CN114801988A - Automobile reversing anti-collision system based on ADAS recognition of automobile rearview mirror - Google Patents
Automobile reversing anti-collision system based on ADAS recognition of automobile rearview mirror Download PDFInfo
- Publication number
- CN114801988A CN114801988A CN202210436110.7A CN202210436110A CN114801988A CN 114801988 A CN114801988 A CN 114801988A CN 202210436110 A CN202210436110 A CN 202210436110A CN 114801988 A CN114801988 A CN 114801988A
- Authority
- CN
- China
- Prior art keywords
- module
- image
- scene
- pixel points
- vehicle
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Q—ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL
- B60Q9/00—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling
- B60Q9/008—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling for anti-collision purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Remote Sensing (AREA)
- Computer Graphics (AREA)
- Human Computer Interaction (AREA)
- Mechanical Engineering (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an automobile reversing anti-collision system based on vehicle rearview mirror ADAS identification, which comprises: the acquisition module is used for acquiring scene information behind and beside the target vehicle; the establishing module is used for establishing a 3D stereoscopic scene model according to the scene information; the filling module is used for filling the established target vehicle model in the 3D stereoscopic scene model and inputting the backing track of the target vehicle; and the alarm module is used for sending an alarm prompt when the reversing track is determined to be crossed with the moving track of the dynamic composition elements in the 3D stereoscopic scene model and the time for the target vehicle to collide is less than a preset time threshold. The method has the advantages that the periphery of the vehicle is comprehensively monitored, the 3D scene model is established, a user can comprehensively and accurately determine the specific conditions around the vehicle, the vehicle backing track is accurately predicted in the process of backing the vehicle, an alarm prompt is timely sent out when the situation that collision is about to occur is determined, and the safety of the vehicle is guaranteed.
Description
Technical Field
The invention relates to the technical field of vehicle safety, in particular to an automobile reversing anti-collision system based on ADAS recognition of a vehicle rearview mirror.
Background
The vehicle has great blind area in the process of backing a car, and traffic accidents easily occur. In the prior art, a reversing radar is arranged behind a vehicle to monitor a target object behind the vehicle. However, the reverse sensor has a small rear area of a detected vehicle, is inaccurate in detecting a moving object, lacks monitoring on the side of the vehicle, and is easy to collide when reversing.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the art described above. Therefore, the invention aims to provide an automobile reversing anti-collision system based on vehicle rearview mirror ADAS recognition, which can comprehensively monitor the periphery of a vehicle and establish a 3D (three-dimensional) scene model, so that a user can comprehensively and accurately determine the specific conditions of the periphery, can accurately predict based on a reversing track in the reversing process, and can timely send out an alarm prompt when collision is determined to happen, thereby ensuring the safety of the vehicle.
In order to achieve the above object, an embodiment of the present invention provides an automobile reverse collision avoidance system based on vehicle rearview mirror ADAS identification, including:
the acquisition module is used for acquiring scene information behind and beside the target vehicle;
the establishing module is used for establishing a 3D stereoscopic scene model according to the scene information;
the filling module is used for filling the established target vehicle model in the 3D stereoscopic scene model and inputting the backing track of the target vehicle;
and the alarm module is used for sending an alarm prompt when the reversing track is determined to be crossed with the moving track of the dynamic composition elements in the 3D stereoscopic scene model and the time for the target vehicle to collide is less than a preset time threshold.
According to some embodiments of the invention, the obtaining module comprises:
the camera module is used for acquiring image signals around the target vehicle;
and the radar module is used for acquiring radar signals around the target vehicle.
According to some embodiments of the invention, the establishing module comprises:
the recognition submodule is used for analyzing the image signal, determining a scene image, inputting the scene image into a pre-trained image recognition model and outputting a recognized combined element;
the first determining submodule is used for determining the size information of each combined element according to the number of pixel points included in each combined element and the actual length information corresponding to the preset pixel points;
the second determining submodule is used for analyzing the radar signal and determining the distance between each combination element;
and the establishing submodule is used for setting a three-dimensional coordinate system and establishing a 3D stereoscopic scene model based on the size information of each combined element and the distance between the combined elements.
According to some embodiments of the invention, the filling module comprises:
the third determining submodule is used for determining the positioning information of the target vehicle in the actual scene;
and the filling sub-module is used for filling the target vehicle model according to the positioning information.
According to some embodiments of the invention, the camera module is a binocular camera.
According to some embodiments of the invention, the radar module is a millimeter wave radar.
According to some embodiments of the invention, the alarm module is an audible and visual alarm.
According to some embodiments of the invention, further comprising: a signal processing module to:
before the second determining module analyzes the radar signal, performing feature extraction on the radar signal to obtain a feature signal and an edge signal;
carrying out low-pass filtering processing on the edge signal to obtain a filtering edge signal;
establishing a feature extraction network model based on a residual block structure in the ResNet-34 network;
inputting the characteristic signal into a characteristic extraction network model, outputting a first characteristic matrix, and acquiring first attribute information of the first characteristic matrix;
inputting the filtering edge signal into a feature extraction network model, outputting a second feature matrix, and acquiring second attribute information of the second feature matrix;
performing matrix flattening and deformation operation according to the first attribute information and the second attribute information, and fusing the first feature matrix and the second feature matrix to obtain a third feature matrix;
and obtaining a noise reduction radar signal according to the third feature matrix.
According to some embodiments of the invention, further comprising:
the abnormal pixel point identification module is used for:
before the recognition submodule inputs the scene image into a pre-trained image recognition model, converting the scene image into an HSV color space to obtain a first characteristic value, wherein the first characteristic value comprises an H channel value, an S channel value and a V channel value;
converting the scene image into an RGB color space to obtain a second characteristic value, wherein the second characteristic value comprises an R channel value, a G channel value and a B channel value;
comparing a first characteristic value and a second characteristic value corresponding to the same pixel point with a preset first characteristic value and a preset second characteristic value respectively to obtain a first difference value and a second difference value;
when the first difference value is determined to be smaller than a preset first difference value and the second difference value is determined to be smaller than a preset second difference value, indicating that the pixel point is normal; otherwise, taking the pixel points as the suspected pixel points to carry out the next detection;
taking a scene image comprising in-doubt pixel points as an in-doubt image, determining scene content corresponding to the in-doubt image, and shooting the scene content for multiple times to obtain a plurality of contrast images;
determining pixel points in the comparison image corresponding to the in-doubt pixel points in the in-doubt image as comparison pixel points;
dynamically monitoring according to the contrast pixel points and the in-doubt pixel points to obtain dynamic pixel difference values, and when the dynamic difference values are determined to be larger than preset difference values, indicating that the in-doubt pixel points are abnormal pixel points;
and the correction module is used for correcting the abnormal pixel points.
According to some embodiments of the invention, further comprising:
the system comprises a driver detection module, a driver detection module and a driver detection module, wherein the driver detection module is used for acquiring a driving image of a driver when a target vehicle backs, identifying the driving image, determining a driving behavior and judging whether the driving behavior is an abnormal driving behavior;
and the alarm module is used for sending out an alarm prompt when the driver detection module determines that the driving behavior is the abnormal driving behavior.
The automobile reversing anti-collision system based on the ADAS recognition of the rearview mirrors of the automobile, provided by the invention, can realize comprehensive monitoring on the periphery of the automobile, and establish a 3D (three-dimensional) scene model, so that a user can comprehensively and accurately determine the specific conditions on the periphery, accurately predict based on the reversing track in the reversing process, and timely send out an alarm prompt when collision is determined to occur, thereby ensuring the safety of the automobile.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
fig. 1 is a block diagram of an automotive reverse collision avoidance system based on vehicle rearview mirror ADAS identification according to one embodiment of the present invention;
FIG. 2 is a block diagram of an acquisition module according to one embodiment of the invention;
FIG. 3 is a block diagram of a setup module according to one embodiment of the invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
As shown in fig. 1, an embodiment of the present invention provides an automobile reverse collision avoidance system based on vehicle rearview mirror ADAS identification, including:
the acquisition module is used for acquiring scene information behind and beside the target vehicle;
the establishing module is used for establishing a 3D stereoscopic scene model according to the scene information;
the filling module is used for filling the established target vehicle model in the 3D stereoscopic scene model and inputting the backing track of the target vehicle;
and the alarm module is used for sending an alarm prompt when the reversing track is determined to be crossed with the moving track of the dynamic composition elements in the 3D stereoscopic scene model and the time for the target vehicle to collide is less than a preset time threshold.
The working principle of the technical scheme is as follows: the acquisition module is used for acquiring scene information behind and beside the target vehicle; the scene information includes the rear and side of the target vehicle, radar signals and image signals. The establishing module is used for establishing a 3D stereoscopic scene model according to the scene information; the filling module is used for filling the established target vehicle model in the 3D stereoscopic scene model and inputting the backing track of the target vehicle; and the alarm module is used for sending an alarm prompt when the reversing track is determined to be crossed with the moving track of the dynamic composition elements in the 3D stereoscopic scene model and the time for the target vehicle to collide is less than a preset time threshold.
The beneficial effects of the above technical scheme are that: the method has the advantages that the periphery of the vehicle is comprehensively monitored, the 3D scene model is established, a user can comprehensively and accurately determine the specific conditions around the vehicle, the vehicle backing track is accurately predicted in the process of backing the vehicle, an alarm prompt is timely sent out when the situation that collision is about to occur is determined, and the safety of the vehicle is guaranteed.
As shown in fig. 2, according to some embodiments of the invention, the obtaining module includes:
the camera module is used for acquiring image signals around the target vehicle;
and the radar module is used for acquiring radar signals around the target vehicle.
The beneficial effects of the above technical scheme are that: based on camera module and radar module accurately obtain the image signal and the radar signal in vehicle rear and side, improve the accuracy of environmental perception.
As shown in fig. 3, according to some embodiments of the invention, the establishing module includes:
the recognition submodule is used for analyzing the image signal, determining a scene image, inputting the scene image into a pre-trained image recognition model and outputting a recognized combined element;
the first determining submodule is used for determining the size information of each combined element according to the number of pixel points included in each combined element and actual length information corresponding to preset pixel points;
the second determining submodule is used for analyzing the radar signal and determining the distance between each combination element;
and the establishing submodule is used for setting a three-dimensional coordinate system and establishing a 3D stereoscopic scene model based on the size information of each combined element and the distance between the combined elements.
The working principle of the technical scheme is as follows: the recognition submodule is used for analyzing the image signal, determining a scene image, inputting the scene image into a pre-trained image recognition model and outputting a recognized combined element; the composition elements include static composition elements and dynamic composition elements. Static composite elements include walls, stone piers, and the like. Dynamic composition elements include moving vehicles, pedestrians, etc. The first determining submodule is used for determining the size information of each combined element according to the number of pixel points included in each combined element and the actual length information corresponding to the preset pixel points; the second determining submodule is used for analyzing the radar signals and determining the distance between each combination element; and the establishing submodule is used for setting a three-dimensional coordinate system and establishing a 3D stereoscopic scene model based on the size information of each combined element and the distance between the combined elements.
The beneficial effects of the above technical scheme are that: and accurately identifying the image signal and determining the size information of each combined element. Accurately analyzing the radar signals and determining the distance between each combination element; the method and the device can accurately acquire the relevant parameters for establishing the 3D stereoscopic scene model, and improve the accuracy for establishing the 3D stereoscopic scene model.
According to some embodiments of the invention, the filling module comprises:
the third determining submodule is used for determining the positioning information of the target vehicle in the actual scene;
and the filling sub-module is used for filling the target vehicle model according to the positioning information.
The working principle of the technical scheme is as follows: the filling module comprises: the third determining submodule is used for determining the positioning information of the target vehicle in the actual scene; and the filling sub-module is used for filling the target vehicle model according to the positioning information.
The beneficial effects of the above technical scheme are that: the simulation of the actual environment of the target vehicle according to the target vehicle model and the 3D scene model is realized, and the user can sense the actual environment more visually and stereoscopically.
According to some embodiments of the invention, the camera module is a binocular camera.
According to some embodiments of the invention, the radar module is a millimeter wave radar.
According to some embodiments of the invention, the alarm module is an audible and visual alarm.
According to some embodiments of the invention, further comprising: a signal processing module to:
before the second determining module analyzes the radar signal, performing feature extraction on the radar signal to obtain a feature signal and an edge signal;
carrying out low-pass filtering processing on the edge signal to obtain a filtering edge signal;
establishing a feature extraction network model based on a residual block structure in the ResNet-34 network;
inputting the characteristic signal into a characteristic extraction network model, outputting a first characteristic matrix, and acquiring first attribute information of the first characteristic matrix;
inputting the filtering edge signal into a feature extraction network model, outputting a second feature matrix, and acquiring second attribute information of the second feature matrix;
performing matrix flattening and deformation operation according to the first attribute information and the second attribute information, and fusing the first feature matrix and the second feature matrix to obtain a third feature matrix;
and obtaining a noise reduction radar signal according to the third feature matrix.
The working principle of the technical scheme is as follows: before the second determining module analyzes the radar signal, performing feature extraction on the radar signal to obtain a feature signal and an edge signal; and extracting the features of the radar signals to obtain feature signals. And taking signals except the characteristic signals in the radar signals as edge signals. The edge signal includes a more noisy signal. Performing low-pass filtering processing on the edge signal to obtain a filtering edge signal; establishing a feature extraction network model based on a residual block structure in the ResNet-34 network; inputting the characteristic signal into a characteristic extraction network model, outputting a first characteristic matrix, and acquiring first attribute information of the first characteristic matrix; the first attribute information includes the size and element association of the first feature matrix. Inputting the filtering edge signal into a feature extraction network model, outputting a second feature matrix, and acquiring second attribute information of the second feature matrix; the second attribute information includes the size and element association of the second feature matrix. Performing matrix flattening and deformation operation according to the first attribute information and the second attribute information, and fusing the first feature matrix and the second feature matrix to obtain a third feature matrix; and obtaining a noise reduction radar signal according to the third feature matrix.
The beneficial effects of the above technical scheme are that: and accurately determining the edge signal, and performing filtering processing to improve the signal-to-noise ratio. The method comprises the steps of constructing a feature extraction network which takes a residual error network as a main framework, improving the effect of feature extraction of signals, reducing the calculated amount and the calculation complexity, accurately outputting corresponding feature matrixes, carrying out corresponding processing based on attribute information of the feature matrixes, realizing matrix fusion, and obtaining fusion signals, namely noise reduction radar signals, according to the fusion matrixes. The accuracy of signal reconstruction is improved, and compared with the signal reconstruction in the prior art, the efficiency is more efficient and simplified. The second determination module analyzes the noise-reduction radar signal, so that the accuracy of analysis is ensured.
According to some embodiments of the invention, further comprising:
an abnormal pixel point identification module, configured to:
before the recognition submodule inputs the scene image into a pre-trained image recognition model, converting the scene image into an HSV color space to obtain a first characteristic value, wherein the first characteristic value comprises an H channel value, an S channel value and a V channel value;
converting the scene image into an RGB color space to obtain a second characteristic value, wherein the second characteristic value comprises an R channel value, a G channel value and a B channel value;
comparing a first characteristic value and a second characteristic value corresponding to the same pixel point with a preset first characteristic value and a preset second characteristic value respectively to obtain a first difference value and a second difference value;
when the first difference value is determined to be smaller than a preset first difference value and the second difference value is determined to be smaller than a preset second difference value, indicating that the pixel point is normal; otherwise, taking the pixel points as the suspected pixel points to carry out the next detection;
taking a scene image comprising in-doubt pixel points as an in-doubt image, determining scene content corresponding to the in-doubt image, and shooting the scene content for multiple times to obtain a plurality of contrast images;
determining pixel points in the comparison image corresponding to the in-doubt pixel points in the in-doubt image as comparison pixel points;
dynamically monitoring according to the contrast pixel points and the in-doubt pixel points to obtain dynamic pixel difference values, and when the dynamic difference values are determined to be larger than preset difference values, indicating that the in-doubt pixel points are abnormal pixel points;
and the correction module is used for correcting the abnormal pixel points.
The working principle of the technical scheme is as follows: an abnormal pixel point identification module, configured to: before the recognition submodule inputs the scene image into a pre-trained image recognition model, converting the scene image into an HSV color space to obtain a first characteristic value, wherein the first characteristic value comprises an H channel value, an S channel value and a V channel value; converting the scene image into an RGB color space to obtain a second characteristic value, wherein the second characteristic value comprises an R channel value, a G channel value and a B channel value; comparing a first characteristic value and a second characteristic value corresponding to the same pixel point with a preset first characteristic value and a preset second characteristic value respectively to obtain a first difference value and a second difference value; when the first difference value is determined to be smaller than a preset first difference value and the second difference value is determined to be smaller than a preset second difference value, indicating that the pixel point is normal; otherwise, taking the pixel points as the suspected pixel points to carry out the next detection; taking a scene image comprising in-doubt pixel points as an in-doubt image, determining scene content corresponding to the in-doubt image, and shooting the scene content for multiple times to obtain a plurality of contrast images; determining pixel points in the comparison image corresponding to the in-doubt pixel points in the in-doubt image as comparison pixel points; dynamically monitoring according to the contrast pixel points and the in-doubt pixel points to obtain dynamic pixel difference values, and when the dynamic difference values are determined to be larger than preset difference values, indicating that the in-doubt pixel points are abnormal pixel points; and the correction module is used for correcting the abnormal pixel points. The dynamic pixel difference value is an average difference value calculated according to the difference value between the pixel value of the contrast pixel point and the pixel value of the in-doubt pixel point.
The beneficial effects of the above technical scheme are that: based on the conversion of the scene image into the HSV color space and the RGB color space, the pixel points are conveniently subjected to primary screening in the chromaticity direction, and the suspected pixel points are determined. And dynamically monitoring the in-doubt pixel points, realizing secondary screening, accurately judging whether the pixel points are abnormal pixel points, and correcting when the pixel points are determined to be abnormal pixel points, so that the accuracy of the scene image of the input image recognition model is ensured.
According to some embodiments of the invention, further comprising:
the driver detection module is used for acquiring a driving image of a driver when the target vehicle backs a car, identifying the driving image, determining a driving behavior and judging whether the driving behavior is an abnormal driving behavior or not;
and the alarm module is used for sending out an alarm prompt when the driver detection module determines that the driving behavior is the abnormal driving behavior.
The working principle of the technical scheme is as follows: the system comprises a driver detection module, a driver detection module and a driver detection module, wherein the driver detection module is used for acquiring a driving image of a driver when a target vehicle backs, identifying the driving image, determining a driving behavior and judging whether the driving behavior is an abnormal driving behavior; and the alarm module is used for sending out an alarm prompt when the driver detection module determines that the driving behavior is the abnormal driving behavior.
The beneficial effects of the above technical scheme are that: the driving behavior of a driver is normalized in the process of backing a car, and collision is avoided.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. The utility model provides an automobile backing anticollision system based on vehicle rearview mirror ADAS discernment which characterized in that includes:
the acquisition module is used for acquiring scene information behind and beside the target vehicle;
the establishing module is used for establishing a 3D stereoscopic scene model according to the scene information;
the filling module is used for filling the established target vehicle model in the 3D stereoscopic scene model and inputting the backing track of the target vehicle;
and the alarm module is used for sending an alarm prompt when the reversing track is determined to be crossed with the moving track of the dynamic composition elements in the 3D stereoscopic scene model and the time for the target vehicle to collide is less than a preset time threshold.
2. The vehicle rearview mirror ADAS identification based automobile reverse collision avoidance system of claim 1, wherein said acquisition module comprises:
the camera module is used for acquiring image signals around the target vehicle;
and the radar module is used for acquiring radar signals around the target vehicle.
3. The vehicle rearview mirror ADAS identification based car reverse collision avoidance system of claim 2, wherein said establishing module comprises:
the recognition submodule is used for analyzing the image signal, determining a scene image, inputting the scene image into a pre-trained image recognition model and outputting a recognized combined element;
the first determining submodule is used for determining the size information of each combined element according to the number of pixel points included in each combined element and the actual length information corresponding to the preset pixel points;
the second determining submodule is used for analyzing the radar signal and determining the distance between each combination element;
and the establishing submodule is used for setting a three-dimensional coordinate system and establishing a 3D stereoscopic scene model based on the size information of each combined element and the distance between the combined elements.
4. The vehicle rearview mirror ADAS identification based automotive reverse collision avoidance system of claim 1, wherein said filling module comprises:
the third determining submodule is used for determining the positioning information of the target vehicle in the actual scene;
and the filling sub-module is used for filling the target vehicle model according to the positioning information.
5. The vehicle rearview mirror ADAS identification based automotive reverse anti-collision system as claimed in claim 2, wherein said camera module is a binocular camera.
6. The vehicle rearview mirror ADAS identification based automotive reverse anti-collision system as claimed in claim 2, wherein said radar module is a millimeter wave radar.
7. The system of claim 1, wherein the alarm module is an audible and visual alarm.
8. The vehicle rearview mirror ADAS identification based car reverse collision avoidance system of claim 3, further comprising: a signal processing module to:
before the second determining module analyzes the radar signal, performing feature extraction on the radar signal to obtain a feature signal and an edge signal;
carrying out low-pass filtering processing on the edge signal to obtain a filtering edge signal;
establishing a feature extraction network model based on a residual block structure in the ResNet-34 network;
inputting the characteristic signal into a characteristic extraction network model, outputting a first characteristic matrix, and acquiring first attribute information of the first characteristic matrix;
inputting the filtering edge signal into a feature extraction network model, outputting a second feature matrix, and acquiring second attribute information of the second feature matrix;
performing matrix flattening and deformation operation according to the first attribute information and the second attribute information, and fusing the first feature matrix and the second feature matrix to obtain a third feature matrix;
and obtaining a noise reduction radar signal according to the third feature matrix.
9. The vehicle rearview mirror ADAS identification based automotive reverse collision avoidance system of claim 1, further comprising:
an abnormal pixel point identification module, configured to:
before the recognition sub-module inputs the scene image into a pre-trained image recognition model, converting the scene image into an HSV color space to obtain a first characteristic value, wherein the first characteristic value comprises an H channel value, an S channel value and a V channel value;
converting the scene image into an RGB color space to obtain a second characteristic value, wherein the second characteristic value comprises an R channel value, a G channel value and a B channel value;
comparing a first characteristic value and a second characteristic value corresponding to the same pixel point with a preset first characteristic value and a preset second characteristic value respectively to obtain a first difference value and a second difference value;
when the first difference value is determined to be smaller than a preset first difference value and the second difference value is determined to be smaller than a preset second difference value, indicating that the pixel point is normal; otherwise, taking the pixel points as the suspected pixel points to carry out the next detection;
taking a scene image comprising in-doubt pixel points as an in-doubt image, determining scene content corresponding to the in-doubt image, and shooting the scene content for multiple times to obtain a plurality of contrast images;
determining pixel points in the comparison image corresponding to the in-doubt pixel points in the in-doubt image as comparison pixel points;
dynamically monitoring according to the contrast pixel points and the in-doubt pixel points to obtain dynamic pixel difference values, and when the dynamic difference values are determined to be larger than preset difference values, indicating that the in-doubt pixel points are abnormal pixel points;
and the correction module is used for correcting the abnormal pixel points.
10. The vehicle rearview mirror ADAS identification based automotive reverse collision avoidance system of claim 1, further comprising:
the system comprises a driver detection module, a driver detection module and a driver detection module, wherein the driver detection module is used for acquiring a driving image of a driver when a target vehicle backs, identifying the driving image, determining a driving behavior and judging whether the driving behavior is an abnormal driving behavior;
and the alarm module is used for sending out an alarm prompt when the driver detection module determines that the driving behavior is the abnormal driving behavior.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210436110.7A CN114801988A (en) | 2022-04-25 | 2022-04-25 | Automobile reversing anti-collision system based on ADAS recognition of automobile rearview mirror |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210436110.7A CN114801988A (en) | 2022-04-25 | 2022-04-25 | Automobile reversing anti-collision system based on ADAS recognition of automobile rearview mirror |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114801988A true CN114801988A (en) | 2022-07-29 |
Family
ID=82507166
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210436110.7A Pending CN114801988A (en) | 2022-04-25 | 2022-04-25 | Automobile reversing anti-collision system based on ADAS recognition of automobile rearview mirror |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114801988A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115675282A (en) * | 2022-10-28 | 2023-02-03 | 重庆长安汽车股份有限公司 | Reversing early warning strategy determination method, early warning method, system, equipment and medium |
-
2022
- 2022-04-25 CN CN202210436110.7A patent/CN114801988A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115675282A (en) * | 2022-10-28 | 2023-02-03 | 重庆长安汽车股份有限公司 | Reversing early warning strategy determination method, early warning method, system, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107392103B (en) | Method and device for detecting road lane line and electronic equipment | |
CN106647776B (en) | Method and device for judging lane changing trend of vehicle and computer storage medium | |
US6690011B2 (en) | Infrared image-processing apparatus | |
US8244027B2 (en) | Vehicle environment recognition system | |
US8855868B2 (en) | Integrated vehicular system for low speed collision avoidance | |
US8437536B2 (en) | Environment recognition system | |
CN110386065B (en) | Vehicle blind area monitoring method and device, computer equipment and storage medium | |
US20010002936A1 (en) | Image recognition system | |
CN112349144B (en) | Monocular vision-based vehicle collision early warning method and system | |
EP2026246A1 (en) | Method and apparatus for evaluating an image | |
US20200285869A1 (en) | Convolutional neural network system for object detection and lane detection in a motor vehicle | |
US20140009614A1 (en) | Apparatus and method for detecting a three dimensional object using an image around a vehicle | |
CN110069990B (en) | Height limiting rod detection method and device and automatic driving system | |
CN107292214B (en) | Lane departure detection method and device and vehicle | |
CN107832788B (en) | Vehicle distance measuring method based on monocular vision and license plate recognition | |
CN114801988A (en) | Automobile reversing anti-collision system based on ADAS recognition of automobile rearview mirror | |
JP5027710B2 (en) | Vehicle environment recognition device and preceding vehicle tracking control system | |
CN114119955A (en) | Method and device for detecting potential dangerous target | |
KR101721442B1 (en) | Avoiding Collision Systemn using Blackbox Rear Camera for vehicle and Method thereof | |
JP2010224936A (en) | Object detection device | |
JPH0927100A (en) | Method and device for recognizing proceding vehicel and read-end collision preventive device equipped there with for vehicle | |
KR20160133386A (en) | Method of Avoiding Collision Systemn using Blackbox Rear Camera for vehicle | |
CN115527395B (en) | Intelligent traffic safety identification device for auxiliary judgment | |
KR101865958B1 (en) | Method and apparatus for recognizing speed limit signs | |
CN112292847A (en) | Image processing apparatus, mobile apparatus, method, and program |
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 |