Urban rail transit outside-station rainfall detection method, device and equipment and storage medium

Abstract

The invention provides an urban rail transit outside-station rainfall detection method, device and equipment and a storage medium, and belongs to the technical field of rail transit, and the method comprises the steps: inputting a to-be-recognized image into a target detection model, and obtaining a target detection result outputted by the target detection model, the to-be-recognized image is obtained by intercepting a traffic station entrance and exit monitoring video, and the target detection model is used for detecting pedestrians and rain gears worn by the pedestrians in the input image; and determining the rainfall state outside the traffic station based on the target detection result. The security cameras are uniformly distributed at the entrance and the exit of the urban rail transit station, and the existing security cameras are used for acquiring monitoring videos and detecting the acquired video images, so that on one hand, additional weather sensing equipment does not need to be added, the rainfall monitoring cost is reduced, and on the other hand, the monitoring videos can be identified and detected in real time; the requirement for continuously monitoring the rainfall condition outside the station in real time is met, and station workers can take corresponding preparation measures in time according to the rainfall condition.

Description

Technical Field

[0001] The present invention relates to the field of rail transit technology, and in particular to a precipitation detection method, device, equipment and storage medium outside an urban rail transit station. Background Art

[0002] When it rains, the security checks, entry and exit gates, platforms and the entire rail transit line in urban rail transit stations will face greater pressure. As an important node for passengers to enter and exit, the operational safety and efficiency of the station are of vital importance. Station staff need to lay anti-slip mats, flood prevention sandbags and other facilities in a timely manner, and be prepared for possible emergencies such as stranded passengers on the platform, water leakage in the station, and slips and falls.

[0003] At present, urban rail stations mainly make preliminary preparations based on weather forecasts, and then make further preparations based on the actual precipitation conditions observed by station staff and the conditions reported by passengers entering the station. However, although weather forecasts can predict precipitation weather to a certain extent, they can generally only make forecasts for larger areas and cannot be detailed to dangerous conditions such as local heavy precipitation. If they rely entirely on weather forecasts, some stations may not experience precipitation, resulting in waste of materials and manpower, or precipitation may exceed expectations, resulting in insufficient materials and manpower for disaster relief. If accurate precipitation information is to be obtained, it is necessary to arrange special personnel to observe or set up special meteorological equipment at each station. However, for underground stations or stations without observation conditions, station staff often cannot continuously and in real time observe the precipitation conditions outside the station, and setting up special meteorological equipment requires corresponding procurement, management and maintenance costs.

[0004] Therefore, how to monitor precipitation conditions outside the station in real time at low cost has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] The present invention provides a method, device, equipment and storage medium for detecting precipitation outside an urban rail transit station, which are used to solve the defect in the prior art that the precipitation condition outside the station cannot be monitored in real time at low cost, and realize low-cost real-time monitoring of the precipitation condition outside the station.

[0006] In a first aspect, the present invention provides a method for detecting precipitation outside an urban rail transit station, comprising: Inputting an image to be identified into a target detection model to obtain a target detection result output by the target detection model, wherein the image to be identified is obtained by intercepting a surveillance video of a traffic station entrance and exit, and the target detection model is used to detect pedestrians and the rain gear they are wearing in the input image; Based on the target detection result, the precipitation status outside the traffic station is determined.

[0007] According to a method for detecting precipitation outside an urban rail transit station provided by the present invention, before determining the precipitation state outside the transportation station based on the target detection result, the method further includes: The image to be identified is input into a water accumulation detection model to obtain a water accumulation detection result output by the water accumulation detection model, and the water accumulation detection model is used to detect water accumulation areas on the ground.

[0008] According to a method for detecting precipitation outside an urban rail transit station provided by the present invention, the method of determining the precipitation state outside the transportation station based on the target detection result includes: Based on the target detection result and the water accumulation detection result, the precipitation status outside the traffic station is determined.

[0009] According to a method for detecting precipitation outside an urban rail transit station provided by the present invention, before determining the precipitation state outside the transportation station based on the target detection result, the method further includes: The image to be identified is input into a precipitation intensity identification model to obtain a precipitation intensity identification result output by the precipitation intensity identification module, wherein the precipitation intensity identification model is used to determine the precipitation intensity based on the visibility in the input image.

[0010] According to a method for detecting precipitation outside an urban rail transit station provided by the present invention, the method of determining the precipitation state outside the transportation station based on the target detection result includes: The precipitation status outside the traffic station is determined based on the target detection result, the water accumulation detection result and the precipitation intensity identification result.

[0011] According to a method for detecting precipitation outside an urban rail transit station provided by the present invention, the method of determining the precipitation state outside the transportation station based on the target detection result, the water accumulation detection result and the precipitation intensity identification result includes: The target detection result, the water accumulation detection result and the precipitation intensity recognition result are input into an information integration model to obtain the precipitation state output by the information integration model, and the information integration model is a fully connected neural network.

[0012] In a second aspect, the present invention further provides a precipitation detection device outside an urban rail transit station, comprising: A target detection module, used to input an image to be identified into a target detection model to obtain a target detection result output by the target detection model, wherein the image to be identified is obtained by intercepting a surveillance video of a traffic station entrance and exit, and the target detection model is used to detect pedestrians and the rain gear they are wearing in the input image; The precipitation detection module is used to determine the precipitation status outside the traffic station based on the target detection result.

[0013] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the above-mentioned methods for detecting precipitation outside an urban rail transit station are implemented.

[0014] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described methods for detecting precipitation outside an urban rail transit station.

[0015] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned methods for detecting precipitation outside an urban rail transit station.

[0016] The method, device, equipment and storage medium for detecting precipitation outside an urban rail transit station provided by the present invention input an image to be identified into a target detection model to obtain a target detection result output by the target detection model. The image to be identified is obtained by intercepting the surveillance video of the entrance and exit of the transportation station. The target detection model is used to detect pedestrians and the rain gear they wear in the input image; based on the target detection result, the precipitation status outside the transportation station is determined. The entrances and exits of urban rail transit stations are equipped with security cameras, and the existing security cameras are used to collect surveillance videos. The precipitation status is determined by detecting the collected video images. On the one hand, there is no need to add additional weather sensing equipment, which reduces the cost of precipitation monitoring. On the other hand, the surveillance video can be identified and detected in real time to meet the needs of continuous and real-time monitoring of the precipitation conditions outside the station, so that station staff can make corresponding preparations in time according to the precipitation status. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a schematic diagram of the flow of the precipitation detection method outside the urban rail transit station provided by the present invention; Figure 2 It is a structural schematic diagram of the target detection model provided by the present invention; Figure 3 It is a structural schematic diagram of the water accumulation detection model provided by the present invention; Figure 4 It is a structural schematic diagram of the precipitation intensity identification model provided by the present invention; Figure 5 It is a structural schematic diagram of the information integration model provided by the present invention; Figure 6 It is a structural schematic diagram of a precipitation detection device outside an urban rail transit station provided by the present invention; Figure 7 It is a structural schematic diagram of the precipitation detection system outside the urban rail transit station provided by the present invention; Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] It should be noted that in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "include one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements. The orientation or position relationship indicated by the terms "upper", "lower" and the like is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0021] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0022] Combine the following Figure 1-Figure 8 The present invention describes a precipitation detection method, device, equipment and storage medium outside an urban rail transit station provided by an embodiment of the present invention.

[0023] Figure 1 : is a flow chart of a method for detecting precipitation outside an urban rail transit station provided by the present invention, such as Figure 1 As shown, including but not limited to the following steps: S110, inputting an image to be identified into a target detection model to obtain a target detection result output by the target detection model, wherein the image to be identified is obtained by intercepting a surveillance video of a traffic station entrance and exit, and the target detection model is used to detect pedestrians and the rain gear they are wearing in the input image; S120: Determine the precipitation status outside the traffic station based on the target detection result.

[0024] It should be noted that the execution subject of the urban rail transit station precipitation detection method provided in the embodiments of the present application may be a server, a computer device, such as a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, etc. For ease of understanding, the following embodiments take the server as the execution subject to illustrate the urban rail transit station precipitation detection method provided by the present invention.

[0025] Here, the surveillance video is taken by the existing security surveillance camera in the urban rail transit station, and the image to be identified is a frame of the surveillance video. In the specific implementation process, the surveillance video is transmitted to the server, and the server can perform precipitation detection on each frame of the surveillance video, or perform precipitation detection at intervals of a certain number of frames, or capture a frame of the image for precipitation detection at intervals of a certain length of time, and there is no specific restriction here.

[0026] There is no restriction on the type of target detection model. For example, commonly used target detection networks such as recurrent neural networks, SSD, and YoLo series can be used.

[0027] The rain gear worn by pedestrians may include but is not limited to umbrellas, raincoats, rain boots, etc.

[0028] Figure 2 It is a schematic diagram of the structure of the target detection model provided by the present invention, such as Figure 2 As shown, a backbone network is used for feature extraction. The network structure used by the backbone network is not limited. For example, a convolutional neural network, a recurrent neural network, and other networks can be used as the backbone network. The neck network and the head network are selected according to actual needs. The network of appropriate size and function, for example, a neck network and a detection head for an umbrella, a raincoat, and rain boots are respectively constructed.

[0029] In some embodiments, the object detection network outputs the detection frame and confidence of each rain gear in the image to be identified. Taking the example that the rain gear in the image to be identified includes an umbrella, a raincoat, and rain boots, the object detection result includes the detection frame of the umbrella and the confidence that the target in the detection frame is the umbrella, the detection frame of the raincoat and the confidence that the target in the detection frame is the raincoat, and the detection frame of the rain boots and the confidence that the target in the detection frame is the rain boots.

[0030] In other embodiments, the target detection network outputs the number of rain gear worn by pedestrians in the image to be identified and their confidence levels. Specifically, the target detection network is constructed to output the number of rain gear worn by each pedestrian and its confidence level, or it can be constructed to output the total number of rain gear in the image and its confidence level.

[0031] In S120, the target detection result is input into the information integration model to obtain the precipitation status outside the traffic station output by the information integration model.

[0032] Preferably, the information integration model takes the target detection result as input and takes the precipitation status that is easy for station staff to understand as output, for example, taking the precipitation status of heavy rain, moderate rain, and light rain as output.

[0033] In some embodiments, the information integration model is a pre-established threshold and logic rule. After the target detection result is input into the information integration model, the information integration model performs a logic judgment based on the target detection result and outputs a corresponding logic result, i.e., the precipitation state outside the traffic station. For example, the precipitation state is determined based on the ratio of the number of people carrying rain gear in the recent period of time. For example, if the target detection result in the last five minutes is that at least 80% of pedestrians carry rain gear, then the current precipitation state is determined to be heavy rain. If the target detection result in the last five minutes is that 20%-80% of pedestrians carry rain gear, then the current precipitation state is determined to be moderate rain. If the target detection result in the last five minutes is that less than 20% of pedestrians carry rain gear, then the current precipitation state is determined to be light rain.

[0034] In other embodiments, the information integration model adopts a neural network model, and after the target detection result is input into the trained information integration model, the information integration model outputs the corresponding precipitation status.

[0035] Optionally, the target detection model is trained based on the following steps: The video data collected by the traffic station surveillance cameras over the years were screened, and a certain number of images with and without precipitation containing passengers were selected, and the images were annotated to build a training data set for detecting what passengers were wearing. According to specific needs, umbrellas, raincoats, rain boots and other targets can be selected as target detection categories, and information on whether there is rainfall can be added.

[0036] For any training data, input the training data into the target detection model and output the corresponding target detection result; Using the preset loss function, the loss value is calculated based on the target detection results and the labeled data corresponding to the training data; Based on the loss value, the model parameters of the target detection model are adjusted until the loss value is less than the preset threshold or the number of training times reaches the preset number, and the final target detection model is obtained.

[0037] The method for detecting precipitation outside an urban rail transit station provided in an embodiment of the present application inputs an image to be identified into a target detection model to obtain a target detection result output by the target detection model. The image to be identified is obtained by intercepting a surveillance video of the entrance and exit of the transportation station. The target detection model is used to detect pedestrians and the rain gear they wear in the input image. Based on the target detection result, the precipitation status outside the transportation station is determined. Security cameras are installed at the entrances and exits of urban rail transit stations. Existing security cameras are used to collect surveillance videos. The precipitation status is determined by detecting the collected video images. On the one hand, there is no need to add additional weather sensing equipment, which reduces the cost of precipitation monitoring. On the other hand, the surveillance video can be identified and detected in real time to meet the needs of continuous and real-time monitoring of precipitation conditions outside the station, so that station staff can make corresponding preparations in time according to the precipitation status.

[0038] In an optional embodiment, before determining the precipitation state outside the traffic station based on the target detection result, the method further includes: The image to be identified is input into a water accumulation detection model to obtain a water accumulation detection result output by the water accumulation detection model, and the water accumulation detection model is used to detect water accumulation areas on the ground.

[0039] Figure 3 Schematic diagram of the structure of the water accumulation detection model provided by the present invention. Figure 3As shown, a backbone network is used for feature extraction. The network structure used by the backbone network is not limited. For example, a convolutional neural network, a recurrent neural network, a transformer network, etc. can be used as the backbone network, and a suitable connection layer (Neck) and a detection head (Head) are selected to build a neural network, and the data set is injected for training.

[0040] Further, determining the precipitation state outside the traffic station based on the target detection result includes: Based on the target detection result and the water accumulation detection result, the precipitation status outside the traffic station is determined.

[0041] The waterlogging detection model uses a semantic segmentation network to identify the ground waterlogging area (Mask) and obtain the waterlogging detection results.

[0042] Preferably, the information integration model takes target detection results and water accumulation detection results as input, and outputs precipitation status that is easy for station staff to understand, such as heavy rain, moderate rain, light rain, and whether there is water accumulation.

[0043] In addition, the surveillance video inside the station can be monitored in real time to detect whether there is water accumulation inside the station.

[0044] In some embodiments, the information integration model is a pre-defined threshold and logic rule. After the target detection result and the water accumulation detection result are input into the information integration model, the information integration model performs a logic judgment based on the target detection result and the water accumulation detection result, and outputs a corresponding logic result, i.e., the precipitation status outside the traffic station. For example, the precipitation status is determined based on the ratio of the number of people carrying rain gear in the recent period and the water accumulation ratio.

[0045] In other embodiments, the information integration model adopts a neural network model.

[0046] Optionally, the water accumulation detection model is obtained by training based on the following steps: A certain number of images without water accumulation and a considerable number of images with water accumulation are selected to construct a training dataset with the water accumulation area as the target.

[0047] For any piece of training data, input the training data into the waterlogging detection model and output the corresponding waterlogging detection result; Using the preset loss function, the loss value is calculated based on the waterlogging detection results and the waterlogging areas marked in the training data. Based on the loss value, the model parameters of the waterlogging detection model are adjusted until the loss value is less than a preset threshold or the number of training times reaches a preset number, thereby obtaining the final waterlogging detection model.

[0048] It is understandable that the target detection model alone cannot distinguish whether pedestrians use umbrellas for sun protection or rain protection. By adding a water accumulation detection model to detect waterlogged areas on the ground, it can be determined that there is precipitation when there is water on the ground and pedestrians are wearing rain gear, thus avoiding misjudgment when pedestrians use umbrellas for sun protection.

[0049] Based on any of the above embodiments, before determining the precipitation state outside the traffic station based on the target detection result, the method further includes: The image to be identified is input into a precipitation intensity identification model to obtain a precipitation intensity identification result output by the precipitation intensity identification module, wherein the precipitation intensity identification model is used to determine the precipitation intensity based on the visibility in the input image.

[0050] The precipitation intensity identification model adopts a classification network, and the specific type of network model used is not restricted here.

[0051] Figure 4 is a schematic diagram of the structure of the precipitation intensity identification model provided by the present invention, such as Figure 4 As shown, a backbone network is used for feature extraction. The network structure used by the backbone network is not limited. For example, convolutional neural networks, recurrent neural networks, transformer networks and other networks can be used as backbone networks. The neck network and the head network are selected according to actual needs. The network of appropriate size and function, for example, the neck network and detection head for sunny days, light rain, moderate rain and heavy rain are respectively constructed.

[0052] In some embodiments, the precipitation intensity recognition model outputs the confidence level of different rainfall intensity categories. Taking the example that the categories recognized by the precipitation intensity recognition model include sunny, light rain, moderate rain and heavy rain, the precipitation intensity recognition result includes the confidence level that the image to be recognized is sunny, the confidence level that the image to be recognized is light rain, the confidence level that the image to be recognized is moderate rain, and the confidence level that the image to be recognized is heavy rain.

[0053] In other embodiments, the precipitation intensity recognition model directly outputs the category of the image to be recognized, that is, the category with the highest confidence.

[0054] Further, determining the precipitation state outside the traffic station based on the target detection result includes: The precipitation status outside the traffic station is determined based on the target detection result, the water accumulation detection result and the precipitation intensity identification result.

[0055] The target detection results are input into the information integration model to obtain the precipitation status outside the traffic station output by the information integration model.

[0056] Preferably, the information integration model takes target detection results, water accumulation detection results and precipitation intensity identification results as input, and outputs precipitation status that is easy for station staff to understand, such as heavy rain, moderate rain, light rain, and whether there is water accumulation.

[0057] In some embodiments, the information integration model is a pre-established threshold and logic rule. After the target detection result, water accumulation detection result and precipitation intensity recognition result are input into the information integration model, the information integration model performs logic judgment according to the target detection result, water accumulation detection result and precipitation intensity recognition result, and outputs the corresponding logic result, i.e., the precipitation status outside the traffic station. For example, the precipitation status is determined according to the ratio of the number of people carrying rain gear, the water accumulation ratio, and the precipitation intensity in the recent period.

[0058] In other embodiments, the information integration model adopts a neural network model.

[0059] Optionally, the precipitation intensity recognition model is obtained by training based on the following steps: A certain number of images without precipitation and a considerable number of images with precipitation are selected to construct a dataset for classifying precipitation intensity according to visibility. According to specific needs, the classification categories can be refined into different precipitation intensities.

[0060] For any piece of training data, the training data is input into the precipitation intensity recognition model, and the corresponding precipitation intensity recognition result is output; Using the preset loss function, the loss value is calculated based on the precipitation intensity recognition result and the labeled data corresponding to the training data; Based on the loss value, the model parameters of the precipitation intensity recognition model are adjusted until the loss value is less than a preset threshold or the number of training times reaches a preset number, and the final precipitation intensity recognition model is obtained.

[0061] It is understandable that based on the target detection model and the water accumulation detection model, it is impossible to distinguish whether pedestrians use umbrellas for this purpose or to keep out the rain when there is no water accumulation area on the ground. By adding a precipitation intensity recognition model, when the weather condition is determined to be sunny, even if the target detection model detects an umbrella, it can be determined that there is no precipitation, thereby avoiding misjudgment and improving the model detection accuracy.

[0062] In an optional embodiment, determining the precipitation state outside the traffic station based on the target detection result, the water accumulation detection result and the precipitation intensity recognition result includes: The target detection result, the water accumulation detection result and the precipitation intensity recognition result are input into an information integration model to obtain the precipitation state output by the information integration model, and the information integration model is a fully connected neural network.

[0063] Figure 5It is a structural diagram of the information integration model provided by the present invention, such as Figure 5 As shown, the target detection results, water accumulation detection results and precipitation intensity recognition results are used as input, and the current precipitation and water accumulation status that is easy for station staff to understand is used as output. It should be understood that Figure 5 The output of the information integration model is only shown as an example. In the specific implementation process, the corresponding output state can be adjusted according to actual usage requirements.

[0064] The following is a description of the precipitation detection device outside the urban rail transit station provided in an embodiment of the present application. The precipitation detection device outside the urban rail transit station described below and the precipitation detection method outside the urban rail transit station described above can be referenced to each other.

[0065] Figure 6 Schematic diagram of the structure of the precipitation detection device outside the urban rail transit station provided by the present invention. Figure 6 As shown, the precipitation detection device outside the urban rail transit station includes: The target detection module 610 is used to input the image to be identified into the target detection model to obtain the target detection result output by the target detection model, wherein the image to be identified is obtained by intercepting the surveillance video of the entrance and exit of the traffic station, and the target detection model is used to detect pedestrians and the rain gear they are wearing in the input image; The precipitation detection module 620 is used to determine the precipitation status outside the traffic station based on the target detection result.

[0066] It should be noted that the precipitation detection device outside the urban rail transit station provided in the embodiment of the present invention can execute the precipitation detection method outside the urban rail transit station described in any of the above embodiments during specific operation, which will not be elaborated in this embodiment.

[0067] The following is a description of the urban rail transit station outside precipitation detection system provided in an embodiment of the present application. The urban rail transit station outside precipitation detection system described below and the urban rail transit station outside precipitation detection method described above can be referenced to each other.

[0068] Figure 7 Schematic diagram of the structure of the precipitation detection system outside the urban rail transit station provided by the present invention. Figure 7As shown, the surveillance cameras use the existing security cameras in the urban rail transit station, which are deployed at multiple entrances and exits of the station. The surveillance videos collected by multiple surveillance cameras are aggregated through the router and uniformly transmitted back to the station server located in the station command center. Preferably, the station server is a high-performance computer equipped with large-capacity storage and large-scale parallel computing devices (such as GPU). The station server decodes the image of the video transmitted back by the surveillance camera via the router and performs neural network algorithm reasoning. After information integration and comprehensive judgment, the calculation results are output. The output results of the station server can be connected to the terminal device wirelessly or wired to inform the station staff of the current precipitation status outside the station. The terminal device can be a screen in the station command center or inside the station, or a mobile device such as a tablet or a mobile phone. There is no specific restriction here.

[0069] It should be understood that the precipitation detection system outside the urban rail transit station should ensure that there is no packet loss in data transmission between multiple monitoring cameras, routers, station servers and multiple terminal devices; the station server should ensure that it has sufficient computing power to meet the specified inference frame rate requirements, and can expand more computing boards when necessary, or use quantitative models to reduce the amount of calculation within an acceptable range of calculation accuracy.

[0070] It should be noted that, for the purpose of description, the system structure depicted is only an example of a suitable environment and does not impose any limitation on the scope of use or function of the present application. Figure 7 There are no dependencies or requirements on any one or combination of components shown.

[0071] The terminal involved in the embodiments of the present application may be a device that provides voice and / or data connectivity to a user, a handheld device with a wireless connection function, or other processing devices connected to a wireless modem, etc. In different systems, the name of the terminal device may also be different. For example, in a 5G system, the terminal device may be called a user equipment (UE).

[0072] Figure 8 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 8As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the precipitation detection method outside the urban rail transit station, the method comprising: inputting the image to be identified into the target detection model, obtaining the target detection result output by the target detection model, the image to be identified is obtained by intercepting the surveillance video of the entrance and exit of the traffic station, the target detection model is used to detect pedestrians in the input image and the rain gear they wear; based on the target detection result, determining the precipitation state outside the traffic station.

[0073] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0074] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the precipitation detection method outside the urban rail transit station provided by the above-mentioned embodiments, and the method includes: inputting an image to be identified into a target detection model to obtain a target detection result output by the target detection model, wherein the image to be identified is obtained by intercepting a surveillance video of the entrance and exit of the transportation station, and the target detection model is used to detect pedestrians and the rain gear they wear in the input image; based on the target detection result, determining the precipitation status outside the transportation station.

[0075] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the precipitation detection method outside the urban rail transit station provided in the above-mentioned embodiments. The method includes: inputting an image to be identified into a target detection model to obtain a target detection result output by the target detection model, wherein the image to be identified is obtained by intercepting a surveillance video of the entrance and exit of the transportation station, and the target detection model is used to detect pedestrians and the rain gear they wear in the input image; based on the target detection result, determining the precipitation status outside the transportation station.

[0076] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0077] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting precipitation outside an urban rail transit station, characterized in that: include: Inputting an image to be identified into a target detection model to obtain a target detection result output by the target detection model, wherein the image to be identified is obtained by intercepting a surveillance video of a traffic station entrance and exit, and the target detection model is used to detect pedestrians and the rain gear they are wearing in the input image; Based on the target detection result, the precipitation status outside the traffic station is determined.

2. The method for detecting precipitation outside an urban rail transit station according to claim 1, characterized in that: Before determining the precipitation state outside the traffic station based on the target detection result, the method further includes: The image to be identified is input into a water accumulation detection model to obtain a water accumulation detection result output by the water accumulation detection model, and the water accumulation detection model is used to detect water accumulation areas on the ground.

3. The method for detecting precipitation outside an urban rail transit station according to claim 2, characterized in that: The determining the precipitation state outside the traffic station based on the target detection result includes: Based on the target detection result and the water accumulation detection result, the precipitation status outside the traffic station is determined.

4. The method for detecting precipitation outside an urban rail transit station according to claim 2, characterized in that: Before determining the precipitation state outside the traffic station based on the target detection result, the method further includes: The image to be identified is input into a precipitation intensity identification model to obtain a precipitation intensity identification result output by the precipitation intensity identification module, wherein the precipitation intensity identification model is used to determine the precipitation intensity based on the visibility in the input image.

5. The method for detecting precipitation outside an urban rail transit station according to claim 4, characterized in that: The determining the precipitation state outside the traffic station based on the target detection result includes: The precipitation status outside the traffic station is determined based on the target detection result, the water accumulation detection result and the precipitation intensity identification result.

6. The method for detecting precipitation outside an urban rail transit station according to claim 5, characterized in that: The determining of the precipitation state outside the traffic station based on the target detection result, the water accumulation detection result and the precipitation intensity identification result includes: The target detection result, the water accumulation detection result and the precipitation intensity recognition result are input into an information integration model to obtain the precipitation state output by the information integration model, and the information integration model is a fully connected neural network.

7. A precipitation detection device outside an urban rail transit station, characterized in that: include: A target detection module, used to input an image to be identified into a target detection model to obtain a target detection result output by the target detection model, wherein the image to be identified is obtained by intercepting a surveillance video of a traffic station entrance and exit, and the target detection model is used to detect pedestrians and the rain gear they are wearing in the input image; The precipitation detection module is used to determine the precipitation status outside the traffic station based on the target detection result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for detecting precipitation outside an urban rail transit station as described in any one of claims 1 to 6 are implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for detecting precipitation outside an urban rail transit station as claimed in any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for detecting precipitation outside an urban rail transit station as claimed in any one of claims 1 to 6 are implemented.

Images