Automatic garage navigation method and system based on machine vision

Abstract

The invention relates to an automatic garage navigation method and system based on machine vision, and the method comprises the steps: obtaining the information of an unoccupied parking space, and carrying out the parking position distribution for a to-be-parked vehicle; after the parking space distribution information is confirmed, vehicle navigation work is carried out; in a vehicle navigation process, performing intersection identification based on driving image data acquired by a vehicle camera; when the intersection is identified, matching the current intersection information with preset map information to obtain corresponding turning information; when the vehicle finishes steering and approaches a target garage in the parking space distribution information, displaying specific azimuth information of a target parking space in the target garage; when the vehicle is close to the target parking space, parking space number identification is carried out based on driving image data collected by a vehicle camera; and judging whether the arrived parking space is the target parking space, and if yes, ending navigation. Compared with the prior art, the method can provide rapid parking space navigation, reduces the operation amount of a user, and is helpful for relieving the urban traffic pressure.

Description

Technical Field

[0001] The present invention relates to the technical field of parking navigation, and in particular to a method and system for automatic parking navigation based on machine vision. Background Art

[0002] Traditional parking management systems usually use manual management, which is not very intelligent and cannot meet the real-time management needs of large-scale parking lots. Satellite signals are severely attenuated in indoor environments, which means that outdoor navigation methods cannot be used directly for indoor positioning. At present, the positioning technology used by mainstream underground garage parking systems is usually wireless positioning technology, including radio frequency indoor positioning technologies such as RFID, Bluetooth and WiFi. These technologies require the deployment of a large number of base stations indoors, which increases costs.

[0003] Some scholars have proposed a low-cost underground garage car search and intelligent navigation system design based on machine vision. Users actively upload pictures of their location, and after positioning through a 2D-3D point matching algorithm, the navigation information is sent back to the user. This method requires users to take photos, which is not conducive to driving safety. At the same time, it is necessary to wait until the recognition and positioning results are sent back, which increases the navigation time and easily causes traffic jams. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a machine vision-based automatic parking navigation method and system that reduces the user operation amount and can provide fast parking space navigation.

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

[0006] A method for automatic garage navigation based on machine vision comprises the following steps:

[0007] Obtaining free parking space information, and allocating parking positions for vehicles to be parked according to the free parking space information;

[0008] After confirming the parking space allocation information of the current vehicle to be parked, vehicle navigation is performed;

[0009] During vehicle navigation, intersection recognition is performed based on driving image data collected by the vehicle camera;

[0010] When an intersection is identified, the current intersection information is matched with the preset map information to obtain the corresponding turning information;

[0011] When the vehicle to be parked completes the turn and approaches the target garage in the parking space allocation information, the specific location information of the target parking space in the target garage is displayed;

[0012] When the vehicle to be parked approaches the target parking space, the parking space number is identified based on the driving image data collected by the vehicle camera;

[0013] The parking space number recognition result determines whether the parking space reached is the target parking space. If so, the navigation ends.

[0014] Furthermore, the method recognizes parking space numbers through a pre-trained MobileNetV2 neural network model.

[0015] Furthermore, the free parking space information is allocated based on current parking resources in the city.

[0016] Furthermore, the method also includes providing voice announcements and display reminders to the driver.

[0017] The present invention also provides a machine vision-based garage automatic navigation system, comprising a core processor, a main control chip, a camera and a display module. The processing process of the garage automatic navigation system comprises the following steps:

[0018] Obtaining free parking space information, and allocating parking positions for vehicles to be parked according to the free parking space information;

[0019] Confirm the parking position allocation information on the vehicle to be parked, and then perform vehicle navigation;

[0020] During vehicle navigation, the core processor uses the driving image data collected by the camera to perform intersection recognition;

[0021] When the core processor identifies an intersection, it sends the information about the intersection it has reached to the main control chip;

[0022] After receiving the intersection information, the main control chip matches it with the preset map information to obtain the corresponding turning information;

[0023] When the vehicle to be parked completes the turn and approaches the target garage in the parking space allocation information, the specific position information of the target parking space in the target garage is displayed through the display module;

[0024] When the vehicle to be parked approaches the target parking space, it enters the number recognition mode;

[0025] In the number recognition mode, the core processor recognizes the parking space number based on the driving image data collected by the camera;

[0026] The core processor communicates the parking space number recognition result with the main control chip to determine whether the parking space reached is the target parking space;

[0027] When the identified parking space number matches the target parking space, navigation ends.

[0028] Furthermore, the core processor recognizes the parking space number through a pre-trained MobileNetV2 neural network model.

[0029] Furthermore, the free parking space information is allocated based on current parking resources in the city.

[0030] Furthermore, the camera is an OpenArt mini camera.

[0031] Furthermore, when an intersection is identified, the core processor sends the information of the arrived intersection to the main control chip via serial port communication.

[0032] Furthermore, the core processor is a single chip microcomputer.

[0033] Compared with the prior art, the present invention has the following advantages:

[0034] (1) The automatic parking navigation method provided by the present invention allocates parking spaces based on the current parking resources of the city before navigation begins, so that the marginal parking spaces are effectively utilized, providing an effective solution for the reasonable allocation of urban parking resources;

[0035] During the navigation process, the system identifies the garage road information, especially the intersection information, and provides navigation guidance information to the target parking space, so that users can quickly find the target garage location, effectively alleviating the problems of traditional cruise parking.

[0036] After arriving at the target garage, the system can quickly and accurately provide a navigation route by introducing the MobilenetV2 neural network model to identify and confirm the parking space number, guide the car to the designated parking space, and confirm the parking space.

[0037] Overall, this solution has low hardware cost, is easy to operate, and can greatly improve parking efficiency.

[0038] (2) The navigation system of the present invention can shorten the time vehicles stay in parking lots by providing fast parking navigation, reduce traffic congestion near parking lot entrances and exits, and help ease urban traffic pressure.

[0039] (3) The present invention also provides a user interactive function, providing voice broadcast and display reminders for the driver, thereby improving user friendliness. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic flow chart of a method for automatic garage navigation based on machine vision provided in an embodiment of the present invention;

[0041] Figure 2The present invention is a schematic diagram of an example flow chart of a method for automatic garage navigation based on machine vision provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0043] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. 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.

[0044] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0045] Example 1

[0046] like Figure 1 As shown, this embodiment provides a garage automatic navigation method based on machine vision, comprising the following steps:

[0047] S1: Obtaining free parking space information, and allocating parking positions for vehicles to be parked according to the free parking space information; wherein the free parking space information is allocated based on the current parking resources of the city;

[0048] Allocate parking spaces so that marginal parking spaces can be effectively utilized, providing an effective solution for the rational allocation of urban parking resources;

[0049] S2: After confirming the parking space allocation information of the current vehicle to be parked, perform vehicle navigation;

[0050] The navigation system can shorten the time vehicles spend in parking lots by providing fast parking navigation, reduce traffic congestion near parking lot entrances and exits, and help ease urban traffic pressure;

[0051] S3: During vehicle navigation, intersection recognition is performed based on driving image data collected by the vehicle camera;

[0052] S4: When an intersection is identified, the current intersection information is matched with the preset map information to obtain corresponding turning information;

[0053] S5: When the vehicle to be parked completes the turn and approaches the target garage in the parking space allocation information, the specific location information of the target parking space in the target garage is displayed;

[0054] S6: When the vehicle to be parked approaches the target parking space, the parking space number is identified based on the driving image data collected by the vehicle camera;

[0055] Preferably, the parking space number recognition is performed through the pre-trained MobileNetV2 neural network model, and the MobilenetV2 neural network model is introduced to recognize and confirm the parking space number. The system can quickly and accurately give a navigation route, guide the car to the designated parking space, and confirm the parking space;

[0056] S7: Determine whether the parking space reached is the target parking space based on the parking space number recognition result. If yes, the navigation ends.

[0057] like Figure 2 As shown, in this embodiment, the working process of the above-mentioned garage automatic navigation method includes:

[0058] S10: Allocate parking space;

[0059] S11: Determine whether the allocated parking space is determined, if so, execute step S11, otherwise return to step S10;

[0060] S12: Start vehicle navigation and identify the intersection through the vehicle camera. If it is determined that the intersection has been reached, execute step S13, otherwise continue to execute step S12;

[0061] S13: After reaching the intersection, matching the current intersection information with the preset map information, obtaining and prompting corresponding turning information;

[0062] S14: Perform intersection turning according to the turning information, and determine whether the turning is completed. If completed, execute the auxiliary S15; otherwise, continue to execute step S14;

[0063] S15: The vehicle continues to move forward;

[0064] S16: Determine whether the second intersection has been reached, if so, execute step S17; otherwise, continue to execute step S16;

[0065] S17: Matching the current intersection information with the preset map information, obtaining and prompting corresponding turning information;

[0066] S18: Perform intersection turning according to the turning information, and determine whether the turning is completed. If completed, execute the auxiliary S19; otherwise, continue to execute step S18;

[0067] S19: When the vehicle to be parked approaches the target garage in the parking space allocation information, the specific location information of the target parking space in the target garage is displayed;

[0068] S20: When the vehicle to be parked approaches the target parking space, the parking space number is identified based on the driving image data collected by the vehicle camera;

[0069] S21: Determine whether the parking space reached is the target parking space. If so, the navigation ends. Otherwise, continue to execute step S21.

[0070] Preferably, the method also includes a user interaction function to provide voice broadcast and display reminders to the driver.

[0071] This embodiment also provides a machine vision-based garage automatic navigation system, including a core processor, a main control chip, a camera and a display module, with a single-chip microcomputer as the core processor, and the main control chip coordinating the collaborative work of various modules in the entire system;

[0072] The process of the garage automatic navigation system includes the following steps:

[0073] After the system starts, it obtains the free parking space information and allocates parking positions for the vehicles to be parked according to the free parking space information;

[0074] After the user confirms the parking space allocation information by pressing a button, the system officially starts navigation;

[0075] During vehicle navigation, the core processor uses the driving image data collected by the OpenArt mini camera to perform intersection recognition;

[0076] When the core processor identifies an intersection, it sends the information of the arrived intersection to the main control chip via serial communication;

[0077] After receiving the intersection information, the main control chip matches it with the preset map information to obtain the corresponding turning information;

[0078] When the vehicle to be parked completes the turn and approaches the target garage in the parking space allocation information, the specific position information of the target parking space in the target garage is displayed through the display module;

[0079] When the vehicle to be parked approaches the target parking space, the camera module will enter the number recognition mode;

[0080] In the number recognition mode, the core processor recognizes the parking space number based on the driving image data collected by the camera. In this embodiment, the parking space number is recognized by the pre-trained MobileNetV2 neural network model;

[0081] The core processor communicates the parking space number recognition result with the main control chip to determine whether the parking space reached is the target parking space;

[0082] When the identified parking space number matches the target parking space, navigation ends.

[0083] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A garage automatic navigation method based on machine vision, characterized in that: The following steps are involved: Obtaining free parking space information, and allocating parking positions for vehicles to be parked according to the free parking space information; After confirming the parking space allocation information of the current vehicle to be parked, vehicle navigation is performed; During vehicle navigation, intersection recognition is performed based on driving image data collected by the vehicle camera; When an intersection is identified, the current intersection information is matched with the preset map information to obtain the corresponding turning information; When the vehicle to be parked completes the turn and approaches the target garage in the parking space allocation information, the specific location information of the target parking space in the target garage is displayed; When the vehicle to be parked approaches the target parking space, the parking space number is identified based on the driving image data collected by the vehicle camera; The parking space number recognition result determines whether the parking space reached is the target parking space. If so, the navigation ends.

2. The method for automatic garage navigation based on machine vision according to claim 1, characterized in that: The method uses a pre-trained MobileNetV2 neural network model to recognize parking space numbers.

3. The automatic garage navigation method based on machine vision according to claim 1 is characterized in that: The free parking space information is allocated based on the current parking resources of the city.

4. The automatic garage navigation method based on machine vision according to claim 1, characterized in that: The method also includes providing a voice announcement and a display reminder to the driver.

5. A garage automatic navigation system based on machine vision, characterized in that: Including a core processor, a main control chip, a camera and a display module, the processing process of the garage automatic navigation system includes the following steps: Obtaining free parking space information, and allocating parking positions for vehicles to be parked according to the free parking space information; Confirm the parking position allocation information on the vehicle to be parked, and then perform vehicle navigation; During vehicle navigation, the core processor uses the driving image data collected by the camera to perform intersection recognition; When the core processor identifies an intersection, it sends the information about the intersection it has reached to the main control chip; After receiving the intersection information, the main control chip matches it with the preset map information to obtain the corresponding turning information; When the vehicle to be parked completes the turn and approaches the target garage in the parking space allocation information, the specific position information of the target parking space in the target garage is displayed through the display module; When the vehicle to be parked approaches the target parking space, it enters the number recognition mode; In the number recognition mode, the core processor recognizes the parking space number based on the driving image data collected by the camera; The core processor communicates the parking space number recognition result with the main control chip to determine whether the parking space reached is the target parking space; When the identified parking space number matches the target parking space, navigation ends.

6. The machine vision-based automatic garage navigation system according to claim 5, characterized in that: The core processor recognizes parking space numbers through a pre-trained MobileNetV2 neural network model.

7. The machine vision-based automatic garage navigation system according to claim 5, characterized in that: The free parking space information is allocated based on the current parking resources of the city.

8. The machine vision-based automatic garage navigation system according to claim 5, characterized in that: The camera is an OpenArt mini camera.

9. The machine vision-based automatic garage navigation system according to claim 5, characterized in that: When an intersection is identified, the core processor sends the information of the arrived intersection to the main control chip via serial port communication.

10. The machine vision-based automatic garage navigation system according to claim 5, characterized in that: The core processor is a single chip microcomputer.

Images