Intelligent parking lot management system and method based on deep learning

Abstract

The invention discloses an intelligent parking lot management system and method based on deep learning, and belongs to the technical field of parking lot management. The method is used for solving the technical problems of poor autonomous supervision and analysis effect and poor active risk management effect of a parking lot for different accidents in an existing scheme. All accidents occurring in different position areas in the parking lot are supervised and subjected to data processing analysis, different area types are dynamically marked, and a second area type obtained through analysis is subjected to data processing calculation in different dimensions and digital representation; and carrying out expansion integration processing on local risk occurrence values obtained by corresponding processing of all types of sub-regions and local risk type values of different local accident types, determining local risk integration states corresponding to different types of sub-regions, and implementing targeted parking risk prevention management. And multi-dimensional expansion mining and analysis management of all accident data and accident position data occurring in the parking lot are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of parking lot management, and in particular to an intelligent parking lot management system and method based on deep learning. Background Art

[0002] Intelligent parking lot management refers to the efficient and scientific management and optimization of parking lots through the application of advanced information technology, automation technology and network technology. It can improve the operational efficiency of parking lots, improve user experience, and contribute to the orderly management of urban traffic.

[0003] When implementing existing parking lot management systems, most of them still remain at the level of single parking data monitoring statistics and visual prompts. They are unable to supervise and handle different accidents occurring in different locations and areas of the parking lot, as well as conduct risk assessments, and implement targeted risk management for different locations and areas and different accident types based on the risk assessment results. This results in poor autonomous supervision and analysis of different accidents in the parking lot and poor proactive risk management. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent parking lot management system and method based on deep learning, which is used to solve the technical problems of poor autonomous supervision and analysis of different accidents in parking lots and poor active risk management in existing solutions.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] Intelligent parking lot management system based on deep learning, including:

[0007] The parking lot accident supervision and processing module is used to supervise and analyze all accidents occurring in different locations in the parking lot, determine the regional risk impact status corresponding to different location areas, and dynamically mark different area types to obtain the first area type or the second area type;

[0008] The parking lot accident multi-dimensional processing and analysis module is used to perform corresponding local risk supervision processing on all types of sub-areas and all local accident types contained in the second area type obtained by analysis, and obtain local risk occurrence values ​​corresponding to different types of sub-areas and local risk type values ​​of different local accident types;

[0009] The parking lot accident risk prevention management module is used to expand and integrate the local risk occurrence values ​​obtained from the corresponding processing of all types of sub-areas and the local risk type values ​​of different local accident types, perform data analysis based on the processing results to determine the local risk integration status corresponding to different types of sub-areas, and implement targeted parking risk prevention management for different target types of sub-areas.

[0010] Preferably, all accidents occurring in all location areas in the parking lot are obtained and the formula Calculate and obtain the regional accident supervision value QJk corresponding to different regional types; where k = 1, 2; k with a value of 1 indicates a public regional type; k with a value of 2 indicates a non-public regional type; Nk is N1 and N2, which are the total number of first accidents and the total number of second accidents corresponding to different regional types, respectively; NBk is NB1 and NB2, which are the total number of first accident standards and the total number of second accident standards corresponding to different regional types, respectively; Sk is S1 and S2, which are the total loss price of the first accident and the total loss price of the second accident corresponding to different regional types, respectively; SBk is SB1 and SB2, which are the standard loss price of the first accident and the standard loss price of the second accident corresponding to different regional types, respectively; α and β are different error correction factors, and their value ranges are both [1, 2); is the floor function.

[0011] Preferably, data analysis is performed on the regional accident supervision values ​​corresponding to different regional types to determine the regional risk impact status corresponding to different regional types;

[0012] If the regional accident supervision value is 0, the regional type is associated with the regional risk impact normal state, and the regional type is marked as the first regional type;

[0013] If the regional accident supervision value is not 0, the regional type is associated with the regional risk impact abnormal state, and the regional type is marked as the second regional type.

[0014] Preferably, all types of sub-regions included in the second region type are obtained, and all local accidents corresponding to different types of sub-regions and the local accident loss prices and local accident types corresponding to different local accidents are obtained;

[0015] By formula Calculate and obtain the local risk occurrence value CFi corresponding to different types of sub-areas; where i is the different types of sub-areas included in the second area type, i=1, 2, 3,..., n; n is a positive integer; JJi is the total price of local accident losses corresponding to all local accidents in different types of sub-areas; YJi is the total impact value of the first local accident type corresponding to all local accidents in different types of sub-areas; Ni is the total number of occurrences corresponding to all local accidents in different types of sub-areas; η is the calculation impact factor, and its value is greater than 1; A is the standard value of local risk occurrence.

[0016] Preferably, by the formula Calculate and obtain the local risk type value LFj of all different local accident types; where j is the different local accident types occurring in the second area type, j=1, 2, 3, ..., m; m is a positive integer; LJj is the total impact value of the second local accident type corresponding to different local accident types; B and C are the impact standard values ​​of the first local accident type and the impact standard values ​​of the second local accident type corresponding to different local accident types, respectively; max() means obtaining the maximum value among several real numbers.

[0017] Preferably, when processing and analyzing the local accident risk states corresponding to all types of sub-regions included in the second area type, the local risk occurrence values ​​and local risk type values ​​obtained by processing all types of sub-regions included in the second area type are subjected to data analysis through a multidimensional risk integration function, and the local risk integration values ​​JZi corresponding to different types of sub-regions are output; the local risk integration values ​​include values ​​of 0 or 1.

[0018] Preferably, the expression of the multidimensional risk integration function is:

[0019]

[0020] Preferably, the sub-area of ​​the corresponding type is marked as a target type sub-area according to the local risk integration value of 1, and targeted parking risk prevention management is implemented for sub-areas of different target types;

[0021] If CFi≤1 and there is a LFj greater than or equal to 1, a local accident type risk prevention management plan is implemented for the target type sub-area;

[0022] If CFi>1 and there is no LFj greater than or equal to 1, or CFi>1 and there is a LFj greater than or equal to 1, the local type sub-area risk prevention management plan will be implemented for the target type sub-area.

[0023] Preferably, the area type includes a public area type and a non-public area type.

[0024] Intelligent parking lot management method based on deep learning, including:

[0025] Supervise and analyze all accidents occurring in different locations in the parking lot, determine the regional risk impact status corresponding to different locations, and dynamically mark different area types to obtain the first area type or the second area type;

[0026] Perform corresponding local risk supervision processing on all types of sub-areas and all local accident types included in the second area type obtained through analysis, and obtain local risk occurrence values ​​corresponding to different types of sub-areas and local risk type values ​​of different local accident types;

[0027] The local risk occurrence values ​​obtained from the corresponding processing of all types of sub-areas and the local risk type values ​​of different local accident types are expanded and integrated, and data analysis is performed based on the processing results to determine the local risk integration status corresponding to different types of sub-areas, and targeted parking risk prevention management is implemented for different target types of sub-areas.

[0028] Compared with the existing solutions, the present invention achieves the following beneficial effects:

[0029] The present invention monitors and performs data processing and analysis on all accidents occurring in different locations in a parking lot, and dynamically marks different area types, thereby achieving the supervision and digital representation of the regional risk impact status corresponding to different locations in the parking lot, thereby improving the modular supervision and processing effect of all accidents in the parking lot.

[0030] The present invention further expands the processing and analysis of all accident data occurring in the parking lot by performing data processing calculations and digital representation of different dimensions on the second area type obtained through analysis, thereby improving the autonomous supervision and analysis effect of the parking lot on different accidents.

[0031] The present invention expands and integrates the local risk occurrence values ​​obtained by corresponding processing of all types of sub-areas and the local risk type values ​​of different local accident types, determines the local risk integration status corresponding to different types of sub-areas and implements targeted parking risk prevention management, realizes multi-dimensional expansion mining and analysis management of all accident data and accident location data occurring in the parking lot, and improves the active risk management effect of the parking lot for different accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The present invention will be further described below in conjunction with the accompanying drawings.

[0033] Figure 1 This is a module block diagram of the intelligent parking lot management system based on deep learning of the present invention.

[0034] Figure 2 It is a flowchart of the steps for implementing targeted parking risk prevention management in the present invention.

[0035] Figure 3 This is a flowchart of the steps of the intelligent parking lot management method based on deep learning of the present invention. DETAILED DESCRIPTION

[0036] 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 only 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.

[0037] Example 1: Figure 1 As shown, the present invention is an intelligent parking lot management system based on deep learning, including a parking lot accident supervision and processing module, a parking lot accident multi-dimensional processing and analysis module, and a parking lot accident risk prevention and management module;

[0038] The parking lot accident supervision and processing module is used to supervise and analyze all accidents occurring in different locations in the parking lot, determine the regional risk impact status corresponding to different location areas, and dynamically mark different area types to obtain the first area type or the second area type; including:

[0039] Get all accidents that occurred in all location areas in the parking lot. The location areas include public areas and non-public areas. The public areas are specifically road areas, and the non-public areas are specifically parking areas. And through the formula Calculate and obtain the regional accident supervision value QJk corresponding to different regional types; regional types include public regional types and non-public regional types; where k=1, 2; k with a value of 1 indicates a public regional type; k with a value of 2 indicates a non-public regional type; Nk is N1 and N2, which are the total number of first accidents and the total number of second accidents corresponding to different regional types, respectively; NBk is NB1 and NB2, which are the total number of first accident standards and the total number of second accident standards corresponding to different regional types, respectively, which can be determined based on the risk prevention design requirements data of the parking lot, or based on the test data of the early trial operation of the parking lot; Sk is S1 and S2, which are the total price of first accident loss and the total price of second accident loss corresponding to different regional types, respectively; SBk is SB1 and SB2, which are the standard price of first accident loss and the standard price of second accident loss corresponding to different regional types, respectively, which can be determined based on the risk prevention design requirements data of the parking lot, or based on the test data of the early trial operation of the parking lot; α and β are different error correction factors, and the value range is [1, 2), α can be 1.21, and β can be 1.74; is the floor rounding function;

[0040] It should be noted that the regional accident supervision value is used to calculate all accident data corresponding to different area types in the parking lot to digitally represent the regional risk impact status corresponding to different area types;

[0041] Conduct data analysis on the regional accident supervision values ​​corresponding to different regional types to determine the regional risk impact status corresponding to different regional types;

[0042] If the regional accident supervision value is 0, the regional type is associated with the regional risk impact normal state, and the regional type is marked as the first regional type;

[0043] If the regional accident supervision value is not 0, the regional type is associated with the regional risk impact abnormal state, and the regional type is marked as the second regional type;

[0044] In the embodiment of the present invention, by supervising and performing data processing and analysis on all accidents occurring in different location areas in the parking lot, and dynamically marking different area types, it is possible to supervise and digitally represent the regional risk impact status corresponding to different location areas in the parking lot, thereby improving the modular supervision and processing effect of all accidents in the parking lot.

[0045] The parking lot accident multi-dimensional processing and analysis module is used to perform corresponding local risk supervision processing on all types of sub-areas and all local accident types contained in the second area type obtained by analysis, and obtain local risk occurrence values ​​corresponding to different types of sub-areas and local risk type values ​​of different local accident types; including:

[0046] Obtain all types of sub-areas included in the second area type, where the type sub-areas are local public areas or local non-public areas at different locations. The specific division rules can be customized according to the actual application requirements of the actual application scenario, and obtain all local accidents that have occurred in history corresponding to different types of sub-areas, as well as local accident loss prices and local accident types corresponding to different local accidents. Local accident types include but are not limited to collision types and scratch types. Local accident damage prices corresponding to different local accidents can be estimated by professional and technical personnel in this field based on work experience and industry standards; in addition, different local accident types are pre-set with a corresponding accident impact value, which is used to digitally represent the negative impact corresponding to the local accident type. The specific values ​​of the accident impact values ​​corresponding to different accident types can be determined according to the median value of all local accident damage prices corresponding to the accident type;

[0047] By formula Calculate and obtain the local risk occurrence value CFi corresponding to different types of sub-areas; where i is the different types of sub-areas included in the second area type, i=1, 2, 3, ..., n; n is a positive integer, indicating the total number of all types of sub-areas included in the second area type; JJi is the total price of local accident losses corresponding to all local accidents in different types of sub-areas; YJi is the total impact value of the first local accident type corresponding to all local accidents in different types of sub-areas, obtained by summing up the accident impact values ​​corresponding to all accident types; Ni is the total number of occurrences corresponding to all local accidents in different types of sub-areas; η is the calculation impact factor, and its value is greater than 1, which can be 1.57; A is the standard value of local risk occurrence, which can be determined based on the risk prevention design requirement data of the parking lot, or based on the test data of the early trial operation of the parking lot;

[0048] It should be noted that the local risk occurrence value is used to calculate all local accident data from the type sub-area dimension, and digitally represent the local risk occurrence status corresponding to different types of sub-areas;

[0049] And, through the formula Calculate and obtain the local risk type value LFj of all different local accident types; where j is the different local accident types that appear in the second area type, j=1, 2, 3, ..., m; m is a positive integer, indicating the total number of all local accident types contained in the second area type; LJj is the total impact value of the second local accident type corresponding to different local accident types, obtained by multiplying the total number of different accident types by the corresponding accident impact value; B and C are the impact standard value of the first local accident type and the impact standard value of the second local accident type corresponding to different local accident types, respectively, which can be determined based on the risk prevention design requirement data of the parking lot, or based on the test data of the early trial operation of the parking lot; max() indicates obtaining the maximum value of several real numbers;

[0050] It should be noted that the local risk type value is used to calculate the occurrence data of all local accident types from the local accident type dimension, and digitally represent the local risk type status corresponding to different local accident types;

[0051] In the embodiment of the present invention, by performing data processing calculations and digital representation of different dimensions on the second area type obtained through analysis, further expanded processing and analysis of all accident data occurring in the parking lot are achieved, thereby improving the autonomous supervision and analysis effect of the parking lot on different accidents.

[0052] The parking lot accident risk prevention management module is used to expand and integrate the local risk occurrence values ​​obtained from the corresponding processing of all types of sub-areas and the local risk type values ​​of different local accident types, perform data analysis based on the processing results to determine the local risk integration status corresponding to different types of sub-areas, and implement targeted parking risk prevention management for different target types of sub-areas; including:

[0053] When processing and analyzing the local accident risk states corresponding to all types of sub-regions included in the second regional type, the local risk occurrence values ​​and local risk type values ​​obtained by processing all types of sub-regions included in the second regional type are analyzed through a multidimensional risk integration function, and the local risk integration values ​​JZi corresponding to different types of sub-regions are output;

[0054] Among them, the expression of the multidimensional risk integration function is:

[0055]

[0056] The local risk integration value contains values ​​of 0 or 1;

[0057] A local risk integration value of 0 indicates that the local risk integration status corresponding to the sub-region of the corresponding type is normal;

[0058] A local risk integration value of 1 indicates that the local risk integration state corresponding to the sub-region of the corresponding type is abnormal;

[0059] It should be noted that the local risk integration value is used to integrate and calculate the risk supervision processing data of different dimensions of the type sub-region to digitally represent the corresponding local risk integration status;

[0060] like Figure 2 As shown, according to the local risk integration value of 1, the sub-area of ​​the corresponding type is marked as the target type sub-area, and targeted parking risk prevention management is implemented for the sub-areas of different target types;

[0061] If CFi≤1 and there is LFj greater than or equal to 1, a local accident type risk prevention management plan is implemented for the target type sub-area to which it belongs; the local accident type risk prevention management plan can specifically be targeted risk prevention for the accident type that occurs most frequently in the target type sub-area to which it belongs, such as adding eye-catching reminders;

[0062] If CFi>1 and there is no LFj greater than or equal to 1, or CFi>1 and there is LFj greater than or equal to 1, the local type sub-area risk prevention management plan will be implemented for the target type sub-area to which it belongs; the local type sub-area risk prevention management plan may specifically be a comprehensive risk prevention reminder for all types of accidents occurring in the target type sub-area, such as adding eye-catching reminders and adding new cameras.

[0063] In an embodiment of the present invention, by expanding and integrating the local risk occurrence values ​​obtained by corresponding processing of all types of sub-areas and the local risk type values ​​of different local accident types, the local risk integration status corresponding to different types of sub-areas is determined and targeted parking risk prevention management is implemented, thereby realizing multi-dimensional expansion mining and analysis management of all accident data and accident location data occurring in the parking lot, thereby improving the active risk management effect of the parking lot for different accidents.

[0064] Example 2: Figure 3 As shown, the intelligent parking lot management method based on deep learning includes:

[0065] Supervise and analyze all accidents occurring in different locations in the parking lot, determine the regional risk impact status corresponding to different locations, and dynamically mark different area types to obtain the first area type or the second area type;

[0066] Perform corresponding local risk supervision processing on all types of sub-areas and all local accident types included in the second area type obtained through analysis, and obtain local risk occurrence values ​​corresponding to different types of sub-areas and local risk type values ​​of different local accident types;

[0067] The local risk occurrence values ​​obtained from the corresponding processing of all types of sub-areas and the local risk type values ​​of different local accident types are expanded and integrated, and data analysis is performed based on the processing results to determine the local risk integration status corresponding to different types of sub-areas, and targeted parking risk prevention management is implemented for different target types of sub-areas.

[0068] In addition, the formulas involved in the above are all dimensionless and numerical calculations. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and simulating it with simulation software.

[0069] In the several embodiments provided by the present invention, it should be understood that the disclosed system can be implemented in other ways. For example, the above-described embodiments of the invention are only illustrative, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.

[0070] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0071] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0072] It is obvious to a person skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the essential characteristics of the present invention.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. Intelligent parking lot management system based on deep learning, characterized by: include: The parking lot accident supervision and processing module is used to supervise and analyze all accidents occurring in different locations in the parking lot, determine the regional risk impact status corresponding to different location areas, and dynamically mark different area types to obtain the first area type or the second area type; The parking lot accident multi-dimensional processing and analysis module is used to perform corresponding local risk supervision processing on all types of sub-areas and all local accident types contained in the second area type obtained by analysis, and obtain local risk occurrence values ​​corresponding to different types of sub-areas and local risk type values ​​of different local accident types; The parking lot accident risk prevention management module is used to expand and integrate the local risk occurrence values ​​obtained from the corresponding processing of all types of sub-areas and the local risk type values ​​of different local accident types, perform data analysis based on the processing results to determine the local risk integration status corresponding to different types of sub-areas, and implement targeted parking risk prevention management for different target types of sub-areas.

2. The deep learning-based intelligent parking lot management system according to claim 1 is characterized in that: Get all the accidents that occurred in all the location areas in the parking lot and use the formula Calculate and obtain the regional accident supervision value QJk corresponding to different regional types; where k = 1, 2; k with a value of 1 indicates a public regional type; k with a value of 2 indicates a non-public regional type; Nk is N1 and N2, which are the total number of first accidents and the total number of second accidents corresponding to different regional types, respectively; NBk is NB1 and NB2, which are the total number of first accident standards and the total number of second accident standards corresponding to different regional types, respectively; Sk is S1 and S2, which are the total loss price of the first accident and the total loss price of the second accident corresponding to different regional types, respectively; SBk is SB1 and SB2, which are the standard loss price of the first accident and the standard loss price of the second accident corresponding to different regional types, respectively; α and β are different error correction factors, and their value ranges are both [1, 2); is the floor function.

3. The deep learning-based intelligent parking lot management system according to claim 2 is characterized in that: Conduct data analysis on the regional accident supervision values ​​corresponding to different regional types to determine the regional risk impact status corresponding to different regional types; If the regional accident supervision value is 0, the regional type is associated with the regional risk impact normal state, and the regional type is marked as the first regional type; If the regional accident supervision value is not 0, the regional type is associated with the regional risk impact abnormal state, and the regional type is marked as the second regional type.

4. The deep learning-based intelligent parking lot management system according to claim 3 is characterized in that: Obtain all types of sub-areas included in the second area type, and obtain all local accidents that have occurred in history corresponding to different types of sub-areas, as well as local accident loss prices and local accident types corresponding to different local accidents; By formula Calculate and obtain the local risk occurrence value CFi corresponding to different types of sub-areas; where i is the different types of sub-areas included in the second area type, i=1, 2, 3,..., n; n is a positive integer; JJi is the total price of local accident losses corresponding to all local accidents in different types of sub-areas; YJi is the total impact value of the first local accident type corresponding to all local accidents in different types of sub-areas; Ni is the total number of occurrences corresponding to all local accidents in different types of sub-areas; η is the calculation impact factor, and its value is greater than 1; A is the standard value of local risk occurrence.

5. The deep learning-based intelligent parking lot management system according to claim 4 is characterized in that: By formula Calculate and obtain the local risk type value LFj of all different local accident types; where j is the different local accident types occurring in the second area type, j=1, 2, 3, ..., m; m is a positive integer; LJj is the total impact value of the second local accident type corresponding to different local accident types; B and C are the impact standard values ​​of the first local accident type and the impact standard values ​​of the second local accident type corresponding to different local accident types, respectively; max() means obtaining the maximum value among several real numbers.

6. The deep learning-based intelligent parking lot management system according to claim 5 is characterized in that: When processing and analyzing the local accident risk states corresponding to all types of sub-regions included in the second area type, the local risk occurrence values ​​and local risk type values ​​obtained by processing all types of sub-regions included in the second area type are subjected to data analysis through a multidimensional risk integration function, and the local risk integration values ​​JZi corresponding to different types of sub-regions are output; the local risk integration values ​​include values ​​of 0 or 1.

7. The deep learning-based intelligent parking lot management system according to claim 6 is characterized in that: The expression of the multidimensional risk integration function is:

8. The deep learning-based intelligent parking lot management system according to claim 6 is characterized in that: When marking the sub-area of ​​the corresponding type as a target type sub-area according to the local risk integration value of 1, and implementing targeted parking risk prevention management for sub-areas of different target types; If CFi≤1 and there is a LFj greater than or equal to 1, a local accident type risk prevention management plan is implemented for the target type sub-area; If CFi>1 and there is no LFj greater than or equal to 1, or CFi>1 and there is a LFj greater than or equal to 1, the local type sub-area risk prevention management plan will be implemented for the target type sub-area.

9. The deep learning-based intelligent parking lot management system according to claim 1 is characterized in that: Area types include public area types and non-public area types.

10. A smart parking lot management method based on deep learning, using the smart parking lot management system based on deep learning as described in any one of claims 1 to 9, characterized in that: include: Supervise and analyze all accidents occurring in different locations in the parking lot, determine the regional risk impact status corresponding to different locations, and dynamically mark different area types to obtain the first area type or the second area type; Perform corresponding local risk supervision processing on all types of sub-areas and all local accident types included in the second area type obtained through analysis, and obtain local risk occurrence values ​​corresponding to different types of sub-areas and local risk type values ​​of different local accident types; The local risk occurrence values ​​obtained from the corresponding processing of all types of sub-areas and the local risk type values ​​of different local accident types are expanded and integrated, and data analysis is performed based on the processing results to determine the local risk integration status corresponding to different types of sub-areas, and targeted parking risk prevention management is implemented for different target types of sub-areas.

Images