CN116309628A - Lane line recognition method and device, electronic equipment and computer readable storage medium - Google Patents
Lane line recognition method and device, electronic equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN116309628A CN116309628A CN202310027786.5A CN202310027786A CN116309628A CN 116309628 A CN116309628 A CN 116309628A CN 202310027786 A CN202310027786 A CN 202310027786A CN 116309628 A CN116309628 A CN 116309628A
- Authority
- CN
- China
- Prior art keywords
- lane line
- fitting
- block area
- result
- block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The application discloses a lane line identification method and device, electronic equipment and a computer readable storage medium, wherein the method comprises the following steps: obtaining a current road image and carrying out lane line segmentation to obtain a lane line segmentation result of the current road image; dividing the lane line dividing result by using a preset dividing strategy to obtain lane line dividing results of a plurality of dividing areas, wherein the dividing areas comprise a first dividing area and a second dividing area; determining a lane line fitting result of the first block area according to the lane line segmentation result of the first block area; and carrying out iterative optimization on the lane line fitting result of the first block region by utilizing a preset iterative optimization strategy according to the lane line segmentation result of the second block region, so as to obtain a lane line identification result of the current road image. According to the method and the device, the lane line fitting result is subjected to iterative optimization, so that the influence of inaccurate lane line example segmentation result on the lane line identification accuracy is reduced, and the lane line identification accuracy and stability are improved.
Description
Technical Field
The present disclosure relates to the field of lane line recognition technologies, and in particular, to a lane line recognition method and apparatus, an electronic device, and a computer readable storage medium.
Background
In the field of automatic driving, lane line detection of a front road image shot by a vehicle-mounted camera is a necessary item in automatic driving perception capability, and the detected lane line can be used for realizing automatic driving functions such as vision auxiliary positioning, lane keeping and the like.
The lane line detection method comprises the steps that a lane line detection target is to output a curve equation of a lane line contained in an image in a 3D space, the lane line in the image is subjected to semantic segmentation mainly through a lane line semantic segmentation model trained in advance, each lane line instance is determined in a clustering mode, and finally the curve equation of each lane line instance in the 3D space is obtained through curve fitting.
However, in an actual scene, the semantic segmentation result of the lane lines and the clustering result of the lane line examples are easy to generate flaws and even have some obvious errors, and especially for a broken line lane line, one broken line lane line may be mistakenly segmented into two lane lines, so that the recognition accuracy of the lane lines is not high.
Disclosure of Invention
The embodiment of the application provides a lane line identification method and device, electronic equipment and storage medium, so as to improve lane line identification precision.
The embodiment of the application adopts the following technical scheme:
in a first aspect, an embodiment of the present application provides a lane line identifying method, where the method includes:
obtaining a current road image and carrying out lane line segmentation on the current road image to obtain a lane line segmentation result of the current road image;
partitioning the lane line segmentation result of the current road image by using a preset partitioning strategy to obtain lane line segmentation results corresponding to a plurality of partitioning areas, wherein the plurality of partitioning areas comprise a first partitioning area and a second partitioning area;
determining a lane line fitting result corresponding to the first block region according to the lane line segmentation result corresponding to the first block region;
and carrying out iterative optimization on the lane line fitting result corresponding to the first block region by utilizing a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block region, so as to obtain a lane line identification result of the current road image.
Optionally, the obtaining the current road image and performing lane line segmentation on the current road image, and obtaining a lane line segmentation result of the current road image includes:
Carrying out semantic segmentation on the current road image by using a preset lane line semantic segmentation model to obtain a lane line semantic segmentation result of the current road image;
and carrying out pixel-level clustering on the lane line semantic segmentation result by using a preset clustering algorithm to obtain a lane line instance segmentation result of the current road image.
Optionally, the lane line segmentation result includes a lane line example, and the determining, according to the lane line segmentation result corresponding to the first block region, the lane line fitting result corresponding to the first block region includes:
determining fitting candidate points corresponding to the lane line examples of the first block area according to the pixel points of the lane line examples corresponding to the first block area;
fitting the fitting candidate points corresponding to the lane line examples of the first block area by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the first block area.
Optionally, the lane line fitting result includes a lane line fitting curve, and the iteratively optimizing the lane line fitting result corresponding to the first block area by using a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block area, so as to obtain a lane line identification result of the current road image includes:
Sampling by using a preset sampling strategy based on a lane line fitting curve corresponding to the first block area to obtain lane line sampling points;
based on the transformation relation between the image coordinate system and the camera coordinate system, projecting the lane line sampling points into the current road image to obtain lane line projection points corresponding to the second block area;
according to the lane line projection points corresponding to the second block area and the lane line segmentation results corresponding to the second block area, carrying out iterative optimization on the lane line fitting curve corresponding to the first block area to obtain lane line fitting results corresponding to the second block area;
and determining a lane line identification result of the current road image according to the lane line fitting result corresponding to the second block area.
Optionally, the lane line segmentation result includes a lane line example, and performing iterative optimization on a lane line fitting curve corresponding to the first segmented region according to the lane line projection point corresponding to the second segmented region and the lane line segmentation result corresponding to the second segmented region, so as to obtain a lane line fitting result corresponding to the second segmented region includes:
determining fitting candidate points corresponding to the lane line examples of the second block areas according to the pixel points of the lane line examples corresponding to the second block areas;
Matching the lane line instance of the second block area with the lane line instance of the first block area based on the fitting candidate point corresponding to the lane line instance of the second block area and the lane line projection point corresponding to the second block area;
and carrying out iterative optimization on the lane line fitting curve corresponding to the first block region according to the matching result to obtain a second lane line fitting result corresponding to the second block region.
Optionally, performing iterative optimization on the lane line fitting curve corresponding to the first block region according to the matching result, and obtaining a second lane line fitting result corresponding to the second block region includes:
obtaining fitting candidate points corresponding to lane line examples of the first block area;
combining fitting candidate points corresponding to the lane line examples of the first block area and fitting candidate points corresponding to the lane line examples of the second block area according to the matching result;
and fitting the combined fitting candidate points by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the second block area.
Optionally, the fitting the combined fitting candidate points by using a preset lane line fitting algorithm, and obtaining a lane line fitting result corresponding to the second block area includes:
If the matching result is that the matching is successful, determining that the lane line instance of the second block area corresponds to the lane line instance of the first block area, namely the same lane line, and fitting the fitting candidate points corresponding to the lane line instances of the same lane line;
if the matching result is that the matching is failed, determining that the lane line instance of the second block area is not the same lane line corresponding to the lane line instance of the first block area, and fitting the fitting candidate points corresponding to the lane line instance of the second block area.
In a second aspect, embodiments of the present application further provide a lane line identifying apparatus, where the apparatus includes:
the lane line segmentation unit is used for acquiring a current road image and carrying out lane line segmentation on the current road image to obtain a lane line segmentation result of the current road image;
the block unit is used for blocking the lane line segmentation result of the current road image by utilizing a preset block strategy to obtain lane line segmentation results corresponding to a plurality of block areas, wherein the plurality of block areas comprise a first block area and a second block area;
the determining unit is used for determining a lane line fitting result corresponding to the first block area according to the lane line segmentation result corresponding to the first block area;
And the optimization unit is used for carrying out iterative optimization on the lane line fitting result corresponding to the first block area by utilizing a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block area, so as to obtain a lane line identification result of the current road image.
In a third aspect, embodiments of the present application further provide an electronic device, including:
a processor; and
a memory arranged to store computer executable instructions which, when executed, cause the processor to perform any of the methods described hereinbefore.
In a fourth aspect, embodiments of the present application also provide a computer-readable storage medium storing one or more programs that, when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform any of the methods described above.
The above-mentioned at least one technical scheme that this application embodiment adopted can reach following beneficial effect: according to the lane line identification method, a current road image is acquired first, lane line segmentation is carried out on the current road image, and a lane line segmentation result of the current road image is obtained; then, dividing the lane line dividing result of the current road image by utilizing a preset dividing strategy to obtain lane line dividing results corresponding to a plurality of dividing areas, wherein the plurality of dividing areas comprise a first dividing area and a second dividing area; then, according to the lane line segmentation result corresponding to the first block region, determining a lane line fitting result corresponding to the first block region; and finally, carrying out iterative optimization on the lane line fitting result corresponding to the first block region by utilizing a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block region, and obtaining a lane line identification result of the current road image. According to the lane line identification method, the lane line segmentation result is segmented by adopting the preset segmentation strategy, the lane line fitting result is subjected to iterative optimization by utilizing the preset iterative optimization strategy, the influence of inaccuracy of the lane line instance segmentation result on the final lane line identification precision is reduced, the lane line identification precision and stability are improved, and particularly, the lane line identification method also has higher identification precision on the broken line lane line identification.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
FIG. 1 is a schematic diagram of a lane line recognition process in the prior art;
FIG. 2 is a schematic diagram of a semantic segmentation result of lane lines in the prior art;
FIG. 3 is a schematic diagram of a lane line example segmentation result in the prior art;
FIG. 4 is a diagram showing the segmentation result of another lane line example in the prior art;
fig. 5 is a flow chart of a lane line recognition method in an embodiment of the present application;
FIG. 6 is a schematic diagram of a partitioning result of a partitioning area according to an embodiment of the present application;
fig. 7 is a schematic diagram of a lane line recognition flow in an embodiment of the present application;
fig. 8 is a schematic structural diagram of a lane line recognition device according to an embodiment of the present application;
fig. 9 is a schematic structural diagram of an electronic device in an embodiment of the present application.
Detailed Description
For the purposes, technical solutions and advantages of the present application, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.
In order to facilitate understanding of the embodiments of the present application, as shown in fig. 1, a schematic diagram of a lane line recognition procedure in the prior art is provided. Firstly, carrying out lane line semantic segmentation by using a lane line semantic segmentation model, and segmenting all pixels predicted as lane lines in an image without distinguishing specific examples of the lane lines, wherein a lane line semantic segmentation result schematic diagram in the prior art is provided, and white parts are lane lines but the lane lines are not distinguished from each other, as shown in fig. 2.
And then, clustering lane line pixels on the basis of the lane line semantic segmentation result, so that the lane line examples are distinguished. As shown in fig. 3, a schematic diagram of a lane line example segmentation result in the prior art is provided, and it can be seen that two lane lines in the image are respectively marked black and white.
And then, according to the lane line example segmentation result, taking a point on each line for each lane line, and converting the points into a camera coordinate system (namely a 3D coordinate system) through camera internal and external parameters, so as to obtain fitting candidate points of each lane line, wherein the camera internal and external parameters can be obtained through camera calibration in advance.
And finally, for each lane line, taking fitting candidate points of the lane line to perform curve fitting, and obtaining a fitting curve of each lane line.
However, in an actual scene, the lane line semantic segmentation result and the lane line example segmentation result may have flaws or even some obvious errors, and particularly for a dotted line lane line, as shown in fig. 4, another lane line example segmentation result schematic diagram in the prior art is provided, and it can be seen that the same dotted line lane line in the right white rectangular frame is mistakenly segmented into two different lane lines, so that the final lane line recognition result generates problems of false detection, jump, substantial drift and the like, and further affects the realization of an automatic driving function depending on the lane recognition result.
Based on this, the embodiment of the application provides a lane line recognition method, as shown in fig. 5, and provides a flow chart of the lane line recognition method in the embodiment of the application, where the method at least includes the following steps S510 to S540:
step S510, obtaining a current road image and carrying out lane line segmentation on the current road image to obtain a lane line segmentation result of the current road image.
When the lane line identification is carried out, the current road image acquired by the vehicle-mounted camera is required to be acquired firstly, then the lane line segmentation is carried out on the current road image by utilizing a certain lane line segmentation algorithm to obtain a lane line segmentation result of the current road image, the lane line segmentation algorithm can be obtained based on the existing lane line semantic segmentation model and clustering algorithm, and specifically, pixels of each lane line instance in the image can be segmented.
And step S520, partitioning the lane line segmentation result of the current road image by utilizing a preset partitioning strategy to obtain lane line segmentation results corresponding to a plurality of partitioned areas, wherein the partitioned areas comprise a first partitioned area and a second partitioned area.
The lane line segmentation result of the current road image obtained in the foregoing step may be regarded as an initial lane line segmentation result, and in order to further improve the accuracy of lane line recognition, a certain policy needs to be adopted to perform optimization, where the policy firstly needs to be a preset segmentation policy, and "segmentation" may be understood as that the current road image including the lane line segmentation result is divided into a plurality of regions, so as to obtain lane line segmentation results corresponding to the plurality of segmentation regions.
Since the direction of the lane lines in the road image, which is shot from the view angle of the vehicle-mounted camera, is generally along the longitudinal direction of the image, the preset block dividing strategy can divide the current road image containing the lane line dividing result into a plurality of areas along the transverse direction of the image at certain intervals, wherein the interval size can be flexibly adjusted according to actual needs, for example, the interval size can be equal, the interval size can also be divided upwards from the bottom of the image according to the principle of near-far-small, and the interval size is smaller. As shown in fig. 6, a schematic diagram of a partitioning result of a partitioning area in the embodiment of the present application is provided, and it can be seen that a current road image including a lane line partitioning result is equally divided into four partitioning areas N01, N02, N03, and N04.
After the above multiple block areas are obtained, it is further required to determine the current first block area and the second block area, where the "first" and "second" in this embodiment of the present application are a relative concept, and may be regarded as two laterally adjacent block areas, and in the initial processing, based on the principle of near-far-small, the first block area may be regarded as the first block area at the bottom of the image, such as the N01 area in fig. 6, and the second block area may be regarded as the N02 area. Because the optimization strategy adopted in the subsequent step of the method is iterative optimization, after the N01 area and the N02 area are processed, the N03 area is continuously processed based on the processing results corresponding to the N01 area and the N02 area, and the N03 area is the current second blocking area until all the blocking areas are iteratively completed.
Step S530, determining a lane line fitting result corresponding to the first block area according to the lane line segmentation result corresponding to the first block area.
After determining the current first block area, determining a lane line fitting result corresponding to the first block area according to a lane line segmentation result corresponding to the first block area, for example, pixel points of each lane line instance segmented in the first block area, wherein the lane line fitting result corresponding to the first block area is the first block area, so that the lane line fitting result can be regarded as an initial or intermediate lane line fitting result as a basis of subsequent iterative optimization.
And S540, carrying out iterative optimization on the lane line fitting result corresponding to the first block region by utilizing a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block region, and obtaining a lane line identification result of the current road image.
The lane line fitting result of the first segmented region obtained by the steps is obtained based on the lane line segmentation result of the first segmented region, and the lane line segmentation result corresponding to the second segmented region and a preset iterative optimization strategy can be further utilized to iteratively optimize the lane line fitting result of the first segmented region until the iterative processing of all segmented regions is completed, so that a final lane line identification result is obtained.
According to the lane line identification method, the lane line segmentation result is segmented by adopting the preset segmentation strategy, the lane line fitting result is subjected to iterative optimization by utilizing the preset iterative optimization strategy, the influence of inaccuracy of the lane line instance segmentation result on the final lane line identification precision is reduced, the lane line identification precision and stability are improved, and particularly, the lane line identification method also has higher identification precision on the broken line lane line identification.
In some embodiments of the present application, the obtaining a current road image and performing lane segmentation on the current road image, where obtaining a lane segmentation result of the current road image includes: carrying out semantic segmentation on the current road image by using a preset lane line semantic segmentation model to obtain a lane line semantic segmentation result of the current road image; and carrying out pixel-level clustering on the lane line semantic segmentation result by using a preset clustering algorithm to obtain a lane line instance segmentation result of the current road image.
When the current road image is subjected to lane line segmentation, the lane line semantic segmentation can be performed on the current road image by utilizing a lane line semantic segmentation model trained in advance, namely, pixel points of all lane lines are segmented from the image, and then the pixel points of all the lane lines can be further clustered by utilizing a preset clustering algorithm, so that each lane line instance contained in the current road image is obtained.
The lane line semantic segmentation model can be obtained by training based on the existing convolutional neural network such as a U-net network, and the preset clustering algorithm can be realized based on the existing DBSCAN (Density-Based Spatial Clustering of Applications with Noise, maximum set of points connected in Density) algorithm, and the like. Of course, the specific manner in which the lane line semantic segmentation and clustering is performed may also be flexibly selected by those skilled in the art in combination with the prior art, and is not specifically limited herein.
In some embodiments of the present application, the lane line segmentation result includes a lane line example, and determining, according to the lane line segmentation result corresponding to the first block region, a lane line fitting result corresponding to the first block region includes: determining fitting candidate points corresponding to the lane line examples of the first block area according to the pixel points of the lane line examples corresponding to the first block area; fitting the fitting candidate points corresponding to the lane line examples of the first block area by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the first block area.
Because the lane lines have a certain width, each pixel row of the divided lane line examples corresponds to a plurality of lane line pixel points, in order to ensure the fitting effect, the pixel points of all the lane line examples do not need to be subjected to lane line fitting, so that candidate points for the lane line fitting can be determined from all the pixel points of the lane line examples corresponding to the first block area, for example, for each lane line example, the middle pixel point of each pixel row corresponding to each lane line can be selected for the lane line fitting.
In addition, since the selected intermediate pixel points are located under the image coordinate system and the fitting of the lane lines is the fitting of the 3D points under the camera coordinate system, the selected intermediate pixel points of each row can be converted under the camera coordinate system based on the camera internal and external parameters calibrated in advance to serve as fitting candidate points corresponding to each lane line example.
And finally, fitting the fitting candidate points corresponding to the lane line examples by using a certain lane line fitting algorithm, so as to obtain lane line fitting curves of the lane line examples. The lane line fitting algorithm may be flexibly selected based on the prior art, and may be implemented by any one of B-spline, cubic spline interpolation, RANSAC (RANdom SAmple Consensus ), least square method, and the like, which is not specifically limited herein.
In some embodiments of the present application, the lane line fitting result includes a lane line fitting curve, and performing iterative optimization on the lane line fitting result corresponding to the first segmented region by using a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second segmented region, where obtaining a lane line identification result of the current road image includes: sampling by using a preset sampling strategy based on a lane line fitting curve corresponding to the first block area to obtain lane line sampling points; based on the transformation relation between the image coordinate system and the camera coordinate system, projecting the lane line sampling points into the current road image to obtain lane line projection points corresponding to the second block area; according to the lane line projection points corresponding to the second block area and the lane line segmentation results corresponding to the second block area, carrying out iterative optimization on the lane line fitting curve corresponding to the first block area to obtain lane line fitting results corresponding to the second block area; and determining a lane line identification result of the current road image according to the lane line fitting result corresponding to the second block area.
Taking the first block area as the first block area of the current road image, namely an N01 area as an example, after a lane line fitting curve corresponding to the N01 area is obtained, sampling lane line points can be carried out by adopting a certain sampling strategy, and then based on internal and external parameters of a camera, the sampling points are projected into the current road image, so that lane line projection points are obtained. The principle of the sampling strategy is to have one projection point per line of pixels in the image for each lane line instance, or at least one projection point per line of pixels in the second segmented region, i.e. the N02 region, in the image.
The purpose of sampling and projecting the lane line points based on the lane line fitting curve of the N01 area is to predict the position of the lane line instance corresponding to the N01 area in the N02 area, and then the relationship between the lane line instance of the N01 area and the lane line instance of the N02 area can be determined by combining the lane line instance originally segmented by the N02 area, so that the lane line fitting curve of the N01 area is subjected to iterative optimization until all the blocks are subjected to iterative optimization, a final lane line recognition result is obtained, the problem that the lane line with the same broken line is mistakenly segmented into different lane lines can be avoided through the iterative optimization processing, and the recognition precision of the lane line is improved.
In some embodiments of the present application, the lane line segmentation result includes a lane line example, and performing iterative optimization on a lane line fitting curve corresponding to the first segmented region according to the lane line projection point corresponding to the second segmented region and the lane line segmentation result corresponding to the second segmented region, to obtain a lane line fitting result corresponding to the second segmented region includes: determining fitting candidate points corresponding to the lane line examples of the second block areas according to the pixel points of the lane line examples corresponding to the second block areas; matching the lane line instance of the second block area with the lane line instance of the first block area based on the fitting candidate point corresponding to the lane line instance of the second block area and the lane line projection point corresponding to the second block area; and carrying out iterative optimization on the lane line fitting curve corresponding to the first block region according to the matching result to obtain a second lane line fitting result corresponding to the second block region.
The lane line segmentation result corresponding to the N02 region comprises all pixel points of the lane line instance segmented by the N02 region, so that fitting candidate points can be selected from the pixel points, for example, the middle pixel point of each pixel row corresponding to each lane line instance can be selected as the fitting candidate point, then a certain matching strategy is adopted to compare the distance between the fitting candidate point corresponding to the lane line instance of the N02 region and the projection point of the lane line sampling point of the N01 region in the N02 region, and accordingly, the matching relation between the lane line instance of the N01 region and the lane line instance of the N02 region, for example, whether the matching relation is the same lane line or not is determined, and further iterative optimization can be carried out on the lane line fitting curve corresponding to the N01 region according to the matching relation between the lane line instance of the N01 region and the lane line instance of the N02 region.
The above-mentioned matching policy may be set, for example, as: if the pixel distance between the fitting candidate points corresponding to the lane line examples of the second block area and the corresponding lane line projection points is smaller than a preset pixel distance threshold value, and the number ratio of the fitting candidate points meeting the requirement reaches a first preset number threshold value, the matching is considered to be successful; and if the pixel distance between the fitting candidate points corresponding to the lane line examples of the second block area and the corresponding lane line projection points is not smaller than the preset pixel distance threshold, and the number ratio of the fitting candidate points which do not meet the requirements reaches a second preset number threshold, the matching is considered to be failed.
In some embodiments of the present application, performing iterative optimization on the lane line fitting curve corresponding to the first segmented region according to the matching result, and obtaining a second lane line fitting result corresponding to the second segmented region includes: obtaining fitting candidate points corresponding to lane line examples of the first block area; combining fitting candidate points corresponding to the lane line examples of the first block area and fitting candidate points corresponding to the lane line examples of the second block area according to the matching result; and fitting the combined fitting candidate points by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the second block area.
The matching result of the lane line instance of the second block area and the lane line instance of the first block area reflects whether the lane line instances of the two areas correspond to the same lane line, the fitting candidate points corresponding to the lane line instance of the first block area and the fitting candidate points corresponding to the lane line instance of the second block area can be combined based on the matching result, the combined fitting candidate points possibly comprise fitting candidate points belonging to the same lane line or fitting candidate points not belonging to the same lane line, and therefore lane line fitting can be carried out according to different combining results to obtain the lane line fitting result corresponding to the second block area.
In some embodiments of the present application, the fitting the combined fitting candidate points by using a preset lane line fitting algorithm, and obtaining a lane line fitting result corresponding to the second block area includes: if the matching result is that the matching is successful, determining that the lane line instance of the second block area corresponds to the lane line instance of the first block area, namely the same lane line, and fitting the fitting candidate points corresponding to the lane line instances of the same lane line; if the matching result is that the matching is failed, determining that the lane line instance of the second block area is not the same lane line corresponding to the lane line instance of the first block area, and fitting the fitting candidate points corresponding to the lane line instance of the second block area.
For example, if the leftmost lane line instance of the second block area is successfully matched with the leftmost lane line instance of the first block area, which means that the lane line instances of the two block areas correspond to the same lane line, the fitting candidate points obtained by merging at this time simultaneously comprise the fitting candidate point of the leftmost lane line instance of the second block area and the fitting candidate point of the leftmost lane line instance of the first block area, and the fitting is performed on the merged fitting candidate points, so that the lane line fitting result after one iteration optimization of the lane line is obtained.
If the rightmost lane line instance of the second block area is not successfully matched with any lane line instance in the first block area, the rightmost lane line instance is possibly segmented in the second block area, so that fitting candidate points corresponding to the rightmost lane line instance of the second block area can be independently fitted, a lane line fitting result of the rightmost lane line instance of the second block area is obtained, and the process is equivalent to gradual reclustering of the fitting candidate points of the lane lines.
It should be noted that, the above process is only illustrated by taking two segmented areas as an example, the processing logic of the subsequent segmented areas is similar to that of the previous process, for example, after the lane line fitting result of the N02 area is obtained, the lane line points can be sampled and projected in combination with all lane line fitting curves corresponding to the N01 area and the N02 area, so as to obtain the lane line projection points corresponding to the N03 area of each lane line instance of the N01 area and the N02 area, and then the distance comparison is performed between the lane line projection points and the fitting candidate points of the lane line instance segmented by the N03 area, so as to determine the matching relationship between each lane line instance segmented by the N03 area and each lane line instance of the N01 area and the N02 area, and update the lane line fitting result corresponding to the N01 area and the N02 area in this way, and continue iteration in this way until all segmented areas are completed in an iteration way, thereby obtaining the final lane line identification result.
In order to facilitate understanding of the above embodiments of the present application, as shown in fig. 7, a schematic diagram of a lane line recognition flow in the embodiment of the present application is provided. Firstly, inputting a current road image into a lane line semantic segmentation model trained in advance to obtain a lane line semantic segmentation result, and clustering the lane line semantic segmentation result by using a preset clustering algorithm to obtain a lane line instance segmentation result of the current road image. And then, partitioning the lane line example partitioning result of the current road image by utilizing a preset partitioning strategy to obtain lane line example partitioning results of a plurality of partitioned areas. And determining a lane line fitting candidate point according to the pixel point of the lane line example of the first block area, performing lane line fitting to obtain a lane line fitting curve of the first block area, sampling the lane line fitting curve of the first block area to obtain a lane line sampling point, projecting the lane line sampling point into the current road image through coordinate transformation to correspondingly obtain a lane line projection point of the second block area, comparing the lane line projection point of the second block area with the fitting candidate point of the lane line example of the second block area, and realizing the reclustering of the fitting candidate point of the lane line example and the updating of the lane line fitting result, so that iteration is performed until all the block iterations are completed, and outputting a final lane line identification result.
The embodiment of the application further provides a lane line recognition device 800, as shown in fig. 8, and a schematic structural diagram of the lane line recognition device in the embodiment of the application is provided, where the device 800 at least includes: a lane line segmentation unit 810, a segmentation unit 820, a determination unit 830, and an optimization unit 840, wherein:
the lane line segmentation unit 810 is configured to obtain a current road image and perform lane line segmentation on the current road image to obtain a lane line segmentation result of the current road image;
the blocking unit 820 is configured to block the lane line segmentation result of the current road image by using a preset blocking strategy, so as to obtain lane line segmentation results corresponding to a plurality of blocking areas, where the plurality of blocking areas include a first blocking area and a second blocking area;
a determining unit 830, configured to determine a lane line fitting result corresponding to the first block area according to the lane line segmentation result corresponding to the first block area;
and the optimizing unit 840 is configured to perform iterative optimization on the lane line fitting result corresponding to the first block area by using a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block area, so as to obtain a lane line identification result of the current road image.
In some embodiments of the present application, the lane line segmentation unit 810 is specifically configured to: carrying out semantic segmentation on the current road image by using a preset lane line semantic segmentation model to obtain a lane line semantic segmentation result of the current road image; and carrying out pixel-level clustering on the lane line semantic segmentation result by using a preset clustering algorithm to obtain a lane line instance segmentation result of the current road image.
In some embodiments of the present application, the lane line segmentation result includes a lane line example, and the determining unit 830 is specifically configured to: determining fitting candidate points corresponding to the lane line examples of the first block area according to the pixel points of the lane line examples corresponding to the first block area; fitting the fitting candidate points corresponding to the lane line examples of the first block area by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the first block area.
In some embodiments of the present application, the lane line fitting result includes a lane line fitting curve, and the optimizing unit 840 is specifically configured to: sampling by using a preset sampling strategy based on a lane line fitting curve corresponding to the first block area to obtain lane line sampling points; based on the transformation relation between the image coordinate system and the camera coordinate system, projecting the lane line sampling points into the current road image to obtain lane line projection points corresponding to the second block area; according to the lane line projection points corresponding to the second block area and the lane line segmentation results corresponding to the second block area, carrying out iterative optimization on the lane line fitting curve corresponding to the first block area to obtain lane line fitting results corresponding to the second block area; and determining a lane line identification result of the current road image according to the lane line fitting result corresponding to the second block area.
In some embodiments of the present application, the lane line segmentation result includes a lane line example, and the optimizing unit 840 is specifically configured to: determining fitting candidate points corresponding to the lane line examples of the second block areas according to the pixel points of the lane line examples corresponding to the second block areas; matching the lane line instance of the second block area with the lane line instance of the first block area based on the fitting candidate point corresponding to the lane line instance of the second block area and the lane line projection point corresponding to the second block area; and carrying out iterative optimization on the lane line fitting curve corresponding to the first block region according to the matching result to obtain a second lane line fitting result corresponding to the second block region.
In some embodiments of the present application, the optimizing unit 840 is specifically configured to: obtaining fitting candidate points corresponding to lane line examples of the first block area; combining fitting candidate points corresponding to the lane line examples of the first block area and fitting candidate points corresponding to the lane line examples of the second block area according to the matching result; and fitting the combined fitting candidate points by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the second block area.
In some embodiments of the present application, the optimizing unit 840 is specifically configured to: if the matching result is that the matching is successful, determining that the lane line instance of the second block area corresponds to the lane line instance of the first block area, namely the same lane line, and fitting the fitting candidate points corresponding to the lane line instances of the same lane line; if the matching result is that the matching is failed, determining that the lane line instance of the second block area is not the same lane line corresponding to the lane line instance of the first block area, and fitting the fitting candidate points corresponding to the lane line instance of the second block area.
It can be understood that the lane line recognition device can implement each step of the lane line recognition method provided in the foregoing embodiment, and the relevant explanation about the lane line recognition method is applicable to the lane line recognition device, which is not repeated herein.
Fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Referring to fig. 9, at the hardware level, the electronic device includes a processor, and optionally an internal bus, a network interface, and a memory. The Memory may include a Memory, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory (non-volatile Memory), such as at least 1 disk Memory. Of course, the electronic device may also include hardware required for other services.
The processor, network interface, and memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or EISA (Extended Industry Standard Architecture ) bus, among others. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in fig. 9, but not only one bus or one type of bus.
And the memory is used for storing programs. In particular, the program may include program code including computer-operating instructions. The memory may include memory and non-volatile storage and provide instructions and data to the processor.
The processor reads the corresponding computer program from the nonvolatile memory into the memory and then runs the computer program to form the lane line identification device on the logic level. The processor is used for executing the programs stored in the memory and is specifically used for executing the following operations:
obtaining a current road image and carrying out lane line segmentation on the current road image to obtain a lane line segmentation result of the current road image;
Partitioning the lane line segmentation result of the current road image by using a preset partitioning strategy to obtain lane line segmentation results corresponding to a plurality of partitioning areas, wherein the plurality of partitioning areas comprise a first partitioning area and a second partitioning area;
determining a lane line fitting result corresponding to the first block region according to the lane line segmentation result corresponding to the first block region;
and carrying out iterative optimization on the lane line fitting result corresponding to the first block region by utilizing a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block region, so as to obtain a lane line identification result of the current road image.
The method executed by the lane line recognition device disclosed in the embodiment shown in fig. 1 of the present application may be applied to a processor or implemented by the processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present application may be embodied directly in hardware, in a decoded processor, or in a combination of hardware and software modules in a decoded processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.
The electronic device may further execute the method executed by the lane line recognition device in fig. 1, and implement the function of the lane line recognition device in the embodiment shown in fig. 1, which is not described herein again.
The embodiments of the present application also provide a computer-readable storage medium storing one or more programs, where the one or more programs include instructions, which when executed by an electronic device that includes a plurality of application programs, enable the electronic device to perform a method performed by the lane line recognition apparatus in the embodiment shown in fig. 1, and specifically are configured to perform:
obtaining a current road image and carrying out lane line segmentation on the current road image to obtain a lane line segmentation result of the current road image;
partitioning the lane line segmentation result of the current road image by using a preset partitioning strategy to obtain lane line segmentation results corresponding to a plurality of partitioning areas, wherein the plurality of partitioning areas comprise a first partitioning area and a second partitioning area;
determining a lane line fitting result corresponding to the first block region according to the lane line segmentation result corresponding to the first block region;
and carrying out iterative optimization on the lane line fitting result corresponding to the first block region by utilizing a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block region, so as to obtain a lane line identification result of the current road image.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.
Claims (10)
1. A lane line identification method, wherein the method comprises:
obtaining a current road image and carrying out lane line segmentation on the current road image to obtain a lane line segmentation result of the current road image;
partitioning the lane line segmentation result of the current road image by using a preset partitioning strategy to obtain lane line segmentation results corresponding to a plurality of partitioning areas, wherein the plurality of partitioning areas comprise a first partitioning area and a second partitioning area;
Determining a lane line fitting result corresponding to the first block region according to the lane line segmentation result corresponding to the first block region;
and carrying out iterative optimization on the lane line fitting result corresponding to the first block region by utilizing a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block region, so as to obtain a lane line identification result of the current road image.
2. The method of claim 1, wherein the obtaining the current road image and performing lane segmentation on the current road image to obtain a lane segmentation result of the current road image comprises:
carrying out semantic segmentation on the current road image by using a preset lane line semantic segmentation model to obtain a lane line semantic segmentation result of the current road image;
and carrying out pixel-level clustering on the lane line semantic segmentation result by using a preset clustering algorithm to obtain a lane line instance segmentation result of the current road image.
3. The method of claim 1, wherein the lane line segmentation result comprises a lane line instance, and the determining a lane line fitting result corresponding to the first segmented region according to the lane line segmentation result corresponding to the first segmented region comprises:
Determining fitting candidate points corresponding to the lane line examples of the first block area according to the pixel points of the lane line examples corresponding to the first block area;
fitting the fitting candidate points corresponding to the lane line examples of the first block area by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the first block area.
4. The method of claim 1, wherein the lane line fitting result includes a lane line fitting curve, and the iteratively optimizing the lane line fitting result corresponding to the first segmented region by using a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second segmented region, to obtain a lane line identification result of the current road image includes:
sampling by using a preset sampling strategy based on a lane line fitting curve corresponding to the first block area to obtain lane line sampling points;
based on the transformation relation between the image coordinate system and the camera coordinate system, projecting the lane line sampling points into the current road image to obtain lane line projection points corresponding to the second block area;
according to the lane line projection points corresponding to the second block area and the lane line segmentation results corresponding to the second block area, carrying out iterative optimization on the lane line fitting curve corresponding to the first block area to obtain lane line fitting results corresponding to the second block area;
And determining a lane line identification result of the current road image according to the lane line fitting result corresponding to the second block area.
5. The method of claim 4, wherein the lane line segmentation result includes a lane line instance, and the iteratively optimizing the lane line fitting curve corresponding to the first segmented region according to the lane line projection point corresponding to the second segmented region and the lane line segmentation result corresponding to the second segmented region includes:
determining fitting candidate points corresponding to the lane line examples of the second block areas according to the pixel points of the lane line examples corresponding to the second block areas;
matching the lane line instance of the second block area with the lane line instance of the first block area based on the fitting candidate point corresponding to the lane line instance of the second block area and the lane line projection point corresponding to the second block area;
and carrying out iterative optimization on the lane line fitting curve corresponding to the first block region according to the matching result to obtain a second lane line fitting result corresponding to the second block region.
6. The method of claim 5, wherein iteratively optimizing the lane line fitting curve corresponding to the first segmented region according to the matching result to obtain a second lane line fitting result corresponding to a second segmented region comprises:
Obtaining fitting candidate points corresponding to lane line examples of the first block area;
combining fitting candidate points corresponding to the lane line examples of the first block area and fitting candidate points corresponding to the lane line examples of the second block area according to the matching result;
and fitting the combined fitting candidate points by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the second block area.
7. The method of claim 6, wherein the fitting the combined fitting candidate points by using a preset lane line fitting algorithm to obtain a lane line fitting result corresponding to the second block region comprises:
if the matching result is that the matching is successful, determining that the lane line instance of the second block area corresponds to the lane line instance of the first block area, namely the same lane line, and fitting the fitting candidate points corresponding to the lane line instances of the same lane line;
if the matching result is that the matching is failed, determining that the lane line instance of the second block area is not the same lane line corresponding to the lane line instance of the first block area, and fitting the fitting candidate points corresponding to the lane line instance of the second block area.
8. A lane line identification apparatus, wherein the apparatus comprises:
the lane line segmentation unit is used for acquiring a current road image and carrying out lane line segmentation on the current road image to obtain a lane line segmentation result of the current road image;
the block unit is used for blocking the lane line segmentation result of the current road image by utilizing a preset block strategy to obtain lane line segmentation results corresponding to a plurality of block areas, wherein the plurality of block areas comprise a first block area and a second block area;
the determining unit is used for determining a lane line fitting result corresponding to the first block area according to the lane line segmentation result corresponding to the first block area;
and the optimization unit is used for carrying out iterative optimization on the lane line fitting result corresponding to the first block area by utilizing a preset iterative optimization strategy according to the lane line segmentation result corresponding to the second block area, so as to obtain a lane line identification result of the current road image.
9. An electronic device, comprising:
a processor; and
a memory arranged to store computer executable instructions which, when executed, cause the processor to perform the method of any of claims 1 to 7.
10. A computer readable storage medium storing one or more programs, which when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310027786.5A CN116309628A (en) | 2023-01-09 | 2023-01-09 | Lane line recognition method and device, electronic equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310027786.5A CN116309628A (en) | 2023-01-09 | 2023-01-09 | Lane line recognition method and device, electronic equipment and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116309628A true CN116309628A (en) | 2023-06-23 |
Family
ID=86796769
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310027786.5A Pending CN116309628A (en) | 2023-01-09 | 2023-01-09 | Lane line recognition method and device, electronic equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116309628A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116580373A (en) * | 2023-07-11 | 2023-08-11 | 广汽埃安新能源汽车股份有限公司 | Lane line optimization method and device, electronic equipment and storage medium |
-
2023
- 2023-01-09 CN CN202310027786.5A patent/CN116309628A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116580373A (en) * | 2023-07-11 | 2023-08-11 | 广汽埃安新能源汽车股份有限公司 | Lane line optimization method and device, electronic equipment and storage medium |
| CN116580373B (en) * | 2023-07-11 | 2023-09-26 | 广汽埃安新能源汽车股份有限公司 | Lane line optimization method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3620981B1 (en) | Object detection method, device, apparatus and computer-readable storage medium | |
| US11113546B2 (en) | Lane line processing method and device | |
| CN111582021B (en) | Text detection method and device in scene image and computer equipment | |
| CN112634209A (en) | Product defect detection method and device | |
| CN112329505A (en) | Method and apparatus for detecting an object | |
| CN111797711A (en) | Model training method and device | |
| CN116309628A (en) | Lane line recognition method and device, electronic equipment and computer readable storage medium | |
| CN114973198A (en) | Course angle prediction method and device of target vehicle, electronic equipment and storage medium | |
| US10803295B2 (en) | Method and device for face selection, recognition and comparison | |
| CN113804214B (en) | Vehicle positioning method and device, electronic equipment and computer readable storage medium | |
| CN112654999B (en) | Method and device for determining labeling information | |
| CN116740145A (en) | Multi-target tracking method, device, vehicle and storage medium | |
| CN116721396A (en) | Lane line detection method, device and storage medium | |
| CN116343148A (en) | Lane line detection method, device, vehicle and storage medium | |
| CN113111872B (en) | Training method and device of image recognition model, electronic equipment and storage medium | |
| CN114066958A (en) | Method and device for predicting depth information of target, electronic device and storage medium | |
| CN113763412B (en) | Image processing method and device, electronic equipment and computer readable storage medium | |
| CN116958954B (en) | License plate recognition method, device and storage medium based on key points and bypass correction | |
| US20230401691A1 (en) | Image defect detection method, electronic device and readable storage medium | |
| CN113807293B (en) | Deceleration strip detection method, deceleration strip detection system, deceleration strip detection equipment and computer readable storage medium | |
| CN113658268B (en) | Verification method and device for camera calibration result, electronic equipment and storage medium | |
| CN113689422B (en) | Image processing method and device and electronic equipment | |
| US20220019797A1 (en) | Methods and systems for agricultural block mapping | |
| US11055852B2 (en) | Fast automatic trimap generation and optimization for segmentation refinement | |
| US10395378B2 (en) | Detecting periodic patterns and aperture problems for motion estimation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |