in the orthogonal fused video, wherein the feature offset is the feature offset data of the edge features of the fusion point relative to the edge features of the center point; and perform offset calibration on the orthogonal fused video based on the feature offset to obtain fused video data.

[0125] In one embodiment, the video fusion module 30 is further configured to determine the offset distance and offset angle of the feature offset; rotate and offset the edge features of the fusion point using the offset angle as a first calibration parameter; after completing the rotation and offset, update the offset direction and the offset distance based on the rotated and offset edge features of the fusion point and the center point edge features, and obtain an offset vector based on the offset direction and the offset distance; adjust the edge features of the fusion point using the offset vector so that the edge features of the fusion point coincide with the edge features of the center point, thereby obtaining fused video data.

[0126] In one embodiment, the fire early warning module 40 is further configured to perform fire early warning detection on the fused video data, determine infrared detection data in the fused video data, determine the infrared imaging image corresponding to the fused video data; determine abnormal high temperature points based on the temperature thresholds of each component of the wind turbine and the infrared imaging image; output fire detection results according to the abnormal high temperature points; and output the fire location according to the fire detection results when the fire detection results are preset fire detection results.

[0127] The wind turbine fire early warning device based on multi-source video data fusion provided in this application, employing the multi-source video data fusion wind turbine fire early warning method described in the above embodiments, can solve the technical problem in the prior art that it is impossible to intervene in advance when a fire occurs. Compared with the prior art, the beneficial effects of the wind turbine fire early warning device based on multi-source video data fusion provided in this application are the same as those of the multi-source video data fusion wind turbine fire early warning method described in the above embodiments, and other technical features in the wind turbine fire early warning device based on multi-source video data fusion are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0128] This application provides a wind turbine fire early warning device based on multi-source video data fusion. The wind turbine fire early warning device based on multi-source video data fusion includes: at least one processor; and a memory communicatively connected to at least one processor; wherein the memory stores instructions executable by at least one processor, and the instructions are executed by at least one processor to enable at least one processor to execute the wind turbine fire early warning method based on multi-source video data fusion in the above embodiment 1.

[0129] The following is for reference. Figure 6 This document illustrates a structural schematic diagram of a wind turbine fire early warning device suitable for implementing multi-source video data fusion in the embodiments of this application. The multi-source video data fusion wind turbine fire early warning device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 6 The wind turbine fire early warning device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0130] like Figure 6As shown, the multi-source video data fusion wind turbine fire early warning device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the multi-source video data fusion wind turbine fire early warning device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the multi-source video data fusion wind turbine fire warning equipment to wirelessly or wiredly communicate with other devices to exchange data. Although the figure shows a multi-source video data fusion wind turbine fire warning equipment with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented or possessed alternatively.

[0131] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0132] The wind turbine fire early warning device based on multi-source video data fusion provided in this application, employing the multi-source video data fusion wind turbine fire early warning method described in the above embodiments, can solve the technical problem in the prior art that it is impossible to intervene in advance when a fire occurs. Compared with the prior art, the beneficial effects of the wind turbine fire early warning device based on multi-source video data fusion provided in this application are the same as those of the wind turbine fire early warning method based on multi-source video data fusion provided in the above embodiments, and other technical features in this wind turbine fire early warning device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0133] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0134] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0135] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the wind turbine fire early warning method of multi-source video data fusion in the above embodiments.

[0136] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0137] The aforementioned computer-readable storage medium may be included in the multi-source video data fusion wind turbine fire early warning device; or it may exist independently and not be assembled into the multi-source video data fusion wind turbine fire early warning device.

[0138] The aforementioned computer-readable storage medium carries one or more programs that, when executed by the multi-source video data fusion fan fire early warning device, enable the multi-source video data fusion fan fire early warning device to:

[0139] When the video data of the wind turbine shows fire warning characteristics, the location of the fire warning is determined based on the video data;

[0140] Multiple target monitoring devices are determined based on the fire warning location, and video data of the multiple target monitoring devices is extracted, wherein the video data of the target monitoring devices includes the fire warning location;

[0141] Using the fire warning location as the center point, the video data is fused with the video data from multiple target monitoring devices to obtain fused video data;

[0142] Fire early warning detection is performed on the fused video data to obtain the fire detection result. When the fire detection result is a preset fire detection result, the fire location is output according to the fire detection result.

[0143] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0144] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0145] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0146] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned multi-source video data fusion wind turbine fire early warning method. This solves the technical problem in the prior art that it is impossible to intervene in advance when a fire occurs. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the multi-source video data fusion wind turbine fire early warning method provided in the above embodiments, and will not be repeated here.

[0147] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the wind turbine fire early warning method as described above, which involves multi-source video data fusion.

[0148] The computer program product provided in this application can solve the technical problem that existing technologies cannot intervene in advance when a fire occurs. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the multi-source video data fusion wind turbine fire early warning method provided in the above embodiments, and will not be repeated here.

[0149] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for early warning of wind turbine fires using multi-source video data fusion, characterized in that, The wind turbine fire early warning method based on multi-source video data fusion includes: When the video data of the wind turbine shows fire warning characteristics, the location of the fire warning is determined based on the video data; Multiple target monitoring devices are determined based on the fire warning location, and video data of the multiple target monitoring devices is extracted, wherein the video data of the target monitoring devices includes the fire warning location; Using the fire warning location as the center point, the video data is fused with the video data from multiple target monitoring devices to obtain fused video data; Fire early warning detection is performed on the fused video data to obtain fire detection results. When the fire detection results are the preset fire detection results, the fire location is output according to the fire detection results. The step of fusing the video data with the video data from multiple target monitoring devices, using the fire warning location as the center point, to obtain fused video data includes: Determine the fire warning location and set the fire warning location as the center point; The fire warning locations in the video data of multiple target monitoring devices are determined respectively, and the fire warning locations in the video data of the target monitoring devices are determined as fusion reference points; The edge features of the center point are determined, and the edge features of the center point are used as the fusion base. Determine the edge features of the fusion reference point and align the fusion reference point with the center point; Based on the fusion base, similar features between the edge features of the fusion point and the edge features of the center point are determined, and the similar features exist simultaneously in the edge features of the fusion point and the edge features of the center point; The fusion reference point and the center point are fused in an orthogonal position based on the video frame to obtain an orthogonally fused video. The feature offset is determined by comparing the edge features of the fusion point with the edge features of the center point in the positive-position fused video. The feature offset is the feature offset data of the edge features of the fusion point with respect to the edge features of the center point. The positive-position fused video is offset-calibrated based on the characteristic offset to obtain fused video data.

2. The method as described in claim 1, characterized in that, When the video data of the wind turbine generator shows fire warning characteristics, the step of determining the location of the fire warning based on the video data includes: Feature recognition is performed on the video data of the wind turbine to determine the temperature feature map of the wind turbine in the video data; Determine the temperature gradient of the temperature characteristic map of the wind turbine generator, and when the temperature gradient is greater than a preset temperature gradient, determine the temperature change vector based on the temperature gradient; The temperature change vectors are aggregated to obtain the vector center, which is then used as the location for fire early warning.

3. The method as described in claim 1, characterized in that, The step of determining multiple target monitoring devices based on the fire warning location and extracting video data from the multiple target monitoring devices, wherein the video data of the target monitoring devices includes the fire warning location, includes: Determine the spatial coordinates of the fire warning location based on the fire warning location; Obtain a map of monitoring locations of the wind turbine monitoring equipment and the monitoring area of ​​each of the wind turbine monitoring equipment; The candidate wind turbine monitoring equipment is determined based on the spatial location coordinates and the monitoring point map of the wind turbine monitoring equipment. Multiple target wind turbine monitoring devices are determined based on the monitoring area of ​​the candidate wind turbine monitoring devices and the spatial location coordinates. Video data from multiple target wind turbine monitoring devices are extracted.

4. The method as described in claim 1, characterized in that, The step of offset calibration of the orthogonal fused video based on the feature offset to obtain fused video data includes: Determine the offset distance and offset angle of the feature offset; Using the offset angle as the first calibration parameter, the edge features of the fusion point are rotated and offset. After completing the rotation offset, the offset direction and the offset distance can be updated based on the edge features of the fusion point after the rotation offset and the edge features of the center point, and the offset vector can be obtained according to the offset direction and the offset distance; The edge features of the fusion point are adjusted using the offset vector so that the edge features of the fusion point coincide with the edge features of the center point, thereby obtaining fused video data.

5. The method as described in claim 1, characterized in that, The step of performing fire early warning detection on the fused video data to obtain the fire detection result, and outputting the fire location based on the fire detection result when the fire detection result is a preset fire detection result, includes: Fire early warning detection is performed on the fused video data to determine the infrared detection data in the fused video data and to determine the infrared imaging image corresponding to the fused video data. Abnormal high temperature points are determined based on the temperature thresholds of each component of the wind turbine and the infrared imaging image. Fire detection results are output based on the abnormal high temperature points; When the fire detection result is a preset fire detection result, the fire location is output according to the fire detection result.

6. A wind turbine fire early warning device based on multi-source video data fusion, characterized in that, The device includes: The location early warning module is used to determine the fire early warning location based on the video data when the video data of the wind turbine unit shows fire early warning characteristics; The video extraction module is used to determine multiple target monitoring devices based on the fire warning location and extract video data from the multiple target monitoring devices, wherein the video data of the target monitoring devices includes the fire warning location; The video fusion module is used to fuse the video data with the video data of multiple target monitoring devices, with the fire warning location as the center point, to obtain fused video data; The fire early warning module is used to perform fire early warning detection on the fused video data, obtain fire detection results, and output the fire location according to the fire detection results when the fire detection results are preset fire detection results. The step of fusing the video data with the video data from multiple target monitoring devices, using the fire warning location as the center point, to obtain fused video data includes: Determine the fire warning location and set the fire warning location as the center point; The fire warning locations in the video data of multiple target monitoring devices are determined respectively, and the fire warning locations in the video data of the target monitoring devices are determined as fusion reference points; The edge features of the center point are determined, and the edge features of the center point are used as the fusion base. Determine the edge features of the fusion reference point and align the fusion reference point with the center point; Based on the fusion base, similar features between the edge features of the fusion point and the edge features of the center point are determined, and the similar features exist simultaneously in the edge features of the fusion point and the edge features of the center point; The fusion reference point and the center point are fused in an orthogonal position based on the video frame to obtain an orthogonally fused video. The feature offset is determined by comparing the edge features of the fusion point with the edge features of the center point in the positive-position fused video. The feature offset is the feature offset data of the edge features of the fusion point with respect to the edge features of the center point. The positive-position fused video is offset-calibrated based on the characteristic offset to obtain fused video data.

7. A wind turbine fire early warning device based on multi-source video data fusion, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the wind turbine fire early warning method based on multi-source video data fusion as described in any one of claims 1 to 5.

8. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the wind turbine fire early warning method based on multi-source video data fusion as described in any one of claims 1 to 5.

Images