Fire-fighting heating and ventilation collaborative data processing method and system

Abstract

The invention provides a fire-fighting heating and ventilation collaborative data processing method and system. The method comprises the following steps: determining an acquisition state of offshore equipment as a foggy state based on an offshore acquisition ship, and carrying out image division on an initial acquisition image based on the offshore equipment to obtain each initial area with acquisition conditions; carrying out image recognition on each initial region, and if any initial region is determined not to have agglomerate fog, determining the initial region as a sea fog region; if it is determined that any initial region has an agglomerate fog region whose corresponding agglomerate fog attribute is a smoke attribute, determining the initial region as a smoke region, and determining each remaining initial region as a target region; and performing video identification on an area acquisition video which is acquired based on the offshore acquisition ship and corresponds to the target area, determining a current diffusion velocity corresponding to an agglomerate fog area located in the target area, and determining an agglomerate fog attribute corresponding to the agglomerate fog area based on the current diffusion velocity. According to the invention, the determination accuracy of the offshore smoke and the sea fog is at least improved.

Description

Technical Field

[0001] This invention relates to data processing technology, and more particularly to a method and system for collaborative data processing of fire protection and HVAC systems. Background Technology

[0002] As the core carriers of marine resource development and shipping, the operational safety of offshore equipment is directly related to the safety of personnel, the marine environment, and property. Fire monitoring is a crucial aspect of the daily operation and maintenance of offshore equipment; however, the unique meteorological environment at sea presents significant challenges to fire monitoring. The smoke produced during a fire is visually similar to sea fog, both appearing as blurred areas formed by the aggregation of suspended particles in the air. Both reduce visibility in the monitoring field. Mistaking sea fog for fire smoke can trigger unnecessary emergency responses and increase maintenance costs; conversely, mistaking fire smoke for sea fog can delay fire response and exacerbate disaster losses.

[0003] To address the challenge of distinguishing between smoke and sea fog in marine environments, existing technologies primarily focus on research and applications based on differences in optical properties and physical property analysis. In terms of optical property monitoring, some technologies utilize dual-spectral imaging systems (visible and infrared) to differentiate between smoke and sea fog by leveraging the differences in their infrared radiation characteristics. Smoke, due to the high temperatures generated by combustion, exhibits significant thermal radiation signals in infrared images, while sea fog, with temperatures close to seawater or ambient temperatures, displays weaker thermal radiation signals. Other technologies measure the light scattering and absorption coefficients of aerosol particles, using the differences in the scattering and absorption capabilities of smoke particles (mostly carbonaceous matter and combustion residue) and sea fog particles (mainly water vapor condensates) to construct optical characteristic parameter models for classification. Regarding physical property analysis, existing technologies deploy temperature and humidity sensors, barometric pressure sensors, and other equipment. By comparing the characteristics of smoke generation, such as sudden temperature increases and localized humidity changes, with the relatively stable temperature and humidity conditions in sea fog environments, these technologies aid in the differentiation between the two.

[0004] Although existing technologies have made some progress in distinguishing between smoke and sea fog, they have not yet been able to accurately distinguish between sea fog and smoke. Existing technologies are prone to misjudgment, mistaking sea fog for smoke and triggering invalid warnings, or mistaking smoke for sea fog and causing fire hazards to be overlooked. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide a fire protection and heating ventilation collaborative data processing method and system that overcomes or at least partially solves the above problems.

[0006] According to one aspect of the present invention, a method for collaborative data processing of fire protection and HVAC systems is provided, comprising the following steps: If the collection state corresponding to any marine equipment is determined to be foggy based on the marine collection vessel, the initial collection image obtained by the marine collection vessel is divided based on the marine equipment to obtain each initial region with collection conditions. For each initial region, perform image recognition. If, based on the recognition results, it is determined that no fog patches exist in any initial region, the initial region is defined as a sea fog region. If, based on the identification results, it is determined that any initial region contains a fog region with a corresponding fog attribute of smoke, then the initial region is determined as a smoke region, and each of the remaining initial regions is determined as a target region. Video recognition is performed on the area video collected by the marine collection vessel corresponding to the target area. Based on the recognition results, the current diffusion speed corresponding to the fog patch area located in the target area is determined, and the fog patch attribute corresponding to the fog patch area is determined based on the current diffusion speed.

[0007] Optionally, in the method according to the present invention, if the collection state corresponding to any marine equipment is determined to be foggy based on the marine collection vessel, the initial collection image acquired based on the marine collection vessel is divided into image segments based on the marine equipment to obtain initial regions with collection conditions, including: When it is determined that the temperature value collected by the temperature sensor of any marine equipment is greater than the preset temperature threshold, the collection point with a collection distance from the marine equipment is determined based on the GPS positioning information of the marine equipment. The control vessel at sea acquires images based on the acquisition point orientation towards the marine equipment, thereby obtaining an initial acquired image that has an acquisition correlation with the marine equipment; Image recognition is performed on the initial acquired image, and the acquisition status of the marine equipment that has an acquisition correlation with the initial acquired image is determined based on the recognition results. The acquisition status includes foggy status and fog-free status. If the acquisition state of the marine equipment that is associated with the initial acquisition image is determined to be foggy, the initial acquisition image is divided based on each marine equipment to obtain each initial region with acquisition conditions.

[0008] Optionally, in the method according to the present invention, image recognition is performed on the initial acquired image, and the acquisition status of the marine equipment having an acquisition association relationship with the initial acquired image is determined based on the recognition result, wherein the acquisition status includes a foggy state and a fog-free state, including: Retrieve preset equipment images corresponding to marine equipment that have a data acquisition relationship with the initially acquired images; Image recognition is performed on the preset device image and the initial acquired image to determine the image overlap between the preset device image and the initial acquired image; The image overlap is compared with the preset overlap. If the comparison result is that the image overlap is less than the preset overlap, the acquisition status of the marine equipment that has an acquisition correlation with the initial acquired image is determined to be a foggy state. Conversely, the acquisition status of marine equipment that has an acquisition correlation with the initial acquired image is determined to be a fog-free state.

[0009] Optionally, in the method according to the present invention, if the acquisition state corresponding to the marine equipment that has an acquisition association with the initial acquired image is determined to be a foggy state, the initial acquired image is divided based on each marine equipment to obtain each initial region with acquisition conditions, including: If it is determined that the acquisition state of the marine equipment that has an acquisition association with the initial acquisition image is foggy, image recognition is performed on the initial acquisition image to determine the equipment area corresponding to each marine equipment in the initial acquisition image. The initial acquired image is processed into a grid to obtain each grid region formed by the obtained horizontal grid lines and vertical grid lines. The grid area that does not include any device area is defined as the initial area with the acquisition conditions, and each remaining grid area is defined as the included area; Determine the area proportion of each device area that contains the region, and compare the area proportion with the retrieved preset proportion threshold. The regions whose proportions are less than the preset proportion threshold will be updated to the initial regions with the conditions for data collection.

[0010] Optionally, in the method according to the present invention, image recognition is performed on each initial region, and if it is determined based on the recognition result that no fog patches exist in any initial region, the initial region is determined as a sea fog region, including: Perform image recognition on each initial region to obtain the image pixel value corresponding to each image pixel that makes up each initial region. The difference between the image pixel values ​​corresponding to each adjacent image pixel in the same initial region is calculated to obtain the pixel difference, and the pixel difference is compared with the preset difference threshold. If no comparison result for the same initial region has a pixel difference greater than a preset difference threshold, it is determined that the initial region does not contain fog, and the initial region is identified as a sea fog region.

[0011] Optionally, in the method according to the invention, the method further includes: If there are image pixels in the same initial region whose pixel difference is greater than the preset difference threshold, then the image pixels whose pixel difference is greater than the preset difference threshold are divided into pixel groups. The image pixel with the maximum pixel value in each pixel group is determined as the target pixel, and the target pixels corresponding to the same initial region are connected to obtain the pixel contour. In the initial region, determine the pixel region corresponding to each pixel contour, and obtain the area of ​​the region corresponding to each pixel region; The area of ​​each region is compared with the preset area threshold. If there is no pixel region in the same initial region whose area is greater than the preset area threshold, it is determined that there is no fog in the initial region and the initial region is determined to be a sea fog region.

[0012] Optionally, in the method according to the present invention, if it is determined based on the identification result that any initial region contains a fog region with a corresponding fog attribute of smoke, the initial region is determined as a smoke region, and each remaining initial region is determined as a target region, including: If any pixel region exists within the same initial region and the comparison result shows that the region area is greater than a preset area threshold, then the pixel region is determined as the comparison region. The average value of each image pixel corresponding to each image pixel point that makes up each comparison area is calculated, and the obtained average pixel value is compared with the preset smoke pixel value respectively. The comparison area corresponding to the pixel mean value of the comparison result being greater than the preset smoke pixel value is determined as the fog area with the corresponding fog attribute being smoke attribute; The initial region that includes any patch of fog with the corresponding fog attribute being smoke is defined as the smoke region, and each of the remaining initial regions is defined as the target region.

[0013] Optionally, in the method according to the present invention, video recognition is performed on the area video collected by the marine collection vessel corresponding to the target area, the current diffusion rate corresponding to the fog patch area located in the target area is determined based on the recognition result, and the fog patch attribute corresponding to the fog patch area is determined based on the current diffusion rate, including: The control vessel at sea performs video acquisition based on the acquisition point orientation towards the equipment at sea, acquires each monitoring image frame that makes up the acquired monitoring video, and performs image cropping on each monitoring image frame based on the target area to obtain each acquired image frame that makes up the regional acquisition video corresponding to the target area. The fog region corresponding to the fog attribute of each acquired image frame is identified as smoke, and the fog area of ​​each fog region is determined. Each acquired image frame with an adjacent relationship is added to the same image group, and the acquired image frame with the minimum acquisition time in each image group is determined as the initial image frame, and the acquired image frame with the maximum acquisition time is determined as the updated image frame. The area of ​​the fog corresponding to the image frame with the minimum acquisition time in each image group is determined as the initial area, and the area of ​​the fog corresponding to the image frame with the maximum acquisition time is determined as the updated area. The area change value is obtained by calculating the difference between the areas of each fog patch corresponding to the minimum and maximum acquisition times of the same image group. If the area change value of any image group is equal to 0, the captured image frame located in the second position of the updated image frame of the image group is determined as the updated image frame corresponding to the image group based on the regional video acquisition. The current diffusion rate corresponding to the fog region located in the target region is determined based on the area change value of each image group corresponding to the target region, and the fog attribute corresponding to the fog region is determined based on the current diffusion rate.

[0014] Optionally, in the method according to the present invention, determining the current diffusion rate corresponding to the fog region located in the target region based on the area change value of each image group corresponding to the target region, and determining the fog attribute corresponding to the fog region based on the current diffusion rate, includes: The difference between the maximum and minimum acquisition times for the same image group is calculated, and the area change value is directly divided by the obtained interval time to obtain the basic diffusion rate. The current wind speed is multiplied by the wind speed influence coefficient, and the wind speed influence increment coefficient is summed with 1 to obtain the diffusion correction coefficient. The current diffusion rate is obtained by multiplying the base diffusion rate by the diffusion correction factor. The current diffusion speed is compared with a preset diffusion threshold. If the comparison result is that the current diffusion speed is greater than the preset diffusion threshold, the retrieved first evaluation value is determined as the group evaluation value corresponding to the image group. If the comparison result is that the current diffusion speed is less than or equal to the preset diffusion threshold, the retrieved second evaluation value will be determined as the group evaluation value corresponding to the image group; If the group evaluation value corresponding to any image group in the same target area is the first evaluation value, it is determined that the target area contains a fog region with the fog attribute as smoke attribute; Conversely, it is determined that the target area contains a fog region with the corresponding fog attribute of sea fog.

[0015] According to another aspect of the present invention, a fire protection and HVAC collaborative data processing system is provided, comprising: The image segmentation module is configured to, if the acquisition state corresponding to any marine equipment is determined to be foggy based on the marine acquisition vessel, segment the initial acquisition image obtained based on the marine acquisition vessel based on the marine equipment to obtain each initial region with acquisition conditions. The image recognition module is configured to perform image recognition on each initial region. If it is determined based on the recognition result that no fog is present in any initial region, the initial region is identified as a sea fog region. The region determination module is configured to, if based on the recognition result, determine that any initial region contains a fog region with a corresponding fog attribute of smoke attribute, determine the initial region as a smoke region, and determine each remaining initial region as a target region; The video recognition module is configured to perform video recognition on the area video corresponding to the target area obtained by the marine collection vessel, determine the current diffusion speed corresponding to the fog patch area located in the target area based on the recognition result, and determine the fog patch attribute corresponding to the fog patch area based on the current diffusion speed.

[0016] According to the present invention, accurate differentiation between marine smoke and sea fog is achieved. The solution fully considers the obstruction and interference of marine equipment on the diffusion path and flow velocity of fog. By selectively dividing the initial acquired images, initial areas that can truly reflect the natural diffusion state of fog are selected, eliminating the deviation in flow velocity data caused by equipment obstruction at the source and ensuring the accuracy of subsequent identification and analysis. Subsequently, fog feature identification is carried out on each initial area to quickly determine sea fog areas without fog patches and clear smoke areas. This not only enables rapid triggering of fire warnings, saving valuable time for marine emergency response, but also avoids the omission of potential smoke risks by marking the remaining target areas for secondary verification. Finally, video is acquired for the target areas, and the essential difference in diffusion velocity between smoke and sea fog is used to complete the final verification of fog patch attributes. Combined with the zoning management strategy, the scope of invalid calculations is greatly reduced, and the analysis efficiency of the server is improved. This solution effectively addresses the technical challenges of visual similarity between smoke and sea fog in complex marine environments, as well as the difficulty in identification due to equipment interference. It avoids ineffective emergency responses caused by misjudging sea fog as smoke and eliminates early warning delays caused by misjudging smoke as sea fog, significantly enhancing the reliability and accuracy of fire monitoring for marine equipment and providing solid technical support for the safe and stable operation of marine equipment. Attached Figure Description

[0017] Figure 1 A flowchart of a fire protection and HVAC collaborative data processing method according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of an initial region according to an embodiment of the present invention is shown; Figure 3A schematic diagram of a region comprising an embodiment of the present invention is shown; Figure 4 A structural block diagram of a fire protection and HVAC collaborative data processing system according to another embodiment of the present invention is shown. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0019] To address the problems existing in the aforementioned background art, the inventors proposed the solution of this invention. One embodiment of this invention provides a method for collaborative data processing of fire protection and HVAC systems, which can be executed in a computing device.

[0020] Figure 1 A flowchart of a fire protection and HVAC collaborative data processing method according to an embodiment of the present invention is shown, the method being adapted to be executed in a computing device.

[0021] like Figure 1 As shown, the fire protection and HVAC collaborative data processing method proposed in this embodiment begins with step S102, which includes the following: If the collection state corresponding to any marine equipment is determined to be foggy based on the marine collection vessel, the initial collection image obtained by the marine collection vessel is divided based on the marine equipment to obtain each initial region with collection conditions.

[0022] For example, in this embodiment, the marine data collection vessel can be understood as a mobile device with mobility and data collection functions at sea. The marine data collection vessel can collect data from the marine equipment. The server can determine the corresponding collection status of the marine equipment based on the data collected by the marine data collection vessel. Here, the collection status refers to whether there is fog in the area where the marine equipment is located. Therefore, the collection status includes foggy status and fog-free status. When the data collection status of any marine device is determined to be foggy based on the marine data collection vessel, it means that there is fog in the area where the marine device is located. In order to ensure the safe operation of the marine device, the server will further determine whether the fog is smoke generated during a fire or sea fog, which is common at sea. Because marine equipment can impede the flow rate of fog, specifically, when fog flows past marine equipment, its diffusion path and flow speed will be locally changed due to the shielding and obstruction of the marine equipment. This causes the flow rate data of fog around the marine equipment to deviate from the true natural diffusion state, which in turn seriously affects the accuracy of the calculation results. Therefore, the server will segment the initial images acquired by the offshore data collection vessel based on the offshore equipment, and select the initial areas that can accurately reflect the movement data such as the flow rate of the fog, that is, the initial areas with the conditions for collection.

[0023] Furthermore, the aforementioned "if the collection state corresponding to any marine equipment is determined to be foggy based on the marine collection vessel, the initial collection image acquired by the marine collection vessel is divided based on the marine equipment to obtain each initial region with collection conditions" also includes the following steps: When it is determined that the temperature value collected by the temperature sensor of any marine equipment is greater than the preset temperature threshold, the collection point with a collection distance from the marine equipment is determined based on the GPS positioning information of the marine equipment. The control vessel at sea acquires images based on the acquisition point orientation towards the marine equipment, thereby obtaining an initial acquired image that has an acquisition correlation with the marine equipment; Image recognition is performed on the initial acquired image, and the acquisition status of the marine equipment that has an acquisition correlation with the initial acquired image is determined based on the recognition results. The acquisition status includes foggy status and fog-free status. If the acquisition state of the marine equipment that is associated with the initial acquisition image is determined to be foggy, the initial acquisition image is divided based on each marine equipment to obtain each initial region with acquisition conditions.

[0024] For example, in this embodiment, each piece of marine equipment corresponds to a temperature sensor. When it is determined that the temperature value collected by the temperature sensor of any piece of marine equipment is greater than the preset temperature threshold, it indicates that the temperature of the equipment corresponding to that piece of marine equipment is too high. In order to determine whether the reason for the high temperature is that the equipment is on fire, the server will control the marine acquisition vessel to collect images of the marine equipment. In order to ensure that the marine data collection vessel can complete the corresponding data collection tasks more comprehensively and safely, the server will first obtain the GPS positioning information corresponding to the marine equipment. After determining the data collection point that is within the data collection distance of the marine equipment based on the GPS positioning information, the server will control the marine data collection vessel to move to the data collection point, so as to collect images from the data collection point towards the marine equipment and obtain the initial data collection image that is related to the data collection of the marine equipment. The acquisition distance can be preset by the server based on the size of the marine equipment, the focal length of the acquisition vessel's camera, and the image clarity requirements. This ensures that the initial acquired image can clearly capture the fog features around the marine equipment, while also preventing the marine acquisition vessel from getting too close to the equipment and causing safety risks. Next, the server will perform image recognition on the initial acquired image and determine the acquisition status of the marine equipment that has an acquisition relationship with the initial acquired image based on the recognition results. Specifically, the acquisition status includes foggy status and fog-free status. If the acquisition status of the marine equipment that is associated with the initial acquisition image is determined to be fog-free, it means that the marine equipment is not currently on fire. At this time, the server will send a high temperature warning signal corresponding to the marine equipment to the management terminal. If it is determined that the acquisition status of the marine equipment that has an acquisition relationship with the initial acquisition image is foggy, the server will divide the initial acquisition image for each marine equipment. This will facilitate further determination of whether the fog in the area where the marine equipment is located is sea fog or smoke based on the initial areas with acquisition conditions.

[0025] Furthermore, the aforementioned "performing image recognition on the initial acquired image and determining the acquisition status of marine equipment that has an acquisition correlation with the initial acquired image based on the recognition results, wherein the acquisition status includes foggy conditions and fog-free conditions" also includes the following steps: Retrieve preset equipment images corresponding to marine equipment that have a data acquisition relationship with the initially acquired images; Image recognition is performed on the preset device image and the initial acquired image to determine the image overlap between the preset device image and the initial acquired image; The image overlap is compared with the preset overlap. If the comparison result is that the image overlap is less than the preset overlap, the acquisition status of the marine equipment that has an acquisition correlation with the initial acquired image is determined to be a foggy state. Conversely, the acquisition status of marine equipment that has an acquisition correlation with the initial acquired image is determined to be a fog-free state.

[0026] For example, in this embodiment, before recognizing the initial acquired image, the server will first retrieve a preset equipment image corresponding to the marine equipment that has a collection association relationship with the initial acquired image from the equipment image database. For example, when the marine equipment that has a collection association relationship with the initial acquired image is a marine crane, the server will retrieve a clear image of the marine crane taken under fog-free and normal working conditions, which is the preset equipment image. Subsequently, the server will use an image matching algorithm to identify and compare the preset equipment image with the initial acquired image, and calculate the image overlap between the two. Here, the image overlap refers to the overlap ratio of the equipment outline of the corresponding marine equipment in the two images, which is used to determine whether the marine equipment is obscured by fog and the corresponding degree of obscuration. Next, the server will retrieve the preset overlap, for example, the preset overlap can be 90%, and the server will compare the image overlap with the preset overlap. One possible comparison result is that the image overlap is less than the preset overlap, indicating that the marine equipment may be obscured by fog, and the degree of obscuration is high. Therefore, the server will determine the acquisition status of the marine equipment that has an acquisition relationship with the initial acquisition image as a foggy state. Another possible comparison result is that the image overlap is greater than or equal to the preset overlap, which means that the marine equipment may not be obscured by fog. Therefore, the server will determine the acquisition status of the marine equipment that has an acquisition relationship with the initial acquisition image as a fog-free state. In this embodiment, the image overlap comparison method is used to determine the acquisition status of the marine equipment. This eliminates the need for complex meteorological sensors, reduces hardware deployment costs, and allows for rapid determination of the acquisition status, meeting the computational needs of real-time server monitoring.

[0027] Furthermore, the aforementioned "if the acquisition state corresponding to the marine equipment that has an acquisition association with the initial acquired image is determined to be a foggy state, the initial acquired image is divided based on each marine equipment to obtain each initial region with acquisition conditions" also includes the following steps: If it is determined that the acquisition state of the marine equipment that has an acquisition association with the initial acquisition image is foggy, image recognition is performed on the initial acquisition image to determine the equipment area corresponding to each marine equipment in the initial acquisition image. The initial acquired image is processed into a grid to obtain each grid region formed by the obtained horizontal grid lines and vertical grid lines. The grid area that does not include any device area is defined as the initial area with the acquisition conditions, and each remaining grid area is defined as the included area; Determine the area proportion of each device area that contains the region, and compare the area proportion with the retrieved preset proportion threshold. The regions whose proportions are less than the preset proportion threshold will be updated to the initial regions with the conditions for data collection.

[0028] For example, in this embodiment, in addition to the marine equipment that has a collection association relationship with the initial collection image, there may be other marine equipment in the initial collection image. Therefore, after the server determines that the collection status of the marine equipment that has a collection association relationship with the initial collection image is a foggy state, it will identify the initial collection image and determine the equipment area in the initial collection image corresponding to each marine equipment. Next, the server performs rasterization processing on the initially acquired image, and then obtains the raster regions formed by the obtained horizontal and vertical raster lines. At this point, the server first determines the raster regions that do not include any device area as the initial regions with acquisition conditions, such as... Figure 2 The polka dot shaded area is shown, and then the server will determine each remaining grid area as the contained area; Next, the server will determine the proportion of the device region to the region of each included region, and then compare the region proportion with a preset proportion threshold value, which can be 10%. One possible comparison result is that the area proportion is less than the preset proportion threshold, indicating that the device area accounts for a small proportion of the included area, and its impact on subsequent monitoring data is weak. Figure 3 The two shaded rectangles shown are the included regions corresponding to the comparison result that the region proportion is less than the preset proportion threshold. Therefore, the server will update the included regions corresponding to the comparison result that the region proportion is less than the preset proportion threshold to the initial regions with the collection conditions. Another possible comparison result is that the area proportion is greater than or equal to the preset proportion threshold. This means that the device area has a large proportion of the area in the included area, which may affect the accuracy of subsequent monitoring data. Therefore, the server will not update the included area corresponding to the comparison result that the area proportion is greater than or equal to the preset proportion threshold to the initial area with the collection conditions.

[0029] Step S104 includes the following: For each initial region, image recognition is performed. If the recognition result determines that no fog patches exist in any initial region, the initial region is identified as a sea fog region.

[0030] For example, in this embodiment, the server performs image recognition on each initial region separately, and determines whether there is fog in the initial region based on the recognition results; Since smoke will generally form patches of fog, if the server does not detect any patches of fog in any initial area, it will directly identify that initial area as a sea fog area.

[0031] Furthermore, the aforementioned "performing image recognition on each initial region, and if it is determined based on the recognition result that no fog patches exist in any initial region, then defining the initial region as a sea fog region" also includes the following steps: Perform image recognition on each initial region to obtain the image pixel value corresponding to each image pixel that makes up each initial region. The difference between the image pixel values ​​corresponding to each adjacent image pixel in the same initial region is calculated to obtain the pixel difference, and the pixel difference is compared with the preset difference threshold. If no comparison result for the same initial region has a pixel difference greater than a preset difference threshold, it is determined that the initial region does not contain fog, and the initial region is identified as a sea fog region.

[0032] For example, in this embodiment, the server will perform image recognition on each initial region to determine the image pixel values ​​corresponding to all image pixels that make up each initial region, and then perform difference calculation on the image pixel values ​​corresponding to adjacent image pixels in the same initial region to obtain the pixel difference. Since the image pixel values ​​in the corresponding fog area are higher than those in other areas, the server will retrieve a preset difference threshold and then compare the difference between each pixel with the preset difference threshold. If no comparison result for the same initial region has a pixel difference greater than the preset difference threshold, it means that there is no fog region in the initial region. Therefore, the server will identify the initial region as a sea fog region.

[0033] Furthermore, the above method also includes the following steps: If there are image pixels in the same initial region whose pixel difference is greater than the preset difference threshold, then the image pixels whose pixel difference is greater than the preset difference threshold are divided into pixel groups. The image pixel with the maximum pixel value in each pixel group is determined as the target pixel, and the target pixels corresponding to the same initial region are connected to obtain the pixel contour. In the initial region, determine the pixel region corresponding to each pixel contour, and obtain the area of ​​the region corresponding to each pixel region; The area of ​​each region is compared with the preset area threshold. If there is no pixel region in the same initial region whose area is greater than the preset area threshold, it is determined that there is no fog in the initial region and the initial region is determined to be a sea fog region.

[0034] For example, in this embodiment, if there are image pixels in the same initial region that have a pixel difference greater than a preset difference threshold, the server will first divide the two image pixels that have a pixel difference greater than the preset difference threshold into the same pixel group, thereby obtaining each pixel group. The image pixel with the maximum pixel value in each pixel group can be understood as the image pixel in the fog region. Therefore, the server will determine the image pixel with the maximum pixel value in each pixel group as the target pixel, and then connect the target pixels in the same initial region to obtain the outline of each pixel. Next, the server will determine the pixel region composed of the outline of each pixel in the initial region, and then obtain the area corresponding to each pixel region. Because sea fog may contain tiny impurities, water droplets, etc., which can cause abnormal interference to local pixels, in order to avoid misjudging normal sea fog as fog, the server will compare the area of ​​each region with the preset area threshold value. If there is no pixel region in the same initial region whose area is greater than the preset area threshold, it means that there is no fog in the initial region, but only interference factors such as tiny impurities or water droplets. Therefore, the server will determine the initial region as a sea fog region. This embodiment calculates the area of ​​each pixel region, which can effectively eliminate abnormal interference from local pixels caused by tiny impurities in sea fog, thus improving the accuracy of judgment.

[0035] Step S106 includes the following: If, based on the identification results, it is determined that any initial region contains a fog region with a corresponding fog attribute of smoke, then the initial region is designated as a smoke region, and each remaining initial region is designated as a target region.

[0036] For example, in this embodiment, if the identification result determines that any initial region has a fog region with a corresponding fog attribute of smoke, the server will identify the initial region as a smoke region and identify each of the remaining initial regions as a target region, so as to facilitate further determination of the fog attribute of the fog region in the target region.

[0037] In this embodiment, the server can quickly trigger a fire warning signal by prioritizing the identification of clearly defined smoke areas, thus gaining valuable time for maritime emergency response. The remaining initial areas are then marked as target areas and subsequently checked a second time. This avoids overlooking potential smoke risks, ensuring the timeliness and accuracy of the warning. Furthermore, the partitioned management makes subsequent analysis more targeted and improves the server's computing efficiency.

[0038] Furthermore, the aforementioned "if, based on the identification results, it is determined that any initial region contains a fog region with a corresponding fog attribute of smoke, the initial region is determined as a smoke region, and each remaining initial region is determined as a target region" also includes the following steps: If any pixel region exists within the same initial region and the comparison result shows that the region area is greater than a preset area threshold, then the pixel region is determined as the comparison region. The average value of each image pixel corresponding to each image pixel point that makes up each comparison area is calculated, and the obtained average pixel value is compared with the preset smoke pixel value respectively. The comparison area corresponding to the pixel mean value of the comparison result being greater than the preset smoke pixel value is determined as the fog area with the corresponding fog attribute being smoke attribute; The initial region that includes any patch of fog with the corresponding fog attribute being smoke is defined as the smoke region, and each of the remaining initial regions is defined as the target region.

[0039] For example, in this embodiment, if there is any pixel region in the same initial region whose area is greater than a preset area threshold, the pixel region may be a fog region. In order to perform a more accurate analysis, the server will first determine the pixel region as the comparison region, and then calculate the average value of each image pixel value corresponding to each image pixel point that makes up each comparison region to obtain the pixel average value. Since the color of the fog produced by smoke is usually dark, the server will compare the average pixel value with the preset smoke pixel value. If the average pixel value is greater than the preset smoke pixel value, it means that the comparison area corresponding to the average pixel value may be fog produced by smoke. Therefore, the server will determine the comparison area corresponding to the comparison result that the average pixel value is greater than the preset smoke pixel value as a fog area with the corresponding fog attribute of smoke. Finally, the server will define the initial region, which includes any patch of fog with the attribute of smoke, as the smoke region, and then define each of the remaining initial regions as the target region, so as to facilitate the further determination of the patch of fog attribute of the patch of fog region in the target region.

[0040] Step S108 includes the following: Video recognition is performed on the area video collected by the marine collection vessel corresponding to the target area. Based on the recognition results, the current diffusion speed corresponding to the fog patch area located in the target area is determined, and the fog patch attribute corresponding to the fog patch area is determined based on the current diffusion speed.

[0041] For example, in this embodiment, after determining the target area, the server will control the marine data collection vessel to collect video, thereby obtaining the area data collection video corresponding to the target area; Because the diffusion rates of smoke and sea fog are significantly different, the server will perform video recognition on the area captured to determine the current diffusion rate of the fog patch area located in the target area, and determine the fog patch attribute corresponding to the fog patch area based on the current diffusion rate.

[0042] Furthermore, the aforementioned "performing video recognition on the area-based video collected by the marine collection vessel corresponding to the target area, determining the current diffusion rate corresponding to the fog patch area located in the target area based on the recognition results, and determining the fog patch attribute corresponding to the fog patch area based on the current diffusion rate" also includes the following steps: The control vessel at sea performs video acquisition based on the acquisition point orientation towards the equipment at sea, acquires each monitoring image frame that makes up the acquired monitoring video, and performs image cropping on each monitoring image frame based on the target area to obtain each acquired image frame that makes up the regional acquisition video corresponding to the target area. The fog region corresponding to the fog attribute of each acquired image frame is identified as smoke, and the fog area of ​​each fog region is determined. Each acquired image frame with an adjacent relationship is added to the same image group, and the acquired image frame with the minimum acquisition time in each image group is determined as the initial image frame, and the acquired image frame with the maximum acquisition time is determined as the updated image frame. The area of ​​the fog corresponding to the image frame with the minimum acquisition time in each image group is determined as the initial area, and the area of ​​the fog corresponding to the image frame with the maximum acquisition time is determined as the updated area. The area change value is obtained by calculating the difference between the areas of each fog patch corresponding to the minimum and maximum acquisition times of the same image group. If the area change value of any image group is equal to 0, the captured image frame located in the second position of the updated image frame of the image group is determined as the updated image frame corresponding to the image group based on the regional video acquisition. The current diffusion rate corresponding to the fog region located in the target region is determined based on the area change value of each image group corresponding to the target region, and the fog attribute corresponding to the fog region is determined based on the current diffusion rate.

[0043] For example, in this embodiment, the server controls the offshore data collection vessel to collect video from the data collection point facing the offshore equipment, thereby obtaining monitoring video; Next, the server will acquire each monitoring image frame that makes up the monitoring video, and perform image cropping on each monitoring image frame based on the target area to obtain each acquired image frame that makes up the area acquisition video corresponding to the target area. That is, each acquired image frame only focuses on the target area, which can not only reduce redundant data, but also improve the server's analysis efficiency. Then, the server will identify the fog region with the fog attribute as smoke in each captured image frame, and determine the fog area of ​​each fog region. Next, the server will add adjacent image frames to the same image group based on the video captured in the area, and determine the image frame with the minimum capture time in each image group as the initial image frame, and determine the image frame with the maximum capture time as the update image frame. For example, if there are three captured image frames in a region, the server will add the first and second captured image frames to one image group, and then add the second and third captured image frames to another image group. This method of overlapping adjacent frames can avoid the deviation in area calculation caused by the error of a single frame image, thereby improving the continuity and accuracy of diffusion speed monitoring. The server will determine the fog area corresponding to the image frame with the minimum acquisition time in each image group as the initial area, and determine the fog area corresponding to the image frame with the maximum acquisition time as the updated area. Next, the server will calculate the difference between the areas of each fog patch corresponding to the minimum and maximum acquisition times of the same image group to obtain the area change value; If the area change value of any image group is equal to 0, that is, the area of ​​the fog patch does not change, the server will determine the captured image frame located in the second position of the updated image frame of that image group as the updated image frame corresponding to that image group based on the area captured video. For example, if the area change value of the image group corresponding to the first and second acquired image frames is equal to 0, the server will determine the third acquired image frame as the updated image frame corresponding to the image group. That is, the updated image group includes the first and third acquired image frames, and then the area change value is recalculated. This avoids calculation errors caused by instantaneous stillness and improves the accuracy of diffusion speed monitoring. Finally, the server determines the current diffusion rate of the fog region located in the target region based on the area change value of each image group corresponding to the target region, and then determines the fog attribute corresponding to the fog region.

[0044] Furthermore, the aforementioned "determining the current diffusion rate corresponding to the fog region located in the target region based on the area change value of each image group corresponding to the target region, and determining the fog attribute corresponding to the fog region based on the current diffusion rate" also includes the following steps: The difference between the maximum and minimum acquisition times for the same image group is calculated, and the area change value is directly divided by the obtained interval time to obtain the basic diffusion rate. The current wind speed is multiplied by the wind speed influence coefficient, and the wind speed influence increment coefficient is summed with 1 to obtain the diffusion correction coefficient. The current diffusion rate is obtained by multiplying the base diffusion rate by the diffusion correction factor. The current diffusion speed is compared with a preset diffusion threshold. If the comparison result is that the current diffusion speed is greater than the preset diffusion threshold, the retrieved first evaluation value is determined as the group evaluation value corresponding to the image group. If the comparison result is that the current diffusion speed is less than or equal to the preset diffusion threshold, the retrieved second evaluation value will be determined as the group evaluation value corresponding to the image group; If the group evaluation value corresponding to any image group in the same target area is the first evaluation value, it is determined that the target area contains a fog region with the fog attribute as smoke attribute; Conversely, it is determined that the target area contains a fog region with the corresponding fog attribute of sea fog.

[0045] For example, in this embodiment, the server first calculates the difference between the maximum and minimum acquisition times corresponding to the same image group to obtain the interval time, and then directly divides the area change value by the interval time to obtain the basic diffusion rate. The basic diffusion rate refers to the fog diffusion rate without considering environmental factors, which is the basis for subsequent correction calculations. Next, the server will multiply the current wind speed and the wind speed influence coefficient, and then sum the obtained wind speed influence increment coefficient with 1 to obtain the diffusion correction coefficient. Here, the wind speed influence coefficient can be a parameter calibrated by the server based on a large amount of experimental data, used to quantify the degree of influence of wind speed on the diffusion of fog. Next, the server will multiply the base diffusion rate by the diffusion correction coefficient to obtain the current diffusion rate. The current diffusion rate refers to the actual diffusion rate after considering environmental factors. Determining the current diffusion rate can ensure the accuracy of subsequent attribute judgments. Subsequently, the server will compare the current diffusion speed with the preset diffusion threshold. If the comparison result is that the current diffusion speed is greater than the preset diffusion threshold, it means that the current diffusion speed is faster. Therefore, the server will determine the first evaluation value that characterizes the smoke diffusion as the group evaluation value corresponding to the image group. If the comparison result shows that the current diffusion speed is less than or equal to the preset diffusion threshold, it means that the current diffusion speed is slow. Therefore, the server will determine the second evaluation value that characterizes the diffusion of sea fog as the group evaluation value corresponding to the image group. If the group evaluation value of any image group in the same target area is the first evaluation value, it means that there is smoke in the target area. Therefore, the server will determine that there is a fog area in the target area with the fog attribute as smoke. If no group evaluation value corresponding to any image group in the same target area is the first evaluation value, it means that there is no smoke in the target area. Therefore, the server will determine that there is a fog region with the fog attribute of sea fog in the target area. In this embodiment, the server can accurately obtain the true diffusion rate of the fog by combining the basic diffusion rate with the diffusion correction coefficient, avoiding misjudgment caused by environmental factors. In addition, the use of grouping to determine the corresponding evaluation value can improve the objectivity and reliability of attribute judgment.

[0046] According to the present invention, accurate differentiation between marine smoke and sea fog is achieved. The solution fully considers the obstruction and interference of marine equipment on the diffusion path and flow velocity of fog. By selectively dividing the initial acquired images, initial areas that can truly reflect the natural diffusion state of fog are selected, eliminating the deviation in flow velocity data caused by equipment obstruction at the source and ensuring the accuracy of subsequent identification and analysis. Subsequently, fog feature identification is carried out on each initial area to quickly determine sea fog areas without fog patches and clear smoke areas. This not only enables rapid triggering of fire warnings, saving valuable time for marine emergency response, but also avoids the omission of potential smoke risks by marking the remaining target areas for secondary verification. Finally, video is acquired for the target areas, and the essential difference in diffusion velocity between smoke and sea fog is used to complete the final verification of fog patch attributes. Combined with the zoning management strategy, the scope of invalid calculations is greatly reduced, and the analysis efficiency of the server is improved. This solution effectively addresses the technical challenges of visual similarity between smoke and sea fog in complex marine environments, as well as the difficulty in identification due to equipment interference. It avoids ineffective emergency responses caused by misjudging sea fog as smoke and eliminates early warning delays caused by misjudging smoke as sea fog, significantly enhancing the reliability and accuracy of fire monitoring for marine equipment and providing solid technical support for the safe and stable operation of marine equipment.

[0047] Another embodiment of the present invention provides a fire protection and HVAC collaborative data processing system. Figure 4 Its corresponding system block diagram includes: The image segmentation module is configured to, if the acquisition state corresponding to any marine equipment is determined to be foggy based on the marine acquisition vessel, segment the initial acquisition image obtained based on the marine acquisition vessel based on the marine equipment to obtain each initial region with acquisition conditions. The image recognition module is configured to perform image recognition on each initial region. If it is determined based on the recognition result that no fog is present in any initial region, the initial region is identified as a sea fog region. The region determination module is configured to, if based on the recognition result, determine that any initial region contains a fog region with a corresponding fog attribute of smoke attribute, determine the initial region as a smoke region, and determine each remaining initial region as a target region; The video recognition module is configured to perform video recognition on the area video corresponding to the target area obtained by the marine collection vessel, determine the current diffusion speed corresponding to the fog patch area located in the target area based on the recognition result, and determine the fog patch attribute corresponding to the fog patch area based on the current diffusion speed.

[0048] In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with the examples of this invention. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing preferred embodiments of the invention.

[0049] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0050] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof.

[0051] Those skilled in the art will understand that modules, units, or components of the devices disclosed in the examples herein can be arranged in the devices described in this embodiment, or alternatively, can be located in one or more devices different from the devices in this example. The modules in the foregoing examples can be combined into a single module or, in addition, can be divided into multiple sub-modules.

[0052] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components.

[0053] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0054] Furthermore, some of the embodiments described herein are methods or combinations of method elements that can be implemented by a processor of a computer system or by other means of performing the functions. Therefore, a processor having the necessary instructions for implementing the methods or method elements forms means for implementing the methods or method elements. Furthermore, the elements described herein in the apparatus embodiments are examples of means for implementing the functions performed by elements for the purposes of carrying out the invention.

[0055] As used herein, unless otherwise specified, the use of ordinal numbers such as “first,” “second,” “third,” etc., to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects being described must have a given order in time, space, ordering, or any other manner.

[0056] Although the invention has been described with respect to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and edibility purposes, and not for the purpose of explaining or limiting the subject matter of the invention.

Claims

1. A method for collaborative data processing of fire protection and HVAC systems, characterized in that, include: If the collection state corresponding to any marine equipment is determined to be foggy based on the marine collection vessel, the initial collection image obtained by the marine collection vessel is divided based on the marine equipment to obtain each initial region with collection conditions. For each initial region, perform image recognition. If, based on the recognition results, it is determined that no fog patches exist in any initial region, the initial region is defined as a sea fog region. If, based on the identification results, it is determined that any initial region contains a fog region with a corresponding fog attribute of smoke, then the initial region is determined as a smoke region, and each of the remaining initial regions is determined as a target region. Video recognition is performed on the area video collected by the marine collection vessel corresponding to the target area. Based on the recognition results, the current diffusion speed corresponding to the fog patch area located in the target area is determined, and the fog patch attribute corresponding to the fog patch area is determined based on the current diffusion speed.

2. The method according to claim 1, characterized in that, If the data collection state corresponding to any marine equipment is determined to be foggy based on the marine data collection vessel, the initial data collection image acquired by the marine data collection vessel is divided based on the marine equipment to obtain initial regions with data collection conditions, including: When it is determined that the temperature value collected by the temperature sensor of any marine equipment is greater than the preset temperature threshold, the collection point with a collection distance from the marine equipment is determined based on the GPS positioning information of the marine equipment. The control vessel at sea acquires images based on the acquisition point orientation towards the marine equipment, thereby obtaining an initial acquired image that has an acquisition correlation with the marine equipment; Image recognition is performed on the initial acquired image, and the acquisition status of the marine equipment that has an acquisition correlation with the initial acquired image is determined based on the recognition results. The acquisition status includes foggy status and fog-free status. If the acquisition state of the marine equipment that is associated with the initial acquisition image is determined to be foggy, the initial acquisition image is divided based on each marine equipment to obtain each initial region with acquisition conditions.

3. The method according to claim 2, characterized in that, Image recognition is performed on the initial acquired images, and based on the recognition results, the acquisition status of marine equipment that has an acquisition correlation with the initial acquired images is determined. The acquisition status includes foggy conditions and fog-free conditions, including: Retrieve preset equipment images corresponding to marine equipment that have a data acquisition relationship with the initially acquired images; Image recognition is performed on the preset device image and the initial acquired image to determine the image overlap between the preset device image and the initial acquired image; The image overlap is compared with the preset overlap. If the comparison result is that the image overlap is less than the preset overlap, the acquisition status of the marine equipment that has an acquisition correlation with the initial acquired image is determined to be a foggy state. Conversely, the acquisition status of marine equipment that has an acquisition correlation with the initial acquired image is determined to be a fog-free state.

4. The method according to claim 3, characterized in that, If the acquisition state of the marine equipment that is associated with the initial acquired image is determined to be foggy, the initial acquired image is divided based on each marine equipment to obtain initial regions with acquisition conditions, including: If it is determined that the acquisition state of the marine equipment that has an acquisition association with the initial acquisition image is foggy, image recognition is performed on the initial acquisition image to determine the equipment area corresponding to each marine equipment in the initial acquisition image. The initial acquired image is processed into a grid to obtain each grid region formed by the obtained horizontal grid lines and vertical grid lines. The grid area that does not include any device area is defined as the initial area with the acquisition conditions, and each remaining grid area is defined as the included area; Determine the area proportion of each device area that contains the region, and compare the area proportion with the retrieved preset proportion threshold. The regions whose proportions are less than the preset proportion threshold will be updated to the initial regions with the conditions for data collection.

5. The method according to claim 1, characterized in that, For each initial region, image recognition is performed. If, based on the recognition results, it is determined that no fog patches exist in any initial region, the initial region is defined as a sea fog region, including: Perform image recognition on each initial region to obtain the image pixel value corresponding to each image pixel that makes up each initial region. The image pixel values ​​corresponding to each adjacent image pixel in the same initial region are calculated to obtain the pixel difference, and the pixel difference is compared with the preset difference threshold. If no comparison result for the same initial region has a pixel difference greater than a preset difference threshold, it is determined that the initial region does not contain fog, and the initial region is identified as a sea fog region.

6. The method according to claim 5, characterized in that, The method further includes: If there are image pixels in the same initial region whose pixel difference is greater than the preset difference threshold, then the image pixels whose pixel difference is greater than the preset difference threshold are divided into pixel groups. The image pixel with the maximum pixel value in each pixel group is determined as the target pixel, and the target pixels corresponding to the same initial region are connected to obtain the pixel contour. In the initial region, determine the pixel region corresponding to each pixel contour, and obtain the area of ​​the region corresponding to each pixel region; The area of ​​each region is compared with the preset area threshold. If there is no pixel region in the same initial region whose area is greater than the preset area threshold, it is determined that there is no fog in the initial region and the initial region is determined to be a sea fog region.

7. The method according to claim 6, characterized in that, If, based on the identification results, it is determined that any initial region contains a fog region with the attribute of smoke, then the initial region is designated as a smoke region, and each of the remaining initial regions is designated as a target region, including: If any pixel region exists within the same initial region and the comparison result shows that the region area is greater than a preset area threshold, then the pixel region is determined as the comparison region. The average value of each image pixel corresponding to each image pixel point that makes up each comparison area is calculated, and the obtained average pixel value is compared with the preset smoke pixel value respectively. The comparison area corresponding to the pixel mean value of the comparison result being greater than the preset smoke pixel value is determined as the fog area with the corresponding fog attribute being smoke attribute; The initial region that includes any patch of fog with the corresponding fog attribute being smoke is defined as the smoke region, and each of the remaining initial regions is defined as the target region.

8. The method according to claim 7, characterized in that, Video recognition is performed on the area video collected by the marine data collection vessel corresponding to the target area. Based on the recognition results, the current diffusion rate of the fog patch area located in the target area is determined, and the fog patch attribute corresponding to the fog patch area is determined based on the current diffusion rate, including: The control vessel at sea performs video acquisition based on the acquisition point orientation towards the equipment at sea, acquires each monitoring image frame that makes up the acquired monitoring video, and performs image cropping on each monitoring image frame based on the target area to obtain each acquired image frame that makes up the regional acquisition video corresponding to the target area. The fog region corresponding to the fog attribute of each acquired image frame is identified as smoke, and the fog area corresponding to each fog region is determined. Each acquired image frame with an adjacent relationship is added to the same image group, and the acquired image frame with the minimum acquisition time in each image group is determined as the initial image frame, and the acquired image frame with the maximum acquisition time is determined as the updated image frame. The area of ​​the fog corresponding to the image frame with the minimum acquisition time in each image group is determined as the initial area, and the area of ​​the fog corresponding to the image frame with the maximum acquisition time is determined as the updated area. The area change value is obtained by calculating the difference between the areas of each fog patch corresponding to the minimum and maximum acquisition times of the same image group. If the area change value of any image group is equal to 0, the captured image frame located in the second position of the updated image frame of the image group is determined as the updated image frame corresponding to the image group based on the regional video acquisition. The current diffusion rate corresponding to the fog region located in the target region is determined based on the area change value of each image group corresponding to the target region, and the fog attribute corresponding to the fog region is determined based on the current diffusion rate.

9. The method according to claim 8, characterized in that, The current diffusion rate corresponding to the fog patch region located in the target region is determined based on the area change value of each image group corresponding to the target region, and the fog patch attribute corresponding to the fog patch region is determined based on the current diffusion rate, including: The difference between the maximum and minimum acquisition times for the same image group is calculated, and the area change value is directly divided by the obtained interval time to obtain the basic diffusion rate. The current wind speed is multiplied by the wind speed influence coefficient, and the wind speed influence increment coefficient is summed with 1 to obtain the diffusion correction coefficient. The current diffusion rate is obtained by multiplying the base diffusion rate by the diffusion correction factor. The current diffusion speed is compared with a preset diffusion threshold. If the comparison result is that the current diffusion speed is greater than the preset diffusion threshold, the retrieved first evaluation value is determined as the group evaluation value corresponding to the image group. If the comparison result is that the current diffusion speed is less than or equal to the preset diffusion threshold, the retrieved second evaluation value will be determined as the group evaluation value corresponding to the image group; If the group evaluation value corresponding to any image group in the same target area is the first evaluation value, it is determined that the target area contains a fog region with the fog attribute as smoke attribute; Conversely, it is determined that the target area contains a fog region with the corresponding fog attribute of sea fog.

10. A fire protection and HVAC collaborative data processing system, characterized in that, include: The image segmentation module is configured to, if the acquisition state corresponding to any marine equipment is determined to be foggy based on the marine acquisition vessel, segment the initial acquisition image obtained based on the marine acquisition vessel based on the marine equipment to obtain each initial region with acquisition conditions. The image recognition module is configured to perform image recognition on each initial region. If it is determined based on the recognition result that no fog is present in any initial region, the initial region is identified as a sea fog region. The region determination module is configured to, if based on the recognition result, determine that any initial region contains a fog region with a corresponding fog attribute of smoke attribute, determine the initial region as a smoke region, and determine each remaining initial region as a target region; The video recognition module is configured to perform video recognition on the area video corresponding to the target area obtained by the marine collection vessel, determine the current diffusion speed corresponding to the fog patch area located in the target area based on the recognition result, and determine the fog patch attribute corresponding to the fog patch area based on the current diffusion speed.

Images