[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN111199177A - Automobile rearview pedestrian detection alarm method based on fisheye image correction - Google Patents

Automobile rearview pedestrian detection alarm method based on fisheye image correction Download PDF

Info

Publication number
CN111199177A
CN111199177A CN201811386242.3A CN201811386242A CN111199177A CN 111199177 A CN111199177 A CN 111199177A CN 201811386242 A CN201811386242 A CN 201811386242A CN 111199177 A CN111199177 A CN 111199177A
Authority
CN
China
Prior art keywords
vehicle
pedestrians
image
pedestrian
fisheye
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
Application number
CN201811386242.3A
Other languages
Chinese (zh)
Inventor
苏航
李召国
张怡
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Research Institute of Sun Yat Sen University
Original Assignee
Shenzhen Research Institute of Sun Yat Sen University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Research Institute of Sun Yat Sen University filed Critical Shenzhen Research Institute of Sun Yat Sen University
Priority to CN201811386242.3A priority Critical patent/CN111199177A/en
Publication of CN111199177A publication Critical patent/CN111199177A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/103Static body considered as a whole, e.g. static pedestrian or occupant recognition
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/80Geometric correction
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/194Segmentation; Edge detection involving foreground-background segmentation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/20Analysis of motion
    • G06T7/254Analysis of motion involving subtraction of images
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30196Human being; Person
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30248Vehicle exterior or interior
    • G06T2207/30252Vehicle exterior; Vicinity of vehicle

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses an automobile rearview pedestrian detection alarm method based on fisheye image correction. The invention comprises the following steps: collecting information behind the vehicle by using a fish-eye lens; extracting an effective region from the collected fisheye image; reprocessing the distortion of the image edge; detecting pedestrians in the image, identifying and segmenting the pedestrians, and predicting moving pedestrians; and judging whether to alarm or not according to the positions of the pedestrians and the tail of the vehicle and the movement trend of the pedestrians. The invention uses the fisheye lens to collect the vehicle running information, thus overcoming the defects of limited visual range and large installation number of the traditional camera; denoising and distorting the fisheye image by combining a plurality of image processing algorithms, so that the pedestrian can be accurately identified; and analyzing the positions and the movement trends of the pedestrians by using a pedestrian recognition algorithm, and finally achieving the effects of prompting and protecting the driving safety of the vehicle.

Description

Automobile rearview pedestrian detection alarm method based on fisheye image correction
Technical Field
The invention relates to the field of computer vision and video image processing, in particular to an automobile rearview pedestrian detection alarm method based on fisheye image correction.
Background
Pedestrian Detection (Pedestrian Detection) is the use of computer vision techniques to determine whether a Pedestrian is present in an image or video sequence and to provide accurate positioning. The technology can be combined with technologies such as pedestrian tracking and pedestrian re-identification, and is applied to the fields of artificial intelligence systems, vehicle driving assistance systems, intelligent robots, intelligent video monitoring, human body behavior analysis, intelligent transportation and the like. As one of the pioneering technologies of unmanned driving, a pedestrian detection system has also become a research and development hotspot in recent years, and is generally integrated into a collision prevention system, a camera and an inductor are used to detect pedestrians, and the pedestrian detection system is decelerated and braked in time so as to reduce accident damage, and the pedestrian detection is also paid more and more attention in the fields of intelligent transportation and the like.
The fisheye lens has the function of wide visual angle range, the visual angle can reach 220 degrees or 230 degrees generally, and conditions are created for shooting a large-range scene at a short distance. The fish-eye lens is an optical imaging system with an ultra-large field of view and a large aperture, and compresses the ultra-large field of view of an object to the field of view range required by a conventional lens. The fisheye lens can realize the full airspace containment and the acquisition of full time domain real-time information, and the fisheye lens plays a full role in the fields of photography, meteorology, measurement technology, medical application, safety monitoring and the like.
Due to the characteristic of the fish-eye lens of an ultra-large wide angle, the imaging of the fish-eye lens also has geometric distortion, and the distortion of the fish-eye image is reduced as much as possible by carrying out distortion correction on the image through an image correction algorithm.
The existing pedestrian detection alarm technology is a system for informing a vehicle approaching by a vehicle sound of Toyota corporation. The method comprises the following specific steps:
1. sensing pedestrians or obstacles around the vehicle by using an ultrasonic sensor;
2. if it is sensed that a pedestrian or other object is less than a given distance from the vehicle, the system automatically sounds a sound (whistle) to notify the pedestrian.
The disadvantages of this method are:
1. the sound informing system detects that the detection object is not only a pedestrian, and when an immovable object (such as a tree and a building) is detected, the noise generated by the vehicle is likely to influence the surrounding environment;
2. in many cities in China, the stipulation of forbidding whistling is provided, and the automatic whistling system easily penalizes drivers;
3. non-fisheye cameras typically do not have a 180 ° viewing angle and are less accurate. A plurality of cameras are generally required to be installed for observing the running environment of the vehicle in an all-around manner, so that not only is the cost increased, but also the attractiveness is influenced to a certain extent;
4. installing multiple cameras increases the lateral volume of the module and increases the complexity of system design and computation.
Disclosure of Invention
The invention aims to overcome the defects of the existing method and provides an automobile rearview pedestrian detection alarm method based on fisheye image correction. The scheme of the invention mainly solves two core problems, firstly, the image distortion caused by the vehicle running condition acquired by the fisheye lens is corrected, and the image distortion and the error are reduced; secondly, detecting and predicting pedestrians behind the vehicle according to denoising and distortion processing of the fisheye image, and judging whether to remind a driver according to the moving direction and the distance of the pedestrians.
In order to solve the above problems, the present invention provides an automobile rearview pedestrian detection alarm method based on fisheye image correction, wherein the method comprises:
collecting information behind the vehicle by using a fish-eye lens;
preprocessing, namely extracting an effective area from the collected fisheye image by using a spherical coordinate positioning algorithm;
the preprocessed image is reprocessed aiming at the distortion of the image edge by using a bilinear interpolation algorithm;
detecting pedestrians in the image by using a background frame difference method, identifying and segmenting the pedestrians, and predicting the moving pedestrians by using a gray model;
and judging whether to alarm or not according to the positions of the pedestrians and the tail of the vehicle and the movement trend of the pedestrians.
Preferably, the judging whether to give an alarm according to the positions of the pedestrians and the vehicle tail and the movement trend of the pedestrians specifically comprises:
if the vehicle is in a reversing state, the distance between the pedestrian and the vehicle is not more than four meters, and the alarm is directly started;
if the vehicle is in a reversing state, if the pedestrian is detected to walk to the tail of the vehicle and the distance is not less than five meters, alarming;
if the vehicle is in a reversing state, the vehicle alarms when the distance from the pedestrian to the tail of the vehicle is not more than three meters, and does not alarm under other conditions.
The invention provides an automobile rearview pedestrian detection alarm method based on fisheye image correction, which uses a fisheye lens to collect vehicle running information and overcomes the defects of limited visual range and large installation number of the traditional camera; denoising and distorting the fisheye image by combining a plurality of image processing algorithms, so that the pedestrian can be accurately identified; and analyzing the positions and the movement trends of the pedestrians by using a pedestrian recognition algorithm, and finally achieving the effects of prompting and protecting the driving safety of the vehicle.
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 is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a pedestrian detection alarm method according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. 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.
Fig. 1 is a flowchart of a pedestrian detection alarm method according to an embodiment of the present invention, and as shown in fig. 1, the method includes:
s1, collecting information behind the vehicle by using the fish-eye lens;
s2, preprocessing, and extracting an effective area of the collected fisheye image by using a spherical coordinate positioning algorithm;
s3, the preprocessed image is reprocessed aiming at the distortion of the image edge by using a bilinear interpolation algorithm;
s4, detecting pedestrians in the image by using a background frame difference method, identifying and segmenting the pedestrians, and predicting the moving pedestrians by using a gray model;
and S5, judging whether to give an alarm according to the positions of the pedestrians and the tail of the vehicle and the movement trend of the pedestrians.
Step S1 is specifically as follows:
the middle part of the tail of the vehicle behind the vehicle is provided with a fisheye lens, and the fisheye lens is used for collecting driving information behind the vehicle. The fisheye lens has a wider visual field, and the distortion effect can be reduced along with the horizontal lines and the number of scenes easy to distinguish, so the fisheye lens is horizontally arranged in the middle of the tail of the vehicle and is parallel to the ground, and the quality of the picture acquired by the fisheye lens is improved.
Step S2 is specifically as follows:
s2-1, establishing a model by using a spherical coordinate positioning algorithm:
the spherical coordinate positioning method is also called longitude mapping method, and is a typical and rapid two-dimensional fisheye image correction algorithm. And after the points on the same longitude are corrected, the corrected coordinates have the same abscissa, and the longitude mapping of the spherical surface is completed by a spherical coordinate positioning method.
S2-2, binarizing the image by using an average value method:
binarization of an image is the process of changing a color RGB image into a pure gray scale image. The average value method is a method in which the luminances of R, G, B three components in a color image are averaged, and the obtained value is output as a gradation value to obtain a gradation image.
S2-3, obtaining an effective area by using a scanning line approximation method:
and obtaining a gray value difference I according to the gray value of each point, and once the value of I is found to be larger than a preset threshold value, considering that the scanning enters an effective area, and stopping searching in the direction.
S2-4, calculating the center and the radius of the effective area:
and after the scanning is finished, calculating to obtain a circumscribed square, and calculating to obtain a center and a radius.
S2-5, obtaining the relation between corresponding pixel points of the target image and the original image:
and (3) performing forward or reverse coordinate transformation by using a spherical coordinate positioning algorithm to obtain the relationship between the initial coordinate and the transformed coordinate:
Figure BDA0001873025890000061
Figure BDA0001873025890000062
xhthe abscissa (which is an integer) of the h point in the original image; r is the radius of the effective area of the fisheye, and Y is the ordinate of the k point in the original image; x is the number ofkAre the corrected coordinates. Since digital images cannot store non-integers. For the point of which the calculated point is not an integer, the non-integer pixel point in the original image can be found through inverse mapping from an established corrected image, one or more accurate points in the original image are positioned through a corresponding interpolation method, and the value of the pixel at the position after correction is obtained through the values of the points. Inverse mapping is performed using the following formula:
Figure BDA0001873025890000063
step S3 is specifically as follows:
after distortion correction is carried out on the fisheye lens image by adopting a spherical coordinate positioning algorithm model, the middle part of the image is completely restored, but the edge part still has certain barrel-shaped distortion. In order to obtain a better correction effect, bilinear gray level interpolation needs to be carried out on the primary corrected image.
Step S4 is specifically as follows:
s4-1, rapidly obtaining a background image by adopting a self-adaptive method:
taking the input 1 st frame image as an original background, carrying out binarization processing on the current inter-frame difference image from the 2 nd frame, and finding a motion area and a non-motion area by using a binary image; the current background image is then updated with the non-moving area portion of the current frame image. Binary image W of current interframe difference imagekIs defined as:
Wk=1(|Ik(i,j)-Ik-1(i,j)|≥g)
Wk=0(|Ik(i,j)-Ik-1(i,j)|<g)
Ik-1(I, j) and Ik(i, j) are respectively the gray values of two continuous frames of images in the image sequence at the image coordinates (i, j), if the maximum peak value in the gray histogram of the frame difference image is M, g is the corresponding gray value at which the right side value of the maximum peak value is 1/10M, and the obtained binary image W is used as the gray valuekFor the original background image Bk-1Updating to obtain the current background image BkComprises the following steps:
Bk(i,j)=Bk-1(i, j) when WkWhen (i, j) is 1
Bk(i,j)=(1-α)Ik(1,j)+αBk-1(i, j) when WkWhen (i, j) is 0
Bk-1(i, j) represents the gray value of the original background image at the coordinate (i, j); wk(i, j) represents the gray scale value of the binary image at the coordinates (i, j), and the background update coefficient α is 0.9
S4-2, segmenting the image, and confirming that the pedestrian:
the height and width of the pedestrian generally conform to a certain proportion, the shape of the pedestrian can be approximated to the aspect ratio of the circumscribed rectangle thereof, and according to the segmentation result, the aspect ratio and the shape dispersion degree of the target are easily calculated, so that the pedestrian and the vehicle in the picture are distinguished. The formula is as follows:
R=H/L
C=P2/A
H. l, P and A are the height, width, circumference and area of the target, respectively; r is the aspect ratio of the target and C is the shape dispersion.
S4-3, using the gray model GM (1,1) as a motion model, predicting the pedestrian motion:
and processing the historical data by using a gray model to find the motion rule of the pedestrian so as to predict the future state of the pedestrian. The gray model is defined as follows:
the initialization variable U is a nonnegative monotonous original data sequence of a pedestrian prediction object, and consists of measurement quantities in n pictures:
U0=(u0(1),u0(2),u0(3),...,u0(n),)
generating a new sequence by accumulation:
U1=(u1(1),u1(2),u1(3),...,u1(n),)
Figure BDA0001873025890000081
the corresponding differential equation for the gray model GM (1,1) model is:
Figure BDA0001873025890000082
and a and b are constant coefficients corresponding to the differential equation.
Figure BDA0001873025890000083
For the parameter sequence to be estimated, solving by using a least square method to obtain:
Figure BDA0001873025890000084
wherein:
Figure BDA0001873025890000085
after the estimation parameters are obtained, the following prediction model can be obtained by solving the differential equation:
Figure BDA0001873025890000086
Figure BDA0001873025890000087
Figure BDA0001873025890000088
according to the formula, the GM (1,1) model can predict the next position of the pedestrian only by three pieces of position information of the pedestrian, an interested area is established according to the predicted position, the new pedestrian position is detected, historical data is updated in time by the obtained position information, the latest change rule of the pedestrian motion can be quickly tracked, and the accurate prediction of the pedestrian position is made.
Step S5 is specifically as follows:
if the vehicle is in a reversing state, the distance between the pedestrian and the vehicle is not more than four meters, and the alarm is directly started; if the vehicle is in a reversing state, if the pedestrian is detected to walk to the tail of the vehicle and the distance is not less than five meters, alarming; if the vehicle is in a reversing state, the vehicle alarms when the distance from the pedestrian to the tail of the vehicle is not more than three meters, and does not alarm under other conditions.
According to the automobile rearview pedestrian detection alarm method based on fisheye image correction, the fisheye lens is used for collecting vehicle running information, and the defects that a traditional camera is limited in visual range and large in installation quantity are overcome; denoising and distorting the fisheye image by combining a plurality of image processing algorithms, so that the pedestrian can be accurately identified; and analyzing the positions and the movement trends of the pedestrians by using a pedestrian recognition algorithm, and finally achieving the effects of prompting and protecting the driving safety of the vehicle.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable storage medium, and the storage medium may include: a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic or optical disk, or the like.
In addition, the automobile rearview pedestrian detection alarm method based on fisheye image correction provided by the embodiment of the invention is described in detail, a specific example is applied in the text to explain the principle and the implementation mode of the invention, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims (2)

1. A car rearview pedestrian detection alarm method based on fisheye image correction is characterized by comprising the following steps:
collecting information behind the vehicle by using a fish-eye lens;
preprocessing, namely extracting an effective area from the collected fisheye image by using a spherical coordinate positioning algorithm;
the preprocessed image is reprocessed aiming at the distortion of the image edge by using a bilinear interpolation algorithm;
detecting pedestrians in the image by using a background frame difference method, identifying and segmenting the pedestrians, and predicting the moving pedestrians by using a gray model;
and judging whether to alarm or not according to the positions of the pedestrians and the tail of the vehicle and the movement trend of the pedestrians.
2. The method as claimed in claim 1, wherein the determining whether to alarm according to the positions of the pedestrians and the vehicle tail and the movement trend of the pedestrians comprises:
if the vehicle is in a reversing state, the distance between the pedestrian and the vehicle is not more than four meters, and the alarm is directly started;
if the vehicle is in a reversing state, if the pedestrian is detected to walk to the tail of the vehicle and the distance is not less than five meters, alarming;
if the vehicle is in a reversing state, the vehicle alarms when the distance from the pedestrian to the tail of the vehicle is not more than three meters, and does not alarm under other conditions.
CN201811386242.3A 2018-11-20 2018-11-20 Automobile rearview pedestrian detection alarm method based on fisheye image correction Pending CN111199177A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811386242.3A CN111199177A (en) 2018-11-20 2018-11-20 Automobile rearview pedestrian detection alarm method based on fisheye image correction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811386242.3A CN111199177A (en) 2018-11-20 2018-11-20 Automobile rearview pedestrian detection alarm method based on fisheye image correction

Publications (1)

Publication Number Publication Date
CN111199177A true CN111199177A (en) 2020-05-26

Family

ID=70746980

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811386242.3A Pending CN111199177A (en) 2018-11-20 2018-11-20 Automobile rearview pedestrian detection alarm method based on fisheye image correction

Country Status (1)

Country Link
CN (1) CN111199177A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111833243A (en) * 2020-09-20 2020-10-27 武汉中科通达高新技术股份有限公司 Data display method, mobile terminal and storage medium
CN112216067A (en) * 2020-09-07 2021-01-12 邢台林樾科技有限公司 Image processing method based on vehicle-mounted wide-angle camera
CN112329552A (en) * 2020-10-16 2021-02-05 爱驰汽车(上海)有限公司 Obstacle detection method and device based on automobile

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101017573A (en) * 2007-02-09 2007-08-15 南京大学 Method for detecting and identifying moving target based on video monitoring
IN2014MU03302A (en) * 2014-10-16 2015-08-07 Kpit Technologies Ltd
CN106920210A (en) * 2017-02-27 2017-07-04 武汉大学 A kind of flake video real time panoramic bearing calibration based on development of a sphere model
CN107133559A (en) * 2017-03-14 2017-09-05 湖北工业大学 Mobile object detection method based on 360 degree of panoramas
CN108638999A (en) * 2018-05-16 2018-10-12 浙江零跑科技有限公司 A kind of collision early warning system and method for looking around input based on 360 degree

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101017573A (en) * 2007-02-09 2007-08-15 南京大学 Method for detecting and identifying moving target based on video monitoring
IN2014MU03302A (en) * 2014-10-16 2015-08-07 Kpit Technologies Ltd
CN106920210A (en) * 2017-02-27 2017-07-04 武汉大学 A kind of flake video real time panoramic bearing calibration based on development of a sphere model
CN107133559A (en) * 2017-03-14 2017-09-05 湖北工业大学 Mobile object detection method based on 360 degree of panoramas
CN108638999A (en) * 2018-05-16 2018-10-12 浙江零跑科技有限公司 A kind of collision early warning system and method for looking around input based on 360 degree

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112216067A (en) * 2020-09-07 2021-01-12 邢台林樾科技有限公司 Image processing method based on vehicle-mounted wide-angle camera
CN112216067B (en) * 2020-09-07 2023-01-06 邢台林樾科技有限公司 Image processing method based on vehicle-mounted wide-angle camera
CN111833243A (en) * 2020-09-20 2020-10-27 武汉中科通达高新技术股份有限公司 Data display method, mobile terminal and storage medium
CN112329552A (en) * 2020-10-16 2021-02-05 爱驰汽车(上海)有限公司 Obstacle detection method and device based on automobile
CN112329552B (en) * 2020-10-16 2023-07-14 爱驰汽车(上海)有限公司 Obstacle detection method and device based on automobile

Similar Documents

Publication Publication Date Title
Faro et al. Adaptive background modeling integrated with luminosity sensors and occlusion processing for reliable vehicle detection
KR100377067B1 (en) Method and apparatus for detecting object movement within an image sequence
JP5297078B2 (en) Method for detecting moving object in blind spot of vehicle, and blind spot detection device
US7418134B2 (en) Method and apparatus for foreground segmentation of video sequences
US20110134245A1 (en) Compact intelligent surveillance system comprising intent recognition
US20060285724A1 (en) Salient motion detection system, method and program product therefor
WO2001033503A1 (en) Image processing techniques for a video based traffic monitoring system and methods therefor
US11436839B2 (en) Systems and methods of detecting moving obstacles
Gangodkar et al. Robust segmentation of moving vehicles under complex outdoor conditions
CN112149458A (en) Obstacle detection method, intelligent driving control method, device, medium, and apparatus
CN105469054B (en) The model building method of normal behaviour and the detection method of abnormal behaviour
CN103366155B (en) Temporal coherence in unobstructed pathways detection
CN111199177A (en) Automobile rearview pedestrian detection alarm method based on fisheye image correction
CN111967396A (en) Processing method, device and equipment for obstacle detection and storage medium
CN108280445B (en) Method for detecting moving objects and raised obstacles around vehicle
CN112561946A (en) Dynamic target detection method
KR101239718B1 (en) System and method for detecting object of vehicle surroundings
Berriel et al. A particle filter-based lane marker tracking approach using a cubic spline model
EP2447912B1 (en) Method and device for the detection of change in illumination for vision systems
Balisavira et al. Real-time object detection by road plane segmentation technique for ADAS
Kanhere et al. Real-time detection and tracking of vehicle base fronts for measuring traffic counts and speeds on highways
JP2002074369A (en) System and method for monitoring based on moving image and computer readable recording medium
Al Okaishi et al. Real-time traffic light control system based on background updating and edge detection
Agrawal et al. Off-road lane detection using superpixel clustering and ransac curve fitting
Denman et al. Multi-view intelligent vehicle surveillance system

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
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20200526