Target fusion method and device based on Hausdorff distance, equipment and medium

Abstract

The invention discloses a Hausdorff distance-based target fusion method, device and equipment and a medium, and relates to the technical field of vehicle-mounted interconnection, and the method comprises the steps: obtaining the coordinate information of a plurality of sensors, obtaining a plurality of pieces of time frame information based on the plurality of pieces of coordinate information, carrying out the calculation according to the plurality of pieces of time frame information, obtaining a plurality of Hausdorff distances, and carrying out the fusion of the plurality of Hausdorff distances. And performing abnormal value judgment according to the plurality of Hausdorff distances to obtain a target fusion result. The Hausdorff distance and abnormal value judgment are obtained through fusion of multi-sensor coordinate calculation, accurate target fusion is achieved, and the accuracy of target fusion is improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle-mounted interconnection technology, and in particular to a target fusion method, device, equipment and medium based on Hausdorff distance. Background Art

[0002] With the development of intelligent transportation systems and autonomous driving technology, accurate perception of the vehicle's surrounding environment has become crucial. Multi-source sensor fusion technology, as a key means to improve environmental perception capabilities, can provide more comprehensive and reliable information about the surrounding environment by integrating data from different sensors, such as radar, LiDAR, and cameras. This technology is of great significance to improving vehicle safety, reducing traffic accidents, and enhancing driving experience.

[0003] At present, multi-source sensor fusion technology has been widely used in target detection, tracking and classification. These technologies usually include data preprocessing, feature extraction, state estimation and data fusion. In the data fusion stage, commonly used methods include time alignment, space alignment, Kalman filtering, particle filtering, etc. These methods can improve the accuracy and robustness of target detection to a certain extent.

[0004] Although the existing multi-source sensor fusion technology has made some progress, it still has obvious shortcomings. Due to differences in their working principles and accuracy, different sensors may have deviations in the measurement results of the same target, especially in long-distance target recognition. At the same time, the acquisition frequency and timestamp of sensor data may be inconsistent, making it difficult to achieve accurate time alignment when performing data fusion. In addition, when processing sensor data, there may be a lack of effective outlier detection and processing mechanisms, resulting in the fusion results being affected by outliers. As the number of sensors and the amount of data increase, the computational complexity of the existing fusion algorithm may become a bottleneck, affecting the real-time performance and efficiency of the system. Therefore, there is an urgent need for a more accurate method to improve the accuracy and robustness of target fusion. Summary of the invention

[0005] The main purpose of this application is to provide a target fusion method, device, equipment and medium based on Hausdorff distance, aiming to solve the technical problem of how to improve the accuracy of target fusion.

[0006] To achieve the above purpose, the present application proposes a target fusion method based on Hausdorff distance, the method comprising:

[0007] Get the coordinate information of multiple sensors;

[0008] Obtaining multiple time frame information based on the multiple coordinate information;

[0009] Calculating based on the multiple time frame information to obtain multiple Hausdorff distances;

[0010] Outlier judgment is performed based on the multiple Hausdorff distances to obtain a target fusion result.

[0011] In one embodiment, the step of calculating based on the plurality of time frame information to obtain the Hausdorff distance includes:

[0012] Selecting the first coordinate point in the plurality of time frame information as a reference point;

[0013] Calculating all points in the same time frame after the reference point to obtain multiple distances;

[0014] Performing error calculation on a plurality of the distances to obtain a plurality of corresponding error terms;

[0015] Obtaining a first scaling factor according to the plurality of distances and the corresponding error terms;

[0016] The plurality of time frame information are scaled according to the first scaling factor to obtain a pre-scaled Hausdorff distance.

[0017] In one embodiment, the step of calculating based on the plurality of time frame information to obtain the Hausdorff distance further includes:

[0018] Selecting middle coordinate points in the plurality of time frame information as reference points;

[0019] Calculating all points in the same time frame before and after the reference point to obtain multiple distances;

[0020] Performing error calculation on a plurality of the distances to obtain a plurality of corresponding error terms;

[0021] Obtaining a second scaling factor according to the plurality of distances and the corresponding error terms;

[0022] The plurality of time frame information are scaled according to the second scaling factor to obtain an intermediate scaled Hausdorff distance.

[0023] In one embodiment, the step of calculating based on the plurality of time frame information to obtain the Hausdorff distance further includes:

[0024] Select the last coordinate point in the multiple time frame information as a reference point;

[0025] Calculating all points in the same time frame before the reference point to obtain multiple distances;

[0026] Performing error calculation on a plurality of the distances to obtain a plurality of corresponding error terms;

[0027] Obtaining a third scaling factor according to the plurality of distances and the corresponding error terms;

[0028] The plurality of time frame information are scaled according to the third scaling factor to obtain a post-scaled Hausdorff distance.

[0029] In one embodiment, the step of performing outlier judgment according to the plurality of Hausdorff distances to obtain a target fusion result includes:

[0030] Calculating a plurality of the Hausdorff distances to obtain a mean and a standard deviation;

[0031] Using the three sigma principle to judge the mean and standard deviation, and obtain a judgment result;

[0032] When the judgment result meets the preset requirements, the target fusion result is target fusion success;

[0033] When the judgment result does not meet the preset requirement, the target fusion result is target fusion failure, and the sensor parameters are further adjusted to improve the target fusion result.

[0034] In one embodiment, the step of obtaining coordinate information of multiple sensors includes:

[0035] Get sensor timestamps and timing information of multiple sensors;

[0036] According to the timestamp, the plurality of timing information are matched to the same time frame to obtain a plurality of synchronized timing information;

[0037] Preprocessing the plurality of synchronized time series information to obtain a plurality of processed data;

[0038] Coordinate information of multiple sensors is extracted from the multiple processed data.

[0039] In one embodiment, the step of corresponding the plurality of timing information to the same time frame according to the timestamp to obtain a plurality of synchronized timing information includes:

[0040] The timestamp is adjusted according to a preset time reference to ensure that the timing information of multiple sensors is aligned in time, thereby obtaining an adjusted timestamp;

[0041] The adjusted timestamp and the corresponding multiple timing information are mapped into the same time frame to obtain multiple synchronized timing information.

[0042] In addition, to achieve the above purpose, the present application also proposes a target fusion device based on Hausdorff distance, and the target fusion device based on Hausdorff distance includes:

[0043] An acquisition module, used to acquire coordinate information of multiple sensors;

[0044] The acquisition module is further used to obtain multiple time frame information based on the multiple coordinate information;

[0045] A calculation module, used for performing calculations based on the plurality of time frame information to obtain a plurality of Hausdorff distances;

[0046] The judgment module is used to judge the outliers according to the multiple Hausdorff distances to obtain the target fusion result.

[0047] In addition, to achieve the above-mentioned purpose, the present application also proposes a medium, which is a computer-readable medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the target fusion method based on Hausdorff distance as described above are implemented.

[0048] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the target fusion method based on Hausdorff distance as described above are implemented.

[0049] This application obtains the coordinate information of multiple sensors, obtains multiple time frame information based on the multiple coordinate information, calculates based on the multiple time frame information, obtains multiple Hausdorff distances, and performs outlier judgment based on the multiple Hausdorff distances to obtain the target fusion result. By fusing the coordinates of multiple sensors to calculate the Hausdorff distance and outlier judgment, accurate target fusion is achieved, which improves the accuracy of target fusion. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0051] Figure 1 This is a flow chart of the first embodiment of the target fusion method based on Hausdorff distance of the present application;

[0052] Figure 2 This is a flow chart of a second embodiment of the target fusion method based on Hausdorff distance of the present application;

[0053] Figure 3This is a schematic diagram of the module structure of a target fusion device based on Hausdorff distance in an embodiment of the present application;

[0054] Figure 4 A schematic diagram of the device structure of the hardware operating environment involved in the target fusion method based on Hausdorff distance in an embodiment of the present application;

[0055] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0056] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0057] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0058] The main solution of the embodiment of the present application is: obtain coordinate information of multiple sensors, obtain multiple time frame information based on the multiple coordinate information, perform calculations based on the multiple time frame information to obtain multiple Hausdorff distances, perform outlier judgment based on the multiple Hausdorff distances, and obtain target fusion results.

[0059] Based on this, the embodiment of the present application provides a target fusion method based on Hausdorff distance, referring to Figure 1 , Figure 1 This is a flowchart of the first embodiment of the target fusion method based on Hausdorff distance of the present application.

[0060] In this embodiment, the target fusion method based on Hausdorff distance includes steps S10 to S40:

[0061] Step S10, obtaining coordinate information of multiple sensors.

[0062] It should be noted that this process involves collecting the position data of the target object from various sensors such as radar, LiDAR, cameras, etc. These sensors work independently, but due to their different characteristics and accuracy, the collected data may be biased and inconsistent. Because the data types and formats provided by cameras and radars may be different: cameras usually provide visual position estimates of a single target, while radars may return a series of reflected signals corresponding to different parts of the target or the results of multiple reflections. This means that within a specific time frame, radar data may contain multiple coordinate points. For camera data, we get a clear position estimate, while for radar data, all radar echo points captured within the time period are included to form a point set.

[0063] Furthermore, sensor timestamps and timing information of multiple sensors are obtained. Data with timestamps are obtained from various sensors, which contain information captured by different sensors at different time points. The timestamps carried in the sensor data provide a time reference, so that the data from different sensors are mapped to the same time frame to achieve time synchronization. These timing information are mapped to the same time frame according to the timestamps. This process involves understanding the sampling frequency and data transmission delay of each sensor in order to accurately adjust the timestamps of each record so that they are aligned on the time axis. The purpose of this is to ensure that the target information recorded by different sensors occurs in the same time period, thereby avoiding fusion errors caused by time deviation.

[0064] Furthermore, multiple time series information are mapped to the same time frame according to the timestamp to obtain multiple synchronized time series information. In the process of multi-sensor target fusion, it is first necessary to ensure that the timestamps of each sensor can be aligned with each other. This is because different sensors have different time bases and different sampling frequencies, which will cause the acquired data to be not completely synchronized in time. In order to eliminate this time deviation, the timestamps of all sensors need to be adjusted according to a preset time base. Specifically, the timestamps are adjusted according to the preset time base to ensure that the time series information of multiple sensors are aligned in time, and the adjusted timestamps are obtained. The adjusted timestamps and the corresponding multiple time series information are mapped to the same time frame to obtain multiple synchronized time series information. The process of adjusting the timestamp involves correcting the time base of each sensor to a unified time line. This is usually achieved by finding the time difference between the start-up times of each sensor or by an external synchronization signal. Once the time base is corrected, the timestamps of all sensors can be corrected so that the time of occurrence of events recorded by different sensors can be accurately aligned. Next, these adjusted timestamps and the corresponding sensor time series information are mapped to the same time frame. This step is the key to ensure that the data of all sensors are compared within the same time interval. By mapping the sensor data to the same time frame, it is ensured that during subsequent processing, such as calculating the Hausdorff distance, all sensor observations are obtained within the same time period, so that the consistency of observations from different sensors can be accurately assessed.

[0065] Further, the multiple synchronized time series information is preprocessed to obtain multiple processed data, and the coordinate information of multiple sensors is extracted from the multiple processed data. Once the time series information is synchronized, the next step is to preprocess the data. The preprocessing stage includes operations such as noise removal, missing value filling, and data smoothing to ensure the quality of the data for subsequent processing. Especially in radar data processing, since the radar will detect multiple target points, these points need to be screened and those with large errors need to be removed to ensure the reliability and validity of the data. After completing the preprocessing, we extract the target coordinate information of each sensor from the processed data. The camera usually provides a single target position estimate, while the radar may return a series of reflection signal points. The data of different sensors are unified into the same coordinate system, such as the vehicle coordinate system or the global coordinate system, and the coordinate information in the same coordinate system is finally obtained.

[0066] In addition, in order to deal with the situation where the number of coordinate values ​​identified by the radar is inconsistent with that of the camera, the radar data needs to be preprocessed to evaluate the reliability of each point by calculating the mean error between each point and the adjacent points, and remove those points whose errors are significantly higher than the average. In order to more accurately reflect the position change of the target object, interpolation or resampling technology can be used to adjust the data of different sensors to a uniform time resolution.

[0067] Step S20, obtaining multiple time frame information based on multiple coordinate information.

[0068] It should be noted that the boundaries of the time frame are defined based on the synchronized coordinate information. These boundaries can be fixed time periods, such as a time frame every 100 milliseconds, or can be dynamically adjusted based on the motion characteristics of the target object. In each time frame, the coordinate information of the corresponding time point is integrated to form a snapshot of the target state in the time frame.

[0069] In addition, in order to improve the accuracy and reliability of the time frame information, the coordinate information in each time frame may need to be further processed. This includes using statistical methods or filtering algorithms to estimate the optimal position of the target object at the center of the time frame, and evaluating and removing possible outliers or data points with large errors.

[0070] The generated time frame information includes not only the position coordinates of the target object, but also dynamic information such as speed and acceleration, providing a rich source of information for subsequent data analysis and target fusion. Through the time frame information, the motion state and behavior pattern of the target object can be more comprehensively understood, and accurate target tracking and recognition can be achieved.

[0071] Step S30, performing calculations based on the multiple time frame information to obtain multiple Hausdorff distances.

[0072] It is important to understand that the consistency of the target position changes captured by different sensors in the same time window is quantified by the Hausdorff distance. The Hausdorff distance is a metric that measures the distance between the farthest points of two sets, which is very suitable for evaluating the consistency of sensor data. In order to adapt to the accuracy differences between different sensors, three scaling strategies are proposed: front scaling, back scaling, and median scaling. The three Hausdorff distances can be calculated using the front, back, and median scaling methods, respectively. These strategies ensure that the data between different sensors can be compared on the same scale by scaling the head, tail, or middle part of the sensor detection value, so as to better analyze the consistency of the trend of the target position over time.

[0073] Further, for the calculated Hausdorff distance before scaling, the first coordinate point in the multiple time frame information is selected as the reference point; all points in the same time frame after the reference point are calculated to obtain multiple distances; errors are calculated for the multiple distances to obtain multiple corresponding error terms; a first scaling factor is obtained according to the multiple distances and the corresponding error terms; the multiple time frame information is scaled according to the first scaling factor to obtain the Hausdorff distance before scaling. Specifically, the first coordinate point in the first time frame is selected as the reference point. Then, the distance between this reference point and all other points in the same time frame is calculated to form a series of distance values. Then, the error calculation is performed on these distance values, that is, the deviation between the actual distance and the theoretical distance between each point and the reference point, thereby obtaining a series of error terms. These error terms reflect the degree of positional deviation of each point relative to the reference point. Based on these error terms, the first scaling factor can be calculated, which will be used to adjust the position of subsequent points so that the relationship between them and the reference point is more consistent. Specifically, the scaling factor is determined based on the ratio of the distance between the reference point and other points to the corresponding error term. This step is to eliminate the scale differences that may exist between different sensors. The coordinate information in each time frame may be scaled using the calculated first scaling factor, and this scaling is called front scaling.

[0074] Further, the middle coordinate point in the multiple time frame information is selected as the reference point, all points in the same time frame before and after the reference point are calculated to obtain multiple distances, the errors of the multiple distances are calculated to obtain multiple corresponding error terms, the second scaling factor is obtained according to the multiple distances and the corresponding error terms, and the multiple time frame information is scaled according to the second scaling factor to obtain the intermediate scaled Hausdorff distance. The middle coordinate point in each time frame information is selected as the reference point. This is done because the middle point can usually better represent the central trend of the target position in the entire time frame, especially when the data distribution is relatively uniform. After the reference point is selected, the distance between the reference point and all other points in the same time frame is calculated. The error calculation is performed on these distance values ​​to obtain a series of error terms. Based on these distances and their corresponding error terms, the second scaling factor can be calculated. Generally, the scaling factor is obtained by analyzing the relationship between the distance and the error term, such as fitting the optimal scaling factor by the least square method or other statistical methods. Finally, the coordinate information in each time frame is scaled using the calculated second scaling factor. This scaling process is called intermediate scaling because it is adjusted based on the middle point in the time frame. After the scaling process, the intermediate scaled Hausdorff distance can be calculated.

[0075] Further, the last coordinate point in the multiple time frame information is selected as the reference point, and all points in the same time frame before the reference point are calculated to obtain multiple distances. Error calculation is performed on the multiple distances to obtain multiple corresponding error terms, and a third scaling factor is obtained according to the multiple distances and the corresponding error terms. The multiple time frame information is scaled according to the third scaling factor to obtain the scaled Hausdorff distance. The last coordinate point in each time frame is selected as the reference point. This selection is based on the assumption that the last point can better represent the state of the target at the end of the time frame. Next, the distances between all points in the same time frame before the reference point and the reference point are calculated. These distance values ​​reflect the spatial position relationship between each point and the reference point. Further, the error calculation is performed on these distance values ​​to obtain the error term between each point and the reference point. Based on these distance values ​​and their corresponding error terms, the third scaling factor is calculated. The scaling factor is calculated by analyzing the relationship between the distance and the error term, which reflects how the data needs to be adjusted to achieve better consistency. The role of the scaling factor is to adjust the data so that the target position change trend detected by different sensors is more consistent. Finally, the coordinate information in each time frame is scaled according to the calculated third scaling factor. This scaling process is called post-scaling, which adjusts the sensor data to ensure the consistency of the data in scale. After the scaling process, the new Hausdorff distance can be calculated, which will be used in the subsequent target fusion process.

[0076] Step S40, performing outlier judgment based on multiple Hausdorff distances to obtain a target fusion result.

[0077] It should be noted that after completing a series of coordinate information acquisition, timing information synchronization, scaling processing and Hausdorff distance calculation, the next step is to perform statistical analysis on these Hausdorff distances to determine whether there are outliers. This process is crucial to determining whether the sensor data is successfully fused. By applying the three-sigma principle, that is, centering on the mean of the Hausdorff distance and adding or subtracting three times the standard deviation (μ±3σ), a normal value range can be constructed. Within this range, the Hausdorff distance calculated under most normal circumstances should fall here.

[0078] If the Hausdorff distance in a time frame exceeds this range, then the data in the time frame is considered to have an outlier, which indicates that the data from different sensors cannot be correctly matched, which may be caused by sensor failure, target recognition error or other factors. At this time, it can be determined that the target has not been effectively fused in the time frame. On the contrary, if all calculated Hausdorff distances are within the range specified by the three sigma principle, it can be considered that the sensor data has been successfully fused, that is, the target has been correctly identified and tracked in the time frame.

[0079] This embodiment provides a target fusion method based on Hausdorff distance, by acquiring coordinate information of multiple sensors, obtaining multiple time frame information based on the multiple coordinate information, performing calculations based on the multiple time frame information to obtain multiple Hausdorff distances, performing outlier judgments based on the multiple Hausdorff distances, and obtaining target fusion results. By fusing the coordinates of multiple sensors to calculate the Hausdorff distance and outlier judgment, accurate target fusion is achieved, thereby improving the accuracy of target fusion.

[0080] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 2 The target fusion method based on Hausdorff distance step S40 also includes steps S401 to S404:

[0081] Step S401, calculating multiple Hausdorff distances to obtain a mean and a standard deviation.

[0082] It should be noted that in the process of completing target fusion, the mean and standard deviation of these distance values ​​can be obtained by statistically analyzing the Hausdorff distances calculated in multiple time frames. The mean reflects the average level of all Hausdorff distances, while the standard deviation indicates the degree to which these distance values ​​fluctuate around the mean. These two statistical indicators together constitute an important basis for evaluating the consistency of sensor data.

[0083] Specifically, the mean of the Hausdorff distance is calculated, which is the sum of all distance values ​​divided by the number of distance values. This mean provides a measure of the central tendency, indicating whether the target position changes detected by different sensors are consistent or significantly different overall. Secondly, the standard deviation of the Hausdorff distance is calculated. The standard deviation reveals the average degree of deviation between each distance value and the mean. The smaller the standard deviation, the closer the target position change trends detected by the sensors in each time frame are, and the better the consistency; conversely, the larger the standard deviation, the greater the difference in the detection results between the sensors, and the poorer the consistency.

[0084] Step S402, using the three sigma principle to judge the mean and standard deviation to obtain a judgment result.

[0085] It is important to understand that the Three Sigma principle is a statistical concept that is often used to identify outliers in data. According to this principle, the vast majority of data points should fall within the range of the mean plus or minus three standard deviations, i.e. (μ-3σ, μ+3σ).

[0086] Step S403: When the judgment result meets the preset requirements, the target fusion result is target fusion success.

[0087] It should be understood that when the result of judging the Hausdorff distance by the three-sigma principle meets the preset requirements, it means that all calculated Hausdorff distances are within the range of (μ-3σ, μ+3σ), and the target fusion can be considered successful. This shows that the target position change trends detected by different sensors in the corresponding time frame are consistent, and there are no significant outliers.

[0088] Step S404: when the judgment result does not meet the preset requirement, the target fusion result is target fusion failure, and the sensor parameters are further adjusted to improve the target fusion result.

[0089] It should be understood that when the judgment result does not meet the preset requirements, that is, the Hausdorff distance is not within the range set by the three sigma principle (μ-3σ, μ+3σ), this means that the sensor data has not been correctly matched and the target fusion has failed. At this time, measures need to be taken to improve the results of target fusion. First, it is necessary to check whether the sensor parameter settings are reasonable. This includes but is not limited to the installation position, angle, operating frequency, etc. of the sensor. For example, if the installation angle or position of the radar and camera is inappropriate, it may cause deviations in the detected target position. In addition, the operating frequency of the sensor may also affect the synchronization of data acquisition and needs to be adjusted to a suitable frequency to ensure time alignment.

[0090] Secondly, you can try to optimize the sensor preprocessing algorithm. For example, perform more detailed noise filtering on radar data, or adjust the camera image processing algorithm to improve the accuracy of target recognition. The logic of removing abnormal points in radar data can be further optimized to ensure that only reliable detection results are retained for fusion calculation.

[0091] Alternatively, you can recalibrate the sensor. Over time, the sensor may become out of calibration due to environmental changes, physical movement, etc. Recalibration can help restore the sensor to optimal performance.

[0092] Finally, considering the differences in detection accuracy of sensors, the data scaling strategy can be adjusted. For example, pre-scaling, post-scaling, or intermediate value scaling methods can be used to better match the data scales between different sensors, thereby improving the consistency of the target position change trend.

[0093] Through the above measures, the problems that lead to target fusion failure can be gradually identified and solved, and ultimately a more accurate target fusion result can be achieved.

[0094] In this embodiment, a plurality of Hausdorff distances are calculated to obtain a mean and a standard deviation, and the mean and the standard deviation are judged by the three sigma principle to obtain a judgment result. When the judgment result meets the preset requirements, the target fusion result is a target fusion success. When the judgment result does not meet the preset requirements, the target fusion result is a target fusion failure. The sensor parameters are further adjusted to improve the target fusion result. The mean and the standard deviation of the Hausdorff distance are judged by the three sigma principle, the sensor parameters are optimized, and the accuracy and reliability of the target fusion are improved.

[0095] This application also provides a target fusion device based on Hausdorff distance, please refer to Figure 3 , the device comprises:

[0096] An acquisition module 10, used to acquire coordinate information of multiple sensors;

[0097] The acquisition module 10 is further used to obtain multiple time frame information based on multiple coordinate information;

[0098] A calculation module 20, configured to perform calculations based on the multiple time frame information to obtain multiple Hausdorff distances;

[0099] The judgment module 30 is used to judge the outliers according to the multiple Hausdorff distances to obtain the target fusion result.

[0100] The target fusion device based on Hausdorff distance provided by the present application adopts the target fusion method based on Hausdorff distance in the above embodiment, which can solve the technical problem of how to improve the accuracy of target fusion. Compared with the prior art, the beneficial effects of the target fusion device based on Hausdorff distance provided by the present application are the same as the beneficial effects of the target fusion method based on Hausdorff distance provided by the above embodiment, and other technical features in the target fusion device based on Hausdorff distance are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0101] In one embodiment, the calculation module 20 is further used to select the first coordinate point in the multiple time frame information as a reference point; calculate all points in the same time frame after the reference point to obtain multiple distances; perform error calculation on the multiple distances to obtain multiple corresponding error terms; obtain a first scaling factor based on the multiple distances and the corresponding error terms; and scale the multiple time frame information according to the first scaling factor to obtain a pre-scaled Hausdorff distance.

[0102] In one embodiment, the calculation module 20 is further used to select the middle coordinate point in the multiple time frame information as the reference point; calculate all points in the same time frame before and after the reference point to obtain multiple distances; perform error calculation on the multiple distances to obtain multiple corresponding error terms; obtain a second scaling factor based on the multiple distances and the corresponding error terms; and scale the multiple time frame information according to the second scaling factor to obtain the intermediate scaled Hausdorff distance.

[0103] In one embodiment, the calculation module 20 is further used to select the last coordinate point in the multiple time frame information as a reference point; calculate all points in the same time frame before the reference point to obtain multiple distances; perform error calculation on the multiple distances to obtain multiple corresponding error terms; obtain a third scaling factor based on the multiple distances and the corresponding error terms; and scale the multiple time frame information according to the third scaling factor to obtain a scaled Hausdorff distance.

[0104] In one embodiment, the judgment module 30 is also used to calculate multiple Hausdorff distances to obtain a mean and a standard deviation; the mean and the standard deviation are judged using the three sigma principle to obtain a judgment result; when the judgment result meets the preset requirements, the target fusion result is a target fusion success; when the judgment result does not meet the preset requirements, the target fusion result is a target fusion failure, and the sensor parameters are further adjusted to improve the target fusion result.

[0105] In one embodiment, the acquisition module 10 is also used to obtain the sensor timestamp and the timing information of multiple sensors; correspond the multiple timing information to the same time frame according to the timestamp to obtain multiple synchronized timing information; pre-process the multiple synchronized timing information to obtain multiple processed data; extract from the multiple processed data to obtain the coordinate information of multiple sensors.

[0106] In one embodiment, the acquisition module 10 is also used to adjust the timestamp according to a preset time reference to ensure that the timing information of multiple sensors is aligned in time, thereby obtaining an adjusted timestamp; and map the adjusted timestamp and the corresponding multiple timing information into the same time frame to obtain multiple synchronized timing information.

[0107] The present application provides a target fusion device based on Hausdorff distance, and the target fusion device based on Hausdorff distance includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the target fusion method based on Hausdorff distance in the above-mentioned embodiment one.

[0108] Reference below Figure 4 , which shows a schematic diagram of the structure of a target fusion device based on Hausdorff distance suitable for implementing the embodiment of the present application. The target fusion device based on Hausdorff distance in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The target fusion device based on Hausdorff distance shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0109] like Figure 4As shown, the target fusion device based on Hausdorff distance may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of the target fusion device based on Hausdorff distance are also stored in RAM1004. The processing device 1001, ROM1002, and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 may allow the target fusion device based on Hausdorff distance to communicate with other devices wirelessly or by wire to exchange data. Although the Hausdorff distance-based target fusion device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have instead.

[0110] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0111] The target fusion device based on Hausdorff distance provided by the present application adopts the target fusion method based on Hausdorff distance in the above embodiment, which can solve the technical problem of how to improve the accuracy of target fusion. Compared with the prior art, the beneficial effects of the target fusion device based on Hausdorff distance provided by the present application are the same as the beneficial effects of the target fusion method based on Hausdorff distance provided by the above embodiment, and other technical features in the target fusion device based on Hausdorff distance are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0112] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0113] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0114] The present application provides a computer-readable medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the target fusion method based on Hausdorff distance in the above-mentioned embodiment.

[0115] The computer-readable medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0116] The computer-readable medium may be included in the target fusion device based on Hausdorff distance; or may exist independently without being assembled into the target fusion device based on Hausdorff distance.

[0117] The computer-readable medium carries one or more programs. When the one or more programs are executed by the target fusion device based on Hausdorff distance, the target fusion device based on Hausdorff distance can be written in one or more programming languages ​​or a combination thereof to perform computer program codes for the operation of the present application. The programming languages ​​include object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0118] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0119] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0120] The readable medium provided by the present application is a computer-readable medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned target fusion method based on Hausdorff distance, and can solve the technical problem of how to improve the accuracy of target fusion. Compared with the prior art, the beneficial effects of the computer-readable medium provided by the present application are the same as the beneficial effects of the target fusion method based on Hausdorff distance provided by the above-mentioned embodiment, and will not be repeated here.

[0121] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned target fusion method based on Hausdorff distance when executed by a processor.

[0122] The computer program product provided by the present application can solve the technical problem of how to improve the accuracy of target fusion. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the target fusion method based on Hausdorff distance provided by the above embodiment, which will not be repeated here.

[0123] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A target fusion method based on Hausdorff distance, characterized in that: The method comprises: Get the coordinate information of multiple sensors; Obtaining multiple time frame information based on the multiple coordinate information; Calculating based on the multiple time frame information to obtain multiple Hausdorff distances; Outlier judgment is performed based on the multiple Hausdorff distances to obtain a target fusion result.

2. The method according to claim 1, characterized in that The step of calculating according to the plurality of time frame information to obtain the Hausdorff distance comprises: Selecting the first coordinate point in the plurality of time frame information as a reference point; Calculating all points in the same time frame after the reference point to obtain multiple distances; Performing error calculation on a plurality of the distances to obtain a plurality of corresponding error terms; Obtaining a first scaling factor according to the plurality of distances and the corresponding error terms; The plurality of time frame information are scaled according to the first scaling factor to obtain a pre-scaled Hausdorff distance.

3. The method according to claim 1, characterized in that The step of calculating according to the plurality of time frame information to obtain the Hausdorff distance further includes: Selecting middle coordinate points in the plurality of time frame information as reference points; Calculating all points in the same time frame before and after the reference point to obtain multiple distances; Performing error calculation on a plurality of the distances to obtain a plurality of corresponding error terms; Obtaining a second scaling factor according to the plurality of distances and the corresponding error terms; The plurality of time frame information are scaled according to the second scaling factor to obtain an intermediate scaled Hausdorff distance.

4. The method according to claim 1, characterized in that The step of calculating according to the plurality of time frame information to obtain the Hausdorff distance further includes: Select the last coordinate point in the multiple time frame information as a reference point; Calculating all points in the same time frame before the reference point to obtain multiple distances; Performing error calculation on a plurality of the distances to obtain a plurality of corresponding error terms; Obtaining a third scaling factor according to the plurality of distances and the corresponding error terms; The plurality of time frame information are scaled according to the third scaling factor to obtain a post-scaled Hausdorff distance.

5. The method according to claim 1, characterized in that The step of performing outlier judgment according to the plurality of Hausdorff distances to obtain a target fusion result includes: Calculating a plurality of Hausdorff distances to obtain a mean and a standard deviation; Using the three sigma principle to judge the mean and standard deviation, and obtain a judgment result; When the judgment result meets the preset requirements, the target fusion result is target fusion success; When the judgment result does not meet the preset requirement, the target fusion result is target fusion failure, and the sensor parameters are further adjusted to improve the target fusion result.

6. The method according to claim 1, characterized in that The step of obtaining coordinate information of multiple sensors includes: Get sensor timestamps and timing information of multiple sensors; According to the timestamp, the plurality of timing information are matched to the same time frame to obtain a plurality of synchronized timing information; Preprocessing the plurality of synchronized time series information to obtain a plurality of processed data; Coordinate information of multiple sensors is extracted from the multiple processed data.

7. The method according to claim 2, characterized in that The step of corresponding the plurality of timing information to the same time frame according to the timestamp to obtain a plurality of synchronized timing information comprises: Adjusting the timestamp according to a preset time reference to ensure that the timing information of multiple sensors is aligned in time, and obtaining an adjusted timestamp; The adjusted timestamp and the corresponding multiple timing information are mapped into the same time frame to obtain multiple synchronized timing information.

8. A target fusion device based on Hausdorff distance, characterized in that: The device comprises: An acquisition module, used to acquire coordinate information of multiple sensors; The acquisition module is further used to obtain multiple time frame information based on the multiple coordinate information; A calculation module, used for performing calculations based on the plurality of time frame information to obtain a plurality of Hausdorff distances; The judgment module is used to judge the outliers according to the multiple Hausdorff distances to obtain the target fusion result.

9. A target fusion method and device based on Hausdorff distance, characterized in that: The device includes: a memory, a processor, and a target fusion method program based on Hausdorff distance stored in the memory and executable on the processor, wherein the target fusion method program based on Hausdorff distance is configured to implement the steps of the target fusion method based on Hausdorff distance as described in any one of claims 1 to 7.

10. A medium, characterized in that The medium stores a target fusion method program based on Hausdorff distance, and when the target fusion method program based on Hausdorff distance is executed by a processor, the steps of the target fusion method based on Hausdorff distance as described in any one of claims 1 to 7 are implemented.

Images