CN210760742U - Intelligent vehicle auxiliary driving system - Google Patents
Intelligent vehicle auxiliary driving system Download PDFInfo
- Publication number
- CN210760742U CN210760742U CN201920824168.2U CN201920824168U CN210760742U CN 210760742 U CN210760742 U CN 210760742U CN 201920824168 U CN201920824168 U CN 201920824168U CN 210760742 U CN210760742 U CN 210760742U
- Authority
- CN
- China
- Prior art keywords
- module
- image
- obstacle
- information
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The utility model discloses an intelligent vehicle driver assistance system to the current development situation in current driver assistance field, combines present vehicle event data recorder to carry out the advantage of a large amount of data acquisition and the characteristics that image processing system unsupervised feature extracted, draws the feature and the orbit is prejudged the application and is dodged the strategy in the vehicle collision. And analyzing and processing a large amount of data returned by processing through Raspberry Pi, extracting the distribution characteristics of the obstacle information around the vehicle, classifying and identifying the obstacle information by using Arduino, finding out the optimized vehicle driving decision under the current obstacle distribution condition around the vehicle, and being beneficial to the driving decision in the driving and collision avoidance strategy of people to avoid the occurrence of collision accidents. By adopting the technology of combining image restoration and recognition prejudgment, the problem that infrared and ultrasonic equipment needs to be additionally added for assistance due to low imaging quality of the conventional auxiliary driving system is solved, and meanwhile, the cost is maximally reduced.
Description
Technical Field
The utility model relates to a car driver assistance control technical field, concretely relates to intelligent vehicle driver assistance system.
Background
With the rapid development of computer technology and artificial intelligence, Advanced Driving Assistance System (ADAS) technology has rapidly emerged, which is a trend to analyze road conditions by computer computing, provide warnings for drivers or assist drivers in driving, and take avoidance measures in dangerous situations. At present, ADAS is basically composed of a series of subsystems, and an Adaptive Cruise Control (ACC) system, a Lane Departure Warning (LDW) system, an Advanced Front Lighting (AFL) system, a pedestrian detection system (PPS), and the like are common. In which a constant distance to a preceding vehicle is maintained, deceleration being performed too close and acceleration being performed too far, by means of an Adaptive Cruise Control (ACC) system. Lane Departure Warning (LDW) systems alert the driver when the vehicle moves out of its lane by detecting the lane, a technique that can help the driver assist in making a lane change to prevent danger. Advanced Front Lighting (AFL) systems allow the beam to be used for different scenes, such as speed and direction of travel, by controlling headlamp parameters.
The input of the current pedestrian detection system is mainly to install radar or infrared sensors on the automobile to detect pedestrians by analyzing the spectrum, and the unmanned vehicle of Google corporation adopts the proposal, but the sensors are very expensive at present and cannot be popularized. And because the image can provide abundant apparent information and motion information, and the price of the camera in the market is lower or the program can be directly integrated into the automobile data recorder, and the real-time performance of detection can be realized based on the analysis of the image. Therefore, the auxiliary driving system for acquiring the images and analyzing the image contents by installing the high-definition camera has great significance for guaranteeing the driving safety of people and completing popularization.
Most of the existing assistant driving systems are connected with terminals through vehicle-mounted sensing equipment and react by depending on the operation of a server, the system has high cost and is difficult to popularize on a large scale, for example, in a patent of 'an assistant driving system', a control script is operated through a vehicle-mounted intelligent terminal according to relevant information acquired by vehicle-mounted sensing, positioning, vision and communication equipment, and assistant driving is realized through a control unit; most of the existing intelligent assistant driving systems only recognize passengers through face recognition so as to perform corresponding adjustment, for example, although an assistant driving system based on face recognition is intelligentized, the recognition is not used for guaranteeing the driving safety of a driver; in the existing intelligent driving obstacle avoidance system utilizing video image information, information collected by an infrared or ultrasonic sensor is often sent to a driving computer or a central controller for operation processing, and then the video information is fed back to a liquid crystal display screen in a vehicle to remind a driver of judging whether braking is needed or not. This technique is high in response to the driver's demand for a vehicle in a high-speed running state and in installation cost. Therefore the utility model discloses a low cost's on-vehicle high definition digtal camera, like equipment such as vehicle event data recorder with the video information of gathering handle and judge the information of gathering through central controller to judge whether will carry out the emergency measure that corresponds, thereby reach driver assistance's function.
Disclosure of Invention
An object of the utility model is to provide an intelligent vehicle driver assistance system to make the driver assistance system who proposes in the above-mentioned background art only need rely on camera and driving computer can calculate and control. The cost is reduced, and the method can be more popularized to the users.
In order to solve the technical problem, the utility model discloses the technical scheme who adopts is:
intelligent vehicle driver assistance system, its characterized in that includes: image acquisition module, image preprocessing module, barrier detection module, barrier classification prejudge module, image display module, early warning processing module and power management module, wherein:
the image acquisition module is used for acquiring image information around the vehicle; the image acquisition module comprises an automobile rear camera and an automobile front camera;
the image preprocessing module is used for preprocessing the image acquired by the image acquisition module to obtain a processed image; the image preprocessing module comprises a data acquisition module, an MTF matrix generation module and an ARM restoration module, wherein the MTF matrix generation module is integrated on a Raspberry type Raspberry Pi, the data acquisition module receives an image acquired by the image acquisition module and transmits the image to the MTF matrix generation module, the MTF matrix generation module converts information transmitted by the data acquisition module into MTF matrix information and transmits the MTF matrix information to the ARM restoration module, and the ARM restoration module processes the obtained information to obtain a restored image;
the obstacle detection module is used for detecting whether an obstacle exists in the obtained image information; the obstacle detection module comprises an obstacle identification module and an obstacle feature extraction module, wherein the obstacle identification module is used for analyzing the image information obtained by preprocessing and judging which of the images are obstacles by using Raspberry Raspberry Pi; the obstacle feature extraction module is used for extracting and carrying out correlation calculation on the motion features of the object identified as the obstacle, and extracting feature information of the identified obstacle by using the feature extraction module integrated on S3C 2440A;
the barrier classification pre-judging module comprises a barrier classification module and a barrier motion track pre-judging module, wherein the barrier classification module adopts ARM classification, obtains characteristic value distribution information of an image by extracting characteristic values of a sample image, takes the characteristic value distribution information as input, outputs the characteristic value distribution information to obtain the type probability of a test image, and the corresponding class with the highest probability is the class of a sample to be tested;
the image display module is used for displaying the processed image and the pre-judging result to be used for reminding the driver and storing and backing up the image;
the early warning processing module is used for monitoring image dynamic in real time, and prompting and making early warning measures when the fact that the movement track of the barrier conflicts with the driving track is judged;
the power management module supplies power to the image preprocessing module, the obstacle detection module, the obstacle classification prejudgment module, the image display module and the early warning processing module.
Furthermore, the rear camera of the automobile adopts a reversing image, and the front camera of the automobile adopts an automobile data recorder.
Furthermore, the early warning processing module comprises a voice prompt, an emergency brake, an intelligent obstacle avoidance, an ejection air bag and an alarm device, and the alarm device comprises a sound generating device for prompting and displaying information on the image display module.
The utility model has the advantages that: the utility model discloses to the current development situation in current driver assistance field, combine present driving recording equipment to carry out the advantage of a large amount of data acquisition and the characteristics that image processing system unsupervised feature extracted down, draw the feature and the orbit is prejudged the application and is dodged the strategy in the vehicle collision. And analyzing and processing a large amount of data returned by processing through Raspberry Pi, extracting the distribution characteristics of the obstacle information around the vehicle, classifying and identifying the obstacle information by using Arduino, finding out the optimized vehicle driving decision under the current obstacle distribution condition around the vehicle, and being beneficial to the driving decision in the driving and collision avoidance strategy of people to avoid the occurrence of collision accidents. By adopting the technology of combining image restoration and recognition prejudgment, the problem that infrared and ultrasonic equipment needs to be additionally added for assistance due to low imaging quality of the conventional auxiliary driving system is solved, and meanwhile, the cost is maximally reduced. The method has very important significance for improving the correctness of the intelligent driving system in judging the obstacles and improving the driving safety of the vehicle in various environments by utilizing image restoration and feature recognition. The utility model discloses assistant driving system's advantage includes:
1. in the active infrared imaging, the front and rear cameras of the automobile are used as image acquisition modules, so that the whole device is smaller in size, easy to install and more universal due to low cost;
2. and the targeted image restoration module is utilized to improve the image quality in a complex environment, so that the detection effect is improved as much as possible under the existing low-cost condition.
3. The Raspberry Pi can be used for more accurately extracting the movement information of the obstacle and calculating the track, the early warning processing module can judge whether the obstacle coincides with the driving track or not according to the information, danger occurs, and early warning reaction can be carried out on the obstacle instead of a driver in an emergency.
Drawings
Fig. 1 is the structure block diagram of the driving assistance system of the intelligent vehicle.
Detailed Description
For a better understanding of the present invention, the following examples are provided to further illustrate the present invention, but the present invention is not limited to the following examples. Various changes or modifications may be effected by one skilled in the art and equivalents may be made thereto without departing from the scope of the invention defined by the claims set forth herein.
As shown in fig. 1, the intelligent vehicle driving assistance system includes: the system comprises an image acquisition module 1 (comprising an automobile rear camera 2 and an automobile front camera 3), an image preprocessing module 4, an obstacle detection module 5, an obstacle classification prejudging module 6, an image display module 7, an early warning processing module 8 and a power management module 9.
The image acquisition module 1 is used for acquiring image information around the vehicle; the image acquisition module comprises a rear camera 2 and a front camera 3 of the automobile, and under the condition that the current automobile is generally configured to be higher, the rear camera can use a reversing image, and the front camera can use a vehicle traveling data recorder to replace the rear camera. The vehicle-mounted image acquisition and transmission system can acquire image information around the vehicle in real time and transmit the image information to the lower module.
The image preprocessing module 4 is configured to preprocess the image acquired by the image acquisition module 1 to obtain a processed image. The image preprocessing module comprises a data acquisition module, an MTF matrix generation module and an ARM restoration module, wherein the MTF matrix generation module and the ARM restoration module are integrated on a Raspberry type Raspberry Pi, the data acquisition module receives an image acquired by the image acquisition module 1 and transmits the image to the MTF matrix generation module, the MTF matrix generation module converts information transmitted by the data acquisition module into MTF matrix information and transmits the MTF matrix information to the ARM restoration module, and the ARM restoration module processes the acquired image information under severe conditions such as night, haze and the like to obtain a restored image.
The obstacle detection module 5 is configured to detect whether an obstacle exists in the obtained image information; the obstacle detection module 5 includes an obstacle recognition module and an obstacle feature extraction module. The obstacle identification module has the function of analyzing the image information obtained by preprocessing, and judging which of the images are obstacles by using Raspberry Raspberry Pi. And the obstacle feature extraction module is used for extracting the motion features of the object which is determined as the obstacle and carrying out correlation calculation. And the feature extraction module integrated on S3C2440A is used for extracting feature information of the identified obstacles, when classification is carried out, the features of the obstacles must be determined firstly, and the selection of the features is important and is the basis for classifying the obstacles. Three features were chosen for this embodiment, namely the symmetry of the obstacle, the levelness of the edge lines and the aspect ratio of the obstacle.
The obstacle classification prejudgment module 6 comprises an obstacle classification module and an obstacle motion trajectory prejudgment module, wherein the obstacle classification module can obtain feature value distribution information of an image by extracting feature values of a sample image, and the feature value distribution information is used as input. And outputting to obtain the type probability of the test image, wherein the corresponding class with the maximum probability is the class of the sample to be tested. The utility model discloses an ARM classifies, can classify unknown picture effectively, and the error rate is lower. And finally, calculating the motion track of the barrier by utilizing an Arduino-based barrier motion track prejudging module through comparing the sample skeleton characteristic information frame by frame, comparing the motion track with the driving track, judging whether a traffic accident occurs or not, and matching emergency measures according to the calculation result.
And the image display module 7 is used for displaying the processed image and the pre-judging result to be used for reminding the driver and storing and backing up the image. The image display module 7 adopts an LCD display screen.
And the early warning processing module 8 is used for monitoring image dynamics in real time, and prompting and making early warning measures when the obstacle motion track is determined to conflict with the driving track. The early warning processing module 8 comprises voice reminding, emergency braking, intelligent obstacle avoidance, an ejection air bag, an alarm device and the like, the current condition is analyzed and judged in real time through the processor, the best early warning method is selected mainly according to the type of the obstacle and the motion track of the obstacle, and the alarm device comprises a sound generating device for reminding and displaying information on the display screen.
Principle of operation
When the intelligent auxiliary automatic driving system works, the power management module 9 supplies power to the image preprocessing module, the obstacle detection module, the obstacle classification pre-judgment module, the image display module and the early warning processing module, the front camera and the rear camera of the automobile start shooting, image information captured in real time is sent to the processor for processing and operation, after the image information is acquired from the image acquisition module, the extracted information is analyzed and processed by the image preprocessing module and other subsequent modules, characteristics beneficial to identifying obstacles around the automobile are extracted from the extracted information, the motion state information of the extracted information is calculated, and the early warning processing module judges and implements the optimal early warning scheme.
The method has the advantages that the method introduces pedestrian tracking, effectively improves detection precision by a method combining tracking and detection, reduces detection frequency for small-scale pedestrians due to high real-time tracking processing, further saves detection time, and simultaneously considers accuracy and real-time performance.
The above embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the present invention, therefore, any modifications, equivalents, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of the claims of the present invention.
Claims (3)
1. Intelligent vehicle driver assistance system, its characterized in that includes: image acquisition module, image preprocessing module, barrier detection module, barrier classification prejudge module, image display module, early warning processing module and power management module, wherein:
the image acquisition module is used for acquiring image information around the vehicle; the image acquisition module comprises an automobile rear camera and an automobile front camera;
the image preprocessing module is used for preprocessing the image acquired by the image acquisition module to obtain a processed image; the image preprocessing module comprises a data acquisition module, an MTF matrix generation module and an ARM restoration module, wherein the MTF matrix generation module is integrated on a Raspberry type Raspberry Pi, the data acquisition module receives an image acquired by the image acquisition module and transmits the image to the MTF matrix generation module, the MTF matrix generation module converts information transmitted by the data acquisition module into MTF matrix information and transmits the MTF matrix information to the ARM restoration module, and the ARM restoration module processes the obtained information to obtain a restored image;
the obstacle detection module is used for detecting whether an obstacle exists in the obtained image information; the obstacle detection module comprises an obstacle identification module and an obstacle feature extraction module, wherein the obstacle identification module is used for analyzing the image information obtained by preprocessing and judging which of the images are obstacles by using Raspberry Raspberry Pi; the obstacle feature extraction module is used for extracting and carrying out correlation calculation on the motion features of the object identified as the obstacle, and extracting feature information of the identified obstacle by using the feature extraction module integrated on S3C 2440A;
the barrier classification pre-judging module comprises a barrier classification module and a barrier motion track pre-judging module, wherein the barrier classification module adopts ARM classification, obtains characteristic value distribution information of an image by extracting characteristic values of a sample image, takes the characteristic value distribution information as input, outputs the characteristic value distribution information to obtain the type probability of a test image, and the corresponding class with the highest probability is the class of a sample to be tested;
the image display module is used for displaying the processed image and the pre-judging result to be used for reminding the driver and storing and backing up the image;
the early warning processing module is used for monitoring image dynamic in real time, and prompting and making early warning measures when the fact that the movement track of the barrier conflicts with the driving track is judged;
the power management module supplies power to the image preprocessing module, the obstacle detection module, the obstacle classification prejudgment module, the image display module and the early warning processing module.
2. The intelligent vehicle driver assistance system of claim 1, wherein the rear camera of the vehicle uses a reverse image, and the front camera of the vehicle uses a vehicle data recorder.
3. The driving assistance system of an intelligent vehicle as claimed in claim 1, wherein the early warning processing module comprises a voice prompt, an emergency brake, an intelligent obstacle avoidance, an ejection airbag and an alarm device, and the alarm device comprises a sound generating device for prompting and displaying information on the image display module.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201821549224 | 2018-09-21 | ||
| CN2018215492248 | 2018-09-21 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN210760742U true CN210760742U (en) | 2020-06-16 |
Family
ID=71043022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201920824168.2U Expired - Fee Related CN210760742U (en) | 2018-09-21 | 2019-06-03 | Intelligent vehicle auxiliary driving system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN210760742U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113077494A (en) * | 2021-04-10 | 2021-07-06 | 山东沂蒙交通发展集团有限公司 | Road surface obstacle intelligent recognition equipment based on vehicle orbit |
| CN115626159A (en) * | 2021-07-01 | 2023-01-20 | 信扬科技(佛山)有限公司 | Vehicle warning system and method and automobile |
-
2019
- 2019-06-03 CN CN201920824168.2U patent/CN210760742U/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113077494A (en) * | 2021-04-10 | 2021-07-06 | 山东沂蒙交通发展集团有限公司 | Road surface obstacle intelligent recognition equipment based on vehicle orbit |
| CN115626159A (en) * | 2021-07-01 | 2023-01-20 | 信扬科技(佛山)有限公司 | Vehicle warning system and method and automobile |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10392009B2 (en) | Automatic parking system and automatic parking method | |
| CN102765365B (en) | Pedestrian detection method based on machine vision and pedestrian anti-collision warning system based on machine vision | |
| CN112389448B (en) | Abnormal driving behavior identification method based on vehicle state and driver state | |
| WO2017029847A1 (en) | Information processing device, information processing method, and program | |
| US11691619B2 (en) | Automatic parking system and automatic parking method | |
| CN109145719B (en) | Driver fatigue state identification method and system | |
| CN110682907B (en) | Automobile rear-end collision prevention control system and method | |
| WO2005036371A2 (en) | Moving object detection using low illumination depth capable computer vision | |
| US20200108717A1 (en) | Apparatus and method for controlling speed | |
| CN112382115B (en) | Driving risk early warning device and method based on visual perception | |
| US20200111362A1 (en) | Method and apparatus for analyzing driving tendency and system for controlling vehicle | |
| JP2013057992A (en) | Inter-vehicle distance calculation device and vehicle control system using the same | |
| CN206383949U (en) | Driving safety system based on the pure image procossings of ADAS | |
| CN210760742U (en) | Intelligent vehicle auxiliary driving system | |
| TWI531499B (en) | Anti-collision warning method and device for tracking moving object | |
| CN109131162A (en) | A kind of driving assistance system based on artificial intelligence | |
| CN112200087B (en) | Obstacle image automatic calibration device for vehicle collision early warning | |
| CN112606804A (en) | Control method and control system for active braking of vehicle | |
| CN114734966B (en) | Automatic emergency braking system and method based on camera and cloud real-time map | |
| JP4848644B2 (en) | Obstacle recognition system | |
| Shirpour et al. | A probabilistic model for visual driver gaze approximation from head pose estimation | |
| JP3562278B2 (en) | Environment recognition device | |
| CN116729261A (en) | Vehicle bottom living body monitoring method and device, terminal equipment and storage medium | |
| CN111028544A (en) | Pedestrian early warning system with V2V technology and vehicle-mounted multi-sensor integration | |
| US20220101025A1 (en) | Temporary stop detection device, temporary stop detection system, and recording medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200616 Termination date: 20210603 |