Traffic marking fusion method, device, storage medium and electronic device
By obtaining local data from the target marking set and generating transform marking sets based on the transformation relationship, determining the translation parameters to fusion marking sets, the large amount of calculation caused by misalignment of traffic marking data is solved, and efficient marking data fusion is achieved .
Patent Information
- Application Number
- CN202110369281.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-04-06
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2041-04-06
AI Technical Summary
In crowdsourcing fusion scenarios with high precision data, there may be misalignment between frames of traffic marking data, resulting in large calculations and low adjustment efficiency when merging linear data, especially when processing crowdsourcing lane line data, it is more significant .
By obtaining the local parts that overlap with the set of converged markers to be fused from the target markers set, and generating the set of transform markers based on the target transformation relationship, determining the translation parameters to fusion markers, and using the reverse parameter to move the set of converged markers to be fused , simplify the calculation process.
It improves the efficiency and accuracy of traffic markings, reduces the calculation amount, and improves the processing efficiency.
Smart Images

Figure CN112906665B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of computer technology, and in particular to a traffic marking fusion method, device, storage medium and electronic device. Background Art
[0002] In the crowdsourcing fusion scenario of high-precision data, when collecting traffic marking data (for example, collecting linear data through environmental perception technologies such as images and lasers), since the data is collected frame by frame, the data collected between frames may not be completely aligned, resulting in misalignment after the linear data of different frames are merged. At present, linear fusion is generally used to fuse the misaligned linear data. In the linear fusion method, it is necessary to calculate the same-name point pairs and use the adjustment method, which has the problem of large amount of calculation. Especially when the processing object is crowdsourced lane line data, the amount of calculation will increase significantly and the adjustment efficiency is extremely low. Summary of the invention
[0003] The purpose of the present disclosure is to provide a traffic marking fusion method, device, storage medium and electronic device to improve the fusion efficiency of traffic markings.
[0004] In order to achieve the above object, according to a first aspect of the present disclosure, a traffic marking fusion method is provided, the method comprising:
[0005] Acquire a first portion from the target mark line set that overlaps with the mark line set to be fused, and acquire a second portion from the mark line set to be fused that overlaps with the target mark line set;
[0006] According to the target transformation relationship, generating a first transformed mark line set obtained by transforming the first part and a second transformed mark line set obtained by transforming the second part;
[0007] determining first translation parameters capable of blending the second transformed reticle set with the first transformed reticle set;
[0008] According to the target transformation relationship, inversely derive the second translation parameter using the first translation parameter;
[0009] The set of marking lines to be fused is moved using the second translation parameter.
[0010] Optionally, the method further comprises:
[0011] For a given target line marking set, a line marking set to be fused corresponding to the target line marking set is determined, wherein the target line marking set and the line marking set to be fused are both taken from line marking sets in road data, and each line marking set includes a group of traffic line markings.
[0012] Optionally, determining a to-be-fused reticle set corresponding to the target reticle set includes:
[0013] According to the area occupied by the target marking set, a marking set whose area in the road data overlaps with the target marking set is determined as the marking set to be fused.
[0014] Optionally, before the step of acquiring a first portion that coincides with the to-be-fused reticle set from the target reticle set and acquiring a second portion that coincides with the target reticle set from the to-be-fused reticle set, the method further comprises:
[0015] Acquire a first number of traffic markings in the target marking set and a second number of traffic markings in the to-be-fused marking set;
[0016] If the first number is inconsistent with the second number, determining the smaller one of the first number and the second number as the target number;
[0017] The traffic markings in the target marking set or the marking set to be merged are deleted so that the number of traffic markings in the target marking set and the marking set to be merged is the target number.
[0018] Optionally, deleting the traffic markings in the target marking set or the marking set to be merged so that the number of traffic markings in the target marking set and the marking set to be merged is the target number, includes:
[0019] Determine a first distance between a traffic marking at a first side edge of the target marking set and a traffic marking at a first side edge of the marking set to be merged;
[0020] Determine a second distance between the traffic markings at the second side edge of the target marking set and the traffic markings at the second side edge of the marking set to be merged;
[0021] taking the side corresponding to the larger one of the first distance and the second distance as the target side;
[0022] On the target side of the target marking set or the marking set to be fused, traffic markings are deleted.
[0023] Optionally, deleting the traffic markings on the target side of the target marking set or the marking set to be fused includes:
[0024] If the first number is greater than the second number, deleting a specified number of traffic markings from the target side of the target marking set;
[0025] If the second number is greater than the first number, deleting a specified number of traffic markings from the target side of the set of markings to be fused;
[0026] The specified number is the difference between the first number and the second number.
[0027] Optionally, the target transformation relationship can make the second part perpendicular to a specified coordinate axis after transformation.
[0028] Optionally, the target transformation relationship is obtained through affine transformation.
[0029] Optionally, the determining a first translation parameter capable of merging the second transformed reticle set with the first transformed reticle set comprises:
[0030] Determining a first bounding rectangle corresponding to the first transformed mark line set;
[0031] The second transformed mark set is translated multiple times, the intersection area between the second bounding rectangle corresponding to the second transformed mark set and the first bounding rectangle in each translation is recorded, and the translation parameter corresponding to the translation with the largest intersection area is used as the first translation parameter.
[0032] Optionally, the step of translating the second transformed mark set multiple times, recording the intersection area between the second bounding rectangle corresponding to the second transformed mark set and the first bounding rectangle in each translation, and using the translation parameter corresponding to a translation with the largest intersection area as the first translation parameter, includes:
[0033] Recording the initial position of the second transformed reticle set before translation, and determining the intersection area of the second bounding rectangle of the second transformed reticle set and the first bounding rectangle at the initial position as the initial intersection area;
[0034] translating the second transformed mark line set in a specified direction by a specified distance so that the second transformed mark line set reaches a target position of this translation;
[0035] Determine the intersection area of the second bounding rectangle of the second transformed marking line set and the first bounding rectangle at the target position of the current translation as the target intersection area;
[0036] comparing the target intersection area to the reference intersection area;
[0037] If the target intersection area is smaller than the reference intersection area, stopping translation of the second transformed reticle set in the specified direction;
[0038] After stopping the translation operation on the second transformed reticle set, a translation parameter of the middle position of the second transformed reticle set compared to the initial position when the intersection area with the first bounding rectangle is the largest during the historical translation process is used as the first translation parameter.
[0039] Optionally, the method further comprises:
[0040] If the target intersection area is greater than the reference intersection area, return to the step of translating the second transformed reticle set by a specified distance in a specified direction so that the second transformed reticle set reaches the target position of this translation.
[0041] Optionally, the designated direction is a direction parallel to the designated coordinate axis, and the designated direction includes a first direction and a second direction that are opposite to each other.
[0042] Optionally, the traffic marking is a lane marking.
[0043] According to a second aspect of the present disclosure, a traffic marking fusion device is provided, the device comprising:
[0044] A first acquisition module is used to acquire a first part that overlaps with the target mark line set from the target mark line set, and acquire a second part that overlaps with the target mark line set from the target mark line set;
[0045] A generating module, configured to generate, according to a target transformation relationship, a first transformed mark line set obtained by transforming the first part and a second transformed mark line set obtained by transforming the second part;
[0046] A first determination module, configured to determine a first translation parameter capable of merging the second transformed reticle set with the first transformed reticle set;
[0047] A derivation module, configured to inversely derive a second translation parameter using the first translation parameter according to the target transformation relationship;
[0048] A moving module is used to move the set of marking lines to be fused using the second translation parameter.
[0049] Optionally, the device further comprises:
[0050] The second determination module is used to determine, for a given target marking set, a marking set to be fused corresponding to the target marking set, wherein the target marking set and the marking set to be fused are both taken from marking sets in road data, and each marking set includes a group of traffic markings.
[0051] Optionally, the second determining module is used to determine, based on the area occupied by the target marking set, a marking set in the road data whose area overlaps with the target marking set as the marking set to be fused.
[0052] Optionally, the device further comprises:
[0053] A second acquisition module, configured to acquire a first number of traffic markings in the target marking set and a second number of traffic markings in the marking set to be merged before the first acquisition module acquires a first portion overlapping with the marking set to be merged from the target marking set and acquires a second portion overlapping with the target marking set from the marking set to be merged;
[0054] a third determining module, configured to determine the smaller one of the first number and the second number as a target number if the first number is inconsistent with the second number;
[0055] The deleting module is used to delete the traffic markings in the target marking set or the marking set to be merged, so that the number of traffic markings in the target marking set and the marking set to be merged is the target number.
[0056] Optionally, the deletion module includes:
[0057] A first determining submodule, configured to determine a first distance between a traffic marking at a first side edge of the target marking set and a traffic marking at a first side edge of the marking set to be fused;
[0058] A second determining submodule, configured to determine a second distance between a traffic marking at a second side edge of the target marking set and a traffic marking at a second side edge of the marking set to be fused;
[0059] A third determination submodule, configured to take a side corresponding to a larger one of the first distance and the second distance as a target side;
[0060] The deleting submodule is used to delete the traffic markings on the target side of the target marking set or the marking set to be fused.
[0061] Optionally, the deletion submodule is used to:
[0062] If the first number is greater than the second number, deleting a specified number of traffic markings from the target side of the target marking set;
[0063] If the second number is greater than the first number, deleting a specified number of traffic markings from the target side of the set of markings to be fused;
[0064] The specified number is the difference between the first number and the second number.
[0065] Optionally, the target transformation relationship can make the second part perpendicular to a specified coordinate axis after transformation.
[0066] Optionally, the target transformation relationship is obtained through affine transformation.
[0067] Optionally, the first determining module includes:
[0068] A fourth determination submodule, configured to determine a first bounding rectangle corresponding to the first transformed marking line set;
[0069] The fifth determination submodule is used to translate the second transformed marking set multiple times, record the intersection area between the second outer bounding rectangle corresponding to the second transformed marking set and the first outer bounding rectangle in each movement, and use the translation parameter corresponding to the movement with the largest intersection area as the first translation parameter.
[0070] Optionally, the fifth determining submodule is used to:
[0071] Recording the initial position of the second transformed reticle set before translation, and determining the intersection area of the second bounding rectangle of the second transformed reticle set and the first bounding rectangle at the initial position as the initial intersection area;
[0072] translating the second transformed mark line set in a specified direction by a specified distance so that the second transformed mark line set reaches a target position of this translation;
[0073] Determine the intersection area of the second bounding rectangle of the second transformed marking line set and the first bounding rectangle at the target position of the current translation as the target intersection area;
[0074] comparing the target intersection area to the reference intersection area;
[0075] If the target intersection area is smaller than the reference intersection area, stopping translation of the second transformed reticle set in the specified direction;
[0076] After stopping the translation operation on the second transformed reticle set, a translation parameter of the middle position of the second transformed reticle set compared to the initial position when the intersection area with the first bounding rectangle is the largest during the historical translation process is used as the first translation parameter.
[0077] Optionally, the fifth determining submodule is further used for:
[0078] If the target intersection area is greater than the reference intersection area, return to the step of translating the second transformed reticle set by a specified distance in a specified direction so that the second transformed reticle set reaches the target position of this translation.
[0079] Optionally, the designated direction is a direction parallel to the designated coordinate axis, and the designated direction includes a first direction and a second direction that are opposite to each other.
[0080] Optionally, the traffic marking is a lane marking.
[0081] According to a third aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps of the method described in the first aspect of the present disclosure are implemented.
[0082] According to a fourth aspect of the present disclosure, there is provided an electronic device, including:
[0083] a memory having a computer program stored thereon;
[0084] A processor is used to execute the computer program in the memory to implement the steps of the method described in the first aspect of the present disclosure.
[0085] Through the above technical solution, the first part that overlaps with the line set to be fused is obtained from the target line set, and the second part that overlaps with the target line set is obtained from the line set to be fused. According to the target transformation relationship, a first transformed line set obtained by transforming the first part and a second transformed line set obtained by transforming the second part are generated, and a first translation parameter that enables the second transformed line set to be fused with the first transformed line set is determined. According to the target transformation relationship, the second translation parameter is inversely deduced using the first translation parameter, and the line set to be fused is moved using the second translation parameter. In this way, in the process of fusing traffic markings, only a simple calculation method is needed, which is conducive to improving the fusion efficiency and accuracy.
[0086] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0087] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:
[0088] Figure 1 is a flow chart of a traffic marking fusion method provided according to an embodiment of the present disclosure;
[0089] Figure 2 is a flow chart of a traffic marking fusion method provided according to another embodiment of the present disclosure;
[0090] Figure 3A is an exemplary schematic diagram of a target marking set and a marking set to be fused in the traffic marking fusion method provided by the present disclosure;
[0091] Figure 3B An exemplary schematic diagram of a target marking set and a marking set to be fused in the traffic marking fusion method provided by the present disclosure;
[0092] FIG. 3C to FIG. 3FAn exemplary schematic diagram of a process of determining a first translation parameter in the traffic marking fusion method provided by the present disclosure;
[0093] Figure 4 is a block diagram of a traffic marking fusion device provided according to an embodiment of the present disclosure;
[0094] Figure 5 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION
[0095] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.
[0096] Traffic markings, that is, road traffic markings, refer to the markings on the road surface that use lines, arrows, text, elevation marks, protruding road signs and contour marks to convey traffic information such as guidance, restrictions, and warnings to traffic participants. Although the traffic markings are stationary, as described in the background technology, in the process of collecting road data, the existing data collection methods (for example, image collection, laser collection) are used for frame-by-frame collection. Due to the difference in viewing angles between frames, the collection of the same group of traffic markings is biased, that is, the same group of traffic markings will be misaligned at the connection. Therefore, in view of this situation, it is necessary to fuse the traffic markings that appear in this situation. At present, the commonly used fusion method in the relevant technology is to perform geometric adjustment based on the principle of adjustment. During the adjustment process, the multi-source data is divided into "adjustment map" and "reference map". Then, the coordinate displacement equation is established according to the corresponding matching point pairs in the "adjustment map" and "reference map". In order to calculate the geometric adjustment amount, it is usually necessary to design the coordinate displacement equation and the shape equation, and then use the constrained least squares method to solve the coordinate value. However, this method has the problem of large amount of calculation and low adjustment efficiency.
[0097] In order to solve the above problems, the present disclosure provides a traffic marking fusion method, device, storage medium and electronic device to improve the fusion efficiency of traffic markings.
[0098] Figure 1 FIG. 1 is a flow chart of a traffic marking fusion method provided according to an embodiment of the present disclosure. Figure 1 As shown, the method may include the following steps 11 to 15:
[0099] In step 11, a first portion that coincides with the target mark line set is obtained from the target mark line set, and a second portion that coincides with the target mark line set is obtained from the target mark line set;
[0100] In step 12, a first transformed mark set obtained by transforming the first part and a second transformed mark set obtained by transforming the second part are generated according to the target transformation relationship;
[0101] In step 13, a first translation parameter is determined that enables the second transformed reticle set to be merged with the first transformed reticle set;
[0102] In step 14, according to the target transformation relationship, the second translation parameter is inversely deduced using the first translation parameter;
[0103] In step 15, the set of reticle lines to be fused is moved using a second translation parameter.
[0104] The method provided by the present disclosure is used to achieve the fusion of traffic markings in the target marking set and the marking set to be fused. Therefore, before executing step 11, it is necessary to determine the marking set to be fused corresponding to the target marking set for a given target marking set.
[0105] Among them, the target marking set and the marking set to be fused are both taken from the marking set in the road data, and each marking set contains a set of traffic markings. Road data is the data about the road that can be collected during the data collection process, and the road data includes a marking set consisting of a set of traffic markings. Here, a set of traffic markings can be considered as several traffic markings with similar positions, for example, traffic markings on the same lane, or traffic markings on the same road, etc. For example, the traffic markings can be lane markings. For another example, the traffic markings can be pedestrian crossing markings.
[0106] As described in the background technology, the misalignment of the same marking set usually comes from the acquisition of different frames. Therefore, in the present disclosure, it is possible to specify which two frames of road data to be fused. For example, in the image acquisition method, the two frames of images acquired earlier and later can be used as the processing objects of the fusion operation.
[0107] In a possible implementation, the following steps may be used to determine the reticle set to be fused corresponding to the target reticle set:
[0108] According to the area occupied by the target marking set, a marking set whose area in the road data overlaps with the target marking set is determined as the marking set to be fused.
[0109] Generally, it is believed that only traffic markings corresponding to the same road or the same lane will be misaligned, and misalignment will cause the area occupied by the markings to partially overlap. Therefore, this can be used as a basis to identify the marking set to be fused corresponding to the target marking set. That is, according to the location of the target marking set, the marking set in the road data that has an area overlap with the target marking set is determined, and the marking set is determined as the marking set to be fused.
[0110] Optionally, in some cases, there may be a phenomenon that the number of traffic markings in the target marking set is inconsistent with the number of traffic markings in the marking set to be fused. In such a case, if subsequent processing is performed directly, data processing errors may occur due to the inconsistency of traffic markings, and even fusion failure may occur. Therefore, in a possible implementation, before executing step 11, Figure 1 Based on the steps shown, the method provided by the present disclosure may also include the following steps, such as Figure 2 As shown:
[0111] In step 21, a first number of traffic markings in the target marking set and a second number of traffic markings in the marking set to be merged are obtained;
[0112] In step 22, if the first quantity and the second quantity are inconsistent, the smaller one of the first quantity and the second quantity is determined as the target quantity;
[0113] In step 23, traffic markings in the target marking set or the marking set to be merged are deleted so that the number of traffic markings in the target marking set and the marking set to be merged is the target number.
[0114] Generally, if the number of traffic markings in the target marking set is inconsistent with the number of traffic markings in the marking set to be merged, the number of traffic markings in the two marking sets can be made consistent by deleting the one with more or adding the one with less. However, in the processing scenario of this solution, adding traffic markings can easily lead to problems such as slowing down data processing speed and reducing accuracy. Therefore, deleting traffic markings is usually used.
[0115] For example, step 23 may include the following steps:
[0116] Determine a first distance between a traffic marking on a first side edge of the target marking set and a traffic marking on a first side edge of the marking set to be fused;
[0117] Determine a second distance between the traffic markings at the second side edge of the target marking set and the traffic markings at the second side edge of the marking set to be fused;
[0118] The side corresponding to the larger one of the first distance and the second distance is taken as the target side;
[0119] Delete the traffic markings on the target side of the target marking set or the marking set to be fused.
[0120] If the first number is greater than the second number, a specified number of traffic markings are deleted from the target side of the target marking set, and if the second number is greater than the first number, a specified number of traffic markings are deleted from the target side of the marking set to be fused. The specified number is the difference between the first number and the second number.
[0121] by Figure 3A For the set of marking lines shown, the first side and the second side can refer to the left side and the right side respectively (the order is not limited). Figure 3A In the example, the target marking set S1 and the to-be-fused marking set S2 are included, wherein the target marking set S1 contains 3 traffic markings, and the to-be-fused marking set S2 contains 2 traffic markings, and the number of traffic markings of the two is inconsistent. Therefore, according to the above steps, the distance D1 between the traffic marking L1 in the target marking set S1 and the traffic marking L2 in the to-be-fused marking set S2 should be determined, and the distance D2 between the traffic marking R1 in the target marking set S1 and the traffic marking R2 in the to-be-fused marking set S2 should be determined. If D1 is greater than D2, the target side is the left side, and L1 is deleted (the specified number is the difference between 3 and 2, 1). If D1 is less than D2, R1 is deleted. Optionally, if D1 is equal to D2, any side can be selected as the target side for deletion.
[0122] Through the above method, the target marking set and the marking set to be fused are preprocessed so that the number of traffic markings contained in the two is the same, which is conducive to improving the subsequent data processing efficiency and can also improve the processing accuracy.
[0123] Reference Figure 1 In step 11, a first portion that overlaps with the mark line set to be fused is obtained from the target mark line set, and a second portion that overlaps with the target mark line set is obtained from the mark line set to be fused.
[0124] According to the overlapping area between the target mark set and the mark set to be fused, a local mark set corresponding to the intersection can be obtained from the target mark set, i.e., the first local area. Similarly, a local mark set corresponding to the intersection can be obtained from the mark set to be fused, i.e., the second local area. Figure 3B In the figure, there are two groups of marking sets, upper and lower. Assume that the upper group of marking sets is the target marking set and the lower group of marking sets is the marking set to be fused. The rectangular shaded part in the figure is the overlapping area of the target marking set and the marking set to be fused. Based on the shaded area, the part of the target marking set falling in the shaded area is the first part, and the part of the marking set to be fused falling in the shaded area is the second part.
[0125] In step 12, a first transformed mark line set obtained by transforming the first part and a second transformed mark line set obtained by transforming the second part are generated according to the target transformation relationship.
[0126] The target transformation relationship enables the second local part to be perpendicular to the specified coordinate axis after transformation. The specified coordinate axis is the horizontal axis or the vertical axis in the preset plane rectangular coordinate system, and the preset plane rectangular coordinate system is constructed according to the earth coordinate system, for example, the horizontal and vertical axes are constructed with reference to longitude and latitude lines.
[0127] For example, the target transformation relationship can be determined by affine transformation, wherein converting a graphic to a desired position by affine transformation is common knowledge in the art and will not be described in detail here.
[0128] After obtaining the target transformation relationship, a first transformation reticle set may be generated by performing transformation based on the first part according to the target transformation relationship, and a second transformation reticle set may be generated by performing transformation based on the second part according to the target transformation relationship.
[0129] The target transformation relationship can make the second part perpendicular to the specified coordinate axis (the horizontal axis or vertical axis in the preset plane rectangular coordinate system) after transformation. Therefore, after the second part is transformed by the target transformation relationship, it can be perpendicular to the above-mentioned specified coordinate axis. In this way, in the subsequent translation process, only the translation attempt in the direction of the specified coordinate axis is required to obtain the optimal solution that can merge the two marking line sets.
[0130] In step 13, a first translation parameter is determined that enables the second transformed reticle set to be merged with the first transformed reticle set.
[0131] In a possible implementation, step 13 may include the following steps:
[0132] Determine a first bounding rectangle corresponding to the first transformed mark line set;
[0133] The second transformed marking set is translated multiple times, the intersection area between the second outer bounding rectangle and the first outer bounding rectangle corresponding to the second transformed marking set in each translation is recorded, and the translation parameter corresponding to the translation with the largest intersection area is used as the first translation parameter.
[0134] Among them, the first enclosing rectangle and the second enclosing rectangle are both the minimum enclosing rectangles of the graphic. The minimum enclosing rectangle is the minimum circumscribed rectangle that encloses the primitive (short for basic graphic element) and whose two adjacent sides are parallel to the horizontal axis and the vertical axis respectively. Correspondingly, the first enclosing rectangle is the minimum circumscribed rectangle that encloses the first transformed marking line set and whose two adjacent sides are parallel to the horizontal axis and the vertical axis in the preset plane rectangular coordinate system respectively. The second enclosing rectangle is the minimum circumscribed rectangle that encloses the second transformed marking line set and whose two adjacent sides are parallel to the horizontal axis and the vertical axis in the preset plane rectangular coordinate system respectively.
[0135] Optionally, the second transformed marking set is translated multiple times, the intersection area between the second outer bounding rectangle and the first outer bounding rectangle corresponding to the second transformed marking set in each movement is recorded, and the translation parameter corresponding to the movement with the largest intersection area is used as the first translation parameter, which can include the following first to sixth steps.
[0136] In the first step, the initial position of the second transformed reticle set before translation is recorded, and the intersection area of the second bounding rectangle and the first bounding rectangle of the second transformed reticle set at the initial position is determined as the initial intersection area.
[0137] In the second step, the second transformed mark line set is translated along the specified direction by a specified distance so that the second transformed mark line set reaches the target position of this translation.
[0138] In the third step, the intersection area of the second outer bounding rectangle and the first outer bounding rectangle of the second transformed marking line set at the target position of this translation is determined as the target intersection area.
[0139] In a fourth step, the target intersection area is compared with a reference intersection area.
[0140] The reference intersection area is the intersection area between the second outer bounding rectangle and the first outer bounding rectangle of the second transformed marking line set before the current translation during the translation along the specified direction, and in the initial case, the reference intersection area is the initial intersection area.
[0141] If the target intersection area is smaller than the reference intersection area, the fifth step is executed.
[0142] In a fifth step, translation of the second transformed reticle set in a specified direction is stopped.
[0143] In the sixth step, after stopping the translation operation on the second transformed reticle set, the translation parameter of the middle position of the second transformed reticle set compared to the initial position when the intersection area with the first bounding rectangle is the largest during the historical translation process is used as the first translation parameter.
[0144] Also, if after comparing the target intersection area with the reference intersection area through the above-mentioned fourth step, it is obtained that the target intersection area is larger than the reference intersection area, then return to the above-mentioned second step, and execute the second step and its subsequent related steps again, that is, perform another translation and intersection area calculation, and compare it with the new reference intersection area again, and determine whether it is necessary to repeat the translation process again based on the comparison result, until it is determined that the fifth step can be executed.
[0145] The specified direction is a direction parallel to the specified coordinate axis, and the specified direction includes a first direction and a second direction that are opposite to each other.
[0146] That is to say, when the second transformed marking set is moved in an attempt to find the most appropriate first translation parameter, the second transformed marking set can be translated in the first direction by a specified distance based on the initial position of the second transformed marking set, and the target intersection area corresponding to this translation is calculated. If the target intersection area corresponding to this translation is increased compared to the target intersection area of the previous translation, then the second transformed marking set is continued to be moved in the first direction based on the target position reached by this translation. On the contrary, if the target intersection area corresponding to this translation is reduced compared to the target intersection area of the previous translation, the second transformed marking set is stopped from being translated in the first direction. The translation processing method in the second direction is the same as that in the first direction, and will not be repeated here. It should be noted that, whether it is the translation in the first direction or the translation in the second direction, the second transformed marking set should be located at the initial position described in the first step, and the translation operations in the first direction and the second direction are independent of each other and do not affect each other, and there is no strict order limit between the two.
[0147] Until both the first direction and the second direction stop translating, a series of intersection areas in the historical movement process are recorded, and the largest intersection area is selected from them, and the translation parameter corresponding to this intersection area is obtained as the first translation parameter.
[0148] Optionally, in order to ensure the accuracy of the first translation parameter, the specified distance may be a unit distance, that is, the unit distance may be moved each time compared to the position before the movement.
[0149] Combine the following FIG. 3C to FIG. 3F The process of determining the first translation parameter is described by an example. In this example, the specified coordinate axis is the horizontal axis of the preset plane rectangular coordinate system, and the specified directions are the positive direction and the negative direction of the horizontal axis, that is, the left direction and the right direction.
[0150] Figure 3C : is an example of the positions of the first transformed reticle set and the second transformed reticle set before translation, wherein the upper reticle set is the first transformed reticle set and the lower reticle set is the second transformed reticle set. Figure 3C The position of the second transformed marking set in is its initial position, Figure 3C The shaded area in is the initial intersection area.
[0151] In this example, the distance change (specified distance) of each translation is the distance between two adjacent lane lines.
[0152] First, a process of performing translation with the left direction as the first direction is described.
[0153] exist Figure 3C Based on this, the second step is performed to move the second transformed marking set to the left by a specified distance (the distance between two adjacent lane lines), and the result is Figure 3D Results shown. Figure 3D The area of the shaded part is the target intersection area of this translation. Since this movement is the first movement in the first direction, the reference intersection area is the initial intersection area. Figure 3C and Figure 3D It can be seen from the shaded part of that the target intersection area of this movement is smaller than the reference intersection area. Therefore, the fifth step can be executed to stop translating the second transformed reticle set to the left. At this point, the translation of the second transformed reticle set in the left direction ends.
[0154] The following describes a process of performing translation with the right direction as the second direction.
[0155] exist Figure 3C Based on this, the second step is performed to move the second transformed marking set to the right by a specified distance (the distance between two adjacent lane lines), and the result is Figure 3E Results shown. Figure 3E The area of the shaded part is the target intersection area of this translation. Since this movement is the first movement in the second direction, the reference intersection area is the initial intersection area. Figure 3C and Figure 3E From the shaded part of , we can see that the target intersection area of this movement is larger than the reference intersection area. Therefore, we need to execute the second step again, that is, continue to move to the right.
[0156] That is, in Figure 3E Based on , the second step is executed again, and the second transformed marking set is moved to the right by a specified distance (the distance between two adjacent lane lines), and the obtained Figure 3F The effect shown. Figure 3F The area of the shaded part is the target intersection area of this translation. Since this movement is not the first movement in the second direction, the reference intersection area of this translation is the intersection area of the second outer bounding rectangle of the second transformed marking set before this translation and the first outer bounding rectangle, that is, Figure 3E The area of the shaded area in the figure. Figure 3E and Figure 3F It can be seen from the shaded part of that the target intersection area of this movement is smaller than the reference intersection area. Therefore, the fifth step can be executed to stop translating the second transformed mark line set to the right. At this point, the translation of the second transformed mark line set in the right direction is completed.
[0157] After the translation in both the left and right directions is completed, the execution condition of the sixth step is satisfied. Therefore, the translation parameter of the middle position of the second transformed reticle set compared to the initial position when the intersection area with the first bounding rectangle is the largest in each historical translation process can be used as the first translation parameter. In this example, the second transformed reticle set is composed of Figure 3C Position in Figure 3EIn the process of position movement, there is an intermediate position where the intersection area of the second outer bounding rectangle of the second transformed marking line set and the first outer bounding rectangle is the largest. The intermediate position is larger than the second transformed marking line set in Figure 3C The parameter of the position change in is the first translation parameter.
[0158] In the above process, since only the graphic movement in the direction parallel to the specified coordinate axis is involved, the amount of calculation during the movement is very small, so the efficiency of obtaining the first translation parameter can be effectively improved.
[0159] In step 14, the second translation parameters are inversely deduced using the first translation parameters according to the target transformation relationship.
[0160] After obtaining the first translation parameter, according to the target transformation relationship, the translation parameter of the first translation parameter before the transformation, i.e., the second translation parameter, can be inversely deduced by using the inverse process of obtaining the target transformation relationship. For example, if the target transformation relationship is obtained by affine transformation, then according to the first translation parameter, an inverse affine transformation is performed to obtain the second translation parameter. The second translation parameter is the translation parameter referenced by the actual movement of the reticle set to be fused.
[0161] In step 15, the set of reticle lines to be fused is moved using a second translation parameter.
[0162] After the second translation parameter is obtained, the reticle set to be fused may be moved using the second translation parameter so that the reticle set to be fused is merged with the target reticle set, thereby obtaining a fused reticle set.
[0163] Through the above technical solution, the first part that overlaps with the line set to be fused is obtained from the target line set, and the second part that overlaps with the target line set is obtained from the line set to be fused. According to the target transformation relationship, a first transformed line set obtained by transforming the first part and a second transformed line set obtained by transforming the second part are generated, and a first translation parameter that enables the second transformed line set to be fused with the first transformed line set is determined. According to the target transformation relationship, the second translation parameter is inversely deduced using the first translation parameter, and the line set to be fused is moved using the second translation parameter. In this way, in the process of fusing traffic markings, only a simple calculation method is needed, which is conducive to improving the fusion efficiency and accuracy.
[0164] Figure 4 is a block diagram of a traffic marking fusion device provided according to an embodiment of the present disclosure. Figure 4 As shown, the device 40 includes:
[0165] A first acquisition module 41 is used to acquire a first portion that coincides with the set of marks to be fused from the target mark set, and to acquire a second portion that coincides with the target mark set from the set of marks to be fused;
[0166] A generating module 42, configured to generate a first transformed marking line set obtained by transforming the first part and a second transformed marking line set obtained by transforming the second part according to the target transformation relationship;
[0167] A first determination module 43, configured to determine a first translation parameter capable of merging the second transformed reticle set with the first transformed reticle set;
[0168] A derivation module 44 is used to inversely deduce the second translation parameter using the first translation parameter according to the target transformation relationship;
[0169] The moving module 45 is used to move the set of marking lines to be fused using the second translation parameter.
[0170] Optionally, the device 40 further includes:
[0171] The second determination module is used to determine a line marking set to be fused corresponding to the target line marking set for a given target line marking set, wherein the target line marking set and the line marking set to be fused are both taken from the line marking set in the road data, and each line marking set includes a group of traffic line markings.
[0172] Optionally, the second determining module is used to determine, based on the area occupied by the target marking set, a marking set in the road data whose area overlaps with the target marking set, as the marking set to be fused.
[0173] Optionally, the device 40 further includes:
[0174] A second acquisition module, configured to acquire a first number of traffic markings in the target line marking set and a second number of traffic markings in the line marking set to be fused before the first acquisition module 41 acquires a first portion that coincides with the line marking set to be fused from the target line marking set and acquires a second portion that coincides with the target line marking set from the line marking set to be fused;
[0175] A third determination module, configured to determine the smaller of the first quantity and the second quantity as the target quantity if the first quantity and the second quantity are inconsistent;
[0176] The deleting module is used to delete the traffic markings in the target marking set or the marking set to be merged, so that the number of traffic markings in the target marking set and the marking set to be merged is the target number.
[0177] Optionally, remove modules, including:
[0178] A first determination submodule, configured to determine a first distance between a traffic marking on a first side edge of a target marking set and a traffic marking on a first side edge of a marking set to be fused;
[0179] A second determination submodule, configured to determine a second distance between a traffic marking on a second side edge of the target marking set and a traffic marking on a second side edge of the marking set to be fused;
[0180] A third determination submodule is used to take the side corresponding to the larger one of the first distance and the second distance as the target side;
[0181] The deletion submodule is used to delete the traffic markings on the target side of the target marking set or the marking set to be fused.
[0182] Optionally, remove submodules, for:
[0183] If the first number is greater than the second number, deleting a specified number of traffic markings from the target side of the target marking set;
[0184] If the second number is greater than the first number, deleting a specified number of traffic markings from the target side of the set of markings to be fused;
[0185] The specified quantity is the difference between the first quantity and the second quantity.
[0186] Optionally, the target transformation relationship can enable the second local portion to be perpendicular to the specified coordinate axis after transformation.
[0187] Optionally, the target transformation relationship is obtained by an affine transformation.
[0188] Optionally, the first determining module 43 includes:
[0189] A fourth determination submodule, used to determine a first outer bounding rectangle corresponding to the first transformed marking line set;
[0190] The fifth determination submodule is used to translate the second transformed mark set multiple times, record the intersection area between the second outer bounding rectangle and the first outer bounding rectangle corresponding to the second transformed mark set in each movement, and use the translation parameter corresponding to the movement with the largest intersection area as the first translation parameter.
[0191] Optionally, the fifth determining submodule is used to:
[0192] Recording the initial position of the second transformed reticle set before translation, and determining the intersection area of the second outer bounding rectangle and the first outer bounding rectangle when the second transformed reticle set is at the initial position as the initial intersection area;
[0193] The second transformed mark set is translated along the specified direction and by a specified distance, so that the second transformed mark set reaches the target position of this translation;
[0194] Determine the intersection area of the second outer bounding rectangle and the first outer bounding rectangle of the second transformed marking line set at the target position of this translation as the target intersection area;
[0195] comparing the target intersection area to the reference intersection area;
[0196] If the target intersection area is smaller than the reference intersection area, stop translating the second transformed reticle set in the specified direction;
[0197] After stopping the translation operation on the second transformed reticle set, a translation parameter of the middle position of the second transformed reticle set compared to the initial position when the intersection area with the first bounding rectangle is the largest during the historical translation process is used as the first translation parameter.
[0198] Optionally, the fifth determining submodule is further used for:
[0199] If the target intersection area is greater than the reference intersection area, return to the step of translating the second transformed reticle set by a specified distance in a specified direction so that the second transformed reticle set reaches the target position of this translation.
[0200] Optionally, the specified direction is a direction parallel to a specified coordinate axis, and the specified direction includes a first direction and a second direction that are opposite to each other.
[0201] Optionally, the traffic marking is a lane marking.
[0202] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.
[0203] Figure 5 1 is a block diagram of an electronic device 1900 according to an exemplary embodiment. For example, the electronic device 1900 may be provided as a server. Figure 5 The electronic device 1900 includes a processor 1922, which may be one or more, and a memory 1932 for storing a computer program executable by the processor 1922. The computer program stored in the memory 1932 may include one or more modules each corresponding to a set of instructions. In addition, the processor 1922 may be configured to execute the computer program to perform the above-mentioned traffic marking fusion method.
[0204] In addition, the electronic device 1900 may further include a power supply component 1926 and a communication component 1950. The power supply component 1926 may be configured to perform power management of the electronic device 1900, and the communication component 1950 may be configured to implement communication of the electronic device 1900, for example, wired or wireless communication. In addition, the electronic device 1900 may further include an input / output (I / O) interface 1958. The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server 2003. TM , Mac OS XTM , Unix TM , Linux TM etc.
[0205] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned traffic marking fusion method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 1932 including program instructions, and the above-mentioned program instructions can be executed by the processor 1922 of the electronic device 1900 to complete the above-mentioned traffic marking fusion method.
[0206] In another exemplary embodiment, a computer program product is further provided. The computer program product includes a computer program executable by a programmable device. The computer program has a code portion for executing the above-mentioned traffic marking fusion method when executed by the programmable device.
[0207] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.
[0208] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.
[0209] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.
Claims
1. A traffic marking fusion method, It is characterized in that The method comprises: Acquire a first portion from the target mark line set that overlaps with the mark line set to be fused, and acquire a second portion from the mark line set to be fused that overlaps with the target mark line set; According to the target transformation relationship, generating a first transformed mark line set obtained by transforming the first part and a second transformed mark line set obtained by transforming the second part; determining first translation parameters capable of blending the second transformed reticle set with the first transformed reticle set; According to the target transformation relationship, inversely derive the second translation parameter using the first translation parameter; The set of marking lines to be fused is moved using the second translation parameter.
2. The method according to claim 1, It is characterized in that The method further comprises: For a given target line marking set, a line marking set to be fused corresponding to the target line marking set is determined, wherein the target line marking set and the line marking set to be fused are both taken from line marking sets in road data, and each line marking set includes a group of traffic line markings.
3. The method according to claim 2, It is characterized in that The step of determining a set of reticle lines to be fused corresponding to a target reticle line set includes: According to the area occupied by the target marking set, a marking set whose area in the road data overlaps with the target marking set is determined as the marking set to be fused.
4. The method according to claim 1, It is characterized in that Before the steps of acquiring a first portion that coincides with the to-be-fused reticle set from the target reticle set and acquiring a second portion that coincides with the target reticle set from the to-be-fused reticle set, the method further includes: Acquire a first number of traffic markings in the target marking set and a second number of traffic markings in the to-be-fused marking set; If the first number is inconsistent with the second number, determining the smaller one of the first number and the second number as the target number; The traffic markings in the target marking set or the marking set to be merged are deleted so that the number of traffic markings in the target marking set and the marking set to be merged is the target number.
5. The method according to claim 4, It is characterized in that The deleting of the traffic markings in the target marking set or the marking set to be merged so that the number of traffic markings in the target marking set and the marking set to be merged is the target number includes: Determine a first distance between a traffic marking at a first side edge of the target marking set and a traffic marking at a first side edge of the marking set to be merged; Determine a second distance between the traffic markings at the second side edge of the target marking set and the traffic markings at the second side edge of the marking set to be merged; taking the side corresponding to the larger one of the first distance and the second distance as the target side; On the target side of the target marking set or the marking set to be fused, traffic markings are deleted.
6. The method according to claim 5, It is characterized in that The deleting of traffic markings on the target side of the target marking set or the marking set to be fused includes: If the first number is greater than the second number, deleting a specified number of traffic markings from the target side of the target marking set; If the second number is greater than the first number, deleting a specified number of traffic markings from the target side of the set of markings to be fused; The specified number is the difference between the first number and the second number.
7. The method according to claim 1, It is characterized in that The target transformation relationship enables the second part to be perpendicular to the specified coordinate axis after transformation.
8. The method according to claim 7, It is characterized in that The target transformation relationship is obtained through affine transformation.
9. The method according to claim 1, It is characterized in that The determining of a first translation parameter capable of merging the second transformed reticle set with the first transformed reticle set comprises: Determining a first bounding rectangle corresponding to the first transformed mark line set; The second transformed mark set is translated multiple times, the intersection area between the second bounding rectangle corresponding to the second transformed mark set and the first bounding rectangle in each translation is recorded, and the translation parameter corresponding to the translation with the largest intersection area is used as the first translation parameter.
10. The method according to claim 9, It is characterized in that The method of translating the second transformed mark set multiple times, recording the intersection area between the second outer bounding rectangle corresponding to the second transformed mark set and the first outer bounding rectangle in each translation, and using the translation parameter corresponding to the translation with the largest intersection area as the first translation parameter, includes: Recording the initial position of the second transformed reticle set before translation, and determining the intersection area of the second bounding rectangle of the second transformed reticle set and the first bounding rectangle at the initial position as the initial intersection area; translating the second transformed mark line set in a specified direction by a specified distance so that the second transformed mark line set reaches a target position of this translation; Determine the intersection area of the second bounding rectangle of the second transformed marking set and the first bounding rectangle at the target position of the current translation as the target intersection area; comparing the target intersection area to the reference intersection area; If the target intersection area is smaller than the reference intersection area, stopping translation of the second transformed reticle set in the specified direction; After stopping the translation operation on the second transformed reticle set, a translation parameter of the middle position of the second transformed reticle set compared to the initial position when the intersection area with the first bounding rectangle is the largest during the historical translation process is used as the first translation parameter.
11. The method according to claim 10, It is characterized in that The method further comprises: If the target intersection area is greater than the reference intersection area, return to the step of translating the second transformed reticle set by a specified distance in a specified direction so that the second transformed reticle set reaches the target position of this translation.
12. The method according to claim 10, It is characterized in that The designated direction is a direction parallel to a designated coordinate axis, and the designated direction includes a first direction and a second direction that are opposite to each other.
13. The method according to any one of claims 1 to 12, It is characterized in that The traffic markings are lane markings.
14. A traffic marking fusion device, It is characterized in that The device comprises: A first acquisition module is used to acquire a first part that overlaps with the to-be-fused mark line set from the target mark line set, and to acquire a second part that overlaps with the target mark line set from the to-be-fused mark line set; A generating module, configured to generate, according to a target transformation relationship, a first transformed mark line set obtained by transforming the first part and a second transformed mark line set obtained by transforming the second part; A first determination module, configured to determine a first translation parameter capable of merging the second transformed reticle set with the first transformed reticle set; A derivation module, configured to inversely derive a second translation parameter using the first translation parameter according to the target transformation relationship; A moving module is used to move the set of marking lines to be fused using the second translation parameter.
15. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the steps of the method described in any one of claims 1 to 13 are implemented.
16. An electronic device, It is characterized in that include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 13.
