CN107203992A - Intelligent identification method for shared bicycles - Google Patents

Abstract

The invention relates to an intelligent identification method for a shared bicycle, which comprises the following steps: determining whether the vertical beam of the sharing bicycle is in a vertical state or not based on the bubble state of the horizontal measuring instrument arranged on the vertical beam of the sharing bicycle; when the vertical beam of the sharing bicycle is determined to be in the vertical state, a first vertical signal is sent out, and when the vertical beam of the sharing bicycle is determined to be in the horizontal state, a first horizontal signal is sent out; when receiving the first lying signal, starting data reception of the remote sensing satellite, and when receiving the first standing signal, closing data reception of the remote sensing satellite; wherein, when receiving first lying signal, open the data reception to the remote sensing satellite, when receiving first upright signal, close the data reception to the remote sensing satellite and include: and using remote sensing image receiving equipment for starting or closing data receiving of the remote sensing satellite when receiving the first lying signal or the first standing signal, wherein the data received from the remote sensing satellite is a remote sensing image.

Description

Intelligent identification method for shared bicycles

technical field

The invention relates to the field of shared bicycles, in particular to an intelligent identification method for shared bicycles.

Background technique

According to the "2016 China Shared Bicycle Market Research Report" recently released by a third-party data research organization, Beda Consulting, by the end of 2016, the overall number of users in China's shared bicycle market had reached 18.86 million. It is expected that in 2017, the scale of users in the shared bicycle market will continue to maintain Substantial growth, will reach 50 million users by the end of the year.

The report pointed out that China's shared bicycle market has gone through three stages of development. The first stage was from 2007 to 2010. The public bicycle model that emerged from abroad began to be introduced into China. The government led the management of cities, and most of them were bicycles with piles. The second stage was from 2010 to 2014. Enterprises specializing in the bicycle market began to appear, but public bicycles were still dominated by piled bicycles. From 2014 to the present is the third stage. With the rapid development of the mobile Internet, Internet shared bicycles led by OFO emerged as the times require, and more convenient dockless bicycles began to replace docked bicycles.

According to the report, at present, OFO and Mobike have obvious advantages in China's shared bicycle market. Among them, OFO has the largest number of bicycles, reaching 800,000 units, with a market share of 51.2%; Mobike has a market share of 600,000 units. 40.1%.

The report also shows that shared bicycles are more popular with young men. Men account for 54.2% of China's shared bicycle users, while women account for 45.8%. In the age distribution of users, people aged 25-35 use the most, followed by people under the age of 25. Among the frequency of use, the users who use it 3-4 times a week are the most.

With the vigorous development of shared bicycles, some problems brought about by them have also surfaced and need to be solved. For example, because the distribution of shared bicycles is too fragmented, it is difficult for its operators to allocate a large amount of manpower and material resources for the repair and maintenance of shared bicycles, resulting in some shared bicycles being down for a long time. At this time, it will not only affect the urban landscape, but also reduce Customers' desire to ride, thus forming a vicious circle.

Contents of the invention

In order to solve the above problems, the present invention provides an intelligent identification method for shared bicycles, which adopts the mode of remote sensing image analysis after detection by a level detector, and reduces the number of times of using cumbersome remote sensing images for analysis without affecting the analysis results, thereby The efficiency and accuracy of dumping state detection are improved, and most importantly, a targeted and high-precision image filtering processing mode is adopted, and adaptive step-by-step filtering processing is performed based on image content, which improves the accuracy of remote sensing images to be detected. clarity.

According to one aspect of the present invention, an intelligent identification method for a shared bicycle is provided, the method comprising:

Using a level measuring instrument arranged on the vertical beam of the shared bicycle to determine whether the vertical beam of the shared bicycle is in an upright state based on the state of the air bubbles therein;

When it is determined that the vertical beam of the shared bicycle is in an upright state, a first erect signal is sent, and when it is determined that the vertical beam of the shared bicycle is in a horizontal state, a first horizontal signal is sent;

When the first lying signal is received, the data reception to the remote sensing satellite is turned on, and when the first upright signal is received, the data reception to the remote sensing satellite is turned off;

Wherein, when the first lying signal is received, the data reception of the remote sensing satellite is started, and when the first erection signal is received, the data reception of the remote sensing satellite is turned off includes: using a remote sensing image receiving device for After receiving the first lying signal or the first erecting signal, the data reception to the remote sensing satellite is turned on or off, and the data received from the remote sensing satellite is a remote sensing image.

More specifically, in the intelligent identification method for shared bicycles, it also includes:

Use the first filtering device, connected with the remote sensing image receiving device, for receiving remote sensing images, and simultaneously perform wavelet filtering processing, Wiener filtering processing, median filtering processing and Gaussian low-pass filtering processing on the remote sensing images, to respectively Obtaining a first filtered image, a second filtered image, a third filtered image and a fourth filtered image, and at the same time SNR analysis is performed to obtain the first SNR, the second SNR, the third SNR and the fourth SNR respectively, and the SNR with the largest numerical value is selected as the target from the four SNRs Signal-to-noise ratio, using the filtered image corresponding to the target signal-to-noise ratio as the target filtered image;

An edge enhancement device is used, connected to the first filtering device, for performing edge enhancement processing on the target filtered image to obtain an edge enhancement image.

More specifically, in the intelligent identification method for shared bicycles, it also includes:

Use a noise analysis device, connected to the edge enhancement device, for analyzing the noise components of the edge enhancement image to obtain various noise types in the edge enhancement image and corresponding noise signal components, and obtain each Selecting the three noise signal components with the largest amplitudes from the noise signal components and sorting them according to the amplitudes from large to small are respectively used as the first noise signal component, the second noise signal component and the third noise signal component;

Use the second filtering device, respectively connected to the edge enhancement device and the noise analysis device, for searching from the image filtering template library corresponding to the first noise signal component, the second noise signal component and the third noise signal component respectively The image filtering template is used as a first filtering template, a second filtering template and a third filtering template, and filtering is performed on the edge-enhanced image based on the first filtering template, the second filtering template and the third filtering template processed to obtain the final filtered image;

Using an image recognition device, connected to the second filtering device, for receiving the final filtered image, and performing target recognition and target state analysis on the final filtered image, to determine whether the shared bicycle in the final filtered image is It is a lying profile, if not, send a second erect signal, if yes, send a second lie signal;

Adaptive pricing equipment is used to connect with the image recognition equipment, and when the second lying signal is received, the reward amount is formulated for the cyclist who chooses to ride the corresponding shared bicycle, and when the second erecting signal is received, for The cyclist who chooses to ride the corresponding shared bicycle sets the charging amount;

Wherein, the second filtering device performs filtering processing on the edge-enhanced image based on the first filtering template, the second filtering template and the third filtering template to obtain a final filtered image includes: first using the first filtering template A filtering template performs filtering processing on the edge-enhanced image to obtain a first intermediate filtering image, and then uses the second filtering template to perform filtering processing on the first intermediate filtering image to obtain a second intermediate filtering image, and finally uses the The third filtering template performs filtering processing on the second intermediate filtering image to obtain a final filtering image.

More specifically, in the intelligent identification method of the shared bicycle, it also includes: using a Big Dipper navigation device to provide current navigation data of the shared bicycle in real time.

More specifically, in the intelligent recognition method for shared bicycles: the image recognition device is connected to the Big Dipper navigation device, and the image recognition device performs target recognition and target state analysis on the final filtered image to determine the Whether the shared bicycle in the final filtered image is a horizontal profile includes: receiving the current navigation data, performing target recognition of the shared bicycle from the final filtered image based on the current navigation data, to determine and segment the final Shared bike sub-images in filtered images.

More specifically, in the intelligent recognition method for shared bicycles: the image recognition device performs target recognition and target state analysis on the final filtered image to determine whether the shared bicycle in the final filtered image is a lying type The outline further includes: based on the matching result of the shared bicycle sub-image and the reference shared bicycle pattern, determining whether the shared bicycle in the final filtered image is a horizontal outline.

Description of drawings

Embodiments of the present invention will be described below in conjunction with the accompanying drawings, wherein:

Fig. 1 is a structural block diagram of an intelligent identification system for shared bicycles according to an embodiment of the present invention.

Fig. 2 is a flow chart showing steps of an intelligent identification method for shared bicycles according to an embodiment of the present invention.

Reference signs: 1 level measuring instrument; 2 bubble state detection equipment; 3 remote sensing start-up equipment; 4 remote sensing image receiving equipment; Whether the vertical beam is in an upright state; S102 sends a first erecting signal when determining that the vertical beam of a shared bicycle is in an upright state, and sends a first lying signal when determining that the vertical beam of a shared bicycle is in a lying state; S103 receives When the first lying signal is used, the data reception of the remote sensing satellite is turned on, and when the first erect signal is received, the data reception of the remote sensing satellite is turned off

detailed description

The implementation of the intelligent identification method for shared bicycles of the present invention will be described in detail below with reference to the accompanying drawings.

At present, due to the rapid development of shared bicycles, it is difficult for their post-management and maintenance to keep up with the growth of their number. At the same time, unlike online car-hailing, the operation of bicycles is also greatly affected by seasonal changes and weather conditions. As for the typhoon, no matter where it is located, the order volume of public bicycle travel will plummet or even return to zero, and the platform will have to face even higher vehicle damage and depreciation costs. Compared with the public bicycles with "stakes", this "stakeless" concept of taking and parking at any time has brought great convenience to citizens, but it has also led to "little red cars" and "little yellow cars". The phenomenon of "occupying roads" is more common, and the management of urban space becomes more difficult.

Due to random parking or strong wind blowing down, once the shared bicycles are in the horizontal state, the mentality of customers willing to ride will change immediately, and they may choose to ride the bicycles that seem to be well maintained and neatly arranged. In this way, After a long time, the shared bicycles placed horizontally will become more obsolete and even scrapped, causing great economic losses to operators.

In order to overcome the above-mentioned shortcomings, the present invention builds an intelligent recognition system and method for shared bicycles, which can immediately determine whether the shared bicycles are in a dumped state, and notify the background in time when they are in a dumped state and provide riding conditions for riders, thereby The effect of self-maintenance and management is achieved, and the economic cost of the operator is reduced.

Fig. 1 is a structural block diagram of a shared bicycle intelligent identification system shown according to an embodiment of the present invention, the system comprising:

The level measuring instrument is arranged on the vertical beam of the shared bicycle, and is used to determine whether the vertical beam of the shared bicycle is in an upright state based on the state of air bubbles therein;

Bubble state detection equipment, connected with the level measuring instrument, is used to send a first erect signal when it is determined that the vertical beam of the shared bicycle is in an upright state, and send a first horizontal signal when it is determined that the vertical beam of the shared bicycle is in a horizontal state. lie signal;

The remote sensing start-up device is connected with the bubble state detection device, and is used to start receiving data from remote sensing satellites when receiving the first lying signal, and turn off data receiving from remote sensing satellites when receiving the first erecting signal. data reception;

The remote sensing image receiving device is connected with the remote sensing enabling device, and is used for enabling or disabling data reception from remote sensing satellites under the control of the remote sensing enabling device, and the data received from the remote sensing satellites are remote sensing images.

Next, continue to further describe the specific structure of the shared bicycle intelligent identification system of the present invention.

In the intelligent identification system for shared bicycles, it also includes:

The first filtering device is connected with the remote sensing image receiving device and is used to receive remote sensing images, and simultaneously perform wavelet filtering, Wiener filtering, median filtering and Gaussian low-pass filtering on the remote sensing images to obtain The first filtered image, the second filtered image, the third filtered image and the fourth filtered image, and the first filtered image, the second filtered image, the third filtered image and the fourth filtered image are simultaneously Signal-to-noise ratio analysis to obtain the first signal-to-noise ratio, the second signal-to-noise ratio, the third signal-to-noise ratio and the fourth signal-to-noise ratio, and select the signal-to-noise ratio with the largest numerical value as the target signal-to-noise ratio from the four signal-to-noise ratios. Noise ratio, using the filtered image corresponding to the target signal-to-noise ratio as the target filtered image;

An edge enhancement device, connected to the first filtering device, configured to perform edge enhancement processing on the target filtered image to obtain an edge enhancement image.

In the intelligent identification system for shared bicycles, it also includes:

A noise analysis device, connected to the edge enhancement device, configured to perform noise component analysis on the edge enhancement image to obtain various noise types in the edge enhancement image and corresponding noise signal components, and obtain each noise Selecting the three noise signal components with the largest amplitudes from the signal components and sorting them according to the amplitudes from large to small are respectively used as the first noise signal component, the second noise signal component and the third noise signal component;

The second filtering device is respectively connected with the edge enhancement device and the noise analysis device, and is used to search the image filtering template library corresponding to the first noise signal component, the second noise signal component and the third noise signal component respectively. an image filtering template as a first filtering template, a second filtering template, and a third filtering template, and performing filtering processing on the edge-enhanced image based on the first filtering template, the second filtering template, and the third filtering template to obtain the final filtered image;

An image recognition device, connected to the second filtering device, for receiving the final filtered image, and performing target recognition and target state analysis on the final filtered image, so as to determine whether the shared bicycle in the final filtered image is Recumbent profile, if not a second erect signal, if yes a second recumbent signal;

The self-adaptive pricing device is connected with the image recognition device, and is used to formulate a reward amount for the rider who chooses to ride the corresponding shared bicycle when receiving the second lying signal; The cyclist who rides the corresponding shared bicycle sets the charging amount;

Wherein, the second filtering device performs filtering processing on the edge-enhanced image based on the first filtering template, the second filtering template and the third filtering template to obtain a final filtered image includes: first using the first filtering template A filtering template performs filtering processing on the edge-enhanced image to obtain a first intermediate filtering image, and then uses the second filtering template to perform filtering processing on the first intermediate filtering image to obtain a second intermediate filtering image, and finally uses the The third filtering template performs filtering processing on the second intermediate filtering image to obtain a final filtering image.

In the intelligent identification system for shared bicycles, it also includes: a Big Dipper navigation device, which is used to provide current navigation data of shared bicycles in real time.

In the shared bicycle intelligent recognition system: the image recognition device is connected to the Big Dipper navigation device, and the image recognition device performs target recognition and target state analysis on the final filtered image to determine the final filtered image Whether the shared bicycle is a horizontal profile includes: receiving the current navigation data, performing target recognition of the shared bicycle from the final filtered image based on the current navigation data, to determine and segment the final filtered image Sharing bicycle images.

In the shared bicycle intelligent recognition system: the image recognition device performs target recognition and target state analysis on the final filtered image to determine whether the shared bicycle in the final filtered image is a horizontal profile and also includes: Based on the matching result of the shared bicycle sub-image and the reference shared bicycle pattern, it is determined whether the shared bicycle in the final filtered image is a horizontal profile.

Fig. 2 is a flow chart of the steps of a shared bicycle intelligent identification method shown according to an embodiment of the present invention, the method comprising:

Using a level measuring instrument arranged on the vertical beam of the shared bicycle to determine whether the vertical beam of the shared bicycle is in an upright state based on the state of the air bubbles therein;

When it is determined that the vertical beam of the shared bicycle is in an upright state, a first erect signal is sent, and when it is determined that the vertical beam of the shared bicycle is in a horizontal state, a first horizontal signal is sent;

When the first lying signal is received, the data reception to the remote sensing satellite is turned on, and when the first upright signal is received, the data reception to the remote sensing satellite is turned off;

Wherein, when the first lying signal is received, the data reception of the remote sensing satellite is started, and when the first erection signal is received, the data reception of the remote sensing satellite is turned off includes: using a remote sensing image receiving device for After receiving the first lying signal or the first erecting signal, the data reception to the remote sensing satellite is turned on or off, and the data received from the remote sensing satellite is a remote sensing image.

Next, the specific flow of the intelligent identification method for shared bicycles of the present invention will be further described.

In the intelligent identification method of shared bicycles, it also includes:

Use the first filtering device, connected with the remote sensing image receiving device, for receiving remote sensing images, and simultaneously perform wavelet filtering processing, Wiener filtering processing, median filtering processing and Gaussian low-pass filtering processing on the remote sensing images, to respectively Obtaining a first filtered image, a second filtered image, a third filtered image and a fourth filtered image, and at the same time SNR analysis is performed to obtain the first SNR, the second SNR, the third SNR and the fourth SNR respectively, and the SNR with the largest numerical value is selected as the target from the four SNRs Signal-to-noise ratio, using the filtered image corresponding to the target signal-to-noise ratio as the target filtered image;

An edge enhancement device is used, connected to the first filtering device, for performing edge enhancement processing on the target filtered image to obtain an edge enhancement image.

In the intelligent identification method of shared bicycles, it also includes:

Use a noise analysis device, connected to the edge enhancement device, for analyzing the noise components of the edge enhancement image to obtain various noise types in the edge enhancement image and corresponding noise signal components, and obtain each Selecting the three noise signal components with the largest amplitudes from the noise signal components and sorting them according to the amplitudes from large to small are respectively used as the first noise signal component, the second noise signal component and the third noise signal component;

Use the second filtering device, respectively connected to the edge enhancement device and the noise analysis device, for searching from the image filtering template library corresponding to the first noise signal component, the second noise signal component and the third noise signal component respectively The image filtering template is used as a first filtering template, a second filtering template and a third filtering template, and filtering is performed on the edge-enhanced image based on the first filtering template, the second filtering template and the third filtering template processed to obtain the final filtered image;

Using an image recognition device, connected to the second filtering device, for receiving the final filtered image, and performing target recognition and target state analysis on the final filtered image, to determine whether the shared bicycle in the final filtered image is It is a lying profile, if not, send a second erect signal, if yes, send a second lie signal;

Adaptive pricing equipment is used to connect with the image recognition equipment, and when the second lying signal is received, the reward amount is formulated for the cyclist who chooses to ride the corresponding shared bicycle, and when the second erecting signal is received, for The cyclist who chooses to ride the corresponding shared bicycle sets the charging amount;

Wherein, the second filtering device performs filtering processing on the edge-enhanced image based on the first filtering template, the second filtering template and the third filtering template to obtain a final filtered image includes: first using the first filtering template A filtering template performs filtering processing on the edge-enhanced image to obtain a first intermediate filtering image, and then uses the second filtering template to perform filtering processing on the first intermediate filtering image to obtain a second intermediate filtering image, and finally uses the The third filtering template performs filtering processing on the second intermediate filtering image to obtain a final filtering image.

In the intelligent identification method of the shared bicycle, it also includes: using a Big Dipper navigation device for providing current navigation data of the shared bicycle in real time.

In the intelligent recognition method for shared bicycles: the image recognition device is connected to the Beidou navigation device, and the image recognition device performs target recognition and target state analysis on the final filtered image to determine the final filtered image Whether the shared bicycle is a horizontal profile includes: receiving the current navigation data, performing target recognition of the shared bicycle from the final filtered image based on the current navigation data, to determine and segment the final filtered image Sharing bicycle images.

In the intelligent recognition method for shared bicycles: the image recognition device performs target recognition and target state analysis on the final filtered image to determine whether the shared bicycle in the final filtered image is a horizontal profile and further includes: Based on the matching result of the shared bicycle sub-image and the reference shared bicycle pattern, it is determined whether the shared bicycle in the final filtered image is a horizontal profile.

In addition, image filtering, that is, to suppress the noise of the target image under the condition of preserving image details as much as possible, is an indispensable operation in image preprocessing, and its processing effect will directly affect the effectiveness of subsequent image processing and analysis. sex and reliability.

Due to the imperfection of imaging system, transmission medium and recording equipment, digital images are often polluted by various noises in the process of formation, transmission and recording. In addition, in some aspects of image processing, when the input object is not as expected, noise will be introduced into the resulting image. These noises often appear as an isolated pixel point or pixel block that causes a strong visual effect on the image. Generally, the noise signal is irrelevant to the object to be studied, it appears in the form of useless information, disturbing the observable information of the image. For digital image signals, the noise table is a large or small extreme value. These extreme values act on the real gray value of the image pixel through addition and subtraction, causing bright and dark spot interference to the image, which greatly reduces the image quality and affects Follow-up work such as image restoration, segmentation, feature extraction, and image recognition. To construct a filter that effectively suppresses noise, two basic issues must be considered: it can effectively remove the noise in the target and background; at the same time, it can well protect the shape, size and specific geometric and topological structure characteristics of the image target.

One of the commonly used image filtering modes is the nonlinear filter. Generally speaking, when the signal spectrum and the noise spectrum are aliased or when the signal contains non-superimposed noise, such as the noise caused by the nonlinearity of the system or the presence of non-linear Gaussian noise, etc.), traditional linear filtering techniques, such as Fourier transform, will always blur image details (such as edges, etc.) reduced sex. The nonlinear filter is based on a nonlinear mapping relationship to the input signal, and can often map a specific noise approximately to zero while retaining the main characteristics of the signal, so it can overcome the limitations of the linear filter to a certain extent. Inadequacies.

Using the shared bicycle intelligent recognition system and method of the present invention, aiming at the technical problem of high maintenance cost of shared bicycles in the prior art, through the dual detection of the level detector and remote sensing image, the accuracy of the situation detection of the shared bicycle itself is improved, and based on The detection results enable the adaptive pricing device to formulate reward amounts for riders who choose to ride the corresponding shared bike when the shared bike is lying down, and set a charging amount for riders who choose to ride the corresponding shared bike when the shared bike is upright. In this way, while providing travel convenience for cyclists, the shared bicycle itself is maintained.

It can be understood that although the present invention has been disclosed above with preferred embodiments, the above embodiments are not intended to limit the present invention. For any person skilled in the art, without departing from the scope of the technical solution of the present invention, the technical content disclosed above can be used to make many possible changes and modifications to the technical solution of the present invention, or be modified into equivalent changes, etc. effective example. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention, which do not deviate from the technical solution of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims (6)

1. A method for intelligent identification of shared bicycles, characterized in that the method comprises: Using a level measuring instrument arranged on the vertical beam of the shared bicycle to determine whether the vertical beam of the shared bicycle is in an upright state based on the state of the air bubbles therein; When it is determined that the vertical beam of the shared bicycle is in an upright state, a first erect signal is sent, and when it is determined that the vertical beam of the shared bicycle is in a horizontal state, a first horizontal signal is sent; When the first lying signal is received, the data reception to the remote sensing satellite is turned on, and when the first upright signal is received, the data reception to the remote sensing satellite is turned off; Wherein, when the first lying signal is received, the data reception of the remote sensing satellite is started, and when the first erection signal is received, the data reception of the remote sensing satellite is turned off includes: using a remote sensing image receiving device for After receiving the first lying signal or the first erecting signal, the data reception to the remote sensing satellite is turned on or off, and the data received from the remote sensing satellite is a remote sensing image. 2. The intelligent identification method for shared bicycles as claimed in claim 1, further comprising: Use the first filtering device, connected with the remote sensing image receiving device, for receiving remote sensing images, and simultaneously perform wavelet filtering processing, Wiener filtering processing, median filtering processing and Gaussian low-pass filtering processing on the remote sensing images, to respectively Obtaining a first filtered image, a second filtered image, a third filtered image and a fourth filtered image, and at the same time SNR analysis is performed to obtain the first SNR, the second SNR, the third SNR and the fourth SNR respectively, and the SNR with the largest numerical value is selected as the target from the four SNRs Signal-to-noise ratio, using the filtered image corresponding to the target signal-to-noise ratio as the target filtered image; An edge enhancement device is used, connected to the first filtering device, for performing edge enhancement processing on the target filtered image to obtain an edge enhancement image. 3. The shared bicycle intelligent identification method as claimed in claim 2, further comprising: Use a noise analysis device, connected to the edge enhancement device, for analyzing the noise components of the edge enhancement image to obtain various noise types in the edge enhancement image and corresponding noise signal components, and obtain each Selecting the three noise signal components with the largest amplitudes from the noise signal components and sorting them according to the amplitudes from large to small are respectively used as the first noise signal component, the second noise signal component and the third noise signal component; Use the second filtering device, respectively connected to the edge enhancement device and the noise analysis device, for searching from the image filtering template library corresponding to the first noise signal component, the second noise signal component and the third noise signal component respectively The image filtering template is used as a first filtering template, a second filtering template and a third filtering template, and filtering is performed on the edge-enhanced image based on the first filtering template, the second filtering template and the third filtering template processed to obtain the final filtered image; Using an image recognition device, connected to the second filtering device, for receiving the final filtered image, and performing target recognition and target state analysis on the final filtered image, to determine whether the shared bicycle in the final filtered image is It is a lying profile, if not, send a second erect signal, if yes, send a second lie signal; Adaptive pricing equipment is used to connect with the image recognition equipment, and when the second lying signal is received, the reward amount is formulated for the cyclist who chooses to ride the corresponding shared bicycle, and when the second erecting signal is received, for The cyclist who chooses to ride the corresponding shared bicycle sets the charging amount; Wherein, the second filtering device performs filtering processing on the edge-enhanced image based on the first filtering template, the second filtering template and the third filtering template to obtain a final filtered image includes: first using the first filtering template A filtering template performs filtering processing on the edge-enhanced image to obtain a first intermediate filtering image, and then uses the second filtering template to perform filtering processing on the first intermediate filtering image to obtain a second intermediate filtering image, and finally uses the The third filtering template performs filtering processing on the second intermediate filtering image to obtain a final filtering image. 4. The shared bicycle intelligent identification method as claimed in claim 3, further comprising: Use Beidou navigation equipment to provide current navigation data of shared bicycles in real time. 5. The shared bicycle intelligent identification method as claimed in claim 4, characterized in that: The image recognition device is connected with the Big Dipper navigation device, and the image recognition device performs target recognition and target state analysis on the final filtered image to determine whether the shared bicycle in the final filtered image is a horizontal profile including : receiving the current navigation data, performing target recognition of the shared bicycle from the final filtered image based on the current navigation data, so as to determine and segment a shared bicycle sub-image in the final filtered image. 6. The shared bicycle intelligent identification method as claimed in claim 5, characterized in that: The image recognition device performs target recognition and target state analysis on the final filtered image to determine whether the shared bicycle in the final filtered image is a horizontal profile. It also includes: based on the shared bicycle sub-image and the reference shared bicycle According to the matching result of the pattern, it is determined whether the shared bicycle in the final filtered image is a horizontal profile.