Intelligent recognition system for vehicle type

Abstract

The application discloses a vehicle model intelligent identification system and relates to the technical field of intelligent car washing, and solves the technical problem that the collection of a vehicle VIN code is difficult and the privacy of a vehicle owner is infringed and the car washing experience of the vehicle owner is affected when a vehicle model is identified through the vehicle VIN code in the prior art; the size features and line features of a vehicle to be washed are extracted based on appearance data of the vehicle to be washed, a target vehicle model is matched and acquired in a database, and intelligent car washing equipment is controlled based on the target vehicle model to perform car washing; the appearance features of the vehicle to be washed are extracted to match the target vehicle model, the matching efficiency of the target vehicle model is improved, and the privacy of the vehicle owner is prevented from being leaked; whether the relative position of the vehicle to be washed and the intelligent car washing equipment is abnormal is judged based on the target vehicle model; if so, the vehicle to be washed or the intelligent car washing equipment is moved; the relative position of the vehicle to be washed and the intelligent car washing equipment is judged based on the size of the vehicle model, and the vehicle to be washed is prevented from being damaged in the car washing process due to measurement errors.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of intelligent car washing, and relates to a car model intelligent identification technology for a car washing device, in particular to a car model intelligent identification system. BACKGROUND

[0002] Most of the automatic car washing devices control the car brush assembly according to a fixed path, but different car models have different shapes and lines. If the car brush assembly is controlled to run according to the same fixed path for all vehicles, the vehicle will be damaged or the cleaning efficiency and quality will be reduced.

[0003] The prior art (invention patent application with publication number CN105205468A) discloses an automatic car washing device, system and method, which collects the vehicle VIN code of the vehicle to be washed through an image collection device, analyzes it according to a predetermined rule, and compares it with the information in the VIN database to obtain the model information of the vehicle. The prior art needs to identify the vehicle according to the vehicle VIN code, but the vehicle VIN code is generally in a certain position inside the vehicle, which is difficult to collect. Moreover, most car owners are very concerned about privacy and are difficult to accept that the vehicle VIN code is used at will.

[0004] The present application provides a car model intelligent identification system to solve the above problems. SUMMARY

[0005] The present application aims to solve at least one of the technical problems existing in the prior art; for this purpose, the present application provides a car model intelligent identification system to solve the technical problem that the collection of the vehicle VIN code is difficult and the privacy of the car owner is infringed when the prior art identifies the car model through the vehicle VIN code, which affects the car owner's car washing experience.

[0006] To achieve the above purpose, the first aspect of the present application provides a car model intelligent identification system, which comprises a central control module, a data interaction module and a car washing control module connected thereto; the data interaction module is connected with a data collection device and an intelligent terminal respectively; the car washing control module is used to control an intelligent car washing device; the central control module obtains the appearance data of the vehicle to be washed through the data collection device; the size features and line features of the vehicle to be washed are extracted based on the appearance data, and the two are integrated into a feature sequence; and the target car model is determined after matching in the database through the feature sequence and verified by the intelligent terminal; it is judged whether the relative position of the vehicle to be washed and the intelligent car washing device is abnormal based on the target car model; if yes, the vehicle to be washed or the intelligent car washing device is moved; if no, the intelligent car washing device is controlled to wash the car based on the target car model.

[0007] The existing intelligent car washing equipment generally moves along a fixed path to complete the car task when washing the car. In order to avoid damage to the vehicle during the washing process, the contour surrounded by the fixed path needs to be as large as possible, so that the brush assembly is difficult to clean fully when cleaning some small size vehicles, which affects the cleaning efficiency. The vehicle model recognition method based on vehicle VIN code disclosed in the prior art may cause leakage of the owner's privacy due to the identification and non-standard use of the vehicle VIN code, and the vehicle VIN code is difficult to obtain through the image acquisition device, thereby affecting the car washing efficiency.

[0008] The present application obtains the appearance data of the vehicle to be washed by the data acquisition device, and identifies the feature sequence of the vehicle to be washed according to the appearance data. The feature sequence is a feature that does not change or changes slightly during the use of the vehicle. The target vehicle model is determined in the database according to the feature sequence, and the intelligent car washing equipment is controlled and adjusted based on the target vehicle model. The present application reasonably extracts the appearance data of the vehicle to be washed according to the control requirements of the intelligent car washing equipment, and realizes accurate and efficient control of the intelligent car washing equipment on the basis of avoiding leakage of the owner's privacy.

[0009] The central control module in the present application is in communication and / or electrical connection with the data interaction module and the car washing control module respectively. The car washing control module controls the intelligent car washing equipment according to the target vehicle model, and monitors the car washing state. The data interaction module is in communication and / or electrical connection with the data acquisition device and the intelligent terminal respectively.

[0010] The central control module is responsible for processing the data collected by the data acquisition device, and the data interaction module is responsible for data acquisition control and necessary data preprocessing. The data acquisition device mainly collects the appearance data of the vehicle to be washed, which includes the appearance image collected by the image acquisition device or the appearance scanning point collected by the three-dimensional scanning device. The intelligent terminal in the present application is used to verify the target vehicle model and display the car washing state, mainly including a smart phone or a computer.

[0011] Preferably, the central control module obtains the appearance data of the vehicle to be washed by the data acquisition device, including: determining that the vehicle to be washed is within the range that can be collected by the data acquisition device; and collecting the appearance data of several sides of the vehicle to be washed by the data acquisition device.

[0012] The appearance data in this invention includes at least a top-side appearance image, a front or rear side appearance image, and a left or right side appearance image; that is, a top-side appearance image of the vehicle to be washed is acquired from a top-down view, a left-side appearance image is acquired from a left-side view, a right-side appearance image is acquired from a right-side view, a front-side appearance image is acquired from a front-side view, and a rear-side appearance image is acquired from a rear-side view. The appearance images can be acquired directly through an image acquisition device or by parsing the appearance scan points. Clearly, acquiring the appearance data of the vehicle to be tested only requires setting up data acquisition devices inside the intelligent car wash equipment or along the washing route. The data acquisition method is simple and the acquisition cost is low; moreover, setting up devices along the washing route can offset data processing time and improve the control efficiency of the intelligent car wash equipment.

[0013] The appearance data of this invention includes at least three side appearance data. The top side appearance image is mandatory, and either the left, right, or front / back side can be selected. The features corresponding to the three side appearance data can accurately identify the corresponding vehicle model, effectively reducing the amount of data processing; moreover, only one of the left, right, front, or back side needs to be selected, which can effectively improve the data error tolerance.

[0014] Preferably, the step of extracting the size and line features of the vehicle to be washed based on the appearance data includes: extracting the appearance data of several sides of the vehicle to be washed, and obtaining the appearance dimensions of several sides after data processing; determining several appearance lines from a top-down perspective based on the appearance data of several sides; and extracting the line features corresponding to the appearance lines.

[0015] The data processing in this invention includes data verification and data correction. Data verification verifies whether the appearance data meets the requirements through pre-set verification rules, and discards data that does not meet the requirements. The appearance data acquired by the data acquisition device may have certain distortions (such as image distortion), which need to be corrected. Based on the processed data, the appearance dimensions corresponding to several sides can be obtained. After obtaining the appearance dimensions of several sides of the vehicle to be washed, several appearance lines are determined with the top view angle as the core. The line features of these appearance lines are then extracted according to the set feature extraction rules.

[0016] This invention extracts feature sequences for vehicle model identification based on the exterior data of the vehicle to be washed. Constructing identification elements from two perspectives—exterior dimensions and line features—improves element extraction efficiency and enhances the matching efficiency and accuracy of the vehicle model to be washed.

[0017] Preferably, the step of determining several exterior lines from a top-down perspective based on several side exterior data includes: extracting the top side exterior image from a top-down perspective and determining the center point of the roof in the top side exterior image as reference point one; determining several reference points two on other side edge lines in the top side exterior image, and connecting reference point one and several reference points two to obtain several exterior lines.

[0018] This invention determines the center point of the vehicle roof in a top-view image of the exterior, and determines several reference points two on other side edge lines in the top-view image. Connecting the center point of the roof with the reference points two yields several exterior lines. Determining the center point of the roof requires first determining the roof outline from the top-view image, and then determining the center point of the roof within the roof outline; of course, in some other preferred embodiments, other points can also be determined as reference points one.

[0019] In this invention, the planes on each side of the vehicle to be washed have a certain slope compared to the planes on adjacent sides. Therefore, the other side edge lines in the top-view image refer to the outlines of other sides (left, right, front, or rear, depending on the appearance data) from a top-down perspective. At least one point is selected as reference point two on each other side. To standardize the determination of reference point two, the point on the outline that contacts the most adjacent sides can be uniformly selected as reference point two. For example, the intersection point of the front, left, and top sides of the vehicle to be washed on the left front outline can be selected as reference point two.

[0020] Preferably, the step of extracting the line features corresponding to several appearance lines includes: calculating the line dimensions of several appearance lines from a top-down view; identifying any contour line of the vehicle to be washed where several appearance lines intersect; dividing the corresponding appearance line into two line segments through the contour line, marking the ratio of the two end line segments as the line ratio; and integrating the line dimensions and line ratios corresponding to each appearance line into line features.

[0021] After determining the appearance lines, this invention allows for the selection of any contour line intersecting the appearance lines, which will divide the appearance lines into two segments. The ratio of the two segment ratios is then used as the line ratio and integrated with the line dimensions of the appearance lines to form the line features.

[0022] The line features in this invention are all calculated based on previously obtained data, without the need for additional data or other auxiliary equipment, which can improve data processing efficiency and is also beneficial for subsequent target vehicle matching.

[0023] Preferably, the matching in the database using feature sequences includes: retrieving the feature sequence of the vehicle to be washed; matching the size and line features in the feature sequence in the database to obtain the matching degree with each model in the database; and marking the model in the database with the highest matching degree as the target model.

[0024] This invention, after determining the feature sequence corresponding to the vehicle to be washed, sequentially extracts the contents of the feature sequence and matches them in a database. The vehicle model with the highest matching degree is the target vehicle model. The database stores a large number of vehicle models, and each vehicle model has a feature sequence associated with it.

[0025] The matching of feature sequences in this invention essentially involves comparing which vehicle model's relevant data in the database is closest to the vehicle to be washed; it only involves data comparison. The entire comparison process is relatively simple, ensuring matching efficiency while avoiding errors in the data processing.

[0026] Preferably, determining whether the relative position between the vehicle to be washed and the intelligent car wash equipment is abnormal based on the target vehicle model includes: extracting the vehicle size corresponding to the target vehicle model; determining several corner points of the vehicle to be washed based on the vehicle size; and determining that the relative position is normal when the relative distance between the several boundary points and the intelligent car wash equipment is within the distance threshold range.

[0027] After matching the target vehicle model to the vehicle under test, this invention uses the dimensions of the target model as the dimensions of the vehicle to be washed. Several points are selected on the boundary of the target model as boundary points. When all these boundary points are within the washing range of the intelligent car wash equipment, the intelligent car wash equipment is controlled to operate according to the target model, completing the washing of the vehicle. This invention uses the dimensions of the target model instead of the actual measured dimensions of the vehicle to be washed to avoid inaccurate measured dimensions due to errors in the measurement process.

[0028] It should be noted that intelligent car wash equipment will automatically adjust its operating path according to different target car models to ensure that the distance between the brush components and the vehicle to be washed is appropriate, so as to ensure the cleaning effect and avoid damage to the vehicle to be washed.

[0029] Compared with the prior art, the beneficial effects of the present invention are:

[0030] 1. This invention extracts the size and line features of the vehicle to be washed based on its appearance data, and integrates them into a feature sequence; the target car model is obtained by matching the feature sequence in a database, and the intelligent car wash equipment is controlled to wash the car based on the target car model; this invention extracts the appearance features of the vehicle to be washed and then matches the target car model, thereby improving the matching efficiency of the target car model and avoiding the leakage of car owner privacy.

[0031] 2. This invention determines whether the relative position between the vehicle to be washed and the intelligent car wash equipment is abnormal based on the target vehicle model; if yes, the vehicle to be washed or the intelligent car wash equipment is moved; if no, the intelligent car wash equipment is controlled to perform car washing based on the target vehicle model; this invention determines the relative position between the vehicle to be washed and the intelligent car wash equipment based on the vehicle size of the target vehicle model, avoiding measurement errors that could damage the vehicle to be washed during the car washing process. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a schematic diagram of the system principle of the present invention;

[0034] Figure 2 This is a schematic diagram of the method steps of the present invention;

[0035] Figure 3 This is an image of the exterior of the top side of the vehicle to be washed according to the present invention. Implementation

[0036] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Please see Figures 1-2 The first aspect of this invention provides an intelligent vehicle model recognition system, including a central control module, and a data interaction module and a car wash control module connected thereto. The data interaction module is connected to a data acquisition device and a smart terminal, respectively. The car wash control module is used to control an intelligent car wash device. The central control module acquires the appearance data of the vehicle to be washed through the data acquisition device; extracts the size features and line features of the vehicle to be washed based on the appearance data, and integrates the two into a feature sequence; matches the feature sequence in a database, and determines the target vehicle model after verification by the smart terminal; based on the target vehicle model, it determines whether the relative position of the vehicle to be washed and the intelligent car wash device is abnormal; if yes, it moves the vehicle to be washed or the intelligent car wash device; if no, it controls the intelligent car wash device to perform a car wash based on the target vehicle model.

[0038] The first step in this embodiment is that the central control module acquires the appearance data of the vehicle to be washed through the data acquisition device; based on the appearance data, the size features and line features of the vehicle to be washed are extracted and integrated into a feature sequence.

[0039] Before the intelligent car wash equipment is installed, data acquisition devices are set up along the route that vehicles must take to reach the intelligent car wash equipment. As the vehicles pass through this route, the data acquisition devices collect the appearance data of the vehicles under test, and send it to the central control module through the data interaction module.

[0040] The data acquisition device in this embodiment has acquisition angles including top-view angle, front-view angle, rear-view angle, left-view angle, and right-view angle. When the vehicle to be washed passes through the necessary road, the top-view angle acquires a top-side appearance image that meets the requirements, and the right-side angle and rear-view angle also need to acquire appearance images that meet the requirements.

[0041] After acquiring the exterior data of the vehicle to be washed, necessary data corrections are performed to ensure the exterior data matches reality. Then, the exterior dimensions corresponding to each acquisition angle can be calculated. These dimensions primarily refer to length, width, and slope. Each vehicle model has different lengths, widths, and slopes, which helps in matching the target vehicle model. In some other preferred embodiments, the exterior dimensions can also be the image area corresponding to the exterior images at each acquisition angle.

[0042] Please see Figure 3 The center point of the roof of the vehicle to be washed is designated as reference point one. Figure 3 Point A is reference point A; the intersection of the right, front, and top sides of the vehicle to be washed is reference point B; the intersection of the right, rear, and top sides of the vehicle to be washed is also reference point C. Connecting reference points A and B, we can obtain the exterior lines AB and AC.

[0043] Next, the lengths of the appearance lines AB and AC need to be calculated as a line feature. Figure 3 It can be seen that the front contour line of the roof intersects with the exterior line AB, and the rear contour line of the roof intersects with the exterior line AC. The exterior line AB is divided into two segments, namely AE and EB; the exterior line AC is divided into two segments, namely AF and FC; therefore, AE / EB and AF / FC can be considered as a single line feature.

[0044] The obtained exterior dimensions, exterior line lengths, and corresponding line proportions can be integrated into a feature sequence for the vehicle under test. Furthermore, the content of this feature sequence varies depending on the size and design language of the vehicle, which helps in accurately identifying the target vehicle model corresponding to the vehicle under test.

[0045] The second step in this embodiment is to match the feature sequence in the database and determine the target car model after verification by the smart terminal; based on the target car model, it is determined whether the relative position between the vehicle to be washed and the smart car wash equipment is abnormal; if yes, the vehicle to be washed or the smart car wash equipment is moved; if no, the smart car wash equipment is controlled to wash the car based on the target car model.

[0046] The feature sequence of the vehicle to be tested is extracted. First, it is matched in the database based on the appearance size to obtain a set of models that meet the requirements. Then, the line size and line features of these models are obtained and compared with the line size and line features of the vehicle to be tested. The model with the smallest error is selected as the target model.

[0047] The selected car model is displayed via the smart terminal. If there is an error, the car owner can manually change the selection. If the car owner does not intervene, the smart car wash equipment can be controlled according to the selected car model. Intervention here includes touch screen operation or voice operation.

[0048] After determining the target vehicle model, extract the vehicle dimensions. First, select two points on the front side of the vehicle to be washed (such as the two headlights) as boundary points. Determine if the distance between the boundary points and the intelligent car wash equipment meets the set range requirements. If yes, determine two points on the rear side of the vehicle to be washed as boundary points based on the vehicle dimensions. Determine if the distance between the rear boundary points and the intelligent car wash equipment is within the set distance range. If the distance between all boundary points and the intelligent car wash equipment is within the set distance range, the vehicle is considered to be in a normal position and can be automatically washed.

[0049] Before the automated car wash begins, the movement path of the intelligent car wash equipment is automatically set according to the target vehicle model. The intelligent car wash equipment is then controlled according to the set movement path to complete the automated car wash. During the wash process, the washing status can be monitored via a smart terminal, and the washing is complete when finished, as displayed on the smart terminal.

[0050] The working principle of this invention is as follows: The exterior data of the vehicle to be washed is acquired; based on the exterior data, the size and line features of the vehicle to be washed are extracted and integrated into a feature sequence. This feature sequence is matched against a database, and after verification by a smart terminal, the target vehicle model is determined; based on the target vehicle model, it is determined whether the relative position between the vehicle to be washed and the smart car wash equipment is abnormal. If yes, the vehicle to be washed or the smart car wash equipment is moved; otherwise, the smart car wash equipment is controlled to perform the car wash based on the target vehicle model.

[0051] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. The vehicle model intelligent recognition system includes: The system includes a central control module, a data interaction module, and a car wash control module connected thereto; the data interaction module is connected to a data acquisition device and a smart terminal respectively; the car wash control module is used to control the intelligent car wash equipment; its features are: The central control module acquires the exterior data of the vehicle to be washed through data acquisition equipment; based on the exterior data, it extracts the size and line features of the vehicle and integrates them into a feature sequence; the data acquisition equipment includes image acquisition equipment or 3D scanning equipment; and The target vehicle model is determined by matching feature sequences in the database and verifying them through a smart terminal. The smart terminal is used to verify the target vehicle model. The selected target vehicle model is displayed on the smart terminal, and the owner can manually change the selection if there is an error. Based on the target vehicle model, it is determined whether the relative position between the vehicle to be washed and the smart car wash equipment is abnormal. If yes, the vehicle to be washed or the smart car wash equipment is moved; otherwise, the smart car wash equipment is controlled to wash the car based on the target vehicle model. The central control module acquires the exterior data of the vehicle to be washed through data acquisition equipment, including: Ensure the vehicle to be washed is within the data acquisition range of the data collection equipment; The exterior data of the vehicle to be washed is collected by a data acquisition device; the exterior data includes at least the top side exterior image, the front or rear side exterior image, and the left or right side exterior image. The extraction of size and line features of the vehicle to be washed based on appearance data includes: Extract the exterior data of several sides of the vehicle to be washed, and obtain the exterior dimensions of several sides after data processing; wherein, data processing includes data verification and data correction; the exterior dimensions include length, width and slope; Based on several side appearance data, several appearance lines are determined from a top-down perspective; the line features corresponding to these appearance lines are extracted; among them, the line features include line length and line segment ratio. The extraction of line features corresponding to several appearance lines includes: Calculate the line dimensions of several exterior lines from a top-down view; Identify any outline of a vehicle to be washed where several exterior lines intersect; divide the corresponding exterior line into two segments using the outline, and mark the proportions of the two segments as the line proportions; The line dimensions and proportions corresponding to each appearance line are integrated into line features; The process of determining several exterior lines from a top-down perspective based on several side exterior data includes: Extract the top-side exterior image from the top-down view and determine the center point of the roof in the top-side exterior image as reference point one; Several reference points 2 are determined on the other side edge lines in the top side appearance image, and several appearance lines are obtained by connecting reference point 1 and several reference points 2.

2. The intelligent vehicle identification system according to claim 1, characterized in that, The matching in the database using feature sequences includes: Retrieve the characteristic sequence of the vehicle to be washed; The system performs matching with the database using size and line features from the feature sequence to obtain the matching degree with each model in the database; the model with the highest matching degree is marked as the target model.

3. The intelligent vehicle identification system according to claim 1, characterized in that, The step of determining whether the relative position between the vehicle to be washed and the intelligent car wash equipment is abnormal, based on the target vehicle model, includes: Extract the vehicle dimensions corresponding to the target vehicle model; Based on the vehicle's dimensions, several corner points of the vehicle to be washed are determined; when the relative distance between these corner points and the intelligent car wash equipment is within the distance threshold range, the relative position is considered normal.

4. The intelligent vehicle identification system according to claim 1, characterized in that, The central control module communicates and / or is electrically connected to the data interaction module and the car wash control module, respectively; the car wash control module controls the intelligent car wash equipment according to the target car model and monitors the car wash status. The data interaction module is communicatively and / or electrically connected to the data acquisition device and the smart terminal, respectively; wherein, the smart terminal is used to verify the target vehicle model and display the car wash status.

Images