Parking lot monitoring method and device based on image recognition, equipment and medium

Abstract

The invention discloses a parking lot monitoring method, device and equipment based on image recognition and a medium, and the method comprises the steps: intercepting a video image from a video stream according to a parking monitoring period, extracting image features, obtaining environment features corresponding to the video image, and storing the environment features corresponding to the video image; and combining the environment features and the image features, inputting the combined features into an anomaly recognition model to obtain a monitoring recognition result of the video image, judging whether the monitoring recognition result is abnormal or not, if not, continuing to execute the step of intercepting the video image, and if yes, sending out alarm prompt information. According to the parking lot monitoring method, the video stream of the existing video monitor can be used for correspondingly intercepting the video image for monitoring analysis, so that the monitoring analysis is not limited by the display size of the video picture, and the clear video image intercepted in the original video stream can be analyzed; and the efficiency and the reliability of abnormal monitoring of the vehicles and the charging facilities in the parking lot are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a parking lot monitoring method, device, equipment and medium based on image recognition. Background Art

[0002] Since the batteries of new energy vehicles need to be charged, abnormal situations such as battery combustion and high-temperature fire of charging facilities may occur in the parking lot; the traditional technical method is usually to monitor the new energy vehicles parked in the parking lot by manual monitoring, but the number of service personnel is limited, and they can only monitor once at a long interval (such as half an hour or an hour), and in some cases, problems cannot be discovered in time. The parking lot is usually equipped with multiple video monitors. The monitoring personnel can monitor the new energy vehicles and related charging facilities parked in the parking lot through the video monitors, and summarize and report the problems found, so as to facilitate the service personnel to deal with the relevant abnormal problems in time to reduce losses. However, due to the large number of video monitors, the video screen displayed by each video monitor is small, and it is impossible to accurately identify the details in the video screen; and a small number of monitoring personnel cannot take into account many video screens at the same time, resulting in the inability to discover abnormal problems in time. Therefore, the existing technical method cannot accurately monitor the abnormal problems of the parking lot by manually viewing the video screen of the monitor. Summary of the invention

[0003] The embodiments of the present invention provide a parking lot monitoring method, device, equipment and medium based on image recognition, aiming to solve the problem in the prior art that abnormal problems in the parking lot cannot be accurately monitored by manually observing the video images of the monitor.

[0004] In a first aspect, an embodiment of the present invention provides a parking lot monitoring method based on image recognition, the method is applied to a management server, the management server establishes a network connection with at least one video monitor to realize data information transmission, the method comprises:

[0005] intercepting corresponding video images from the video stream input by the video monitor according to a preset parking monitoring cycle;

[0006] Extracting corresponding image features from each of the video images according to a preset image feature extraction model;

[0007] Acquire the environmental features corresponding to the video monitors to which the video images belong according to preset environmental feature acquisition rules;

[0008] Combining the image features and environmental features of each of the video images and inputting them into a preset anomaly recognition model to perform anomaly recognition on each of the video images to obtain a corresponding monitoring recognition result;

[0009] Determining whether there is an abnormality in the monitoring and recognition results of each of the video images;

[0010] If the monitoring recognition results of each of the video images are normal, return to the step of intercepting the corresponding video image from the video stream input by the video monitor according to the preset parking monitoring cycle;

[0011] If any of the monitoring and recognition results of the video images is abnormal, a corresponding alarm prompt message is issued.

[0012] In a second aspect, an embodiment of the present invention further provides a parking lot monitoring device based on image recognition, wherein the device is configured in a management server, the management server establishes a network connection with at least one video monitor to realize data information transmission, and the device applies the parking lot monitoring method based on image recognition as described in the first aspect above, and the device includes:

[0013] A video image capture unit, used to capture corresponding video images from the video stream input by the video monitor according to a preset parking monitoring cycle;

[0014] An image feature extraction unit, used to extract corresponding image features from each of the video images according to a preset image feature extraction model;

[0015] An environmental feature acquisition unit, used to acquire environmental features corresponding to the video monitor to which each of the video images belongs according to a preset environmental feature acquisition rule;

[0016] A monitoring recognition result acquisition unit, used for combining the image features and environmental features of each of the video images and inputting them into a preset abnormality recognition model to perform abnormality recognition on each of the video images to obtain a corresponding monitoring recognition result;

[0017] An abnormality judgment unit, used to judge whether there is an abnormality in the monitoring recognition result of each of the video images;

[0018] A return execution unit, configured to return to the step of intercepting corresponding video images from the video stream input by the video monitor according to a preset parking monitoring cycle if the monitoring identification results of each of the video images are normal;

[0019] The alarm prompt unit is used to issue a corresponding alarm prompt message if there is an abnormality in the monitoring and recognition result of any of the video images.

[0020] In a third aspect, an embodiment of the present invention further provides a computer device, wherein the device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0021] Memory, used to store computer programs;

[0022] The processor is used to implement the steps of the parking lot monitoring method based on image recognition as described in the first aspect when executing the program stored in the memory.

[0023] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the parking lot monitoring method based on image recognition as described in the first aspect are implemented.

[0024] The embodiment of the present invention provides a parking lot monitoring method, device, equipment and medium based on image recognition, the method comprising: intercepting a video image from a video stream according to a parking monitoring cycle and extracting image features, obtaining environmental features corresponding to the video image, combining the environmental features and the image features and inputting them into an abnormality recognition model to obtain a monitoring recognition result of the video image, judging whether the monitoring recognition result is abnormal, if there is no abnormality, continuing to execute the step of intercepting the video image, and if there is an abnormality, issuing an alarm prompt message. The above-mentioned parking lot monitoring method can monitor and analyze the intercepted video image corresponding to the video stream of an existing video monitor, so that it is not limited by the display size of the video screen, and can analyze the clearer video image intercepted from the original video stream, which greatly improves the efficiency and reliability of abnormal monitoring of vehicles and charging facilities in the parking lot, and can accurately monitor abnormal problems in the parking lot using the video stream of the video monitor. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0026] Figure 1 A schematic diagram of a flow chart of a parking lot monitoring method based on image recognition provided by an embodiment of the present invention;

[0027] Figure 2 A schematic diagram of an application scenario of a parking lot monitoring method based on image recognition provided by an embodiment of the present invention;

[0028] Figure 3 A schematic block diagram of a parking lot monitoring device based on image recognition provided by an embodiment of the present invention;

[0029] Figure 4 A schematic block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments 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.

[0031] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0032] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0033] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0034] See also Figure 1 , Figure 2 As shown in the figure, the embodiment of the present invention provides a parking lot monitoring method based on image recognition, which is applied to the management server 10. Figure 2 As shown, the management server 10 establishes a network connection with at least one video monitor 20 to realize the transmission of data information. The management server 10 is a server side installed in the parking lot for obtaining the video streams collected by each video monitor 20 and performing abnormal monitoring. The management server 10 can be a terminal device with video image processing and information display, such as a desktop computer, a monitoring management terminal, etc. The video monitor 20 is a surveillance camera device installed inside the parking lot, at the entrance of the parking lot, etc. The video monitor 20 can obtain the real-time monitoring screen to obtain the video stream and transmit it to the management server 10. Figure 1 As shown, the method includes steps S110 to S170.

[0035] S110. Capturing corresponding video images from the video stream input by the video monitor according to a preset parking monitoring cycle.

[0036] The corresponding video images are intercepted from the video stream input by the video monitor according to the preset parking monitoring cycle. The video monitor transmits the collected video screen to the management server in real time, and the management server receives a group of video streams, each group of video streams corresponding to a video monitor. The management server intercepts the corresponding video images from the video stream according to the preset parking monitoring cycle, and each group of video streams can intercept a corresponding video image in one monitoring. For example, the parking monitoring cycle can be set to 1 minute, and a video image is intercepted from each video stream every 1 minute.

[0037] S120: extract corresponding image features from each of the video images according to a preset image feature extraction model.

[0038] According to the preset image feature extraction model, the corresponding image features are extracted from each of the video images. Furthermore, image features can be extracted from each video image according to the image feature extraction model, and the image features can be used to characterize the features of the video image. Specifically, the image feature extraction model is composed of a convolution layer and a pooling layer, and the convolution layer and the pooling layer are alternately configured; the pixel values ​​of the pixels in the video image are upsampled by the convolution layer to obtain data values ​​with lower resolution, and the number of horizontal pixels and the number of vertical pixels are reduced after one upsampling process; the dimension of the data value obtained by upsampling is reduced by the pooling layer, that is, the data value is reduced in dimension, and the number of horizontal pixels remains unchanged and the number of vertical pixels is reduced after one dimensionality reduction process. For example, a video image with a resolution of 1920×1080 is extracted by the image feature extraction model, and finally a set of 1×256 arrays are obtained as image features; then the image features corresponding to each video image can be obtained through this image feature extraction rule.

[0039] S130: Acquire environmental features corresponding to the video monitors to which the video images belong according to preset environmental feature acquisition rules.

[0040] The environmental features corresponding to the video monitors to which each of the video images belongs are obtained according to the preset environmental feature acquisition rules. Further, the environmental features corresponding to each of the video images can be obtained according to the environmental feature acquisition rules. The environmental features are used to characterize the environment in which each of the video images is located. In this process, the video monitors to which the video images belong can be determined. According to the environmental feature acquisition rules, the environmental features corresponding to each of the video monitors can be obtained.

[0041] In a specific embodiment, step S130 specifically includes sub-steps: determining corresponding place type characteristics according to the installation place of the video monitor to which the video image belongs; acquiring matching lighting characteristics from the environmental characteristic acquisition rules according to the current time point and the installation place corresponding to the video image, as the place lighting characteristics corresponding to the video image; combining the place type characteristics and the place lighting characteristics of the video image into the environmental characteristics of the video image.

[0042] Specifically, the corresponding place type can be determined according to the installation place of the video monitor to which the video image belongs. The place type can be an entrance, a ground parking lot, an underground parking lot, an underground equipment room, etc. Each place type corresponds to a place type feature, and the place type feature is a value. If the place type corresponds to one or more installation places, the corresponding place type can be determined according to the installation place of the video monitor, and the place type feature can be obtained according to the place type corresponding to each video image.

[0043] Furthermore, according to the current time point and the installation site corresponding to the video image, the matching site lighting features are obtained from the environmental feature acquisition rules, and the site lighting features are used to characterize the lighting of the site where the video image is collected. Specifically, the lighting type of the installation site is obtained. The lighting types configured in the environmental feature acquisition rules include closed type, semi-closed type and open type. Each lighting type corresponds to one or more site types. The closed type includes underground parking lots, underground equipment rooms, etc., the semi-closed type includes entrances and exits, etc., and the open type includes ground parking lots, etc. The external environment light is determined according to the current time point. The intensity of sunlight corresponding to different time points is different, so the intensity of external environment light corresponding to different time points is different; the environmental feature acquisition rule also includes the external environment light intensity corresponding to different time points, such as the current time point is 8:00 to 16:29, the external environment light intensity is determined to be 10; the current time point is 16:30 to 17:59, the external environment light intensity is determined to be 8; the current time point is 18:00 to 19:29, the external environment light intensity is determined to be 5; the current time point is 19:30 to 5:29 the next day, the external environment light intensity is determined to be 2. If the lighting type of the installation site is a closed type, the site lighting feature corresponding to the installation site is determined to be "1"; if the lighting type of the installation site is a semi-closed type, the site lighting feature corresponding to the installation site is determined to be [r / 2+1], r is the external environment light intensity corresponding to the current time, and [x] is the calculation symbol for rounding up x. If the lighting type of the installation location is an open type, the location lighting feature corresponding to the installation location is determined to be the external environment lighting intensity corresponding to the current time.

[0044] By combining the place type features of the video image with the place lighting features, the environmental features of the video image can be obtained.

[0045] In a specific embodiment, after combining the place type feature and the place lighting feature of the video image into the environmental feature of the video image, it also includes: obtaining weather information corresponding to the current time point; obtaining weather features matching the weather information from the environmental feature acquisition rules; judging whether the installation place corresponding to the video image is an open place; if the installation place of the video image is an open place, adding the weather features to the environmental features of the video image; if the installation place of the video image is not an open place, adding pre-stored fill-in features to the environmental features of the video image.

[0046] Furthermore, in order to improve the richness of the environmental features of the video image, the weather information corresponding to the current time point can also be obtained, and the environmental feature acquisition rules also include weather features that match the weather information; for example, the weather feature corresponding to a sunny day in the environmental feature acquisition rules is "10", the weather feature corresponding to a cloudy day in the environmental feature acquisition rules is "8", the weather feature corresponding to a cloudy day in the environmental feature acquisition rules is "6", the weather feature corresponding to a rainy day in the environmental feature acquisition rules is "4", and the weather feature corresponding to a snowy day in the environmental feature acquisition rules is "2". Further, it is determined whether the installation place corresponding to the video image is an open place. If it is an open place, the weather features are added to the environmental features of the video image. If the installation place of the video image is not an open place, the fill-in feature is added to the environmental features of the video image. For example, the fill-in feature can be "0".

[0047] S140: Combining the image features and environmental features of each of the video images and inputting them into a preset anomaly recognition model to perform anomaly recognition on each of the video images to obtain a corresponding monitoring recognition result.

[0048] The image features and environmental features of each of the video images are combined and input into a preset anomaly recognition model to perform anomaly recognition on each of the video images to obtain a corresponding monitoring recognition result. Furthermore, the image features and environmental features of the video images can be combined, and the combined data information can be input into the anomaly recognition model; a set of data information corresponding to each video image is respectively input into the anomaly recognition model, and the anomaly recognition model can perform anomaly recognition on a set of data information of each video image, thereby obtaining a monitoring recognition result.

[0049] In a specific embodiment, step S140 specifically includes the sub-steps of: combining the image feature with the environmental feature to obtain a combined feature; inputting the feature value contained in the combined feature into the corresponding input node in the abnormality recognition model, so as to perform abnormality recognition through the abnormality recognition model to obtain a corresponding abnormality probability value as the corresponding monitoring recognition result.

[0050] Specifically, the image feature and the environmental feature can be combined, the image feature is a 1×M dimensional array (for example, a 1×256 dimensional array), and the environmental feature is composed of multiple feature values; after the combination, a 1×(M+N) array can be obtained, which is also the combined feature, and N is the number of feature values ​​in the environmental feature. The feature values ​​contained in the combined feature are input to the input nodes in the abnormal recognition model, and the number of feature values ​​contained in the combined feature is equal to the number of input nodes. The abnormal recognition model includes an intermediate layer and an output node, the intermediate layer includes at least one layer of intermediate nodes, and a layer of intermediate nodes is composed of multiple intermediate nodes. The intermediate layer performs correlation analysis on the feature values ​​input by the input node, so that the output node outputs the corresponding abnormal probability value, which is the probability value obtained by abnormal recognition to reflect the existence of abnormalities in the video image; the higher the abnormal probability value, the greater the possibility of the existence of abnormalities in the video image.

[0051] In a specific embodiment, before combining the image features and environmental features of each of the video images and inputting them into a preset abnormality recognition model, it also includes: training the initial abnormality recognition model according to a plurality of preset training image sets to obtain alternative recognition models corresponding to each of the training image sets; verifying each of the alternative recognition models according to a preset verification image set to obtain the accuracy of each alternative recognition model; and selecting an alternative recognition model with the highest accuracy as the trained abnormality recognition model.

[0052] Specifically, before using the abnormality recognition model for abnormality analysis, the initial abnormality recognition model can also be trained according to a plurality of pre-set training image sets. The initial abnormality recognition model is also an artificial intelligence model that is initially constructed and not trained. Then, there is only one initial abnormality recognition model constructed. The initial abnormality recognition model can be trained according to a plurality of training image sets respectively. The training image set contains a plurality of training images, and the number of training images in each training image set can be equal or unequal. Each training image corresponds to a set of combined features and an abnormal label. The combined features of the training images in the training image set are sequentially input into the initial abnormality recognition model, and the output value of the output node is obtained; the abnormal label of the training image is combined with the output value of the output node to perform a gradient descent operation, so as to optimize the parameter value in the initial abnormality recognition model based on the gradient descent operation. This parameter tuning process is also the training of the initial abnormality recognition model. The initial abnormality recognition model can be trained once by one training image, and the initial abnormality recognition model can be iteratively trained by multiple training images in the training image set. After each training image in the training image set completes the training of the initial abnormality recognition model, the candidate recognition model corresponding to the training image set can be obtained. Then each training image set can obtain a corresponding candidate recognition model.

[0053] Each candidate recognition model is verified according to the verification image set, which contains multiple verification images; each verification image corresponds to a set of combined features and an abnormal label. Specifically, the combined features of the verification images contained in the verification image set can be input into the candidate recognition model in sequence, and the output results of each verification image are obtained through the alternative recognition model, and it is determined whether the output results of the verification image match the abnormal label of the verification image. The ratio between the number of output results that match the abnormal label and the total number of verification images is obtained as the accuracy rate corresponding to each candidate recognition model, and an accuracy rate can be obtained for each candidate image. The candidate recognition model with the highest accuracy is selected as the trained abnormal recognition model for use.

[0054] S150: Determine whether there is any abnormality in the monitoring and recognition results of each of the video images.

[0055] Determine whether the monitoring and recognition results of each of the video images are abnormal. Further, the abnormal probability value in the monitoring and recognition results of the video images can be determined to distinguish whether the monitoring and recognition results are abnormal.

[0056] In a specific embodiment, step S150 specifically includes the steps of: determining whether the abnormal probability value in the monitoring identification result is not less than a pre-stored abnormal probability threshold, thereby determining whether there is an abnormality in the monitoring identification result.

[0057] Specifically, it can be determined whether the abnormal probability value in the monitoring and identification result is not less than a pre-stored abnormal probability threshold. For example, the abnormal probability threshold can be set to 60%; if the abnormal probability value in the monitoring and identification result is not less than the abnormal probability threshold, then it is determined that there is an abnormality in the monitoring and identification result; if the abnormal probability value in the monitoring and identification result is less than the abnormal probability threshold, then it is determined that there is no abnormality in the monitoring and identification result.

[0058] S160: If the monitoring and identification results of the video images are normal, return to the step of intercepting the corresponding video image from the video stream input by the video monitor according to the preset parking monitoring cycle.

[0059] If the monitoring and recognition results of the video images are all normal, the process returns to the step of intercepting the corresponding video image from the video stream input by the video monitor according to the preset parking monitoring cycle. If the monitoring and recognition results of the video images are all normal, the video images in the video stream are intercepted again according to the parking monitoring cycle for abnormality recognition, that is, the process returns to step S110.

[0060] S170: If any of the monitoring and recognition results of the video images is abnormal, a corresponding alarm prompt message is issued.

[0061] If the monitoring recognition result of any video image is abnormal, the corresponding alarm prompt information will be generated according to the installation location of the video monitor corresponding to the video image, and the alarm prompt information will include the installation location corresponding to the video image and the time when the prompt was generated. Users can view the alarm prompt information to obtain the specific location of the abnormality (that is, the installation location corresponding to the video image) and the time when the abnormal problem was discovered (that is, the time when the prompt was generated).

[0062] In a specific embodiment, step S170 further includes the following steps: generating an exception handling work order corresponding to the monitoring identification result; and storing the exception handling work order in a pre-configured monitoring work order table.

[0063] Furthermore, after the alarm prompt information is issued, the corresponding exception handling work order can be generated according to the monitoring identification results. The exception handling work order is the work order information used to record the generation of abnormal problems, the processing status and subsequent processing operations; the generated exception handling work order can be distributed to property service personnel, and the property service personnel can perform exception processing based on the exception handling work order. Furthermore, in order to integrate and record the exception handling work order and improve the efficiency of viewing and processing the exception handling work order, the generated exception handling work order can be stored in a pre-configured monitoring work order table. Then the user can quickly view the processing progress and information of the handler of each exception handling work order by viewing the monitoring work order table, thereby improving the efficiency of users viewing, processing and counting exception handling work orders.

[0064] In the parking lot monitoring method based on image recognition provided in the embodiment of the present invention, according to the parking monitoring cycle, video images are intercepted from the video stream and image features are extracted to obtain environmental features corresponding to the video images, and the environmental features and image features are combined and input into the abnormal recognition model to obtain the monitoring recognition results of the video images, and it is determined whether the monitoring recognition results are abnormal. If there is no abnormality, the step of intercepting the video image is continued, and if there is an abnormality, an alarm prompt message is issued. The above-mentioned parking lot monitoring method can monitor and analyze the intercepted video images corresponding to the video stream of the existing video monitor, so that it is not limited by the display size of the video screen, and the clearer video images intercepted from the original video stream can be analyzed, which greatly improves the efficiency and reliability of abnormal monitoring of vehicles and charging facilities in the parking lot, and can accurately monitor abnormal problems in the parking lot using the video stream of the video monitor.

[0065] The embodiment of the present invention also provides a parking lot monitoring device based on image recognition, which can be configured in a management server, and is used to execute any embodiment of the aforementioned parking lot monitoring method based on image recognition. Figure 3 , Figure 3 A schematic block diagram of a parking lot monitoring device based on image recognition provided by an embodiment of the present invention.

[0066] like Figure 3 As shown, the parking lot monitoring device 100 based on image recognition includes a video image capture unit 110, an image feature extraction unit 120, an environmental feature acquisition unit 130, a monitoring recognition result acquisition unit 140, an abnormality judgment unit 150, a return execution unit 160 and an alarm prompt unit 170.

[0067] The video image capture unit 110 is used to capture corresponding video images from the video stream input by the video monitor according to a preset parking monitoring cycle.

[0068] The image feature extraction unit 120 is used to extract corresponding image features from each of the video images according to a preset image feature extraction model.

[0069] The environment feature acquisition unit 130 is used to acquire the environment feature corresponding to the video monitor to which each of the video images belongs according to a preset environment feature acquisition rule.

[0070] The monitoring recognition result acquisition unit 140 is used to combine the image features and environmental features of each of the video images and input them into a preset abnormality recognition model to perform abnormality recognition on each of the video images to obtain a corresponding monitoring recognition result.

[0071] The abnormality determination unit 150 is used to determine whether there is an abnormality in the monitoring and recognition results of each of the video images.

[0072] Return to the execution unit 160, which is used to return to the step of intercepting the corresponding video image from the video stream input by the video monitor according to the preset parking monitoring cycle if the monitoring and identification results of each of the video images are normal.

[0073] The alarm prompt unit 170 is used to issue a corresponding alarm prompt message if there is an abnormality in the monitoring and recognition result of any of the video images.

[0074] The parking lot monitoring device based on image recognition provided in the embodiment of the present invention applies the parking lot monitoring method based on image recognition, and the method includes: intercepting video images from the video stream according to the parking monitoring cycle and extracting image features, obtaining environmental features corresponding to the video images, merging environmental features and image features and inputting them into the abnormal recognition model to obtain the monitoring recognition results of the video images, judging whether the monitoring recognition results are abnormal, if there is no abnormality, continuing to execute the step of intercepting the video image, and if there is an abnormality, issuing an alarm prompt message. The above-mentioned parking lot monitoring method can monitor and analyze the intercepted video images corresponding to the video stream of the existing video monitor, so as not to be limited by the display size of the video screen, and can analyze the clearer video images intercepted from the original video stream, which greatly improves the efficiency and reliability of abnormal monitoring of vehicles and charging facilities in the parking lot, and can accurately monitor abnormal problems in the parking lot using the video stream of the video monitor.

[0075] The above-mentioned parking lot monitoring device based on image recognition can be implemented in the form of a computer program, and the computer program can be used in Figure 4 Runs on the computer device shown.

[0076] See also Figure 4 , Figure 4 1 is a schematic block diagram of a computer device provided by an embodiment of the present invention. The computer device may be a management server for executing a parking lot monitoring method based on image recognition to obtain a video stream and perform abnormal monitoring.

[0077] See also Figure 4 The computer device 500 includes a processor 502 , a memory and a communication interface 505 connected via a communication bus 501 , wherein the memory may include a storage medium 503 and an internal memory 504 .

[0078] The storage medium 503 may store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 may execute a parking lot monitoring method based on image recognition, wherein the storage medium 503 may be a volatile storage medium or a non-volatile storage medium.

[0079] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0080] The internal memory 504 provides an environment for the operation of the computer program 5032 in the storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a parking lot monitoring method based on image recognition.

[0081] The communication interface 505 is used for network communication, such as providing data information transmission, etc. Those skilled in the art will appreciate that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device 500 to which the solution of the present invention is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0082] The processor 502 is used to run the computer program 5032 stored in the memory to implement the corresponding functions in the above-mentioned parking lot monitoring method based on image recognition.

[0083] Those skilled in the art will understand that Figure 4 The embodiments of the computer device shown in the figure do not constitute a limitation on the specific composition of the computer device. In other embodiments, the computer device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, in some embodiments, the computer device may only include a memory and a processor. In such embodiments, the structure and function of the memory and the processor are the same as those of the embodiment of the present invention. Figure 4 The embodiments shown are consistent and will not be described again here.

[0084] It should be understood that in the embodiment of the present invention, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0085] In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps included in the above-mentioned parking lot monitoring method based on image recognition are implemented.

[0086] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the above-described equipment, devices and units can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. Those of ordinary skill in the art can appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0087] In the several embodiments provided by the present invention, it should be understood that the disclosed equipment, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. Units with the same function may also be combined into one unit. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices or units, or may be an electrical, mechanical or other form of connection.

[0088] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0089] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0090] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a computer-readable storage medium, including a number of instructions for 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 computer-readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a magnetic disk or an optical disk.

[0091] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A parking lot monitoring method based on image recognition, the method is applied to a management server, the management server establishes a network connection with at least one video monitor to realize data information transmission, characterized in that: The method comprises: intercepting corresponding video images from the video stream input by the video monitor according to a preset parking monitoring cycle; Extracting corresponding image features from each of the video images according to a preset image feature extraction model; Acquire the environmental features corresponding to the video monitors to which the video images belong according to preset environmental feature acquisition rules; Combining the image features and environmental features of each of the video images and inputting them into a preset anomaly recognition model to perform anomaly recognition on each of the video images to obtain a corresponding monitoring recognition result; Determining whether there is an abnormality in the monitoring and recognition results of each of the video images; If the monitoring recognition results of each of the video images are normal, return to the step of intercepting the corresponding video image from the video stream input by the video monitor according to the preset parking monitoring cycle; If any of the monitoring and recognition results of the video images is abnormal, a corresponding alarm prompt message is issued.

2. The parking lot monitoring method based on image recognition according to claim 1, characterized in that: The step of acquiring the environmental features corresponding to the video monitors to which the video images belong according to the preset environmental features acquisition rules includes: Determine a corresponding location type feature according to an installation location of a video monitor to which the video image belongs; Acquire matching lighting features from the environmental feature acquisition rule according to the current time point and the installation location corresponding to the video image as the location lighting features corresponding to the video image; The place type feature and the place lighting feature of the video image are combined into the environment feature of the video image.

3. The parking lot monitoring method based on image recognition according to claim 2 is characterized in that: After combining the place type feature and the place lighting feature of the video image into the environment feature of the video image, the method further includes: Obtaining weather information corresponding to the current time point; Acquire weather characteristics matching the weather information from the environmental characteristics acquisition rule; Determining whether the installation place corresponding to the video image is an open place; If the installation location of the video image is an open place, adding the weather characteristics to the environmental characteristics of the video image; If the installation place of the video image is not an open place, the pre-stored filling feature is added to the environmental feature of the video image.

4. The parking lot monitoring method based on image recognition according to any one of claims 1 to 3, characterized in that: The combining of the image features and the environmental features of each of the video images and inputting the combined features into a preset anomaly recognition model to perform anomaly recognition on each of the video images to obtain a corresponding monitoring recognition result includes: Combining the image feature with the environment feature to obtain a combined feature; The feature value contained in the combined feature is input into the corresponding input node in the abnormality recognition model, so as to perform abnormality recognition through the abnormality recognition model to obtain the corresponding abnormality probability value as the corresponding monitoring recognition result.

5. The parking lot monitoring method based on image recognition according to claim 4 is characterized in that: Before combining the image features and environmental features of each of the video images and inputting them into a preset abnormality recognition model, the method further includes: The initial abnormality recognition model is trained according to a plurality of preset training image sets respectively to obtain candidate recognition models corresponding to the respective training image sets; Verifying each candidate recognition model according to a preset verification image set to obtain the accuracy of each candidate recognition model; The candidate recognition model with the highest accuracy is selected as the trained anomaly recognition model.

6. The parking lot monitoring method based on image recognition according to any one of claims 1 to 3, characterized in that: The determining whether the monitoring recognition result of each of the video images is abnormal includes: It is determined whether the abnormal probability value in the monitoring and identification result is not less than a pre-stored abnormal probability threshold value, thereby determining whether there is an abnormality in the monitoring and identification result.

7. The parking lot monitoring method based on image recognition according to claim 6 is characterized in that: After the corresponding alarm prompt information is issued, the method further includes: Generate an exception handling work order corresponding to the monitoring identification result; The exception handling work order is stored in a pre-configured monitoring work order table.

8. A parking lot monitoring device based on image recognition, characterized in that: The device is configured in a management server, and the management server establishes a network connection with at least one video monitor to realize data information transmission. The device applies the parking lot monitoring method based on image recognition according to any one of claims 1 to 7, and the device includes: A video image capture unit, used to capture corresponding video images from the video stream input by the video monitor according to a preset parking monitoring cycle; An image feature extraction unit, used to extract corresponding image features from each of the video images according to a preset image feature extraction model; An environmental feature acquisition unit, used to acquire environmental features corresponding to the video monitor to which each of the video images belongs according to a preset environmental feature acquisition rule; A monitoring recognition result acquisition unit, used for combining the image features and environmental features of each of the video images and inputting them into a preset abnormality recognition model to perform abnormality recognition on each of the video images to obtain a corresponding monitoring recognition result; An abnormality judgment unit, used to judge whether there is an abnormality in the monitoring recognition result of each of the video images; A return execution unit, configured to return to the step of intercepting corresponding video images from the video stream input by the video monitor according to a preset parking monitoring cycle if the monitoring identification results of each of the video images are normal; The alarm prompt unit is used to issue a corresponding alarm prompt message if there is an abnormality in the monitoring and recognition result of any of the video images.

9. A computer device, characterized in that: The device includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the parking lot monitoring method based on image recognition described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A 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 parking lot monitoring method based on image recognition as described in any one of claims 1 to 7 are implemented.

Images