Ultrasonic-based vehicle detection method, apparatus, and system

Abstract

The present application relates to the field of ultrasonic data detection, and particularly relates to a vehicle detection method, device and system based on ultrasonic waves, comprising: analyzing change stability information of data between different environment ultrasonic data segments and environment ultrasonic data sequences to which the different environment ultrasonic data segments belong, and obtaining a vehicle reference environment data sequence; comparing matching of ultrasonic information between the vehicle reference environment data sequence and the environment ultrasonic data sequence, and combining the environment ultrasonic data segments to obtain an environment reference ultrasonic data segment; obtaining a parking unobstructed ultrasonic data segment according to matching differences of ultrasonic information between different environment reference ultrasonic data segments and a parking ultrasonic data sequence; analyzing attenuation progress of vehicle ultrasonic information in different parking unobstructed ultrasonic data segments to obtain an actual vehicle parking ultrasonic data segment; and then performing ultrasonic vehicle detection. The present application makes the ultrasonic signals collected by the ultrasonic device more real, and improves the efficiency of detecting the real parking of vehicles.

Description

Technical Field

[0001] The present invention relates to the field of ultrasonic data detection, and in particular to an ultrasonic-based vehicle detection method, device and system. Background Art

[0002] As the number of cars continues to grow, urban parking is becoming a widespread problem. Due to limited underground space, many vehicles are parked outdoors. Cameras in these areas are significantly affected by the weather, resulting in poor detection of vehicle movement. Therefore, existing technologies typically employ an underground vehicle detector. This device, placed in a roadside parking space, uses both an ultrasonic module and a geomagnetic module to determine whether a vehicle is parked in the space.

[0003] When a vehicle is inspected using ultrasonic equipment, the road vibrations caused by the movement of surrounding vehicles and debris such as mud adhering to the vehicle's bottom chassis will interfere with the propagation of the ultrasonic waves emitted by the ultrasonic equipment, resulting in a certain degree of distortion in the peak performance of the ultrasonic signal recorded by the ultrasonic equipment, thereby reducing the accuracy of vehicle status detection. Summary of the Invention

[0004] The present invention provides an ultrasonic-based vehicle detection method, device, and system to solve an existing problem: when an ultrasonic device is used to detect a vehicle, road vibrations generated by the movement of surrounding vehicles and debris such as mud adhering to the vehicle's bottom base will cause certain interference with the propagation of the ultrasonic waves emitted by the ultrasonic device, resulting in certain distortion in the peak performance of the ultrasonic signal received and recorded by the ultrasonic device.

[0005] The ultrasonic vehicle detection method of the present invention adopts the following technical solutions:

[0006] The following steps are involved:

[0007] Acquire a parking ultrasonic data sequence and a plurality of environmental ultrasonic data sequences;

[0008] Each environmental ultrasonic data sequence is divided into a number of environmental ultrasonic data segments within a preset number range; information on data changes and stability between different environmental ultrasonic data segments and the environmental ultrasonic data sequences to which they belong is analyzed, and a vehicle reference environmental data sequence is selected from all environmental ultrasonic data sequences; ultrasonic information matching between the vehicle reference environmental data sequence and each environmental ultrasonic data sequence is compared, and cumulative smoothing processing is performed on the environmental ultrasonic data sequence in combination with the environmental ultrasonic data segments to obtain a number of environmental reference ultrasonic data segments;

[0009] Based on the differences in ultrasonic information matching between different environmental baseline ultrasonic data segments and the parking ultrasonic data sequence, a number of parking unobstructed ultrasonic data segments are divided and processed from the parking ultrasonic data sequence; the attenuation progress of the vehicle ultrasonic information in the different parking unobstructed ultrasonic data segments is analyzed, and the attenuation of the ultrasonic information is simulated for each parking unobstructed ultrasonic data segment. The simulated parking unobstructed ultrasonic data segment is used as the actual vehicle parking ultrasonic data segment;

[0010] Ultrasonic vehicle detection is performed based on actual vehicle parking ultrasonic data segments.

[0011] Preferably, the method for acquiring the vehicle reference environment data sequence is:

[0012] For any ambient ultrasonic data sequence, the fluctuation stability of the data in each ambient ultrasonic data segment in the ambient ultrasonic data sequence is analyzed as the data segment stability of each ambient ultrasonic data segment; the ultrasonic data stabilities of all ambient ultrasonic data segments in the ambient ultrasonic data sequence are comprehensively accumulated as the sequence stability of the ambient ultrasonic data sequence; the proportional relationship between the stability of each data segment and the sequence stability is analyzed to obtain the ultrasonic benchmark preference of each ambient ultrasonic data segment; the ultrasonic benchmark preference of the comprehensively accumulated ambient ultrasonic data segments is used as the ambient ultrasonic benchmark preference of the ambient ultrasonic data sequence; and according to the ambient ultrasonic benchmark preference, a vehicle benchmark ambient data sequence is screened out from all ambient ultrasonic data sequences.

[0013] Preferably, the method for obtaining the data segment stability is:

[0014] For any ambient ultrasound data segment in the ambient ultrasound data sequence, an inverse proportional value of a standard deviation of data in the ambient ultrasound data segment is used as the data segment stability of the ambient ultrasound data segment.

[0015] Preferably, the method for obtaining the ultrasonic reference preference is:

[0016] The ratio of the stability of each data segment to the sequence stability is used as the ultrasonic benchmark preference of each environmental ultrasonic data segment.

[0017] Preferably, the method for acquiring the environmental reference ultrasonic data segment is:

[0018] For any environmental ultrasonic data sequence, information matching is performed between the vehicle reference environmental data sequence and the environmental ultrasonic data sequence to obtain an environmental ultrasonic information matching segment of the environmental ultrasonic data sequence; based on the ultrasonic matching information in the environmental ultrasonic information matching segment, the data in the environmental ultrasonic data sequence is accumulated as the ultrasonic information smoothing weight of the environmental ultrasonic data sequence; based on the ultrasonic information smoothing weight, each environmental ultrasonic data segment in the environmental ultrasonic data sequence is filtered and smoothed, and the smoothed environmental ultrasonic data segment is used as the environmental reference ultrasonic data segment.

[0019] Preferably, the method for obtaining the matching segment of the environmental ultrasonic information is:

[0020] Perform DTW matching on the vehicle reference environment data sequence and the environment ultrasonic data sequence to obtain a number of DTW matching pairs; and use the data segments of the DTW matching pairs that match in the environment ultrasonic data sequence as the environment ultrasonic information matching segments.

[0021] Preferably, the method for acquiring the parking unblocking ultrasonic data segment is:

[0022] The environmental reference ultrasonic data segment is integrated to obtain the environmental interference removal reference sequence; information matching is performed on the parking ultrasonic data sequence to obtain the parking ultrasonic information matching segment of the parking ultrasonic data sequence; the data difference between the parking ultrasonic information matching segment and the environmental interference removal reference sequence is used as the ultrasonic information matching difference; the data segment after data correction of the parking ultrasonic information matching segment is performed using the ultrasonic information matching difference as the parking de-occlusion ultrasonic data segment.

[0023] Preferably, the method for acquiring the actual vehicle parking ultrasonic data segment is:

[0024] For any parking deobstruction ultrasonic data segment, the parking deobstruction ultrasonic data segment contains multiple ultrasonic data peaks; within the parking deobstruction ultrasonic data segment, the ultrasonic data peaks are taken as starting points in turn, and the numerical relationship between the subsequent ultrasonic data peaks and the starting points is used as the judgment condition for traversal, all ultrasonic data peaks are traversed to obtain several ultrasonic information attenuation judgment sequences; the stable fluctuation of the data difference between each ultrasonic information attenuation judgment sequence and the parking deobstruction ultrasonic data segment is used as the attenuation confidence weight of each ultrasonic information attenuation judgment sequence; the parking deobstruction ultrasonic data segment is corrected according to the attenuation confidence weight, and the corrected parking deobstruction ultrasonic data segment is used as the actual vehicle parking ultrasonic data segment.

[0025] The ultrasonic-based vehicle detection system includes a memory, a processor, and a computer program stored in the memory and running on the processor. The processor executes the computer program to implement the steps of the above method.

[0026] Ultrasonic-based vehicle detection equipment, the equipment comprising:

[0027] A data sequence acquisition module, an actual vehicle parking ultrasonic data segment acquisition module and an ultrasonic vehicle detection module; wherein the data sequence acquisition module is used to acquire a parking ultrasonic data sequence and several environmental ultrasonic data sequences, the actual vehicle parking ultrasonic data segment acquisition module obtains the actual vehicle parking ultrasonic data segment by calling a computer program to implement the steps of the ultrasonic-based vehicle detection method, and the ultrasonic vehicle detection module performs ultrasonic vehicle detection based on the actual vehicle parking ultrasonic data segment.

[0028] The beneficial effects of the technical solution of the present invention are: analyzing the change stability information of the data between different environmental ultrasonic data segments and the environmental ultrasonic data sequences to which they belong, and screening out the vehicle benchmark environmental data sequences; comparing the matching of ultrasonic information between the vehicle benchmark environmental data sequences and each environmental ultrasonic data sequence, and combining the environmental ultrasonic data segments, performing cumulative smoothing processing on the environmental ultrasonic data sequences to obtain environmental benchmark ultrasonic data segments; wherein the environmental benchmark ultrasonic data segments are used to describe the data segments that reduce the influence of vehicle road vibration on ultrasonic signals, and effectively reduce the influence of vehicle road vibration on ultrasonic signals without destroying the effective information of the parking environment; obtaining the parking unblocking ultrasonic data segments based on the matching differences of ultrasonic information between different environmental benchmark ultrasonic data segments and the parking ultrasonic data sequences; analyzing the vehicle ultrasonic information in different parking unblocking ultrasonic data segments. Attenuation progress, the attenuation simulation of ultrasonic information is performed on each parking unobstructed ultrasonic data segment, and the simulated parking unobstructed ultrasonic data segment is used as the actual vehicle parking ultrasonic data segment; wherein the actual vehicle parking ultrasonic data segment is used to describe the data segment that reduces the influence of the vehicle base obstruction on the ultrasonic signal, so that the vehicle status of the vehicle in the parking space is easier to identify and observe; the present invention compares the matching of the environment to the ultrasonic information at different times, performs smoothing processing to obtain the environmental reference ultrasonic data segment, and removes the interference of the vehicle road vibration on the ultrasonic signal; then analyzes the matching between the environmental reference ultrasonic data segment and the parking ultrasonic data sequence, performs attenuation simulation, obtains the actual vehicle parking ultrasonic data segment, and removes the interference of the vehicle base obstruction on the ultrasonic signal; makes the ultrasonic signal collected by the ultrasonic equipment more realistic, and improves the efficiency of detecting the actual parking of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1The figure is a flowchart of the steps of the ultrasonic-based vehicle detection method, device and system of the present invention. DETAILED DESCRIPTION

[0031] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the ultrasonic vehicle detection method, device, and system proposed in accordance with the present invention, including its specific implementation, structure, features, and effectiveness. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0032] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0033] The specific solutions of the ultrasonic-based vehicle detection method, device, and system provided by the present invention are described in detail below with reference to the accompanying drawings.

[0034] See also Figure 1 , which shows a flowchart of a vehicle detection method based on ultrasound provided by an embodiment of the present invention, the method comprising the following steps:

[0035] Step S001: Acquire a parking ultrasonic data sequence and several environmental ultrasonic data sequences.

[0036] It should be noted that when an ultrasonic device is used to detect a vehicle, the road vibrations caused by the movement of surrounding vehicles and debris such as mud adhering to the vehicle's bottom base will cause certain interference to the propagation of the ultrasonic waves emitted by the ultrasonic device, resulting in a certain distortion in the peak performance of the ultrasonic signal recorded by the ultrasonic device, thereby reducing the accuracy of vehicle status detection.

[0037] In a specific implementation of an embodiment of the present invention, a method for obtaining a parking ultrasonic data sequence and an environmental ultrasonic data sequence is as follows: taking any parking space as an example, ultrasonic data when a vehicle is parked in the parking space is obtained from the ultrasonic data platform, and each ultrasonic data is used as parking ultrasonic data, and the sequence composed of the parking ultrasonic data is used as a parking ultrasonic data sequence; then, ultrasonic data with the same number of data as that contained in the parking ultrasonic data sequence when the parking space is not open is obtained from the ultrasonic data platform, and each ultrasonic data is used as environmental ultrasonic data, and the sequence composed of the environmental ultrasonic data is used as an environmental ultrasonic data sequence.

[0038] It should be noted that the ultrasonic data platform in this embodiment acquires data at a frequency of 1 time / second by default; this embodiment is described by taking the acquisition of 5 environmental ultrasonic data sequences as an example, and each environmental ultrasonic data sequence corresponds to environmental ultrasonic data collected in different time periods.

[0039] It should be noted that the ultrasonic data when a vehicle is parked in the parking space in this embodiment includes ultrasonic data from the beginning of parking in the parking space to the first time the vehicle leaves the parking space. The parking status of the vehicle is first determined by the ultrasonic reflection conditions preset by the ultrasonic equipment itself. This is a well-known technology and will not be described in detail in this embodiment.

[0040] So far, a parking ultrasonic data sequence and several environmental ultrasonic data sequences are obtained through the above method.

[0041] Step S002: Divide each environmental ultrasonic data sequence into a number of environmental ultrasonic data segments within a preset number range; analyze the data change stability information between different environmental ultrasonic data segments and the environmental ultrasonic data sequences to which they belong, and screen out a vehicle baseline environmental data sequence from all environmental ultrasonic data sequences; compare the ultrasonic information matching between the vehicle baseline environmental data sequence and each environmental ultrasonic data sequence, and perform cumulative smoothing processing on the environmental ultrasonic data sequence in combination with the environmental ultrasonic data segments to obtain a number of environmental benchmark ultrasonic data segments.

[0042] It should be noted that in a parking lot equipped with ultrasonic equipment, vehicles will move in the parking lot during its opening period. Under the subjective influence of their drivers, different routes will be formed, and they will randomly pass through different parking spaces. When vehicles move in the parking lot, they will transmit different degrees of vibration energy to the road surface. These vibration energies have different energy intensities due to factors such as the distance between the vehicle and the parking space, the speed of the vehicle, and the weight of the vehicle, and randomly cause different sizes of interference to the ultrasonic equipment in the parking space. Therefore, each environmental ultrasonic data sequence can be divided into several environmental ultrasonic data segments within a preset number range; the change stability information of the data between different environmental ultrasonic data segments and the environmental ultrasonic data sequences to which they belong is analyzed, and the vehicle baseline environmental data sequence is screened out from all environmental ultrasonic data sequences; the matching of ultrasonic information between the vehicle baseline environmental data sequence and each environmental ultrasonic data sequence is compared, and the environmental ultrasonic data sequence is cumulatively smoothed in combination with the environmental ultrasonic data segments to obtain several environmental benchmark ultrasonic data segments.

[0043] Preset a data quantity , take any environmental ultrasonic data sequence as an example, The data segment consisting of the environmental ultrasonic data is used as the environmental ultrasonic data segment; all the environmental ultrasonic data segments in the environmental ultrasonic data sequence are obtained. This example is described as an example, and this embodiment is not specifically limited. It may depend on the specific implementation situation.

[0044] It is particularly noted that if the number of remaining environmental ultrasonic data in the environmental ultrasonic data sequence does not meet the preset , then the data segment composed of the remaining environmental ultrasonic data is used as the environmental ultrasonic data segment.

[0045] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the vehicle reference environment data sequence is as follows: for any environmental ultrasonic data sequence, analyze the fluctuation stability of the data in each environmental ultrasonic data segment in the environmental ultrasonic data sequence as the data segment stability of each environmental ultrasonic data segment; comprehensively accumulate the ultrasonic data stability of all environmental ultrasonic data segments in the environmental ultrasonic data sequence as the sequence stability of the environmental ultrasonic data sequence; analyze the proportional relationship between the stability of each data segment and the sequence stability to obtain the ultrasonic reference preference of each environmental ultrasonic data segment; comprehensively accumulate the ultrasonic reference preference of the environmental ultrasonic data segments as the environmental ultrasonic reference preference of the environmental ultrasonic data sequence; and screen out the vehicle reference environment data sequence from all environmental ultrasonic data sequences based on the environmental ultrasonic reference preference. The specific process is as follows:

[0046] Preferably, in some implementations of the present invention, the method for obtaining data segment stability is: for any ambient ultrasound data segment in the ambient ultrasound data sequence, the inverse proportional value of the standard deviation of the data in the ambient ultrasound data segment is used as the data segment stability of the ambient ultrasound data segment. The specific process is as follows:

[0047] Taking any ambient ultrasound data segment in the ambient ultrasound data sequence as an example, an inversely proportional normalized value of the standard deviation of the ambient ultrasound data in the ambient ultrasound data segment is used as the data segment stability of the ambient ultrasound data segment. The data segment stability of all ambient ultrasound data segments in the ambient ultrasound data sequence is obtained.

[0048] It is particularly noted that the embodiment adopts Model to present inverse proportional relationship and normalization processing, As the input of the model, the implementer can choose the inverse proportional function and normalization function according to the actual situation.

[0049] Furthermore, the accumulated value of the data segment stability of all the ambient ultrasonic data segments in the ambient ultrasonic data sequence is used as the sequence stability of the ambient ultrasonic data sequence.

[0050] Preferably, in some implementations of the present invention, the method for obtaining the ultrasonic reference preference is to use the ratio of the stability of each data segment to the stability of the sequence as the ultrasonic reference preference of each environmental ultrasonic data segment. The specific process is as follows:

[0051] The ratio between the data segment stability of each ambient ultrasonic data segment and the sequence stability of the ambient ultrasonic data sequence is used as the ultrasonic reference preference of each ambient ultrasonic data segment.

[0052] Furthermore, the value obtained by multiplying the ultrasonic baseline preferences of all ambient ultrasonic data segments in the ambient ultrasonic data sequence is used as the ambient ultrasonic baseline preference of the ambient ultrasonic data sequence; the ambient ultrasonic baseline preference of each ambient ultrasonic data sequence is obtained; and the ambient ultrasonic data sequence with the largest ambient ultrasonic baseline preference is used as the vehicle baseline ambient data sequence.

[0053] It should be noted that the greater the environmental ultrasonic baseline priority, the more parking lot environment content is contained in the ultrasonic information in the corresponding environmental ultrasonic data sequence, reflecting that the ultrasonic wave sent by the ultrasonic device within the corresponding time range has a stronger correlation with the surrounding environment.

[0054] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the environmental reference ultrasonic data segment is as follows: for any environmental ultrasonic data sequence, the vehicle reference environmental data sequence is matched with the environmental ultrasonic data sequence to obtain the environmental ultrasonic information matching segment of the environmental ultrasonic data sequence; based on the ultrasonic matching information in the environmental ultrasonic information matching segment, the data in the environmental ultrasonic data sequence is accumulated as the ultrasonic information smoothing weight of the environmental ultrasonic data sequence; based on the ultrasonic information smoothing weight, each environmental ultrasonic data segment in the environmental ultrasonic data sequence is filtered and smoothed, and the smoothed environmental ultrasonic data segment is used as the environmental reference ultrasonic data segment. The specific process is as follows:

[0055] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the ambient ultrasound information matching segment is as follows: performing DTW matching on the vehicle reference environment data sequence and the ambient ultrasound data sequence to obtain a plurality of DTW matching pairs; and using the data segments of the DTW matching pairs that match within the ambient ultrasound data sequence as the ambient ultrasound information matching segment. The specific process is as follows:

[0056] The environmental ultrasonic data sequences other than the vehicle reference environmental ultrasonic data sequence are used as reference environmental ultrasonic data sequences. Taking any reference environmental ultrasonic data sequence as an example, DTW matching is performed between the vehicle reference environmental ultrasonic data sequence and the reference environmental ultrasonic data sequence to obtain a number of DTW matching pairs. A data segment consisting of the environmental ultrasonic data corresponding to each DTW matching pair in the reference environmental ultrasonic data sequence is used as an environmental ultrasonic information matching segment. Each DTW matching pair contains corresponding ultrasonic data in both the vehicle reference environmental ultrasonic data sequence and the reference environmental ultrasonic data sequence. Each environmental ultrasonic data in the environmental ultrasonic information matching segment corresponds to a shortest DTW distance.

[0057] It is particularly noted that the process of obtaining the DTW matching pair and the shortest DTW distance is a well-known content of the DTW (Dynamic Time Warping) dynamic time warping algorithm, and will not be described in detail in this embodiment.

[0058] Furthermore, as an example, the ultrasound information smoothing weight can be calculated by the following formula:

[0059]

[0060] Where, represents the ultrasonic information smoothing weight of the reference environment ultrasonic data sequence; Indicates the total number of ambient ultrasonic data contained in all ambient ultrasonic information matching segments in the reference ambient ultrasonic data sequence; Indicates the Environmental ultrasound data; Indicates the The shortest dtw distance of the environmental ultrasonic data; represents a normalization function, which normalizes the ultrasonic information smoothing weights of all reference environment ultrasonic data sequences.

[0061] It should be noted that the larger the ultrasonic smoothing weight, the greater the random influence of vehicle and road vibration on the ultrasonic information contained in the reference environment ultrasonic data sequence, and the greater the intensity of smoothing required for the data in the reference environment ultrasonic data sequence.

[0062] Furthermore, the ultrasonic information smoothing weight is considered a filtering weight, and filtering processing is performed on each ambient ultrasonic data segment in the reference ambient ultrasonic data sequence based on the filtering weight. The processed ambient ultrasonic data segment is used as the ambient reference ultrasonic data segment. All ambient reference ultrasonic data segments are obtained. The process of filtering the data sequence based on the filtering weight is a well-known aspect of smoothing filtering algorithms and will not be further described in this embodiment.

[0063] So far, several environmental reference ultrasonic data segments have been obtained through the above method.

[0064] Step S003: Based on the matching differences of ultrasonic information between different environmental reference ultrasonic data segments and the parking ultrasonic data sequence, a number of parking unobstructed ultrasonic data segments are divided and processed from the parking ultrasonic data sequence; the attenuation progress of the vehicle ultrasonic information in different parking unobstructed ultrasonic data segments is analyzed, and the attenuation of the ultrasonic information is simulated for each parking unobstructed ultrasonic data segment. The simulated parking unobstructed ultrasonic data segment is used as the actual vehicle parking ultrasonic data segment.

[0065] It should be noted that the interference of the external environment on the ultrasonic signal of the ultrasonic equipment is mainly divided into two types, one is the random interference caused by road vibration, and the other is the continuous interference caused by the obstruction of the vehicle base; after eliminating the random interference caused by road vibration, it is also necessary to eliminate the continuous interference caused by the obstruction of the vehicle base; under normal circumstances, the obstruction of the vehicle base is relatively small, generally small objects such as branches, mud, fallen leaves, etc. When the vehicle remains almost stopped, it will continue to generate interference signals with weaker energy to interfere with the stable information of the ultrasonic signal change; therefore, according to the matching difference of the ultrasonic information between the different environmental baseline ultrasonic data segments and the parking ultrasonic data sequence, several parking unobstructed ultrasonic data segments can be divided and processed from the parking ultrasonic data sequence; the attenuation progress of the vehicle ultrasonic information in different parking unobstructed ultrasonic data segments is analyzed, and the attenuation simulation of the ultrasonic information is performed on each parking unobstructed ultrasonic data segment, and the simulated parking unobstructed ultrasonic data segment is used as the actual vehicle parking ultrasonic data segment.

[0066] Preferably, in some implementations of the present invention, the method for obtaining the parking ultrasonic data segment for de-obstruction is as follows: synthesizing the environmental reference ultrasonic data segment to obtain the environmental interference removal reference sequence; performing information matching on the parking ultrasonic data sequence to obtain the parking ultrasonic information matching segment of the parking ultrasonic data sequence; using the data difference between the parking ultrasonic information matching segment and the environmental interference removal reference sequence as the ultrasonic information matching difference; and using the ultrasonic information matching difference to correct the parking ultrasonic information matching segment for the data segment, obtaining the data segment as the parking ultrasonic data segment for de-obstruction. The specific process is as follows:

[0067] The average of the environmental ultrasonic data of all reference environmental ultrasonic data sequences and the vehicle benchmark environmental data sequence at the same sequence number is used as the comprehensive benchmark environmental ultrasonic data at the same sequence number; all comprehensive benchmark environmental ultrasonic data with the same sequence number are obtained; and the sequence constructed by all comprehensive benchmark environmental ultrasonic data with the same sequence number is used as the environmental interference removal benchmark sequence.

[0068] Furthermore, referring to the method for obtaining the environmental ultrasonic data segment, the environmental ultrasonic data sequence is replaced with the environmental interference removal reference sequence to obtain the environmental ultrasonic data segment of the environmental interference removal reference sequence and use it as the environmental reference data segment of the environmental interference removal reference sequence.

[0069] Furthermore, referring to the method for obtaining matching segments of reference environmental ultrasonic information, the reference environmental ultrasonic data sequence is replaced with the parking ultrasonic data sequence, and the vehicle reference environmental data sequence is replaced with the environmental reference data segments. The environmental ultrasonic information matching segments of the parking ultrasonic data sequence are obtained and used as the parking ultrasonic information matching segments of the parking ultrasonic data sequence. Each parking ultrasonic information matching segment corresponds to an environmental reference data segment.

[0070] Furthermore, the average of all ultrasonic data differences between the parking ultrasonic information matching segment and the corresponding environmental reference data segment is used as the ultrasonic information matching difference of the parking ultrasonic information matching segment; within the parking ultrasonic information matching segment, the absolute value after subtracting the ultrasonic information matching difference from each parking ultrasonic data is used as the parking corrected ultrasonic data; the parking ultrasonic information matching segment after processing all parking ultrasonic data is used as the parking unobstructed ultrasonic data segment.

[0071] Preferably, in some implementations of the present invention, the method for obtaining the actual vehicle parking ultrasonic data segment is as follows: for any parking de-occlusion ultrasonic data segment, the parking de-occlusion ultrasonic data segment contains multiple ultrasonic data peaks; within the parking de-occlusion ultrasonic data segment, the ultrasonic data peaks are sequentially used as starting points, and the numerical value relationship between the subsequent ultrasonic data peaks and the starting point is used as the judgment condition for traversal, all ultrasonic data peaks are traversed to obtain several ultrasonic information attenuation judgment sequences; the stable fluctuation of the data difference between each ultrasonic information attenuation judgment sequence and the parking de-occlusion ultrasonic data segment is used as the attenuation confidence weight of each ultrasonic information attenuation judgment sequence; the parking de-occlusion ultrasonic data segment is corrected according to the attenuation confidence weight, and the corrected parking de-occlusion ultrasonic data segment is used as the actual vehicle parking ultrasonic data segment. The specific process is as follows:

[0072] Taking any parking and de-obstruction ultrasonic data segment as an example, all ultrasonic data peaks in the parking and de-obstruction ultrasonic data segment are obtained; in the parking and de-obstruction ultrasonic data segment, the first ultrasonic data peak is used as the starting point of the ultrasonic information attenuation sequence, with a step size of 1, and the ultrasonic data peaks are traversed, and the ultrasonic data peaks that are smaller than the starting point are incorporated into the ultrasonic information attenuation sequence until all ultrasonic data peaks are traversed; then, in the parking and de-obstruction ultrasonic data segment, the second ultrasonic data peak is used as the starting point of the ultrasonic information attenuation sequence, with a step size of 1, and the ultrasonic data peaks are traversed, and the ultrasonic data peaks that are smaller than the starting point are incorporated into the ultrasonic information attenuation sequence until all ultrasonic data peaks are traversed; in the parking and de-obstruction ultrasonic data segment, the third ultrasonic data peak is used as the starting point of the ultrasonic information attenuation sequence, with a step size of 1, and the ultrasonic data peaks are traversed, and the ultrasonic data peaks that are smaller than the starting point are incorporated into the ultrasonic information attenuation sequence until all ultrasonic data peaks are traversed; and so on, all ultrasonic information attenuation sequences are obtained.

[0073] Furthermore, any ultrasonic information attenuation sequence is used as the target ultrasonic information attenuation sequence. If any ultrasonic information attenuation sequence contains all the ultrasonic data peaks in the target ultrasonic information attenuation sequence, then the target ultrasonic information attenuation sequence is eliminated. Similarly, the remaining ultrasonic information attenuation sequence after elimination is obtained and used as the ultrasonic information attenuation judgment sequence.

[0074] Furthermore, taking any ultrasonic information attenuation determination sequence and any parking ultrasonic data in the parking de-obstruction ultrasonic data segment as an example, the absolute value of the difference between each ultrasonic data peak value and the parking ultrasonic data in the ultrasonic information attenuation determination sequence is used as the difference value between each ultrasonic data peak value and the parking ultrasonic data; the difference value between each ultrasonic data peak value and each parking ultrasonic data is obtained, and the inverse proportional normalized value of the standard deviation of the difference values ​​between all ultrasonic data peak values ​​and all parking ultrasonic data is used as the attenuation confidence weight of the ultrasonic information attenuation determination sequence for the parking de-obstruction ultrasonic data segment.

[0075] Furthermore, the average of the attenuation confidence weights of all ultrasonic information attenuation judgment sequences for the parking unobstructed ultrasonic data segment is used as the reference weight of the fitting line for the parking unobstructed ultrasonic data segment; the fitting line reference weight is regarded as the weight during straight line fitting, and a straight line fitting is performed on the parking unobstructed ultrasonic data segment according to the weight to obtain the fitting straight line of the parking unobstructed ultrasonic data segment after fitting; the parking unobstructed ultrasonic data segment after the fitting straight line is used as the actual vehicle parking ultrasonic data segment.

[0076] It is particularly noted that the process of performing straight line fitting according to weights is a well-known content of the straight line fitting algorithm and will not be described in detail in this embodiment.

[0077] It should be noted that the fitted straight line has a mathematical expression in the form of ,in is the independent variable, is the dependent variable, is the slope, is a constant term.

[0078] So far, several actual vehicle parking ultrasonic data segments have been obtained through the above method.

[0079] Step S004: performing ultrasonic vehicle detection based on the actual vehicle parking ultrasonic data segment.

[0080] In a specific implementation of the embodiment of the present invention, the specific process of ultrasonic vehicle detection is as follows:

[0081] The normalized value of the slope of the fitting straight line of the actual vehicle parking ultrasonic data segment is taken as the real parking state judgment degree of the actual vehicle parking ultrasonic data segment.

[0082] Further, a real parking state judgment degree threshold value is preset If the real parking state judgment degree of the actual vehicle parking ultrasonic data segment is greater than or equal to , the actual vehicle parking ultrasonic data segment is taken as a corrected parking ultrasonic data segment; if the real parking state judgment degree of the actual vehicle parking ultrasonic data segment is less than , the actual vehicle parking ultrasonic data segment is taken as a parking error ultrasonic data segment. The embodiment is described by taking as an example, and the embodiment is not specifically limited, wherein may be determined according to specific implementation conditions.

[0083] It should be noted that the corrected parking ultrasonic data segment is the ultrasonic data segment corresponding to the actual parking position when the actual parking position is real, which indicates that the result of the parking detection by the ultrasonic wave device itself preset ultrasonic wave reflection condition in the corresponding time range is real and reliable; and the parking error ultrasonic data segment is the ultrasonic data segment corresponding to the actual parking position when the actual parking position does not conform to the real parking position, the vehicle in the corresponding time range may transport objects to block the corresponding parking position, or may temporarily use the corresponding parking position for reversing operation in order to change the driving route, which indicates that the result of the parking detection by the ultrasonic wave device itself preset ultrasonic wave reflection condition in the corresponding time range is not real and reliable, and the device needs to be adjusted.

[0084] Thus, the embodiment is completed.

[0085] Another embodiment of the application provides an ultrasonic wave-based vehicle detection system, which comprises a memory and a processor, and the processor executes the computer program stored in the memory to perform the above-mentioned method steps S001 to S004.

[0086] Still another embodiment of the application provides an ultrasonic wave-based vehicle detection device, which comprises a data sequence acquisition module, an actual vehicle parking ultrasonic data segment acquisition module and an ultrasonic wave vehicle detection module; and the device executes the above-mentioned method steps S001 to S004 by calling a computer program.

[0087] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. Ultrasonic vehicle detection method, characterized in that: The method comprises the following steps: Acquire a parking ultrasonic data sequence and a plurality of environmental ultrasonic data sequences; Each environmental ultrasonic data sequence is divided into a number of environmental ultrasonic data segments within a preset number range; information on data changes and stability between different environmental ultrasonic data segments and the environmental ultrasonic data sequences to which they belong is analyzed, and a vehicle reference environmental data sequence is selected from all environmental ultrasonic data sequences; ultrasonic information matching between the vehicle reference environmental data sequence and each environmental ultrasonic data sequence is compared, and cumulative smoothing processing is performed on the environmental ultrasonic data sequence in combination with the environmental ultrasonic data segments to obtain a number of environmental reference ultrasonic data segments; Based on the differences in ultrasonic information matching between different environmental baseline ultrasonic data segments and the parking ultrasonic data sequence, a number of parking unobstructed ultrasonic data segments are divided and processed from the parking ultrasonic data sequence; the attenuation progress of the vehicle ultrasonic information in the different parking unobstructed ultrasonic data segments is analyzed, and the attenuation of the ultrasonic information is simulated for each parking unobstructed ultrasonic data segment. The simulated parking unobstructed ultrasonic data segment is used as the actual vehicle parking ultrasonic data segment; The method for obtaining the actual vehicle parking ultrasonic data segment is: For any parking unobstructed ultrasonic data segment, the parking unobstructed ultrasonic data segment includes multiple ultrasonic data peaks; within the parking unobstructed ultrasonic data segment, the ultrasonic data peaks are sequentially used as starting points, and the numerical relationship between the subsequent ultrasonic data peaks and the starting points is used as a traversal judgment condition, traversing all ultrasonic data peaks to obtain multiple ultrasonic information attenuation judgment sequences; the stable fluctuation of the data difference between each ultrasonic information attenuation judgment sequence and the parking unobstructed ultrasonic data segment is used as the attenuation confidence weight of each ultrasonic information attenuation judgment sequence; the parking unobstructed ultrasonic data segment is corrected according to the attenuation confidence weight, and the corrected parking unobstructed ultrasonic data segment is used as the actual vehicle parking ultrasonic data segment; Ultrasonic vehicle detection is performed based on actual vehicle parking ultrasonic data segments.

2. The ultrasonic vehicle detection method according to claim 1, characterized in that: The method for obtaining the vehicle reference environment data sequence is: For any ambient ultrasonic data sequence, the fluctuation stability of the data in each ambient ultrasonic data segment in the ambient ultrasonic data sequence is analyzed as the data segment stability of each ambient ultrasonic data segment; the ultrasonic data stabilities of all ambient ultrasonic data segments in the ambient ultrasonic data sequence are comprehensively accumulated as the sequence stability of the ambient ultrasonic data sequence; the proportional relationship between the stability of each data segment and the sequence stability is analyzed to obtain the ultrasonic benchmark preference of each ambient ultrasonic data segment; the ultrasonic benchmark preference of the comprehensively accumulated ambient ultrasonic data segments is used as the ambient ultrasonic benchmark preference of the ambient ultrasonic data sequence; and according to the ambient ultrasonic benchmark preference, a vehicle benchmark ambient data sequence is screened out from all ambient ultrasonic data sequences.

3. The ultrasonic vehicle detection method according to claim 2, characterized in that: The method for obtaining the stability of the data segment is: For any ambient ultrasound data segment in the ambient ultrasound data sequence, an inverse proportional value of a standard deviation of data in the ambient ultrasound data segment is used as the data segment stability of the ambient ultrasound data segment.

4. The ultrasonic vehicle detection method according to claim 2, characterized in that: The method for obtaining the ultrasonic reference optimization degree is: The ratio of the stability of each data segment to the sequence stability is used as the ultrasonic benchmark preference of each environmental ultrasonic data segment.

5. The ultrasonic vehicle detection method according to claim 1, characterized in that: The method for obtaining the environmental reference ultrasonic data segment is: For any environmental ultrasonic data sequence, information matching is performed between the vehicle reference environmental data sequence and the environmental ultrasonic data sequence to obtain an environmental ultrasonic information matching segment of the environmental ultrasonic data sequence; based on the ultrasonic matching information in the environmental ultrasonic information matching segment, the data in the environmental ultrasonic data sequence is accumulated as the ultrasonic information smoothing weight of the environmental ultrasonic data sequence; based on the ultrasonic information smoothing weight, each environmental ultrasonic data segment in the environmental ultrasonic data sequence is filtered and smoothed, and the smoothed environmental ultrasonic data segment is used as the environmental reference ultrasonic data segment.

6. The ultrasonic vehicle detection method according to claim 5, characterized in that: The method for obtaining the environmental ultrasonic information matching segment is: Perform DTW matching on the vehicle reference environment data sequence and the environment ultrasonic data sequence to obtain a number of DTW matching pairs; and use the data segments of the DTW matching pairs that match in the environment ultrasonic data sequence as the environment ultrasonic information matching segments.

7. The ultrasonic vehicle detection method according to claim 1, characterized in that: The method for obtaining the parking unblocking ultrasonic data segment is as follows: The environmental reference ultrasonic data segment is integrated to obtain the environmental interference removal reference sequence; information matching is performed on the parking ultrasonic data sequence to obtain the parking ultrasonic information matching segment of the parking ultrasonic data sequence; the data difference between the parking ultrasonic information matching segment and the environmental interference removal reference sequence is used as the ultrasonic information matching difference; the data segment after data correction of the parking ultrasonic information matching segment is performed using the ultrasonic information matching difference as the parking de-occlusion ultrasonic data segment.

8. An ultrasonic-based vehicle detection system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the ultrasonic-based vehicle detection method according to any one of claims 1 to 7 are implemented.

9. Ultrasonic vehicle detection equipment, characterized in that The device comprises: A data sequence acquisition module, an actual vehicle parking ultrasonic data segment acquisition module and an ultrasonic vehicle detection module; wherein the data sequence acquisition module is used to acquire a parking ultrasonic data sequence and several environmental ultrasonic data sequences, the actual vehicle parking ultrasonic data segment acquisition module implements the steps of the ultrasonic-based vehicle detection method as described in any one of claims 1-7 by calling a computer program to obtain the actual vehicle parking ultrasonic data segment, and the ultrasonic vehicle detection module performs ultrasonic vehicle detection based on the actual vehicle parking ultrasonic data segment.

Images