CN116168317A - Weather identification method, weather identification device, electronic equipment and storage medium - Google Patents

Abstract

The present disclosure provides a weather recognition method, which relates to the technical field of artificial intelligence, in particular to the technical field of image recognition and video analysis technology, and can be applied to scenarios such as smart cities, urban governance, and emergency management. The specific implementation scheme is: determine at least one scene information corresponding to at least one initial video, wherein at least one initial video is related to the target area; use at least one video processing strategy corresponding to the at least one scene information to process at least one initial video respectively, and obtain at least one target video; respectively performing first recognition processing on the at least one target video to obtain at least one first recognition result; and obtaining a target weather recognition result corresponding to the target area according to the at least one first recognition result. The present disclosure also provides a weather recognition device, electronic equipment and a storage medium.

Description

Weather recognition method, device, electronic device and storage medium

technical field

The present disclosure relates to the field of artificial intelligence technology, in particular to the field of image recognition technology and video analysis technology, and can be applied in scenarios such as smart cities, urban governance, and emergency management. More specifically, the present disclosure provides a weather recognition method, device, electronic device and storage medium.

Background technique

With the development of artificial intelligence technology, the application scenarios of deep learning technology continue to increase. The image or video collected by the camera can be recognized to determine the weather conditions in the area where the camera is located.

Contents of the invention

The present disclosure provides a weather recognition method, device, equipment and storage medium.

According to an aspect of the present disclosure, a weather recognition method is provided, the method includes: determining at least one scene information corresponding to at least one initial video, wherein at least one initial video is related to a target area; The at least one video processing strategy processes at least one initial video respectively to obtain at least one target video; respectively performs first recognition processing on at least one target video to obtain at least one first recognition result; and according to at least one first recognition result, obtains the same as The target weather recognition result corresponding to the target area.

According to another aspect of the present disclosure, there is provided a weather recognition device, which includes: a determination module configured to determine at least one piece of scene information corresponding to at least one initial video, wherein at least one initial video is related to a target area; processing The model is configured to use at least one video processing strategy corresponding to at least one scene information to process at least one initial video to obtain at least one target video; the identification processing module is configured to perform first identification processing on at least one target video to obtain at least one target video. A first recognition result; and an obtaining module, configured to obtain a target weather recognition result corresponding to the target area according to at least one first recognition result.

According to another aspect of the present disclosure, an electronic device is provided, including: at least one processor; and a memory connected in communication with the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are At least one processor executes, so that the at least one processor can execute the method provided according to the present disclosure.

According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method provided according to the present disclosure.

According to another aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the method provided according to the present disclosure.

It should be understood that what is described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will be readily understood through the following description.

Description of drawings

The accompanying drawings are used to better understand the present solution, and do not constitute a limitation to the present disclosure. in:

FIG. 1 is a schematic diagram of an exemplary system architecture to which a weather recognition method and device can be applied according to an embodiment of the present disclosure;

FIG. 2 is a flowchart of a weather recognition method according to an embodiment of the present disclosure;

Figure 3A is a schematic diagram of an initial video frame according to one embodiment of the present disclosure;

3B is a schematic diagram of an initial video frame according to another embodiment of the present disclosure;

3C is a schematic diagram of an initial video frame according to another embodiment of the present disclosure;

Fig. 4 is a schematic diagram of obtaining target weather recognition results according to an embodiment of the present disclosure;

FIG. 5 is a block diagram of a weather recognition device according to an embodiment of the present disclosure;

FIG. 6 is a block diagram of an electronic device to which a weather recognition method can be applied according to one embodiment of the present disclosure.

Detailed ways

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and they should be regarded as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

Weather phenomena may include various natural phenomena such as rainfall, snowfall, and hailstone. Taking rainfall as an example, rainfall is a natural phenomenon with obvious regularity in local areas, and it is an important part of weather forecast. The observation of rainfall by the meteorological department is mainly based on the data of meteorological observation stations.

The recognition result of the rainfall recognition can indicate whether a rainfall phenomenon has occurred in the relevant area. The distribution density of meteorological observation stations is low, and the measurement tools are limited, so fine-grained (such as streets or roads) rainfall identification cannot be achieved, and the accuracy requirements for rainfall identification may not be met. Rainfall may include torrential rain. Torrential rain can refer to continuous heavy rainfall that may become a meteorological disaster. Torrential rain may pose a threat to the safety of the masses and infrastructure, and may also cause geological disasters (such as mudslides, etc.).

In order to improve the accuracy of rainfall recognition, image recognition or video recognition technology can be used to recognize the images or videos collected by the camera, and obtain the recognition result indicating whether the area where the camera is located is rainy or not. The rain recognition technology based on image recognition can use the scene information of the camera to a certain extent, but it is difficult to use the action information related to the falling of rainwater, resulting in low recognition accuracy. The rain recognition technology based on video recognition can use the scene information of the camera and the action information of the rainwater to a certain extent. However, in some examples, the rain recognition technology based on video recognition is mostly based on the data of a single camera, without considering the relationship between different cameras, and it is difficult to carry out rainfall recognition based on larger-scale non-homogeneous data.

Fig. 1 is a schematic diagram of an exemplary system architecture to which a weather recognition method and device can be applied according to an embodiment of the present disclosure. It should be noted that, what is shown in FIG. 1 is only an example of the system architecture to which the embodiments of the present disclosure can be applied, so as to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other device, system, environment or scenario.

As shown in FIG. 1 , a system architecture 100 according to this embodiment may include terminal devices 101 , 102 , 103 , a network 104 and a server 105 . The network 104 is used as a medium for providing communication links between the terminal devices 101 , 102 , 103 and the server 105 . Network 104 may include various connection types, such as wired and/or wireless communication links, among others.

Users can use terminal devices 101 , 102 , 103 to interact with server 105 via network 104 to receive or send messages and the like. The terminal devices 101, 102, 103 may be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers and the like.

The server 105 may be a server that provides various services, such as a background management server that provides support for websites browsed by users using the terminal devices 101 , 102 , 103 (just an example). The background management server can analyze and process received data such as user requests, and feed back processing results (such as webpages, information, or data obtained or generated according to user requests) to the terminal device.

It should be noted that, generally, the weather recognition method provided by the embodiment of the present disclosure may be executed by the server 105 . Correspondingly, the weather identification device provided by the embodiment of the present disclosure can generally be set in the server 105 . The weather identification method provided by the embodiments of the present disclosure may also be executed by a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101 , 102 , 103 and/or the server 105 . Correspondingly, the weather recognition apparatus provided by the embodiments of the present disclosure may also be set in a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101 , 102 , 103 and/or the server 105 .

FIG. 2 is a flowchart of a weather recognition method according to an embodiment of the present disclosure.

As shown in FIG. 2, the method 200 may include operation S210 to operation S240.

In operation S210, at least one scene information corresponding to at least one initial video is determined.

In an embodiment of the present disclosure, at least one initial video is related to the target area. For example, the target area may correspond to a street. The street may include roads and features (such as buildings, etc.). For another example, at least one video collection device (such as a camera) may be deployed in the target area. The video captured by the video capture device can be used as the initial video.

In an embodiment of the present disclosure, the scene information may represent a scene related to the initial video. For example, scene information may indicate that the original video is from a scene with high visibility. It is understandable that the average brightness of video frames is higher in the original video from a scene with high visibility. It can also be understood that in this embodiment, the scene information representing a scene with high visibility is taken as an example, but the disclosure is not limited thereto, and the scene information may also represent other scenes.

In operation S220, at least one initial video is respectively processed using at least one video processing policy corresponding to at least one scene information to obtain at least one target video.

In the embodiment of the present disclosure, the scene information corresponds to the video processing strategy. For example, the scene information may represent a scene with high visibility, and the video processing strategy corresponding to the scene information may indicate the following operation: take the initial video as the target video. When using the video processing strategy to process the initial video, the initial video can be used as the target video.

In operation S230, first recognition processing is performed on at least one target video to obtain at least one first recognition result.

In the embodiments of the present disclosure, various manners may be used to perform the first identification process. For example, the first recognition process can be performed using a trained deep learning model.

In an embodiment of the present disclosure, the first identification result may indicate whether there is a preset weather phenomenon in the target area. For example, the preset weather phenomenon may include phenomena such as rain, snow, and hail. For another example, the first first identification result may indicate that there is rainfall in the target area.

In operation S240, a target weather recognition result corresponding to the target area is obtained according to at least one first recognition result.

For example, taking 3 initial videos as an example, the 3 initial videos may correspond to 3 first recognition results. The first first recognition result may indicate that there is rainfall in the target area. The second first identification result may also indicate that there is rainfall in the target area. The third first identification result may indicate that there is no rain phenomenon in the target area. The number of recognition results indicating the presence of the rainfall phenomenon is larger than the number of recognition results indicating the absence of the rainfall phenomenon. Thus, the target weather identification result may indicate that there is rainfall in the target area.

Through the embodiments of the present disclosure, at least one video is identified, and the fine-grained features of different videos can be fully utilized, which helps to improve the accuracy of identification. In addition, at least one initial video is processed using a video processing strategy corresponding to the scene information, so that recognition can be performed based on data from at least one camera, which helps to expand the application scenarios of weather recognition.

It can be understood that the method flow of the present disclosure has been described above, and the scene information in the above operation S210 will be described below.

In some embodiments, the initial video includes a plurality of initial video frames, and at least one piece of scene information includes at least one of first scene information, second scene information, and third scene information.

FIG. 3A is a schematic diagram of an initial video frame according to one embodiment of the present disclosure.

In an embodiment of the present disclosure, the first scene information is used to indicate that a proportion of a preset object in at least one video frame of the initial video is greater than or equal to a preset proportion threshold. For example, the preset objects may be objects such as tree branches and baffles. The preset object may block the video capture device, causing some image areas of the initial video frame to be occupied by the preset object, which may lead to a decrease in recognition accuracy. For another example, target detection may be performed on multiple initial video frames respectively to obtain multiple target detection results. The object detection result may indicate the image area where the preset object is located in the video frame. According to the image area where the preset object is located, the proportion of the preset object in the initial video frame can be determined. If the proportion is greater than the preset proportion threshold, it may be determined that the initial video corresponds to the first scene information. As shown in FIG. 3A , the proportion of branches in the initial video frame 301 is large, and it can be determined that the initial video including the initial video frame 301 corresponds to the first scene information.

FIG. 3B is a schematic diagram of an initial video frame according to another embodiment of the present disclosure.

In an embodiment of the present disclosure, the second scene information is used to indicate that the average brightness of at least one video frame of the initial video is less than or equal to a preset average brightness threshold. For example, visibility is low at night or in hazy weather. In the case of deep learning models for recognition, the models can be trained using video samples from scenes with high visibility. Furthermore, in the inference stage, the recognition accuracy of the deep learning model for videos with low visibility is not high. For another example, multiple initial video frames may be converted into multiple grayscale images. The respective average luminances of the plurality of grayscale images are respectively determined. If the average brightness of at least one gray-scale image in the plurality of gray-scale images is less than or equal to the preset average brightness threshold, it may be determined that the initial video corresponds to the second scene information. As shown in FIG. 3B , the average brightness of the initial video frame 302 is low, and it can be determined that the initial video including the initial video frame 302 corresponds to the second scene information.

FIG. 3C is a schematic diagram of an initial video frame according to another embodiment of the present disclosure.

In the embodiment of the present disclosure, the third scene information is used to indicate that there is no image area corresponding to the road in at least one video frame of the initial video. For example, multiple video capture devices may be distributed across a street with a road area. The recognition accuracy is higher for videos with image regions corresponding to roads. This may be because in the absence of a road area, there may be more trees or ponds in the target area, and the action information of weather phenomena such as rainfall will be weakened. In one example, if there are trees and no roads in the target area during low rainfall, the trees may sway due to the rain. The motion information generated by the swing can affect the motion information of the rain itself. For another example, various methods may be used to determine whether there is a road area in the initial video frame. For example, image segmentation may be performed on the initial video frame to determine whether there is an image region corresponding to a road in the initial video frame. As shown in FIG. 3C , there is no image area corresponding to the road in the initial video frame 303 , and there is an area corresponding to the pond. It can be determined that the initial video including the initial video frame 303 corresponds to the third scene information.

In an embodiment of the present disclosure, at least one piece of scene information may further include preset scene information. For example, scene information other than the first scene information, the second scene information, and the third scene information may be used as preset scene information.

It can be understood that the scene information of the present disclosure has been described above, and the video processing strategy in the above operation S220 will be described below.

In some embodiments, the at least one video processing policy includes at least one of a first video processing policy, a second video processing policy, and a third video processing policy.

In the embodiment of the present disclosure, the first video processing strategy corresponds to the first scene information. The first video processing policy is used to indicate: use the adjusted video capture device to capture the target video. For example, as mentioned above, in the initial video corresponding to the first scene information, there may be preset objects with a large proportion. When processing the initial video using the first video processing strategy, the video capture device may be adjusted. The adjusted video is captured using the adjusted video capture device. Next, use the adjusted video as the target video to replace the initial video. Through the embodiments of the present disclosure, in the case of occlusion, the video can be re-collected, which helps to improve the accuracy of weather recognition.

In the embodiment of the present disclosure, the second video processing strategy corresponds to the second scene information. The second video processing strategy is used to indicate: delete at least one video frame whose average brightness is less than or equal to a preset average brightness threshold from the initial video. For example, as mentioned above, in the initial video corresponding to the second scene information, there may be initial video frames with low average brightness, and these initial video frames may be deleted. The video with these initial video frames removed can be used as the target video. Through the embodiments of the present disclosure, video frames with low brightness are deleted, which can reduce the computing pressure of weather recognition and help improve the accuracy of weather recognition.

In the embodiment of the present disclosure, the third video processing strategy corresponds to the third scene information. The third video processing strategy is used to indicate: crop the multiple initial video frames of the initial video according to a preset clipping strategy. For example, as mentioned above, there may not be a road area in the initial video corresponding to the third scene information. A preset clipping strategy can be set for at least one video capture device, so that the clipped video frame can be similar to the road area. The target video can be obtained according to the cropped video frame. Through the embodiments of the present disclosure, video frames are clipped, and weather recognition can be performed in various regions, which improves the application range of weather recognition and helps to improve the accuracy of weather recognition.

It can be understood that the video processing strategy of the present disclosure has been described above, and the first identification process in operation S230 will be described below.

In some embodiments, videos may have temporal dimension information compared to images. Various deep learning models can be utilized to implement the first recognition process.

In the embodiment of the present disclosure, a recurrent neural network (Recurrent Neural Network, RNN) may be used for time series modeling to implement the first recognition process. For example, the first recognition process can be realized by using an attention-based long-short-term memory network (AttentionLSTM). It can be appreciated that recurrent neural networks can process video features.

In the embodiments of the present disclosure, a three-dimensional neural network may also be used to extract timing information of a video. For example, the fast (Fast) network and the slow (Slow) network of the SlowFast model can be used to obtain the action information and appearance information (information that can be directly obtained from the video) of the video respectively. It can be understood that the fast and slow models can be used to perform video classification tasks.

In the embodiments of the present disclosure, other neural networks may also be used to extract the timing information of the video. For example, time sequence information may be extracted by using a temporal segment network (Temporal Segment Networks, TSN) or a temporal shift model (Temporal ShiftModule, TSM). Different from the time segmentation network, the time shift model can improve the recognition accuracy without increasing the amount of calculation, and is more suitable for weather recognition scenarios. After training, the accuracy of weather recognition based on the time-shift model can exceed that of the fast-slow model. In addition, in the inference stage, the time-shift model has faster inference speed and stronger inference performance.

It can be understood that the manner of implementing the first identification process has been described above, and the above operation S230 will be further described below.

In some embodiments, performing first recognition processing on at least one target video respectively, and obtaining at least one first recognition result may include: determining the collection time of at least one target video according to the collection time of at least one initial video. According to the collection time of at least one target video, at least one first identification threshold is determined.

In the embodiment of the present disclosure, when the acquisition time is within the first period, the first preset value is used as the first identification threshold. For example, the first time period may be an early morning time period. For another example, at 6 o'clock in the morning, two first initial videos may be collected. The collection time of both is 6 o'clock in the morning. After the first first initial video is processed by using the related video processing strategy, the first first target video can be obtained. After the second first initial video is processed by using the related video processing strategy, the second first target video can be obtained. The acquisition time of the two first target videos may also be exactly 6 o'clock in the morning. The first preset value can be used as the first identification threshold.

In the embodiment of the present disclosure, when the acquisition time is within the second period, the second preset value is used as the first identification threshold. For example, the second time period may be an evening time period. For another example, at 20 o'clock in the afternoon, two second initial videos may be collected. The collection time of both is 20 o'clock in the evening. After the first second initial video is processed by using the related video processing strategy, the first second target video can be obtained. After the second second initial video is processed by using the related video processing strategy, the second second target video can be obtained. The acquisition time of the two second target videos may also be exactly 17:00 in the afternoon. The second preset value can be used as the first identification threshold. It can be understood that the first preset value may be different from the second preset value. The lighting conditions at different times are also different. According to the embodiments of the present disclosure, the recognition thresholds are different depending on the acquisition time of the video. Therefore, under different time periods and different lighting conditions, weather recognition can be performed efficiently and accurately, further improving the ability of cross-domain weather recognition.

In some embodiments, performing first recognition processing on at least one target video to obtain at least one first recognition result may include: performing first recognition processing on at least one target video to obtain at least one first recognition value. According to at least one first identification value and at least one first identification threshold, at least one first identification result is obtained.

In an embodiment of the present disclosure, the first identification result is used to indicate whether there is a preset weather phenomenon in the target area.

In the embodiment of the present disclosure, if the first identification value is greater than the first identification threshold, it may be determined that there is a preset weather phenomenon in the target area.

It can be understood that some manners of weather recognition under different lighting conditions have been described above, and the above operation S240 will be further described below.

In some embodiments, there may be M at least one first recognition result. For example, M is an integer greater than or equal to 1.

In some embodiments, according to at least one first identification result, obtaining the target weather identification result corresponding to the target area may include: in response to determining that N first identification results indicate that there is a preset weather phenomenon in the target area, after determining and N When there are preset image regions in the target video frames of the N target videos respectively corresponding to the first recognition results, the N target videos are respectively subjected to the second recognition process to obtain N second recognition results.

For example, N is an integer greater than or equal to 1 and less than M.

For another example, the preset image area may correspond to stagnant water, tree shade, etc. in the target area. Image segmentation may be performed on the target video frame to determine whether there is an image region corresponding to areas such as ponding water or tree shade in the target video frame.

For another example, in the case that there are preset image regions in the target video frames of the N target videos, another trained model may be used to perform the second recognition processing on the N target videos respectively to obtain the second recognition results. The second recognition result may indicate whether a preset weather phenomenon exists in the target area. Through the embodiments of the present disclosure, it is possible to perform weather recognition more accurately when there is ponding or tree shade in the target area.

In some embodiments, obtaining the target weather recognition result corresponding to the target area according to at least one first recognition result may include: obtaining the target weather recognition result according to N second recognition results and M-N first recognition results. For example, among these identification results, if the number of identification results indicating that there is a preset weather phenomenon is greater than the number of identification results indicating that there is no preset weather phenomenon, one identification result indicating that there is a preset weather phenomenon can be used as the target weather identification result.

Other ways to obtain target weather recognition results will be further described below in conjunction with FIG. 4 .

Fig. 4 is a schematic diagram of obtaining target weather recognition results according to an embodiment of the present disclosure.

In an embodiment of the present disclosure, obtaining a target weather recognition result corresponding to the target area according to at least one first recognition result may include: determining a first number of first recognition results indicating that a preset weather phenomenon exists in the target area. A second number of first identification results indicating absence of the predetermined weather phenomenon in the target area is determined. Any first recognition result corresponding to the larger value of the first quantity and the second quantity is used as the target weather recognition result.

As shown in FIG. 4 , the video capture device 410 can capture the first initial video. The first target video can be obtained after processing the first initial video by using relevant video processing strategies. The first recognition process is performed on the first target video to obtain the first first recognition result 411 . The video capture device 420 can capture the second initial video. After the second initial video is processed by the related video processing strategy, the second target video can be obtained. The first recognition process is performed on the second target video to obtain a second first recognition result 421 . The video capture device 430 can capture the third initial video. After processing the third initial video by using relevant video processing strategies, the third target video can be obtained. The first recognition process is performed on the third target video to obtain a third first recognition result 431 .

Taking the first recognition result 411 indicating that there is rainfall in the target area, the first recognition result 421 indicating that there is rain in the target area, and the first recognition result 431 indicating that there is no rain in the target area as an example, it can be determined that there is The first number of the first recognition result of the rainfall phenomenon is two. It may be determined that the second number of first recognition results indicating that no rain phenomenon exists in the target area is 1. The first number is greater than the second number, and any one of the first recognition result 411 or the first recognition result 421 may be used as the target weather recognition result 440 . The target weather recognition result 440 may indicate that there is a rain phenomenon in the target area.

FIG. 5 is a block diagram of a weather recognition device according to one embodiment of the present disclosure.

As shown in FIG. 5 , the apparatus 500 may include a determination module 510 , a processing module 520 , an identification processing module 530 and an obtaining module 540 .

A determining module 510, configured to determine at least one piece of scene information corresponding to at least one initial video. For example, at least one initial video is associated with a target region.

The processing module 520 is configured to use at least one video processing strategy corresponding to the at least one scene information to process at least one initial video to obtain at least one target video.

The recognition processing module 530 is configured to respectively perform first recognition processing on at least one target video to obtain at least one first recognition result.

The obtaining module 540 is configured to obtain a target weather recognition result corresponding to the target area according to at least one first recognition result.

In some embodiments, the initial video includes a plurality of initial video frames, at least one piece of scene information includes at least one of first scene information, second scene information, and third scene information, and the first scene information is used to characterize at least one of the initial video The proportion of the preset object in an initial video frame is greater than or equal to the preset proportion threshold; the second scene information is used to represent the average brightness of at least one initial video frame of the initial video is less than or equal to the preset average brightness threshold; the third scene The information is used to characterize that there is no image region corresponding to the road in at least one initial video frame of the initial video.

In some embodiments, at least one video processing strategy includes at least one of a first video processing strategy, a second video processing strategy, and a third video processing strategy, the first video processing strategy corresponds to the first scene information, and the first video processing strategy The strategy is used to indicate: use the adjusted video capture device to capture the adjusted video, and use the adjusted video to replace the original video; the second video processing strategy corresponds to the second scene information, and the second video processing strategy is used to indicate: from Delete at least one initial video frame whose average brightness is less than or equal to the preset average brightness threshold in the initial video; the third video processing strategy corresponds to the third scene information, and the third video processing strategy is used to indicate: according to the preset clipping strategy, the initial video multiple initial video frames for cropping.

In some embodiments, the recognition processing module includes: a first determination unit, configured to determine the acquisition time of at least one target video according to the acquisition time of at least one initial video; a second determination unit, configured to determine the acquisition time of at least one target video according to the acquisition time of at least one target video time, determine at least one first recognition threshold; the first recognition unit is used to perform first recognition processing on at least one target video respectively to obtain at least one first recognition value; the first obtaining unit is used to obtain at least one first recognition value based on at least one first recognition value and at least one first identification threshold to obtain at least one first identification result, wherein the first identification result is used to indicate whether there is a preset weather phenomenon in the target area.

In some embodiments, at least one first recognition result is M, and M is an integer greater than or equal to 1, and the obtaining module includes: a second recognition unit, configured to respond to determining that the N first recognition results indicate that there are Preset weather phenomena, when it is determined that there is a preset image area in the respective target video frames of the N target videos corresponding to the N first recognition results, the N target videos are respectively subjected to the second recognition process to obtain N second identification results, wherein N is an integer greater than or equal to 1 and less than M; the second obtaining unit is configured to obtain target weather identification results according to the N second identification results and the M-N first identification results.

In some embodiments, the obtaining module includes: a third determining unit, configured to determine the first number of first recognition results indicating that there is a preset weather phenomenon in the target area; a fourth determining unit, configured to determine that indicating that there is no preset weather phenomenon in the target area There is a second number of first recognition results of preset weather phenomena; a third obtaining unit is configured to use any first recognition result corresponding to the larger value of the first number and the second number as the target weather recognition result .

In the technical solution of this disclosure, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved are all in compliance with relevant laws and regulations, and do not violate public order and good customs.

According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

FIG. 6 shows a schematic block diagram of an example electronic device 600 that may be used to implement embodiments of the present disclosure. Electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are by way of example only, and are not intended to limit implementations of the disclosure described and/or claimed herein.

As shown in FIG. 6, the device 600 includes a computing unit 601 that can execute according to a computer program stored in a read-only memory (ROM) 602 or loaded from a storage unit 608 into a random-access memory (RAM) 603. Various appropriate actions and treatments. In the RAM 603, various programs and data necessary for the operation of the device 600 can also be stored. The calculation unit 601 , the ROM 602 and the RAM 603 are connected to each other through a bus 604 . An input/output (I/O) interface 605 is also connected to the bus 604 .

Multiple components in the device 600 are connected to the I/O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, an optical disk, etc. ; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 609 allows the device 600 to exchange information/data with other devices over a computer network such as the Internet and/or various telecommunication networks.

The computing unit 601 may be various general-purpose and/or special-purpose processing components having processing and computing capabilities. Some examples of computing units 601 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 executes various methods and processes described above, such as a weather recognition method. For example, in some embodiments, the weather identification method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 608 . In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 600 via the ROM 602 and/or the communication unit 609 . When the computer program is loaded into RAM 603 and executed by computing unit 601, one or more steps of the weather recognition method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to execute the weather recognition method in any other suitable manner (for example, by means of firmware).

Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips Implemented in a system of systems (SOC), complex programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpreted on a programmable system including at least one programmable processor, the programmable processor Can be special-purpose or general-purpose programmable processor, can receive data and instruction from storage system, at least one input device, and at least one output device, and transmit data and instruction to this storage system, this at least one input device, and this at least one output device an output device.

Program codes for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special purpose computer, or other programmable data processing devices, so that the program codes, when executed by the processor or controller, make the functions/functions specified in the flow diagrams and/or block diagrams Action is implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer discs, hard drives, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) display or an LCD (liquid crystal display)) for displaying information to the user and a keyboard and pointing device (eg, a mouse or a trackball) through which the user may provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and can be in any form (including Acoustic input, speech input or, tactile input) to receive input from the user.

The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., as a a user computer having a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system can be interconnected by any form or medium of digital data communication, eg, a communication network. Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN) and the Internet.

A computer system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, each step described in the present disclosure may be executed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in the present disclosure can be achieved, no limitation is imposed herein.

The specific implementation manners described above do not limit the protection scope of the present disclosure. It should be apparent to those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made depending on design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims (15)

1. A weather identification method, comprising:

determining at least one scene information corresponding to at least one initial video, wherein at least one of the initial videos is related to a target area;

respectively processing at least one initial video by utilizing at least one video processing strategy corresponding to at least one scene information to obtain at least one target video;

respectively carrying out first identification processing on at least one target video to obtain at least one first identification result; and

and obtaining a target weather identification result corresponding to the target area according to at least one first identification result.

2. The method of claim 1, wherein the initial video comprises a plurality of initial video frames, at least one of the scene information comprises at least one of first scene information, second scene information, and third scene information,

The first scene information is used for representing that the duty ratio of a preset object in at least one initial video frame of the initial video is larger than or equal to a preset duty ratio threshold value;

the second scene information is used for representing that the average brightness of at least one initial video frame of the initial video is smaller than or equal to a preset average brightness threshold value;

the third scene information is used for representing that no image area corresponding to the road exists in at least one initial video frame of the initial video.

3. The method of claim 2, wherein at least one of the video processing policies comprises at least one of a first video processing policy, a second video processing policy, and a third video processing policy,

the first video processing policy corresponds to the first scene information, and the first video processing policy is used for indicating: collecting the adjusted video by using the adjusted video collecting equipment, and replacing the initial video by using the adjusted video;

the second video processing policy corresponds to the second scene information, and the second video processing policy is used for indicating: deleting at least one initial video frame with average brightness less than or equal to a preset average brightness threshold value from the initial video;

The third video processing policy corresponds to the third scene information, and the third video processing policy is used for indicating: and clipping the plurality of initial video frames of the initial video according to a preset clipping strategy.

4. The method of claim 1, wherein the performing the first recognition processing on the at least one target video to obtain at least one first recognition result includes:

determining the acquisition time of at least one target video according to the acquisition time of at least one initial video;

determining at least one first recognition threshold according to the acquisition time of at least one target video;

respectively carrying out first identification processing on at least one target video to obtain at least one first identification value;

and obtaining at least one first identification result according to at least one first identification value and at least one first identification threshold, wherein the first identification result is used for indicating whether a preset weather phenomenon exists in the target area.

5. The method of claim 1, wherein at least one of the first recognition results is M, M is an integer greater than or equal to 1,

the obtaining, according to at least one first recognition result, a target weather recognition result corresponding to the target area includes:

In response to determining that N first recognition results indicate that a preset weather phenomenon exists in the target area, respectively performing second recognition processing on N target videos under the condition that the fact that preset image areas exist in target video frames of N target videos respectively corresponding to the N first recognition results is determined, so as to obtain N second recognition results, wherein N is an integer which is greater than or equal to 1 and smaller than M;

and obtaining the target weather identification result according to the N second identification results and the M-N first identification results.

6. The method of claim 1, wherein the obtaining, according to at least one of the first recognition results, a target weather recognition result corresponding to the target area includes:

determining a first number of the first recognition results indicating the existence of a preset weather phenomenon in the target area;

determining a second number of first recognition results indicating that the preset weather phenomenon does not exist in the target area;

and taking any one of the first recognition results corresponding to the larger value in the first quantity and the second quantity as the target weather recognition result.

7. A weather identification device, comprising:

A determining module, configured to determine at least one scene information corresponding to at least one initial video, where at least one initial video is related to a target area;

the processing module is used for respectively processing at least one initial video by utilizing at least one video processing strategy corresponding to at least one scene information to obtain at least one target video;

the identification processing module is used for respectively carrying out first identification processing on at least one target video to obtain at least one first identification result; and

and the obtaining module is used for obtaining a target weather identification result corresponding to the target area according to at least one first identification result.

8. The apparatus of claim 7, wherein the initial video comprises a plurality of initial video frames, at least one of the scene information comprises at least one of first scene information, second scene information, and third scene information,

the first scene information is used for representing that the duty ratio of a preset object in at least one initial video frame of the initial video is larger than or equal to a preset duty ratio threshold value;

the second scene information is used for representing that the average brightness of at least one initial video frame of the initial video is smaller than or equal to a preset average brightness threshold value;

The third scene information is used for representing that no image area corresponding to the road exists in at least one initial video frame of the initial video.

9. The apparatus of claim 8, wherein at least one of the video processing policies comprises at least one of a first video processing policy, a second video processing policy, and a third video processing policy,

the first video processing policy corresponds to the first scene information, and the first video processing policy is used for indicating: collecting the adjusted video by using the adjusted video collecting equipment, and replacing the initial video by using the adjusted video;

the second video processing policy corresponds to the second scene information, and the second video processing policy is used for indicating: deleting at least one initial video frame with average brightness less than or equal to a preset average brightness threshold value from the initial video;

the third video processing policy corresponds to the third scene information, and the third video processing policy is used for indicating: and clipping the plurality of initial video frames of the initial video according to a preset clipping strategy.

10. The apparatus of claim 7, wherein the identification processing module comprises:

The first determining unit is used for determining the acquisition time of at least one target video according to the acquisition time of at least one initial video;

the second determining unit is used for determining at least one first recognition threshold according to the acquisition time of at least one target video;

the first identification unit is used for respectively carrying out first identification processing on at least one target video to obtain at least one first identification value;

the first obtaining unit is used for obtaining at least one first identification result according to at least one first identification value and at least one first identification threshold value, wherein the first identification result is used for indicating whether a preset weather phenomenon exists in the target area.

11. The apparatus of claim 7, wherein at least one of the first recognition results is M, M being an integer greater than or equal to 1,

the obtaining module includes:

the second recognition unit is used for responding to the fact that N first recognition results indicate that a preset weather phenomenon exists in the target area, and respectively carrying out second recognition processing on N target videos under the condition that the fact that preset image areas exist in target video frames of N target videos respectively corresponding to the N first recognition results is determined, so that N second recognition results are obtained, wherein N is an integer which is larger than or equal to 1 and smaller than M;

The second obtaining unit is used for obtaining the target weather identification result according to the N second identification results and the M-N first identification results.

12. The apparatus of claim 7, wherein the obtaining means comprises:

a third determining unit configured to determine a first number of the first recognition results indicating the existence of a preset weather phenomenon in the target area;

a fourth determining unit configured to determine a second number of first recognition results indicating that the preset weather phenomenon does not exist in the target area;

a third obtaining unit configured to use, as the target weather identification result, any one of the first identification results corresponding to a larger value of the first number and the second number.

13. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 6.

14. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1 to 6.

15. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images