Intelligent parking lot identification system and identification method

Abstract

The present application provides an intelligent identification system and identification method for a smart parking lot. The method first collects an access control vehicle image set of the current access control vehicle; performs feature extraction on the access control vehicle image set, and determines whether the current access control vehicle is an identified vehicle according to the access control vehicle features; if the current access control vehicle is an identified vehicle, obtains a developed image corresponding to the current access control vehicle, identifies the current access control vehicle according to the developed image, extracts identification vectors from the developed image and the access control vehicle image set, and obtains a vehicle identification enthalpy vector; performs empty sequence extraction on the access control vehicle image set to obtain an empty correction sequence, and updates the license plate recognition entropy of the identified vehicle according to the empty correction sequence and the vehicle identification enthalpy vector; and performs development identification on the remaining access control vehicles in the hospital parking lot according to the updated license plate recognition entropy, which greatly improves the vehicle identification speed in the hospital underground parking lot and reduces the traffic pressure in the hospital parking lot.

Description

Technical Field

[0001] The present application relates to the field of vehicle identification technology, and more specifically, to an intelligent identification system and identification method for a smart parking lot. Background Art

[0002] Against the backdrop of high-quality development of hospitals, the government has continuously increased the scale of high-quality medical resource construction, expanded the supply of high-quality medical resources, and provided high-level medical and health services to citizens. During the transition period of high-quality medical resource construction, hospital operations and renovation and expansion coexist. Especially in the context of rapid resumption of work and production in hospitals across the country, patients' demand for medical treatment has increased dramatically, and parking demand has also been increasing. The lack of parking resources and traffic pressure continue to affect the medical experience of patients.

[0003] In hospital parking lots, vehicles are usually identified and released after their license plates are recognized. However, in the prior art, due to weather reasons or camera aging, some information is inevitably missing when collecting vehicle images at a long distance for license plate number identification, resulting in the vehicle's license plate number identification distance being too short and the vehicle identification speed being slow. When parking spaces in the hospital are tight, it is easy to cause vehicle queues to pile up, greatly increasing the traffic pressure in the hospital's underground parking spaces. Summary of the invention

[0004] The present application provides an intelligent identification system and identification method for a smart parking lot to increase the vehicle identification speed in a hospital underground parking lot and reduce the traffic pressure in the parking lot.

[0005] On the first aspect, the present application provides a smart identification method for a smart parking lot. The method can be executed by a network device, or can also be executed by a chip configured in the network device, and the present application does not limit this.

[0006] Specifically, the method includes:

[0007] Start the hospital parking lot intelligent identification and collect the access control vehicle image set of the current access control vehicles;

[0008] Extracting features from the access control vehicle image set to obtain access control vehicle features, and determining whether the current access control vehicle is an identified vehicle based on the access control vehicle features;

[0009] When the current access control vehicle is an identified vehicle, a developed image corresponding to the current access control vehicle is obtained, and after the current access control vehicle is identified according to the developed image, an identification vector is extracted from the developed image and the access control vehicle image set to obtain a vehicle identification enthalpy vector;

[0010] Extracting an open space sequence from the access control vehicle image set to obtain an open space correction sequence, and updating the license plate recognition entropy of the marked vehicle according to the open space correction sequence and the vehicle identification enthalpy vector;

[0011] The remaining access-controlled vehicles in the hospital parking lot are identified using the updated license plate recognition entropy.

[0012] In combination with the first aspect, in certain implementations of the first aspect, extracting features from the access control vehicle image set to obtain access control vehicle features specifically includes:

[0013] Performing color histogram conversion on the access control vehicle image set to obtain color features of the access control vehicle image set;

[0014] After the access control vehicle image set is standardized, a feature vector of the access control vehicle image set is extracted through a convolutional neural network;

[0015] The color features and feature vectors of the access control vehicle image set are combined to obtain access control vehicle features.

[0016] In combination with the first aspect, in some implementations of the first aspect, determining whether the current access control vehicle is an identified vehicle according to the access control vehicle feature specifically includes:

[0017] Acquire the access control vehicle characteristics;

[0018] Acquire all the image features stored in the vehicle information system, and measure the similarity with the access control vehicle features respectively, to obtain the similarity between the access control vehicle features and each image feature;

[0019] When the similarity between the developed feature and the access control vehicle feature is higher than a preset similarity threshold, the current access control vehicle is determined to be a marked vehicle; otherwise, the current access control vehicle is determined to be an unmarked vehicle.

[0020] In combination with the first aspect, in certain implementations of the first aspect, extracting identification vectors from the developed image and the access control vehicle image set to obtain the vehicle identification enthalpy vector specifically includes:

[0021] aligning the scales of the access control vehicle images in the access control vehicle image set according to the developed image to obtain a corrected vehicle image set;

[0022] The vehicle identification enthalpy bases corresponding to the respective corrected vehicle images in the corrected vehicle image set are determined according to the developed images, and the vehicle identification enthalpy vectors are determined according to the vehicle identification enthalpy bases corresponding to the respective corrected vehicle images.

[0023] In combination with the first aspect, in certain implementations of the first aspect, performing development and identification of remaining access-controlled vehicles in the hospital parking lot using the updated license plate recognition entropy specifically includes:

[0024] Obtain updated license plate recognition entropy; obtain a set of access control vehicle images of the current access control vehicle;

[0025] Performing development fusion on the access control vehicle image set of the current access control vehicle according to the updated license plate recognition entropy to obtain a license plate development image;

[0026] Perform character segmentation and label recognition on the label development image

[0027] In combination with the first aspect, in certain implementations of the first aspect, identifying the current access control vehicle based on the developed image specifically includes: cropping the developed image to obtain the license plate area image of the current access control vehicle; performing character segmentation and sign recognition on the license plate area image to obtain the license plate characters of the current access control vehicle.

[0028] In combination with the first aspect, in certain implementations of the first aspect, collecting the access control vehicle image set of the current access control vehicle is performed by collecting the access control vehicle images of the current access control vehicle at equal intervals through a high-definition camera to obtain the access control vehicle image set of the current access control vehicle.

[0029] In a second aspect, the present application provides an intelligent identification system for a smart parking lot, including a vehicle identification unit, wherein the vehicle identification unit includes:

[0030] A collection module is used to start the hospital parking lot intelligent identification and collect the access control vehicle image set of the current access control vehicle;

[0031] A processing module extracts features from the access control vehicle image set to obtain access control vehicle features, and determines whether the current access control vehicle is a marked vehicle based on the access control vehicle features;

[0032] The processing module is further configured to, when the current access control vehicle is an identified vehicle, obtain a developed image corresponding to the current access control vehicle, identify the current access control vehicle according to the developed image, extract an identification vector from the developed image and the access control vehicle image set, and obtain a vehicle identification enthalpy vector;

[0033] The processing module extracts an open sequence from the access control vehicle image set to obtain an open correction sequence, and updates the license plate recognition entropy of the marked vehicle according to the open correction sequence and the vehicle identification enthalpy vector;

[0034] The execution module uses the updated license plate recognition entropy to develop and identify the remaining access control vehicles in the hospital parking lot.

[0035] In a third aspect, the present application provides a computer terminal device, which includes a memory and a processor, wherein the memory stores a code, and the processor is configured to obtain the code and execute the above-mentioned intelligent identification method for a smart parking lot.

[0036] In a fourth aspect, the present application provides a computer-readable storage medium, which stores at least one computer program, and the computer program is loaded and executed by a processor to implement the operations performed by the above-mentioned intelligent identification method for a smart parking lot.

[0037] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0038] In an intelligent identification system and identification method for a smart parking lot provided by the present application, first, an access control vehicle image set of the current access control vehicle is collected; feature extraction is performed on the access control vehicle image set to obtain access control vehicle features, and whether the current access control vehicle is an identified vehicle is determined based on the access control vehicle features; when the current access control vehicle is an identified vehicle, a developed image corresponding to the current access control vehicle is obtained, and after the current access control vehicle is identified based on the developed image, identification vectors are extracted from the developed image and the access control vehicle image set to obtain a vehicle identification enthalpy vector; an empty sequence is extracted from the access control vehicle image set to obtain an empty correction sequence, and the license plate recognition entropy of the identified vehicle is updated based on the empty correction sequence and the vehicle identification enthalpy vector; and the remaining access control vehicles in the hospital parking lot are developed and identified based on the updated license plate recognition entropy.

[0039] In the present application, the license plate recognition entropy of the identified vehicle is updated according to the empty correction sequence and the vehicle identification enthalpy vector. The updated license plate recognition entropy records the environmental interference characteristics of the real-time process image of the identified vehicle from far to near. The image of the unidentified vehicle is enhanced and identified according to the updated license plate recognition entropy. The license plate features of the current access control vehicle can be highlighted when the vehicle is crowded, so that the license plate area is more prominent in the image, thereby improving the recognition algorithm's positioning and recognition capabilities for the license plate, and helping to accurately identify the license plate number at a longer distance or in more blurred conditions. In addition, by correcting the vehicle identification enthalpy vector, the ability of the vehicle identification to adapt to images of different empty spaces is improved, which greatly improves the vehicle identification speed in the hospital's underground parking lot and reduces the traffic pressure in the hospital parking lot. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is an exemplary flow chart of a smart parking lot intelligent identification method according to some embodiments of the present application;

[0041] Figure 2is an exemplary flow chart for obtaining a vehicle identification enthalpy vector in some embodiments of the present application;

[0042] Figure 3 is a schematic diagram of exemplary hardware and / or software of a vehicle identification unit according to some embodiments of the present application;

[0043] Figure 4 It is a structural diagram of a computer terminal device for implementing a smart parking lot intelligent identification method as shown in some embodiments of the present application. DETAILED DESCRIPTION

[0044] The present application collects an access control vehicle image set of the current access control vehicle; performs feature extraction on the access control vehicle image set to obtain access control vehicle features, and determines whether the current access control vehicle is an identified vehicle based on the access control vehicle features; when the current access control vehicle is an identified vehicle, obtains a developed image corresponding to the current access control vehicle, identifies the current access control vehicle based on the developed image, extracts identification vectors from the developed image and the access control vehicle image set to obtain a vehicle identification enthalpy vector; performs open space sequence extraction on the access control vehicle image set to obtain an open space correction sequence, and updates the license plate recognition entropy of the identified vehicle based on the open space correction sequence and the vehicle identification enthalpy vector; and performs development and identification of the remaining access control vehicles in the hospital parking lot based on the updated license plate recognition entropy, thereby greatly improving the vehicle identification speed in the hospital underground parking lot and reducing the traffic pressure in the hospital parking lot.

[0045] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of a smart parking lot smart identification method according to some embodiments of the present application. The smart parking lot smart identification method 100 mainly includes the following steps:

[0046] In step S101, the hospital parking lot intelligent identification is started to collect a set of access control vehicle images of the current access control vehicle.

[0047] It should be noted that the current access control vehicle is the current vehicle at the hospital access control entrance. In specific implementation, a high-definition camera located at the hospital access control passage can be used to collect access control vehicle images of the current access control vehicle at equal intervals to obtain an access control vehicle image set of the current access control vehicle. The access control vehicle image set contains the collection time corresponding to each access control vehicle image. In some embodiments, other devices or equipment that can realize vehicle image collection can also be used, which is not limited here.

[0048] In step S102, feature extraction is performed on the access control vehicle image set to obtain access control vehicle features, and it is determined whether the current access control vehicle is a marked vehicle based on the access control vehicle features.

[0049] Preferably, in some embodiments, feature extraction is performed on the access control vehicle image set to obtain access control vehicle features in the following manner:

[0050] Performing color histogram conversion on the access control vehicle image set to obtain color features of the access control vehicle image set;

[0051] After the access control vehicle image set is standardized, a feature vector of the access control vehicle image set is extracted through a convolutional neural network;

[0052] The color features and feature vectors of the access control vehicle image set are combined to obtain access control vehicle features.

[0053] In specific implementation, each access control vehicle image in the access control vehicle image set can be converted from the RGB color space to the HSV color space, and then the color histogram in the HSV color space is calculated, and then the histogram features of different channels are merged to obtain a complete color feature vector as the color feature of the access control vehicle image set.

[0054] Optionally, in some embodiments, standardizing the access control vehicle image set specifically includes: resizing the access control vehicle image set and normalizing the images.

[0055] The following is a specific embodiment of extracting the feature vector of the access control vehicle image set through a convolutional neural network in the present application: first, a pre-trained convolutional neural network model (such as ResNet50, VGG16, etc.) is used to extract image features, and then the pre-trained ResNet50 model is loaded using the Keras library, and the fully connected layer of the model is removed. Finally, feature extraction is performed on the entire image set, and the feature vectors of all images are stored together to obtain the feature vector of the access control vehicle image set.

[0056] In a specific implementation, the color feature vector serving as the color feature may be combined with the feature vector of the access control vehicle image set by row vector combination to obtain a feature matrix serving as the access control vehicle feature.

[0057] It should be noted that the developed feature is a historical access control vehicle feature of a vehicle that has been identified in the past and stored in the parking lot control center. By comparing the access control vehicle feature with the developed feature, it can be determined whether the current access control vehicle is an unidentified vehicle entering for the first time. Therefore, when the current access control vehicle enters repeatedly, the developed image of the current access control vehicle is obtained and identified, thereby achieving the purpose of shortening the vehicle identification time. Optionally, in some embodiments, judging whether the current access control vehicle is an identified vehicle based on the access control vehicle feature can be specifically determined in the following manner:

[0058] Acquire the access control vehicle characteristics;

[0059] Acquire all the image features stored in the vehicle information system, and measure the similarity with the access control vehicle features respectively, to obtain the similarity between the access control vehicle features and each image feature;

[0060] When the similarity between the developed feature and the access control vehicle feature is higher than a preset similarity threshold, the current access control vehicle is determined to be a marked vehicle; otherwise, the current access control vehicle is determined to be an unmarked vehicle.

[0061] In a specific implementation, the cosine similarity between the image feature and the access control vehicle feature can be determined respectively, thereby achieving similarity measurement between the image feature and the access control vehicle feature, and the similarity threshold is calibrated as a constant based on experience.

[0062] In step S103, when the current access control vehicle is an identified vehicle, a developed image corresponding to the current access control vehicle is obtained, and after the current access control vehicle is identified according to the developed image, an identification vector is extracted from the developed image and the access control vehicle image set to obtain a vehicle identification enthalpy vector.

[0063] Optionally, in some embodiments, when the current access control vehicle is an unidentified vehicle, the license plate recognition entropy of an identified vehicle is obtained, and the current access control vehicle is visually identified based on the license plate recognition entropy of the identified vehicle.

[0064] It should be noted that the vehicle identification enthalpy vector is composed of vehicle identification enthalpy bases corresponding to each corrected vehicle image. The vehicle identification enthalpy base is the relative difference between the vehicle surface scene brightness and the stored vehicle brightness characteristics when the access control vehicle is driving from far to near. Since this relative difference is usually caused by interference from environmental factors, the vehicle images of subsequent vehicles can be developed and enhanced according to the vehicle identification enthalpy vector, thereby reducing the impact of environmental factors on the accuracy of vehicle identification in long-distance scenes, and then increasing the recognition distance of the vehicle's license plate number during the in-car identification process, thereby improving the vehicle identification speed in the hospital's underground parking lot and reducing the traffic pressure in the parking lot.

[0065] It should be noted that when the access control vehicle feature is a marked feature, indicating that the current access control vehicle is not a vehicle that enters the parking lot for the first time for identification, the corresponding vehicle license plate and historical images can be retrieved according to the development feature to complete the vehicle identification of the current access control vehicle. In some embodiments, the access control vehicle image with the highest image brightness in the historical access control vehicle image set of the current access control vehicle stored in the parking lot control center can be used as the development image corresponding to the current access control vehicle.

[0066] Optionally, in some embodiments, the current access control vehicle may be identified according to the developed image in the following manner, namely:

[0067] Performing image cropping on the developed image to obtain a license plate area image of the current access control vehicle;

[0068] The license plate area image is segmented into characters and the plate recognition is performed to obtain the license plate characters of the current access control vehicle.

[0069] In some other embodiments, the corresponding vehicle license plate may also be retrieved according to the development feature, thereby completing the vehicle identification of the current access control vehicle.

[0070] Optionally, in some embodiments, reference Figure 2 As shown in FIG. 1 , this figure is an exemplary flow chart of obtaining a vehicle identification enthalpy vector in some embodiments of the present application. The identification vector is extracted from the developed image and the access control vehicle image set to obtain the vehicle identification enthalpy vector, which can be specifically implemented by the following steps, namely:

[0071] In step S1031, scale alignment is performed on each access control vehicle image in the access control vehicle image set according to the developed image to obtain a corrected vehicle image set;

[0072] In step S1032, the vehicle identification enthalpy bases corresponding to the respective corrected vehicle images in the corrected vehicle image set are determined according to the developed images, and the vehicle identification enthalpy vectors are determined according to the vehicle identification enthalpy bases corresponding to the respective corrected vehicle images.

[0073] In a specific implementation, the corrected vehicle image set can be obtained by performing feature point matching on each access control vehicle image in the access control vehicle image set, such as feature point matching on edge feature points of windows and doors, and then performing background separation and scale alignment on the image of the current access control vehicle in the dream vehicle image.

[0074] In specific implementation, the ratio between the image grayscale mean corresponding to each corrected vehicle image and the image grayscale mean corresponding to the developed image can be used as the vehicle identification enthalpy basis corresponding to the corrected vehicle image, and the vehicle identification enthalpy basis corresponding to each corrected vehicle image can be used as a vector element to form the vehicle identification enthalpy vector.

[0075] In step S104, an open space sequence is extracted from the access control vehicle image set to obtain an open space correction sequence, and the license plate recognition entropy of the marked vehicle is updated according to the open space correction sequence and the vehicle identification enthalpy vector.

[0076] It should be noted that the more other vehicles there are in the access control vehicle image, the greater the interference factors will be when identifying the license plate number of the current access control vehicle. Therefore, it is necessary to update the license plate recognition entropy of the identified vehicle according to the degree of openness around the current access control vehicle at different times, so as to improve the accuracy of the license plate number identification process.

[0077] The empty correction sequence described in the present application is composed of empty coefficients corresponding to different acquisition moments, and the empty coefficient reflects the image empty degree of the access control vehicle image corresponding to the acquisition moment. Optionally, in some embodiments, the empty sequence is extracted from the access control vehicle image set to obtain the empty correction sequence, which can be specifically implemented in the following manner, namely:

[0078] Obtain the number of vehicle objects corresponding to each access control vehicle image in the access control vehicle image set;

[0079] The open space correction sequence is determined according to the number of vehicle objects corresponding to each access control vehicle object.

[0080] In specific implementation, the number of vehicle objects corresponding to each access control vehicle object can be mapped to the corresponding empty space coefficient according to a preset mapping table according to the threshold interval of the number of vehicle objects, and the empty space coefficients are sequentially composed of the empty space correction sequence.

[0081] It should be noted that the license plate recognition entropy described in the present application is a coefficient sequence composed of various license plate recognition fusion coefficients. The license plate recognition fusion coefficient is used to enhance the image when performing license plate recognition on unidentified vehicles. The license plate recognition fusion coefficient reflects the environmental interference characteristics of the identified vehicle in the image of the driving process from far to near, and then the image of the access control vehicle at different distances is enhanced and recognized according to the license plate recognition fusion coefficient, which can reduce the environmental light and shadow interference in the long-distance recognition process. According to the open correction sequence and the vehicle identification enthalpy vector, the license plate recognition entropy of the identified vehicle is updated specifically including:

[0082] According to each openness index in the openness correction sequence, each vector element in the vehicle identification enthalpy vector is corrected and fused respectively, and the license plate recognition entropy of the identified vehicle is updated according to each license plate recognition fusion coefficient obtained after fusion.

[0083] In specific implementation, the sequence length in the open correction sequence is consistent with the number of vector elements in the vehicle identification enthalpy vector. Therefore, each vector element in the vehicle identification enthalpy vector can be multiplied by the open coefficient at the same position in the open correction sequence to obtain a new license plate recognition fusion coefficient group. The new license plate recognition fusion coefficient group is used to update the license plate recognition fusion coefficient group in the original license plate recognition entropy, thereby updating the license plate recognition entropy of the identified vehicle and obtaining the updated license plate recognition entropy.

[0084] Optionally, in some embodiments, after determining the license plate recognition entropy of the identified vehicle, the method further includes: storing the license plate recognition entropy of the identified vehicle in a parking lot control center.

[0085] It should be noted that, according to the empty correction sequence and the vehicle identification enthalpy vector, the license plate recognition entropy of the identified vehicle is updated, and the remaining access control vehicles in the hospital parking lot are developed and identified according to the updated license plate recognition entropy, that is, the access control vehicle images of the remaining unidentified vehicles are enhanced according to the license plate recognition entropy, which can highlight the license plate features of the unidentified vehicle when the vehicle is crowded, making the license plate area more prominent in the image, thereby improving the recognition algorithm's positioning and recognition capabilities for the license plate, and helping to accurately identify the license plate number at a farther distance or in more blurred conditions, and by correcting the vehicle identification enthalpy vector, the ability of the vehicle identification to adapt to images of different degrees of empty space is improved, and ultimately the recognition distance of the license plate number is improved.

[0086] In step S105, the remaining access-controlled vehicles in the hospital parking lot are identified using the updated license plate recognition entropy.

[0087] It should be noted that when the access control vehicle is characterized as an unidentified vehicle, it indicates that the current access control vehicle is a vehicle entering the parking lot for the first time. At this time, the environmental noise has a greater impact on the vehicle identification process. Therefore, it is necessary to identify the current access control vehicle based on the updated license plate recognition entropy, thereby reducing the impact of environmental noise on the vehicle identification process during long-distance identification and improving the identification distance of the current access control vehicle.

[0088] It should be noted that the updated license plate recognition entropy is the license plate recognition entropy corresponding to the identification moment with the smallest time difference relative to the current moment. The current access control vehicle is identified by the license plate recognition entropy of the nearest identification moment, which can minimize the environmental interference caused by scene changes, thereby increasing the recognition distance of the vehicle's license plate number, improving the vehicle identification speed in the hospital's underground parking lot, and reducing the traffic pressure in the parking lot.

[0089] Optionally, in some embodiments, the updated license plate recognition entropy may be used to identify other access-controlled vehicles in the hospital parking lot in the following manner:

[0090] Obtain updated license plate recognition entropy; obtain a set of access control vehicle images of the current access control vehicle;

[0091] Performing development fusion on the access control vehicle image set of the current access control vehicle according to the updated license plate recognition entropy to obtain a license plate development image;

[0092] Character segmentation and sign recognition are performed on the sign development image.

[0093] Optionally, in some embodiments, the access control vehicle image set of the current access control vehicle is developed and fused according to the updated license plate recognition entropy to obtain the license plate development image, which specifically includes: performing pixel development enhancement on the license plate area image in each access control vehicle image in the access control vehicle image set according to each license plate recognition fusion coefficient in the license plate recognition entropy, and then performing image fusion after scaling.

[0094] In specific implementation, the YOLO detection algorithm can be used to extract the license plate area images in each access control vehicle image in the access control vehicle image set, and then the grayscale of different license plate area images is enhanced respectively. The enhancement ratio is determined according to the license plate recognition fusion coefficient of the corresponding position, and then the average image of each license plate area image after grayscale enhancement is used as the sign development image.

[0095] In a specific implementation, the EasyOCR database may be used to perform sign recognition on the sign development image.

[0096] In addition, in another aspect of the present application, in some embodiments, the present application provides an intelligent parking lot intelligent identification system, the system includes a vehicle identification unit, reference Figure 3 , which is a schematic diagram of exemplary hardware and / or software of a vehicle identification unit according to some embodiments of the present application, the vehicle identification unit 200 includes: a collection module 201, a processing module 202 and an execution module 203, which are described as follows:

[0097] In some specific embodiments of the present application, the acquisition module 201 is mainly used to start the hospital parking lot smart identification and collect the access control vehicle image set of the current access control vehicle;

[0098] Processing module 202, in some specific embodiments of the present application, processing module 202 is used to extract features from the access control vehicle image set to obtain access control vehicle features, and determine whether the current access control vehicle is an identified vehicle based on the access control vehicle features;

[0099] It should be noted that the processing module 202 is also used for, when the current access control vehicle is an identified vehicle, obtaining a development image corresponding to the current access control vehicle, identifying the current access control vehicle according to the development image, extracting an identification vector from the development image and the access control vehicle image set, and obtaining a vehicle identification enthalpy vector;

[0100] In addition, the processing module 202 is further used to extract an open sequence from the access control vehicle image set to obtain an open correction sequence, and update the license plate recognition entropy of the marked vehicle according to the open correction sequence and the vehicle identification enthalpy vector;

[0101] Execution module 203, in some specific embodiments of the present application, execution module 203 is mainly used to develop and identify the remaining access control vehicles in the hospital parking lot by using the updated license plate recognition entropy.

[0102] The above describes in detail an example of an intelligent identification system and identification method for a smart parking lot provided in an embodiment of the present application. It can be understood that in order to realize the above functions, the corresponding device includes a hardware structure and / or software module corresponding to executing each function.

[0103] Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function in the application is executed in hardware or in a manner driven by computer software depends on the specific application and design constraints of the technical solution. Therefore, 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 be beyond the scope of the present application.

[0104] In addition, the present application also provides a computer terminal device, which includes a memory and a processor, the memory stores a code, and the processor is configured to obtain the code and execute the above-mentioned intelligent identification method for a smart parking lot.

[0105] In some embodiments, reference Figure 4 , which is a schematic diagram of the structure of a computer terminal device using a smart parking lot smart identification method according to some embodiments of the present application. Figure 4 The computer terminal device 300 shown in the figure is implemented, and the computer terminal device 300 includes at least one communication bus 301, a communication interface 302, a processor 303 and a memory 304.

[0106] The processor 303 can be a general-purpose central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more for controlling the execution of a smart parking lot intelligent identification method in the present application.

[0107] The communication bus 301 may include a path for transmitting information between the above-mentioned components.

[0108] The memory 304 may be a read only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read only memory (EEPROM), a compact disc read only memory (CD ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 304 may exist independently and be connected to the processor 303 via the communication bus 301. The memory 304 may also be integrated with the processor 303.

[0109] The memory 304 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 303. The processor 303 is used to execute the program code stored in the memory 304. The program code may include one or more software modules. The determination of the license plate recognition entropy in the above embodiment can be implemented by the processor 303 and one or more software modules in the program code in the memory 304.

[0110] The communication interface 302 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0111] Optionally, the computer terminal device 300 may further include a power supply 305 for providing power to various devices or circuits in the real-time computer terminal device.

[0112] In a specific implementation, as an embodiment, a computer terminal device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).

[0113] The above-mentioned computer terminal device may be a general-purpose computer terminal device or a dedicated computer terminal device. In a specific implementation, the computer terminal device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer terminal device.

[0114] In addition, in other aspects of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores at least one computer program, and the computer program is loaded and executed by a processor to implement the operations performed by the above-mentioned intelligent identification method for a smart parking lot.

[0115] In summary, in an intelligent identification system and identification method for a smart parking lot disclosed in an embodiment of the present application, first, an access control vehicle image set of the current access control vehicle is collected; feature extraction is performed on the access control vehicle image set to obtain access control vehicle features, and whether the current access control vehicle is an identified vehicle is determined based on the access control vehicle features; when the current access control vehicle is an identified vehicle, a developed image corresponding to the current access control vehicle is obtained, and after the current access control vehicle is identified based on the developed image, an identification vector is extracted from the developed image and the access control vehicle image set to obtain a vehicle identification enthalpy vector; an empty sequence is extracted from the access control vehicle image set to obtain an empty correction sequence, and the license plate recognition entropy of the identified vehicle is updated based on the empty correction sequence and the vehicle identification enthalpy vector; the remaining access control vehicles in the hospital parking lot are developed and identified based on the updated license plate recognition entropy, which ultimately improves the vehicle identification speed in the hospital underground parking lot and reduces the traffic pressure in the hospital parking lot.

[0116] The above is only an embodiment of the present application, and the common knowledge such as the specific technical scheme or characteristics in the scheme is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical scheme of the present application, several modifications and improvements can be made, which should also be regarded as the scope of protection of the present application, and these will not affect the effect of the implementation of the present application and the practicality of the patent.

[0117] The scope of protection claimed by this application shall be based on the content of its claims. The specific implementation methods and other records in the specification can be used to interpret the content of the claims. Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include these modifications and variations.

Claims

1. A smart parking lot intelligent identification method, characterized in that: include: Start the hospital parking lot intelligent identification and collect the access control vehicle image set of the current access control vehicles; Extracting features from the access control vehicle image set to obtain access control vehicle features, and determining whether the current access control vehicle is an identified vehicle based on the access control vehicle features; When the current access control vehicle is an identified vehicle, a developed image corresponding to the current access control vehicle is obtained, and after the current access control vehicle is identified according to the developed image, an identification vector is extracted from the developed image and the access control vehicle image set to obtain a vehicle identification enthalpy vector; Extracting an open space sequence from the access control vehicle image set to obtain an open space correction sequence, and updating the license plate recognition entropy of the marked vehicle according to the open space correction sequence and the vehicle identification enthalpy vector; The remaining access-controlled vehicles in the hospital parking lot are developed and identified using the updated license plate recognition entropy; The step of extracting identification vectors from the developed image and the access control vehicle image set to obtain the vehicle identification enthalpy vector specifically includes: aligning the scales of the access control vehicle images in the access control vehicle image set according to the developed image to obtain a corrected vehicle image set; Determining the vehicle identification enthalpy bases corresponding to the respective corrected vehicle images in the corrected vehicle image set according to the developed images, and determining the vehicle identification enthalpy vector according to the vehicle identification enthalpy bases corresponding to the respective corrected vehicle images, wherein the vehicle identification enthalpy vector is composed of the vehicle identification enthalpy bases corresponding to the respective corrected vehicle images; Among them, the updated license plate recognition entropy is used to identify the remaining access control vehicles in the hospital parking lot, including: Obtain updated license plate recognition entropy; obtain a set of access control vehicle images of the current access control vehicle; Performing development fusion on the access control vehicle image set of the current access control vehicle according to the updated license plate recognition entropy to obtain a license plate development image; Performing character segmentation on the signboard development image and performing signboard recognition; Among them, updating the license plate recognition entropy of the identified vehicle according to the open correction sequence and the vehicle identification enthalpy vector specifically includes: correcting and fusing each vector element in the vehicle identification enthalpy vector according to each open correction sequence, and updating the license plate recognition entropy of the identified vehicle according to each license plate recognition fusion coefficient obtained after fusion.

2. The method according to claim 1, characterized in that The feature extraction is performed on the access control vehicle image set to obtain the access control vehicle features, which specifically include: Performing color histogram conversion on the access control vehicle image set to obtain color features of the access control vehicle image set; After the access control vehicle image set is standardized, a feature vector of the access control vehicle image set is extracted through a convolutional neural network; The color features and feature vectors of the access control vehicle image set are combined to obtain access control vehicle features.

3. The method according to claim 1, characterized in that Determining whether the current access control vehicle is a marked vehicle according to the access control vehicle characteristics specifically includes: Acquire the access control vehicle characteristics; Acquire all the image features stored in the vehicle information system, and measure the similarity with the access control vehicle features respectively, to obtain the similarity between the access control vehicle features and each image feature; When the similarity between the developed feature and the access control vehicle feature is higher than a preset similarity threshold, the current access control vehicle is determined to be a marked vehicle; otherwise, the current access control vehicle is determined to be an unmarked vehicle.

4. The method according to claim 1, characterized in that Identifying the current access control vehicle according to the developed image specifically includes: cropping the developed image to obtain the license plate area image of the current access control vehicle; segmenting the license plate area image and performing sign recognition to obtain the license plate characters of the current access control vehicle.

5. The method according to claim 1, characterized in that Acquiring the access control vehicle image set of the current access control vehicle involves acquiring the access control vehicle images of the current access control vehicle at equal intervals through a high-definition camera to obtain the access control vehicle image set of the current access control vehicle.

6. An intelligent identification system for a smart parking lot, which uses the method described in any one of claims 1 to 5 to identify vehicles, and the system includes a vehicle identification unit, characterized in that: The vehicle identification unit comprises: A collection module is used to start the hospital parking lot intelligent identification and collect the access control vehicle image set of the current access control vehicle; A processing module extracts features from the access control vehicle image set to obtain access control vehicle features, and determines whether the current access control vehicle is a marked vehicle based on the access control vehicle features; The processing module is further configured to, when the current access control vehicle is an identified vehicle, obtain a developed image corresponding to the current access control vehicle, identify the current access control vehicle according to the developed image, extract an identification vector from the developed image and the access control vehicle image set, and obtain a vehicle identification enthalpy vector; The processing module extracts an open sequence from the access control vehicle image set to obtain an open correction sequence, and updates the license plate recognition entropy of the marked vehicle according to the open correction sequence and the vehicle identification enthalpy vector; The execution module uses the updated license plate recognition entropy to develop and identify the remaining access control vehicles in the hospital parking lot.

7. A computer terminal device, characterized in that: The computer terminal device includes a memory and a processor, the memory stores a code, and the processor is configured to obtain the code and execute the intelligent identification method for a smart parking lot as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing at least one computer program, characterized in that: The computer program is loaded and executed by a processor to implement the operations performed by the intelligent identification method for a smart parking lot as described in any one of claims 1 to 5.

Images