Vehicle distance estimation method and system

Abstract

A vehicle distance estimation method implemented by a vehicle distance estimation system, the vehicle distance estimation system is suitable for being arranged on a first vehicle, and the method comprises: the vehicle distance estimation system performs image analysis on a dynamic image at least according to a correction weight parameter to generate an estimated distance, the estimated distance indicating the interval distance between the first vehicle and a second vehicle presented by the dynamic image. The vehicle distance estimation system also calculates a standard reference distance according to a first geographical location coordinate related to the first vehicle and a second geographical location coordinate related to the second vehicle, and updates the correction weight parameter according to the size relationship between the standard reference distance and the estimated distance, and generates another estimated distance according to the updated correction weight parameter.

Description

[Technical field]

[0001] The present invention relates to a vehicle distance estimation method, and more particularly to a vehicle distance estimation method that uses image recognition and geographic location coordinates simultaneously. The present invention also relates to a vehicle distance estimation system that can implement the vehicle distance estimation method. [Background technology]

[0002] In the prior art, many driving assistance systems perform image analysis on real-time video images to estimate the distance between the vehicle and the vehicle in front, and issue a warning when the distance is too close, thereby reminding the driver to avoid collision with the vehicle in front.

[0003] However, the algorithm for estimating vehicle distance using image analysis is inevitably prone to errors. Furthermore, the installation position and angle of the camera lens may vary depending on the vehicle model and the driver's preference. Therefore, for the vehicle distance estimation result obtained using image analysis, the uncertainty of the camera lens posture may further amplify the estimation error. Therefore, there is still room for further improvement in the existing technology. [Summary of the invention]

[0004] The technical problem to be solved by the present invention is to provide a vehicle distance estimation method which can improve the prior art.

[0005] To solve the above technical problems, a vehicle distance estimation method of the present invention is implemented by a vehicle distance estimation system, which is suitable for being set on a first vehicle, and the vehicle distance estimation method comprises: the vehicle distance estimation system performs image analysis on a dynamic image at least according to a correction weight parameter to generate an estimated distance, wherein the estimated distance indicates the interval distance between the first vehicle and a second vehicle presented by the dynamic image, and the vehicle distance estimation system also calculates a standard reference distance according to a first geographical location coordinate related to the first vehicle and a second geographical location coordinate related to the second vehicle, and updates the correction weight parameter according to the size relationship between the standard reference distance and the estimated distance, and generates another estimated distance according to the updated correction weight parameter.

[0006] Preferably, the vehicle distance estimation system stores a plurality of location feature data each indicating a specific location, and a plurality of geographical location coordinates corresponding to the location feature data. The vehicle distance estimation system performs image analysis on the dynamic image to generate the estimated distance only when it determines that a location matching condition is met, and the location matching condition represents that the vehicle distance estimation system determines that the surrounding environment of the second vehicle matches one of the matching location feature data according to the dynamic image. The vehicle distance estimation system obtains the second geographical location coordinate by using the geographical location coordinate corresponding to the matching location feature data as the second geographical location coordinate.

[0007] Preferably, the first geographic location coordinate is generated by positioning using satellite positioning technology when the vehicle distance estimation system determines that both a satellite signal quality condition related to satellite signal strength and the location matching condition are met.

[0008] Preferably, the location of the second vehicle in the environment presented by the dynamic image is taken as a target location, and the estimated distance indicates the distance between the first vehicle and the second vehicle located at the target location. The first geographical location coordinate is generated by the vehicle distance estimation system using satellite positioning technology for positioning, and the first geographical location coordinate represents the location of the first vehicle when the second vehicle is located at the target location. The second geographical location coordinate is generated by the vehicle distance estimation system using satellite positioning technology for positioning when the vehicle distance estimation system determines that the first vehicle has moved to the target location based on the dynamic image.

[0009] Preferably, the first geographical location coordinate and the second geographical location coordinate are generated by positioning using satellite positioning technology when the vehicle distance estimation system determines that a satellite signal quality condition related to satellite signal strength is met.

[0010] Preferably, after calculating the standard reference distance, the vehicle distance estimation system first determines whether the difference between the standard reference distance and the estimated distance is within a preset allowable range, and only if the judgment result is no, updates the correction weight parameter according to the size relationship between the standard reference distance and the estimated distance.

[0011] Another technical problem to be solved by the present invention is to provide a vehicle distance estimation system capable of implementing the vehicle distance estimation method.

[0012] To solve the above-mentioned another technical problem, the vehicle distance estimation system of the present invention is suitable for being set in a first vehicle, and the vehicle distance estimation system is used to perform image analysis on a dynamic image at least according to a correction weight parameter to generate an estimated distance, wherein the estimated distance indicates the interval distance between the first vehicle and a second vehicle presented by the dynamic image, and the vehicle distance estimation system is also used to calculate a standard reference distance according to a first geographical location coordinate related to the first vehicle and a second geographical location coordinate related to the second vehicle, and update the correction weight parameter according to the size relationship between the standard reference distance and the estimated distance, and generate another estimated distance according to the updated correction weight parameter.

[0013] Preferably, the vehicle distance estimation system stores a plurality of location feature data each indicating a specific location, and a plurality of geographical location coordinates corresponding to the location feature data. The vehicle distance estimation system performs image analysis on the dynamic image to generate the estimated distance only when it determines that a location matching condition is met, and the location matching condition represents that the vehicle distance estimation system determines that the surrounding environment of the second vehicle matches one of the matching location feature data according to the dynamic image. The vehicle distance estimation system obtains the second geographical location coordinate by using the geographical location coordinate corresponding to the matching location feature data as the second geographical location coordinate.

[0014] Preferably, the first geographic location coordinate is generated by positioning using satellite positioning technology when the vehicle distance estimation system determines that both a satellite signal quality condition related to satellite signal strength and the location matching condition are met.

[0015] Preferably, the location of the second vehicle in the environment presented by the dynamic image is taken as a target location, and the estimated distance indicates the distance between the first vehicle and the second vehicle located at the target location. The first geographical location coordinate is generated by the vehicle distance estimation system using satellite positioning technology for positioning, and the first geographical location coordinate represents the location of the first vehicle when the second vehicle is located at the target location. The second geographical location coordinate is generated by the vehicle distance estimation system using satellite positioning technology for positioning when the vehicle distance estimation system determines that the first vehicle has moved to the target location based on the dynamic image.

[0016] Preferably, the first geographical location coordinate and the second geographical location coordinate are generated by positioning using satellite positioning technology when the vehicle distance estimation system determines that a satellite signal quality condition related to satellite signal strength is met.

[0017] Preferably, after calculating the standard reference distance, the vehicle distance estimation system first determines whether the difference between the standard reference distance and the estimated distance is within a preset allowable range, and only if the judgment result is no, updates the correction weight parameter according to the size relationship between the standard reference distance and the estimated distance.

[0018] Another technical problem to be solved by the present invention is to provide a computer program product.

[0019] To solve another technical problem mentioned above, the computer program product of the present invention includes an application program. When the application program is loaded and executed by an electronic device, the electronic device can implement the vehicle distance estimation method as described in any of the aforementioned implementations.

[0020] Compared with the prior art, the vehicle distance estimation system of the present invention can calculate the standard reference distance using the first geographical location coordinates and the second geographical location coordinates, and update the correction weight parameter according to the size relationship between the standard reference distance and the estimated distance, and then recalculate another estimated distance according to the updated correction weight parameter. In this way, the vehicle distance estimation system can effectively correct the estimation error caused by image analysis (including the error caused by the setting method of the shooting unit and / or the error of the image ranging algorithm itself), thereby obtaining a more accurate estimated distance.

Brief Description of the Drawings

[0021] Other features and effects of the present invention will be clearly presented in the embodiments with reference to the accompanying drawings, in which:

[0022] Figure 1 is a block diagram schematically illustrating a first embodiment of a vehicle distance estimation system of the present invention, and a first vehicle on which the first embodiment is disposed;

[0023] Figure 2 is a schematic diagram exemplarily illustrating a frame of a dynamic image generated by a shooting unit of the first embodiment;

[0024] Figure 3 is a flow chart for exemplarily illustrating how the first embodiment implements a vehicle distance estimation method; and

[0025] Figure 4 is a flow chart for exemplarily illustrating how a second embodiment of the vehicle distance estimation system of the present invention implements another vehicle distance estimation method. [Specific implementation method]

[0026] Before the present invention is described in detail, it should be noted that similar components are denoted by the same reference numerals in the following description.

[0027] If not specifically defined, the "electrical connection" described in this patent specification generally refers to the "wired electrical connection" achieved by connecting multiple electronic devices / devices / components to each other through conductive materials, and the "radio connection" for single / bidirectional wireless signal transmission through wireless communication technology. In addition, if not specifically defined, the "electrical connection" described in this patent specification also generally refers to the "direct electrical connection" formed by directly connecting multiple electronic devices / devices / components to each other, and the "indirect electrical connection" formed by indirectly connecting multiple electronic devices / devices / components to each other through other electronic devices / devices / components.

[0028] See also Figure 1The present invention provides a first embodiment of a vehicle distance estimation system 1. In this embodiment, the vehicle distance estimation system 1 is suitable for being arranged in a vehicle, and for the convenience of description, the vehicle distance estimation system 1 is arranged in the vehicle as shown in this embodiment. Figure 1 A first vehicle 51.

[0029] In the present embodiment, the vehicle distance estimation system 1 may be equivalent to a driving assistance system with image recording and satellite positioning functions, and the vehicle distance estimation system 1 may be manufactured and sold separately, and may be installed on the first vehicle 51 after the first vehicle 51 leaves the factory. More specifically, for example, the vehicle distance estimation system 1 may be implemented as a digital video recorder (DVR) with satellite positioning function, that is, a driving recorder. However, in other embodiments, the vehicle distance estimation system 1 is not limited to being implemented as a digital video recorder, and may be built into the first vehicle 51 before the first vehicle 51 leaves the factory, so the specific implementation of the vehicle distance estimation system 1 is not limited to the present embodiment.

[0030] More specifically, in this embodiment, the vehicle distance estimation system 1 includes, for example, a storage unit 11, a photographing unit 12, a positioning unit 13, an input / output unit 14, and a processing unit 15 electrically connected to the storage unit 11, the photographing unit 12, the positioning unit 13 and the input / output unit 14.

[0031] The storage unit 11 in this embodiment may be implemented as a memory module for storing digital data, and may be implemented by a flash memory or a hard disk, but is not limited thereto.

[0032] In this embodiment, the photographing unit 12 may be implemented as a photographing lens module including a lens group and a photosensitive component, and may be configured to photograph the front side of the first vehicle 51, but the present invention is not limited thereto.

[0033] The positioning unit 13 is implemented as a satellite positioning module based on satellite positioning technology, for example. More specifically, the positioning unit 13 can receive satellite signals and immediately address the current location of the positioning unit 13 itself. More specifically, in the present embodiment, the satellite signal can be, for example, from the Global Positioning System (GPS), that is, the positioning unit 13 is, for example, a GPS satellite positioning module in the present embodiment. However, in other embodiments, the satellite signal can also be, for example, from other systems that use satellite signals to achieve immediate addressing functions, such as the Beidou Satellite Navigation System (BDS), the Galileo Positioning System (GALILEO) and the GLONASS System (GLONASS), etc. Therefore, the specific implementation of the positioning unit 13 is not limited to the present embodiment.

[0034] The input / output unit 14 may include, for example, one or more of a display screen, a button assembly, a prompt light assembly, and a speaker, and the display screen may be implemented as a touch display screen, but is not limited thereto.

[0035] The processing unit 15 can be implemented as a central processing unit, or a control circuit board having a central processing unit, but is not limited thereto.

[0036] In this embodiment, the storage unit 11 stores, for example, a specific location database and an image ranging model.

[0037] The specific location database includes a plurality of location feature data and a plurality of geographical location coordinates respectively corresponding to the location feature data.

[0038] In the present embodiment, each location feature data indicates a specific location corresponding to the location feature data itself, and each specific location can be, for example, a lane under a specific traffic light (such as a traffic light), a lane in front of a specific store, a specific intersection, a lane next to a specific road sign, a specific ramp, a lane under a specific sign, or an entrance to a specific parking lot, etc., but is not limited thereto. In more detail, each location feature data can, for example, include one or more images corresponding to the specific location, or include one or more environmental features extracted from the image corresponding to the specific location. Specifically, each location feature data is used by the processing unit 15 as an environmental template of the corresponding specific location, and can be used by the processing unit 15 to compare with the image captured by the shooting unit 12, so that the processing unit 15 can determine whether the environment presented in the image is the specific location indicated by the location feature data. It should be understood that each location feature data can be implemented as long as the processing unit 15 can identify the specific location corresponding to the location feature data from the image. Therefore, the implementation of each location feature data is not limited to this embodiment.

[0039] On the other hand, the geographic location coordinates corresponding to each location feature data may be implemented as a latitude and longitude coordinate, and each geographic location coordinate may be generated through satellite positioning technology (such as GPS satellite positioning), or may be calculated based on multiple known satellite positioning coordinates, but is not limited to this.

[0040] In the present embodiment, the image ranging model includes, for example, an image ranging algorithm, and a correction weight parameter corresponding to the image ranging algorithm and indicating a value, and by running the image ranging algorithm of the image ranging model, the processing unit 15 can perform image analysis on the image captured by the shooting unit 12 to identify the object (e.g., another vehicle) presented in the image, and estimate the distance between the object and the shooting unit 12 according to the pixel size of the object presented in the image and the focal length of the shooting unit 12. The pixel size includes, for example, the maximum pixel width and the maximum pixel height of the object in the image, but is not limited thereto. Since the details of the image ranging algorithm are not the technical focus of this patent specification, they are not described in detail here.

[0041] See also Figure 1 and Figure 2 The photographing unit 12 is adapted to continuously record the video under the control of the processing unit 15 during the driving process of the first vehicle 51, thereby generating a dynamic image M, and Figure 21 shows one frame of the dynamic image M. In other words, the dynamic image M is a real-time video produced by the camera unit 12. In addition, in order to facilitate the description of the operation of the vehicle distance estimation system 1 of this embodiment, it is assumed that during the driving process of the first vehicle 51, there is a second vehicle 52 (shown as an example in FIG. Figure 2 ) is traveling in front of the first vehicle 51 and is located within the shooting range of the shooting unit 12. Therefore, the dynamic image M is, for example, Figure 2 As shown, the second vehicle 52 located in front of the first vehicle 51 is presented. In addition, the dynamic image M may also present a portion of the front of the first vehicle 51, but is not limited thereto.

[0042] When the dynamic image M shows the second vehicle 52, the vehicle distance estimation system 1 may repeatedly implement a vehicle distance estimation method, and further refer to Figure 3 The following is an illustrative description of how the vehicle distance estimation system 1 of this embodiment implements the vehicle distance estimation method.

[0043] First, in step S11, on the one hand, the processing unit 15 performs image comparison between the dynamic image M and at least a portion of the location feature data to determine whether a location matching condition is met, and the location matching condition is related to whether the current location of the second vehicle 52 is a specific location corresponding to one of the location feature data. On the other hand, the processing unit 15 also determines whether a satellite signal quality condition is met based on the satellite signal received by the positioning unit 13, and the satellite signal quality condition is related to whether the average signal strength of the satellite signal is sufficient.

[0044] In this embodiment, before determining whether the location matching condition is met, the processing unit 15 may, for example, first confirm the current location of the vehicle distance estimation system 1 itself (equivalent to confirming the current location of the first vehicle 51) by means of the positioning function of the positioning unit 13, and select N geographical location coordinates closest to the current location from the geographical location coordinates stored in the specific location database, and use the N location feature data corresponding to the N geographical location coordinates as the N adjacent location feature data in this embodiment. Then, the processing unit 15 compares each adjacent location feature data with the surrounding environment of the second vehicle 52 presented in the dynamic image M in an image recognition manner, and determines whether the location matching condition is met based on the comparison result. Wherein, N may, for example, be implemented as an integer greater than or equal to 1, and by first selecting the N adjacent location feature data, this embodiment can reduce the number of location feature data that the processing unit 15 needs to compare, thereby shortening the time required for the processing unit 15 to perform the comparison.

[0045] More specifically, in this embodiment, the location matching condition, for example, represents that the processing unit 15 determines that the surrounding environment of the second vehicle 52 matches one of the matching location feature data of the adjacent location feature data according to the dynamic image M. In other words, in this embodiment, the location matching condition represents that the processing unit 15 determines that the current location of the second vehicle 52 is a specific location corresponding to one of the location feature data (i.e., the matching location feature data) according to the dynamic image M. And, for the convenience of subsequent description, the specific location corresponding to the matching location feature data is referred to as a matching specific location.

[0046] If Figure 2 For example, the matching specific location may be Figure 2 That is, when the second vehicle 52 drives under the specific traffic sign 6 and is photographed by the shooting unit 12, the processing unit 15 determines that the location matching condition is met according to the dynamic image M.

[0047] On the other hand, in the present embodiment, the satellite signal quality condition, for example, represents that the average signal strength of the satellite signal received by the positioning unit 13 within a period of time (for example, within the last three seconds) is greater than or equal to a signal strength threshold. In more detail, the processing unit 15 can, for example, determine whether the satellite signal quality condition is met based on the historical record of the positioning unit 13 receiving the satellite signal, and the satellite signal quality condition is used for the processing unit 15 to determine whether the strength and stability of the satellite signal currently received by the positioning unit 13 are sufficient. In other words, the satellite signal quality condition is used to evaluate whether the positioning result of the positioning unit 13 in the current environment has sufficient accuracy. It should be noted that since the method of determining whether the strength / stability of the satellite signal is sufficient can be freely changed according to different determination criteria, the specific implementation of the satellite signal quality condition is not limited to the present embodiment.

[0048] In addition, in this embodiment, the processing unit 15 can, for example, perform the determination of the location matching condition and the satellite signal quality condition simultaneously in a multi-task processing manner. However, in other implementations, the processing unit 15 can, for example, first determine whether the satellite signal quality condition is met, and then determine whether the location matching condition is met, or the processing unit 15 can, for example, first determine whether the location matching condition is met, and then determine whether the satellite signal quality condition is met. In other words, the timing of the processing unit 15 determining the location matching condition and the satellite signal quality condition is not limited to this embodiment.

[0049] If the processing unit 15 determines that both the location matching condition and the satellite signal quality condition are met, the process proceeds to step S12 and step S13 at the same time. On the other hand, if the processing unit 15 determines that either the location matching condition or the satellite signal quality condition is not met, the process of this vehicle distance estimation method ends.

[0050] In step S12 following step S11, once the processing unit 15 determines that both the location matching condition and the satellite signal quality condition are met, the processing unit 15 performs image analysis on the dynamic image M by running the image ranging algorithm of the image ranging model to obtain an estimated distance as a preliminary estimated result, and then the processing unit 15 calculates another estimated distance as a weighted estimated result based on the preliminary estimated result and the correction weight parameter. More specifically, in the present embodiment, the processing unit 15 calculates the estimated distance as the weighted estimated result by, for example, multiplying the value of the preliminary estimated result and the value of the correction weight parameter, but the present invention is not limited thereto.

[0051] Specifically, the preliminary estimation result in this embodiment represents the distance between the first vehicle 51 and the second vehicle 52 at the moment when the processing unit 15 determines that the location matching condition is met. In other words, the preliminary estimation result in this embodiment represents the distance between the first vehicle 51 and the second vehicle 52 at the moment when the second vehicle 52 is located at the matching specific location (for example, the lane below the specific traffic sign 6).

[0052] It is particularly noted that the preliminary estimation result is generated by the processing unit 15 only by running the image ranging algorithm, that is, the preliminary estimation result is generated only based on image analysis. Therefore, compared with the actual distance between the first vehicle 51 and the second vehicle 52, the preliminary estimation result may have errors, and the errors may include errors caused by the angle and position of the camera unit 12 on the first vehicle 51, and errors of the image ranging algorithm of the image ranging model itself.

[0053] Based on the possibility of errors in the preliminary estimation result, the correction weight parameter in the image ranging model is used by the processing unit 15 to correct the preliminary estimation result to obtain the weighted estimation result. In other words, the weighted estimation result also represents the interval distance between the first vehicle 51 and the second vehicle 52 when the second vehicle 52 is located at the matching specific location, and the difference between the weighted estimation result and the preliminary estimation result is that the weighted estimation result is the result of the processing unit 15 weightedly correcting the preliminary estimation result with the correction weight parameter, and is not generated only based on image analysis.

[0054] However, it is further explained that if the vehicle distance estimation system 1 implements the vehicle distance estimation method for the first time, the value of the correction weight parameter may only be a default value (e.g., "1") that has no substantial correction effect. On the other hand, if the vehicle distance estimation system 1 does not implement the vehicle distance estimation method for the first time, then during the implementation of the vehicle distance estimation method this time, the value of the correction weight parameter will be the value determined by the vehicle distance estimation system 1 during the previous implementation of the vehicle distance estimation method. It is worth noting that the error of the preliminary estimation result may vary with the position / angle of the shooting unit 12 and the current shooting environment. Therefore, the correction weight parameter determined by the processing unit 15 during the previous implementation of the vehicle distance estimation method may not be suitable for correcting the preliminary estimation result generated during the current implementation of the vehicle distance estimation method. Therefore, the value of the correction weight parameter will also need to be dynamically adjusted as the error of the preliminary estimation result changes, so as to ensure that the correction weight parameter has a substantial correction effect. In other words, although the value indicated by the weighted estimation result has been corrected by the processing unit 15 with the correction weight parameter, the correction weight parameter itself is not necessarily suitable for correcting the preliminary estimation result generated in this vehicle distance estimation method. Therefore, the weighted estimation result itself may still have errors.

[0055] In step S13 following step S11, once the processing unit 15 determines that both the location matching condition and the satellite signal quality condition are met, the processing unit 15 controls the positioning unit 13 to perform positioning in real time, so as to obtain a positioning coordinate generated by the positioning unit 13 from the positioning unit 13, and the positioning coordinate is used by the processing unit 15 as a first geographical location coordinate corresponding to the first vehicle 51 in this embodiment. It is particularly noted that in this embodiment, the first geographical location coordinate is obtained by, for example, controlling the positioning unit 13 to perform positioning in real time when the processing unit 15 determines that both the satellite signal quality condition and the location matching condition are met. Therefore, in this embodiment, the first geographical location coordinate is equivalent to the result of the immediate addressing performed by the positioning unit 13 when the second vehicle 52 is located at the matching specific location. In other words, the first geographical location coordinate is equivalent to the location of the first vehicle 51 at the time when the second vehicle 52 is located at the matching specific location in this embodiment. In addition, since the first geographical location coordinate is generated when the processing unit 15 determines that the satellite signal quality condition is met, this embodiment can ensure that the first geographical location coordinate has sufficient accuracy.

[0056] In step S14 following step S13, the processing unit 15 calculates a standard reference distance according to the first geographical location coordinate and the geographical location coordinate corresponding to the matching location feature data. More specifically, the geographical location coordinate corresponding to the matching location feature data is used by the processing unit 15 as the second geographical location coordinate corresponding to the second vehicle 52 in this embodiment, and the standard reference distance is, for example, the straight-line distance between the first geographical location coordinate and the second geographical location coordinate, but is not limited thereto.

[0057] In step S15 following step S12 and step S14, the processing unit 15 determines whether the difference between the standard reference distance and the weighted estimation result is within a preset allowable range. If the result of the processing unit 15 is yes, the processing unit 15 maintains the current value of the correction weight parameter, and the process of this vehicle distance estimation method ends. On the other hand, if the result of the processing unit 15 is no, the process proceeds to step S16. Specifically, for example, the preset allowable range can be implemented as "90% to 110%", and the processing unit 15 can, for example, divide the weighted estimation result by the standard reference distance and convert it into a percentage, thereby calculating a magnification percentage (for example, 125%) of the weighted estimation result relative to the standard reference distance, and then determine whether the magnification percentage is within the preset allowable range. In other words, in this embodiment, the processing unit 15 can, for example, use the magnification percentage as the difference between the standard reference distance and the weighted estimation result. However, in other embodiments, the degree of difference between the standard reference distance and the weighted estimation result may also be achieved in different ways, and the preset allowable range may also be implemented as different values ​​or forms, and is not limited to this embodiment.

[0058] In step S16 following step S15, once the processing unit 15 determines that the degree of difference between the standard reference distance and the weighted estimation result is not within the preset allowable range, the processing unit 15, for example, first determines a correction value based on the degree of difference between the standard reference distance and the weighted estimation result, and then updates (adjusts) the value of the correction weight parameter based on the correction value, and the processing unit 15, for example, also calculates another estimated distance used as a correction estimation result based on the updated (adjusted) correction weight parameter.

[0059] More specifically, in the present embodiment, the processing unit 15 updates the correction weight parameter by, for example, multiplying the value of the correction weight parameter before updating by the correction value, and using the result of the multiplication as the value of the correction weight parameter after updating (that is, replacing the value of the correction weight parameter before updating with the result of the multiplication), but the present invention is not limited thereto. For example, assuming that the value of the correction weight parameter before updating is X (X is a real number), and assuming that the value of the weighted estimation result is 1.25 times (that is, 125%) of the standard reference distance, the correction value determined by the processing unit 15 may be, for example, the reciprocal of 1.25, that is, 0.8. Therefore, in this example, the processing unit 15 may, for example, update the value of the correction weight parameter from the original "X" to "0.8X", but the present invention is not limited thereto.

[0060] Furthermore, in the present embodiment, the processing unit 15 calculates the corrected estimation result by, for example, multiplying the preliminary estimation result by the updated correction weight parameter to calculate the corrected estimation result. However, in another embodiment, the processing unit 15 may also, for example, calculate the corrected estimation result by re-correcting the weighted estimation result according to the updated correction weight. Continuing with the previous example, since the value of the correction weight parameter has been updated from the original "X" to "0.8X", the processing unit 15 calculates the corrected estimation result by, for example, multiplying the value of the weighted estimation result by 0.8, and is not limited to the present embodiment.

[0061] It is additionally noted that the correction estimation result may be used, for example, for the processing unit 15 to compare it with a collision warning threshold value, and decide whether to output a front vehicle collision warning for reminding the driver through the input / output unit 14 according to the comparison result. Moreover, in other embodiments, the correction estimation result may also be used, for example, as input data for triggering an automatic driving assistance mechanism (such as an automatic deceleration mechanism), or as recorded data for evaluating the driving habits of the driver, and is not limited to being used for the processing unit 15 to decide whether to output the front vehicle collision warning. Therefore, the application of the correction estimation result is not limited to this embodiment.

[0062] The above is an example description of the vehicle distance estimation method of this embodiment.

[0063] By implementing the vehicle distance estimation method, the vehicle distance estimation system 1 of the present embodiment can use the first geographical location coordinates and the second geographical location coordinates generated based on satellite positioning technology to calculate the standard reference distance that is more accurate than the weighted estimation result, and update the correction weight parameter according to the degree of difference between the standard reference distance and the weighted estimation result. In this way, the updated correction weight parameter can not only enable the vehicle distance estimation system 1 to calculate the correction estimation result, but also has the opportunity to make the weighted estimation result generated by the vehicle distance estimation system 1 next time more accurate without the need to update the correction weight parameter again.

[0064] The above is an exemplary description of the first embodiment of the vehicle distance estimation system 1 of the present invention.

[0065] The present invention further provides a second embodiment of the vehicle distance estimation system 1 , and the second embodiment of the vehicle distance estimation system 1 will be described below.

[0066] In the second embodiment, the vehicle distance estimation system 1 also includes the storage unit 11, the photographing unit 12, the positioning unit 13, the input / output unit 14 and the processing unit 15, and the implementation of the storage unit 11, the photographing unit 12, the positioning unit 13, the input / output unit 14 and the processing unit 15 can be the same as the first embodiment, so they are not repeated here. However, different from the first embodiment, in the second embodiment, the storage unit 11 does not need to store Figure 1 In other words, the vehicle distance estimation system 1 in the second embodiment does not need to store the location feature data and the geographic location coordinates.

[0067] In the second embodiment, the camera unit 12 continuously records the first vehicle 51 during its travel as described in the first embodiment, and generates the dynamic image M (shown in FIG. 1 ) as described in the first embodiment. Figure 2 ). In order to facilitate the description of the operation of the vehicle distance estimation system 1 of the second embodiment, it is still assumed that the dynamic image M is, for example, Figure 2 The diagram shows a partial front portion of the first vehicle 51 and the second vehicle 52 located in front of the first vehicle 51 .

[0068] When the dynamic image M shows the second vehicle 52, the vehicle distance estimation system 1 of the second embodiment also repeatedly implements a vehicle distance estimation method. However, the vehicle distance estimation method implemented by the vehicle distance estimation system 1 in the second embodiment is different from that in the first embodiment. Figure 1 , Figure 2 Also refer to Figure 4The following uses the dynamic image M as an example to illustrate how the vehicle distance estimation system 1 of the second embodiment implements another vehicle distance estimation method different from the first embodiment.

[0069] First, in step S21, the processing unit 15 determines whether the satellite signal quality condition is met according to the satellite signal received by the positioning unit 13. If the result of the processing unit 15 is yes, the process proceeds to step S22. On the other hand, if the result of the processing unit 15 is no, the process of this vehicle distance estimation method ends. It is additionally noted that the way in which the processing unit 15 determines whether the satellite signal quality condition is met is the same as that of the first embodiment, so it will not be repeated here.

[0070] In step S22 following step S21, when the processing unit 15 determines that the satellite signal quality condition is met, the processing unit 15 uses the location of the second vehicle 52 in the environment presented by the dynamic image M as a target location. Figure 2 For example, the target location may be Figure 2 The lane below the specific traffic sign 6 in the figure, but is not limited to this.

[0071] Furthermore, while the processing unit 15 uses the location of the second vehicle 52 as the target location, the processing unit 15 also performs image analysis on the dynamic image M by running the image ranging algorithm of the image ranging model to obtain an estimated distance as a preliminary estimated result. Then, the processing unit 15 calculates another estimated distance as a weighted estimated result based on the preliminary estimated result and the correction weight parameter. In the second embodiment, the preliminary estimated result represents the interval distance between the first vehicle 51 and the second vehicle 52 when the second vehicle 52 is located at the target location, and is generated by the processing unit 15 only by running the image ranging algorithm. Therefore, similar to the preliminary estimated result described in the first embodiment, the preliminary estimated result in the second embodiment may also have errors. On the other hand, the processing unit 15 calculates the weighted estimated result in the same manner as the first embodiment, and, as described in the first embodiment, since the value of the correction weight parameter is not necessarily applicable to correcting the preliminary estimated result generated in the current vehicle distance estimation method, the weighted estimated result itself may still have errors.

[0072] Further, while the processing unit 15 uses the location of the second vehicle 52 as the target location, the processing unit 15 also obtains a first geographical location coordinate corresponding to the first vehicle 51. In addition, in the second embodiment, the processing unit 15 obtains the first geographical location coordinate by, for example, controlling the positioning unit 13 to perform positioning while the target location is set, so as to obtain the first geographical location coordinate from the positioning unit 13. Therefore, in the second embodiment, the first geographical location coordinate is equivalent to the result of the immediate addressing performed by the positioning unit 13 when the second vehicle 52 is located at the target location. In other words, the first geographical location coordinate is equivalent to the location of the first vehicle 51 when the second vehicle 52 is located at the target location in the second embodiment.

[0073] In step S23 following step S22, the processing unit 15 obtains a second geographical location coordinate corresponding to the second vehicle 52. In the second embodiment, the processing unit 15 obtains the second geographical location coordinate by, for example, continuously determining whether the first vehicle 51 has moved to the target location based on the dynamic image M, and controlling the positioning unit 13 to perform positioning in real time when the determination result is yes, so as to obtain the second geographical location coordinate from the positioning unit 13. That is, after the second vehicle 52 leaves the target location, once the first vehicle 51 moves forward to the target location, the processing unit 15 controls the positioning unit 13 to perform positioning in real time to obtain the second geographical location coordinate. Therefore, in the second embodiment, the second geographical location coordinate is equivalent to the result of the immediate addressing performed by the positioning unit 13 when the first vehicle 51 is located at the target location.

[0074] In more detail, in the second embodiment, the processing unit 15 may, for example, determine whether the first vehicle 51 has moved to the target position based on the relative position change between the front of the first vehicle 51 and the target position presented in the dynamic image M. Specifically, for example, the processing unit 15 may, for example, determine whether the first vehicle 51 has moved to the target position when the target position presented by the dynamic image M is blocked by the front of the first vehicle 51. However, it should be understood that in other embodiments, the processing unit 15 may also use other image recognition methods to determine whether the first vehicle 51 has moved to the target position. Therefore, the processing unit 15 is not limited to the above-mentioned example in determining whether the first vehicle 51 has moved to the target position.

[0075] It is additionally noted that in other implementations, the processing unit 15 may also select one of the multiple positioning coordinates included in a positioning record whose positioning time matches the time when the target position is set as the first geographical location coordinate, and select another one of the positioning coordinates whose positioning time matches the time when the first vehicle 51 moves to the target position as the second geographical location coordinate. In other words, the way in which the processing unit 15 obtains the first geographical location coordinate and the second geographical location coordinate is not limited to the second embodiment. In addition, since the first geographical location coordinate and the second geographical location coordinate are both generated when the processing unit 15 determines that the satellite signal quality condition is met, the second embodiment can ensure that the first geographical location coordinate and the second geographical location coordinate have sufficient accuracy.

[0076] In step S24 following step S23, the processing unit 15 calculates a standard reference distance according to the first geographic location coordinate and the second geographic location coordinate. The standard reference distance is, for example, a straight-line distance between the first geographic location coordinate and the second geographic location coordinate, but is not limited thereto.

[0077] In step S25 following step S24, the processing unit 15 determines whether the difference between the standard reference distance and the weighted estimation result is within a preset allowable range, and the manner in which the processing unit 15 determines whether the difference between the standard reference distance and the weighted estimation result is within the preset allowable range is the same as that of the first embodiment, so it is not repeated here. If the result of the judgment by the processing unit 15 is yes, the processing unit 15 maintains the current value of the correction weight parameter, and the process of this vehicle distance estimation method ends. On the other hand, if the result of the judgment by the processing unit 15 is no, the process proceeds to step S26.

[0078] In step S26 following step S25, once the processing unit 15 determines that the difference between the standard reference distance and the weighted estimation result is not within the preset allowable range, the processing unit 15 first determines a correction value according to the difference between the standard reference distance and the weighted estimation result, and then updates the correction weight parameter according to the correction value, and calculates another estimated distance as a correction estimation result according to the updated correction weight parameter. In the second embodiment, the processing unit 15 updates the correction weight parameter and calculates the correction estimation result in the same manner as in the first embodiment, so it is not repeated here.

[0079] The above is an exemplary description of the second embodiment of the vehicle distance estimation system 1 of the present invention.

[0080] It should be noted that although Figure 3 and Figure 4 , respectively, show the flow charts of the vehicle distance estimation method of the first embodiment and the second embodiment. However, it should be understood that Figure 3 and Figure 4 The flowchart in is only an exemplary representation of the vehicle distance estimation method of the present invention, that is, the sequential relationship between step S11 to step S16 and the sequential relationship between step S21 to step S26 are not absolute, nor are they used to limit the scope of implementation of the present invention.

[0081] The present invention also provides a computer program product, which includes an application program, and when an electronic device with hardware functions equal to those of the vehicle distance estimation system 1 loads and executes the application program, the application program enables the electronic device to implement the vehicle distance estimation method described in any of the aforementioned embodiments. In some embodiments, the computer program product can be implemented as, for example, an optical disk, a flash drive, or a memory module storing the application program, that is, the computer program product can be implemented as, for example, a physical computer-readable recording medium. However, in other embodiments, the computer program product can also be implemented as, for example, a set of software products stored in a network space and available for download and installation, and is not limited to being implemented in the form of a physical product.

[0082] In summary, by implementing the vehicle distance estimation method, the vehicle distance estimation system 1 can calculate the standard reference distance using the first geographical location coordinates and the second geographical location coordinates generated based on the satellite positioning technology, and update the correction weight parameter according to the size relationship between the standard reference distance and the weighted estimation result, and then calculate the correction estimation result according to the updated correction weight parameter. In this way, the vehicle distance estimation system 1 can effectively correct the error of the preliminary estimation result (including the error caused by the setting method of the shooting unit and / or the error of the image ranging algorithm itself), so the purpose of the present invention can be achieved.

[0083] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A vehicle distance estimation method, implemented by a vehicle distance estimation system, wherein the vehicle distance estimation system is suitable for being arranged on a first vehicle, characterized in that: The vehicle distance estimation method comprises: the vehicle distance estimation system stores a plurality of location feature data each indicating a specific location, and a plurality of geographical location coordinates respectively corresponding to the location feature data; The vehicle distance estimation system is to perform image analysis on a dynamic image at least according to a correction weight parameter to generate an estimated distance when determining that a location matching condition is met, wherein the estimated distance indicates the interval distance between the first vehicle and a second vehicle presented by the dynamic image, and the vehicle distance estimation system also calculates a standard reference distance based on a first geographical location coordinate related to the first vehicle and a geographical location coordinate corresponding to a matching location feature data related to the second vehicle as a second geographical location coordinate, and updates the correction weight parameter according to the size relationship between the standard reference distance and the estimated distance, and generates another estimated distance according to the updated correction weight parameter.

2. The vehicle distance estimation method according to claim 1, characterized in that: The location matching condition represents that the vehicle distance estimation system determines, based on the dynamic image, that the surrounding environment of the second vehicle matches the matching location feature data among the location feature data.

3. The vehicle distance estimation method according to claim 2, characterized in that: The first geographical location coordinate is generated by positioning using satellite positioning technology when the vehicle distance estimation system determines that both a satellite signal quality condition related to satellite signal strength and the location matching condition are met.

4. The vehicle distance estimation method according to claim 1, characterized in that: The location of the second vehicle in the environment presented by the dynamic image is taken as a target location, and the estimated distance indicates the interval distance between the first vehicle and the second vehicle located at the target location; The first geographical location coordinates are generated by the vehicle distance estimation system using satellite positioning technology, and the first geographical location coordinates represent the location of the first vehicle when the second vehicle is located at the target location; and The second geographical location coordinates are generated by the vehicle distance estimation system using satellite positioning technology to locate the first vehicle when the first vehicle is determined to move to the target location based on the dynamic image.

5. The vehicle distance estimation method according to claim 4, characterized in that: The first geographical location coordinate and the second geographical location coordinate are generated by positioning using satellite positioning technology when the vehicle distance estimation system determines that a satellite signal quality condition related to the satellite signal strength is met.

6. The vehicle distance estimation method according to claim 1, characterized in that: After calculating the standard reference distance, the vehicle distance estimation system first determines whether the difference between the standard reference distance and the estimated distance is within a preset allowable range, and only if the judgment result is no, updates the correction weight parameter according to the size relationship between the standard reference distance and the estimated distance.

7. A vehicle distance estimation system, suitable for being arranged on a first vehicle, characterized in that: The vehicle distance estimation system is used to implement the vehicle distance estimation method according to any one of claims 1 to 6.

8. A computer program product, comprising an application program, characterized in that: When the application is loaded and executed by an electronic device, the electronic device can implement the vehicle distance estimation method according to any one of claims 1 to 6.

Images