CN113706622B - Road surface fitting method and system based on binocular stereo vision and intelligent terminal - Google Patents
Road surface fitting method and system based on binocular stereo vision and intelligent terminal Download PDFInfo
- Publication number
- CN113706622B CN113706622B CN202111270327.7A CN202111270327A CN113706622B CN 113706622 B CN113706622 B CN 113706622B CN 202111270327 A CN202111270327 A CN 202111270327A CN 113706622 B CN113706622 B CN 113706622B
- Authority
- CN
- China
- Prior art keywords
- road surface
- particles
- points
- motion state
- grid
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/579—Depth or shape recovery from multiple images from motion
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a road surface fitting method, a road surface fitting system and an intelligent terminal based on binocular stereo vision, wherein the method comprises the following steps: acquiring continuous multi-frame left and right views of the same road scene and corresponding disparity maps; extracting feature points based on two adjacent frames of images, establishing a rotation and translation matrix based on each feature point, and predicting the motion state of the particles through the rotation and translation matrix; carrying out grid projection on the disparity map, and calculating the occupancy rate and the vacancy rate in the grid projection range to obtain the posterior probability; and correcting the predicted particle motion state according to the posterior probability, and fitting a road surface model based on the corrected particle motion state. The method can acquire a relatively accurate road surface equation through effective high filtering, so that the accuracy of road surface fitting is improved.
Description
Technical Field
The invention relates to the technical field of automatic driving assistance, in particular to a road surface fitting method and system based on binocular stereoscopic vision and an intelligent terminal.
Background
With the development of automatic driving technology, people have increasingly higher requirements on safety and comfort of vehicles for assisting driving. The road surface detection is a key technology of an intelligent vehicle visual navigation system, and for realizing automatic driving and auxiliary driving in visual navigation, an algorithm needs to be dynamically adjusted according to the change condition of the road surface so as to complete the detection of various obstacles in a lane. The accurate road surface fitting effect can provide a high-quality detection basis for subsequent obstacle detection. The road surface fitting is usually completed by using a binocular disparity map, but due to the diversity of an unstructured road and the limitation of the parallax precision of the road, the road surface fitting is difficult to obtain a high-precision result. The most obvious problem is that the road surface obtained by fitting is not stable in height, because the input disparity map has certain error fluctuation, it is more intuitive that the concave-convex change condition of the road surface disparity far exceeds the fluctuation of the actual road surface, and the fluctuation change is difficult to eliminate by simply adopting multi-frame data fitting.
Therefore, a road surface fitting method based on binocular stereo vision is provided, so that a relatively accurate road surface equation is obtained through effective height filtering, the accuracy of road surface fitting is improved, and the problem to be solved by technical personnel in the field is solved urgently.
Disclosure of Invention
Therefore, the embodiment of the invention provides a road surface fitting method, a road surface fitting system and an intelligent terminal based on binocular stereo vision, so that a relatively accurate road surface equation is obtained through effective height filtering, and the accuracy of road surface fitting is improved.
In order to achieve the above object, the embodiments of the present invention provide the following technical solutions:
a road surface fitting method based on binocular stereo vision, the method comprises the following steps:
acquiring continuous multi-frame left and right views of the same road scene and corresponding disparity maps;
extracting feature points based on two adjacent frames of images, establishing a rotation and translation matrix based on each feature point, and predicting the motion state of the particles through the rotation and translation matrix;
carrying out grid projection on the disparity map, and calculating the occupancy rate and the vacancy rate in the grid projection range to obtain the posterior probability;
and correcting the predicted particle motion state according to the posterior probability, and fitting a road surface model based on the corrected particle motion state.
Further, the extracting of the feature points based on the adjacent front and rear frame images specifically includes:
respectively extracting feature points of a left eye view and a right eye view of two adjacent frames by using an SIFT algorithm to obtain feature points of four images;
and sequentially comparing the characteristics of the current frame left view and the previous frame left view, the previous frame left view and the previous frame right view, the previous frame right view and the current frame right view, and the current frame right view and the current frame left view to obtain the common matching characteristic points of the four images.
Further, the rotational-translation matrix is established using the following equation:
wherein the content of the first and second substances,
which represents the feature points of the current frame,
the feature points of the previous frame are represented,
is the rotation and translation matrix between two frames to be solved.
Further, the occupancy within the grid projection range is calculated using the following formula:
wherein the content of the first and second substances,
is the occupancy rate;
Further, the vacancy rate within the grid projection range is calculated using the following formula:
wherein the content of the first and second substances,
the empty rate.
Further, the modifying the predicted particle motion state according to the posterior probability specifically includes:
theory is calculated using the following formulaNumber of posterior particles
：
Wherein the content of the first and second substances,
an upper limit for the number of particles per grid;
calculating the ratio of the posterior particle number to the actual particle number by using the following formula
：
Wherein the content of the first and second substances,
is the number of particles on the actual grid;
if the ratio is
If the number of the particles is more than 1, all the particles in the corresponding grid are copied
Doubling;
if the ratio is
If the number of particles is less than 1, the existing particles are randomly deleted according to the uniform distribution U (0, 1).
Further, the road surface model is:
randomly selecting two groups of point pairs from all the road points, and calculating a road straight line;
calculating the distance from all points to the road surface straight line, recording the points with the distance less than a preset threshold value as inner points, and recording the number of the inner points of the straight line equation;
and repeating the steps for N times, selecting the point with the largest number of inner points, and solving again by using a least square method to obtain the pavement model.
The invention also provides a road surface fitting system based on binocular stereo vision, which comprises:
the scene image acquisition unit is used for acquiring continuous multi-frame left and right views of the same road scene and corresponding disparity maps;
the particle matrix calculation unit is used for extracting feature points based on adjacent front and back frames of images, establishing a rotation and translation matrix based on each feature point, and predicting the motion state of particles through the rotation and translation matrix;
the posterior probability calculation unit is used for carrying out grid projection on the disparity map and calculating the occupancy rate and the vacancy rate in the grid projection range to obtain the posterior probability;
and the road surface model fitting unit is used for correcting the predicted particle motion state according to the posterior probability and fitting the road surface model based on the corrected particle motion state.
According to the road surface fitting method, the road surface fitting system and the intelligent terminal based on the binocular stereo vision, the transformation matrix of the camera between two frames is calculated according to the characteristic points of the front frame image and the rear frame image, and therefore the motion state of the vehicle between the two frames is indirectly acquired. Based on the probability statistical principle, the incredible road surface height points are removed by utilizing the motion grids so as to improve the accuracy of the road surface fitting and achieve the aim of filtering the road surface height. Therefore, through effective high-speed filtering, a relatively accurate road surface equation is obtained, and the accuracy of road surface fitting is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It should be apparent that the drawings in the following description are merely exemplary, and that other embodiments can be derived from the drawings provided by those of ordinary skill in the art without inventive effort.
The structures, ratios, sizes, and the like shown in the present specification are only used for matching with the contents disclosed in the specification, so as to be understood and read by those skilled in the art, and are not used to limit the conditions that the present invention can be implemented, so that the present invention has no technical significance, and any structural modifications, changes in the ratio relationship, or adjustments of the sizes, without affecting the effects and the achievable by the present invention, should still fall within the range that the technical contents disclosed in the present invention can cover.
FIG. 1 is a flow chart of a road surface fitting method based on binocular stereo vision according to an embodiment of the present invention;
fig. 2 is a block diagram of a road surface fitting system based on binocular stereo vision according to an embodiment of the present invention.
Detailed Description
The present invention is described in terms of particular embodiments, other advantages and features of the invention will become apparent to those skilled in the art from the following disclosure, and it is to be understood that the described embodiments are merely exemplary of the invention and that it is not intended to limit the invention to the particular embodiments disclosed. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
According to the binocular stereoscopic vision-based road surface fitting method, the road surface points with low reliability are mainly deleted from the viewpoint of probability reliability according to the characteristics of the road surface height error points, so that a road surface equation with high precision is obtained, and the aim of road surface height filtering is fulfilled.
In one embodiment, as shown in fig. 1, the method for fitting a road surface based on binocular stereo vision provided by the invention comprises the following steps:
s1: and acquiring continuous multi-frame left and right views of the same road scene and corresponding disparity maps.
A series of left and right images and corresponding parallax images can be obtained through the binocular stereo camera so as to provide data support for the extraction and calculation of subsequent feature points, wherein the calculation of the feature points needs the left and right images and the left and right images of the previous frame.
S2: extracting feature points based on two adjacent frames of images, establishing a rotation and translation matrix based on each feature point, predicting the motion state of the particles through the rotation and translation matrix,
the feature point extraction based on two adjacent front and rear frame images specifically comprises the following steps:
respectively extracting feature points of a left eye view and a right eye view of two adjacent frames by using an SIFT algorithm to obtain feature points of four images;
and sequentially comparing the characteristics of the current frame left view and the previous frame left view, the previous frame left view and the previous frame right view, the previous frame right view and the current frame right view, and the current frame right view and the current frame left view to obtain the common matching characteristic points of the four images.
That is, in the actual algorithm, Feature points are first extracted using the SIFT (Scale-innovative Feature Transform) algorithm. SIFT is used for detecting local features in an image, searching extreme points in a spatial scale, and extracting the positions and the scales of the points. The algorithm has strong robustness to light, noise and slight visual angle change, and is very suitable for feature point extraction in the field of daily intelligent driving. After obtaining the feature points of the four images, feature point matching needs to be performed on every two images one by one, and the matching and calculation sequence is that the current frame left image and the previous frame left image, the previous frame left image and the previous frame right image, the previous frame right image and the current frame right image, and the current frame right image and the current frame left image. And after the common matching feature points of the four images are obtained, the next calculation can be carried out.
The core frame randomly selects three groups of point pairs from all matching points based on a RASAC (random sample consensus) algorithm, calculates a rotation translation matrix, transforms all matching points of a previous frame by the matrix, then calculates the Euclidean distance between the transformed points and the corresponding points of the current frame, records the points with the distance smaller than a certain threshold as interior points, and records the number of the interior points corresponding to the rotation translation matrix. And repeating the steps for N times, selecting the point pairs with the largest number of inner points, and solving again to obtain a more accurate rotation and translation matrix.
Thus, the rotational-translation matrix may be established using the following equation:
wherein the content of the first and second substances,
which represents the feature points of the current frame,
the feature points of the previous frame are represented,
is the rotation and translation matrix between two frames to be solved. An optimal roto-translation matrix needs to be optimized so that the above formula takes a minimum. The solution is generally performed using newton gaussians.
In step S2, the following embodiment is used for the calculation of the particle motion state.
Before the disparity map is particlized, the disparity map needs to be subjected to grid projection, and the grid projection can be understood as top view projection compression of raw data. For example, the vehicle is used as the origin, the left and right sides are wide by ± 6 meters, and the longitudinal direction is 60 meters as the detection range; the grid pattern is set to have a length and width of 0.2m per cell, and the pixel size of the grid pattern is 24 pixels 120 pixels. And (3) placing data points (marked as particles) into corresponding grids according to x and y coordinate information (the height information is not temporarily considered) in the parallax point three-dimensional information, and obtaining a grid map.
When focusing on successive frame images, the macroscopic manifestation of particle motion is the motion of the object. Due to the grid division, we only need to pay attention to the motion state of the grid. The motion state of each grid depends on the sum of its internal particle motion states. The algorithm already obtains the rotation and translation matrixes of the previous frame and the next frame, and then the position of the particle of the current frame in the next frame can be calculated according to the matrixes.
S3: and carrying out grid projection on the disparity map, and calculating the occupancy rate and the vacancy rate in the grid projection range to obtain the posterior probability.
According to the principle of epipolar geometry, the three-dimensional reconstruction system error of the binocular stereo camera is as follows:
wherein, x, z: respectively representing coordinate values of the space three-dimensional point p in the directions of the x axis and the z axis;
baseline: a baseline for a binocular camera;
focus: camera focal length (pixel).
Specifically, the occupancy rate within the grid projection range is calculated using the following formula:
wherein the content of the first and second substances,
is the occupancy rate;
Calculating the vacancy rate within the projection range of the grid using the following formula:
wherein the content of the first and second substances,
the empty rate.
S4: and correcting the predicted particle motion state according to the posterior probability, and fitting a road surface model based on the corrected particle motion state. The particle state prediction is calculated by the prior of the system, and the probability calculation reflects the posterior probability of the actual system. And (3) correcting the deviation of the particle motion according to the posterior probability, so that the whole hypothesis model (motion grid) gradually approaches to a real system. The method of correcting the particle motion deviation is to adjust the number of particles on the motion grid to be the same as the number of particles expressed by the posterior probability.
Specifically, the modifying the predicted particle motion state according to the posterior probability specifically includes:
the theoretical posterior particle number is calculated by the following formula
：
Wherein the content of the first and second substances,
an upper limit for the number of particles per grid;
calculating the ratio of the posterior particle number to the actual particle number by using the following formula
：
Wherein the content of the first and second substances,
is the number of particles on the actual grid;
if the ratio is
If the number of the particles is more than 1, all the particles in the corresponding grid are copied
Doubling;
if the ratio is
If the number of particles is less than 1, the existing particles are randomly deleted according to the uniform distribution U (0, 1).
Further, the road surface model is:
randomly selecting two groups of point pairs from all the road points, and calculating a road straight line;
calculating the distance from all points to the road surface straight line, recording the points with the distance less than a preset threshold value as inner points, and recording the number of the inner points of the straight line equation;
and repeating the steps for N times, selecting the point with the largest number of inner points, and solving again by using a least square method to obtain the pavement model.
After probability resampling, most wrong road surface parallax points are deleted, the number of correct and reliable parallax points is increased, and at the moment, a RASAC algorithm is used for performing road surface straight line fitting.
The specific method comprises the following steps: randomly selecting two groups of point pairs from all the road surface points, calculating a road surface straight line, then calculating the distance from all the points to the road surface straight line, recording the points with the distance less than a certain threshold value as inner points, and recording the number of the inner points of the straight line equation. And repeating the steps for N times, selecting the point with the largest number of inner points, and solving again by using a least square method to obtain a more accurate road surface linear equation.
The road surface equation is of the general formula y = kx + b. The least square method comprises the following calculation modes:
in the above embodiment, the road surface fitting method, system and intelligent terminal based on binocular stereo vision provided by the invention calculate the transformation matrix of the camera between two frames according to the feature points of the front and rear frame images, thereby indirectly acquiring the motion state of the vehicle between the two frames. Based on the probability statistical principle, the incredible road surface height points are removed by utilizing the motion grids so as to improve the accuracy of the road surface fitting and achieve the aim of filtering the road surface height. Therefore, through effective high-speed filtering, a relatively accurate road surface equation is obtained, and the accuracy of road surface fitting is improved.
In addition to the above method, the present invention also provides a road surface fitting system based on binocular stereo vision, as shown in fig. 2, the system includes:
a scene image acquiring unit 100, configured to acquire multiple continuous frames of left and right views of the same road scene and corresponding disparity maps;
the particle matrix calculation unit 200 is configured to perform feature point extraction based on two adjacent frames of images, establish a rotation-translation matrix based on each feature point, and predict a particle motion state through the rotation-translation matrix;
the posterior probability calculating unit 300 is configured to perform grid projection on the disparity map, and calculate an occupancy rate and a vacancy rate within the grid projection range to obtain a posterior probability;
and a road surface model fitting unit 400, configured to correct the predicted particle motion state according to the posterior probability, and fit a road surface model based on the corrected particle motion state.
In the above embodiment, the road surface fitting system based on binocular stereo vision provided by the invention calculates the transformation matrix of the camera between two frames according to the feature points of the front and rear frame images, thereby indirectly acquiring the motion state of the vehicle between the two frames. Based on the probability statistical principle, the incredible road surface height points are removed by utilizing the motion grids so as to improve the accuracy of the road surface fitting and achieve the aim of filtering the road surface height. Therefore, through effective high-speed filtering, a relatively accurate road surface equation is obtained, and the accuracy of road surface fitting is improved.
The present invention also provides an intelligent terminal, including: the device comprises a data acquisition device, a processor and a memory;
the data acquisition device is used for acquiring data; the memory is to store one or more program instructions; the processor is configured to execute one or more program instructions to perform the method as described above.
In correspondence with the above embodiments, embodiments of the present invention also provide a computer storage medium containing one or more program instructions therein. Wherein the one or more program instructions are for executing the method as described above by a binocular camera depth calibration system.
In an embodiment of the invention, the processor may be an integrated circuit chip having signal processing capability. The Processor may be a general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete Gate or transistor logic device, discrete hardware component.
The various methods, steps and logic blocks disclosed in the embodiments of the present invention 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 the method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The processor reads the information in the storage medium and completes the steps of the method in combination with the hardware.
The storage medium may be a memory, for example, which may be volatile memory or nonvolatile memory, or which may include both volatile and nonvolatile memory.
The nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash Memory.
The volatile Memory may be a Random Access Memory (RAM) which serves as an external cache. By way of example and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), SLDRAM (SLDRAM), and Direct Rambus RAM (DRRAM).
The storage media described in connection with the embodiments of the invention are intended to comprise, without being limited to, these and any other suitable types of memory.
Those skilled in the art will appreciate that the functionality described in the present invention may be implemented in a combination of hardware and software in one or more of the examples described above. When software is applied, the corresponding functionality may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
The above embodiments are only for illustrating the embodiments of the present invention and are not to be construed as limiting the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made on the basis of the embodiments of the present invention shall be included in the scope of the present invention.
Claims (8)
1. A road surface fitting method based on binocular stereo vision is characterized by comprising the following steps:
acquiring continuous multi-frame left and right views of the same road scene and corresponding disparity maps;
extracting feature points based on two adjacent frames of images, establishing a rotation and translation matrix based on each feature point, and predicting the motion state of the particles through the rotation and translation matrix;
carrying out grid projection on the disparity map, and calculating the occupancy rate and the vacancy rate in the grid projection range to obtain the posterior probability;
correcting the predicted particle motion state according to the posterior probability, and fitting a road surface model based on the corrected particle motion state;
obtaining the number of posterior particles according to the occupancy rate and the vacancy rate in the parallax map grid projection range, correcting the actual particles according to the acquired number of posterior particles, and eliminating the incredible road height points by utilizing the motion grids;
predicting the motion state of the particles through the rotation and translation matrix specifically comprises the following steps:
before the granulation of the disparity map, grid projection is needed to be carried out on the disparity map, and the grid projection is top view projection compression of original data; obtaining a grid image according to x and y coordinate information in the parallax point three-dimensional information, namely, the particles are placed in the corresponding grids;
when focusing on successive frame images, the macroscopic representation of the particle motion is the motion of the object, the motion state of each grid depends on the motion state of its internal particles, and the position of the particle of the current frame in the next frame is calculated according to the rotation-translation matrix.
2. The road surface fitting method according to claim 1, wherein the feature point extraction based on two adjacent frames of images comprises:
respectively extracting feature points of a left eye view and a right eye view of two adjacent frames by using an SIFT algorithm to obtain feature points of four images;
and sequentially comparing the characteristics of the current frame left view and the previous frame left view, the previous frame left view and the previous frame right view, the previous frame right view and the current frame right view, and the current frame right view and the current frame left view to obtain the common matching characteristic points of the four images.
3. A road surface fitting method according to claim 1, wherein the vacancy rate in the grid projection range is calculated using the following formula:
wherein the content of the first and second substances,
in order to obtain the empty rate,
represents the occupancy of the grid points (r, c).
4. The road surface fitting method according to claim 1, wherein the correcting the predicted particle motion state according to the posterior probability specifically includes:
if the ratio is
If the number of particles is less than 1, the existing particles are randomly deleted according to the uniform distribution U (0, 1).
5. A road surface fitting method according to claim 1, wherein the road surface model is:
randomly selecting two groups of point pairs from all the road points, and calculating a road straight line;
calculating the distance from all points to the road surface straight line, recording the points with the distance less than a preset threshold value as inner points, and recording the number of the inner points of the road surface straight line;
and repeating the steps for N times, selecting the point with the largest number of inner points, and solving again by using a least square method to obtain the pavement model.
6. A road surface fitting system based on binocular stereo vision is characterized in that the system comprises:
the scene image acquisition unit is used for acquiring continuous multi-frame left and right views of the same road scene and corresponding disparity maps;
the particle matrix calculation unit is used for extracting feature points based on adjacent front and back frames of images, establishing a rotation and translation matrix based on each feature point, and predicting the motion state of particles through the rotation and translation matrix;
the posterior probability calculation unit is used for carrying out grid projection on the disparity map and calculating the occupancy rate and the vacancy rate in the grid projection range to obtain the posterior probability;
the road surface model fitting unit is used for correcting the predicted particle motion state according to the posterior probability and fitting a road surface model based on the corrected particle motion state;
obtaining the number of posterior particles according to the occupancy rate and the vacancy rate in the parallax map grid projection range, correcting the actual particles according to the acquired number of posterior particles, and eliminating the incredible road height points by utilizing the motion grids;
before the granulation of the disparity map, grid projection is needed to be carried out on the disparity map, and the grid projection is top view projection compression of original data; obtaining a grid image according to x and y coordinate information in the parallax point three-dimensional information, namely, the particles are placed in the corresponding grids;
when focusing on successive frame images, the macroscopic representation of the particle motion is the motion of the object, the motion state of each grid depends on the motion state of its internal particles, and the position of the particle of the current frame in the next frame is calculated according to the rotation-translation matrix.
7. An intelligent terminal, characterized in that, intelligent terminal includes: the device comprises a data acquisition device, a processor and a memory;
the data acquisition device is used for acquiring data; the memory is to store one or more program instructions; the processor, configured to execute one or more program instructions to perform the method of any of claims 1-5.
8. A computer-readable storage medium having one or more program instructions embodied therein for performing the method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111270327.7A CN113706622B (en) | 2021-10-29 | 2021-10-29 | Road surface fitting method and system based on binocular stereo vision and intelligent terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111270327.7A CN113706622B (en) | 2021-10-29 | 2021-10-29 | Road surface fitting method and system based on binocular stereo vision and intelligent terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113706622A CN113706622A (en) | 2021-11-26 |
| CN113706622B true CN113706622B (en) | 2022-04-19 |
Family
ID=78647670
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111270327.7A Active CN113706622B (en) | 2021-10-29 | 2021-10-29 | Road surface fitting method and system based on binocular stereo vision and intelligent terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113706622B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112906449A (en) * | 2020-12-02 | 2021-06-04 | 北京中科慧眼科技有限公司 | Dense disparity map-based road surface pothole detection method, system and equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11069082B1 (en) * | 2015-08-23 | 2021-07-20 | AI Incorporated | Remote distance estimation system and method |
| WO2018058356A1 (en) * | 2016-09-28 | 2018-04-05 | 驭势科技（北京）有限公司 | Method and system for vehicle anti-collision pre-warning based on binocular stereo vision |
| CN107945220B (en) * | 2017-11-30 | 2020-07-10 | 华中科技大学 | Binocular vision-based reconstruction method |
-
2021
- 2021-10-29 CN CN202111270327.7A patent/CN113706622B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112906449A (en) * | 2020-12-02 | 2021-06-04 | 北京中科慧眼科技有限公司 | Dense disparity map-based road surface pothole detection method, system and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113706622A (en) | 2021-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11010924B2 (en) | Method and device for determining external parameter of stereoscopic camera | |
| CN113240632B (en) | Pavement detection method and system based on semantic segmentation network and intelligent terminal | |
| CN111160232B (en) | Front face reconstruction method, device and system | |
| US11461911B2 (en) | Depth information calculation method and device based on light-field-binocular system | |
| US20200077073A1 (en) | Real-time stereo calibration by direct disparity minimization and keypoint accumulation | |
| US20220044433A1 (en) | Method and system for distance measurement based on binocular camera, device and computer-readable storage medium | |
| CN114495043B (en) | Method and system for detecting up-and-down slope road conditions based on binocular vision system and intelligent terminal | |
| CN112465831B (en) | Bend scene sensing method, system and device based on binocular stereo camera | |
| CN111295667A (en) | Image stereo matching method and driving assisting device | |
| CN111225201B (en) | Parallax correction method and device, and storage medium | |
| CN113240631B (en) | Road surface detection method and system based on RGB-D fusion information and intelligent terminal | |
| CN113781543B (en) | Binocular camera-based height limiting device detection method and system and intelligent terminal | |
| CN113965742A (en) | Dense disparity map extraction method and system based on multi-sensor fusion and intelligent terminal | |
| CN113140002B (en) | Road condition detection method and system based on binocular stereo camera and intelligent terminal | |
| CN114119777A (en) | Stereo matching method and system based on deep learning | |
| CN113706622B (en) | Road surface fitting method and system based on binocular stereo vision and intelligent terminal | |
| CN113674275B (en) | Dense disparity map-based road surface unevenness detection method and system and intelligent terminal | |
| CN113763303B (en) | Real-time ground fusion method and system based on binocular stereo vision and intelligent terminal | |
| CN115100621A (en) | Ground scene detection method and system based on deep learning network | |
| CN115908551A (en) | Vehicle distance measuring method and device, electronic equipment and storage medium | |
| CN114049307A (en) | Road surface height detection method and system based on binocular stereoscopic vision and intelligent terminal | |
| CN110148086B (en) | Depth filling method and device for sparse depth map and three-dimensional reconstruction method and device | |
| CN113657205B (en) | Obstacle detection method and system based on motion grid and intelligent terminal | |
| CN116030145A (en) | Stereo matching method and system for binocular lenses with different focal lengths | |
| CN113129313A (en) | Dense matching algorithm, system and intelligent terminal based on superpixel |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |