CN110084086A - A kind of automatic driving vehicle drivable region detection method of view-based access control model sensor - Google Patents
A kind of automatic driving vehicle drivable region detection method of view-based access control model sensor Download PDFInfo
- Publication number
- CN110084086A CN110084086A CN201811511924.2A CN201811511924A CN110084086A CN 110084086 A CN110084086 A CN 110084086A CN 201811511924 A CN201811511924 A CN 201811511924A CN 110084086 A CN110084086 A CN 110084086A
- Authority
- CN
- China
- Prior art keywords
- image
- automatic driving
- driving vehicle
- access control
- detection method
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
A kind of automatic driving vehicle drivable region detection method of view-based access control model sensor, the image that input is only obtained by visual sensor；Reduction and size adjusting image preprocessing are carried out to image；Image after pretreatment is input to network to feedover to it, the boundary for detecting barrier obtains can travel area coordinate point；To travelable area coordinate point, it is computed and restores more accurate coordinate value result, calculate the travelable region in former described image, it can only pass through the visual sensor system in automatic driving vehicle sensory perceptual system, the travelable region detection of structured road vehicle, the statement in region and its output be can travel to lower layer, i.e. Vehicle Decision Method planning module, carry out vehicle running path planning.
Description
Technical field
The present invention proposes a kind of automatic driving vehicle drivable region detection method of visual sensor.
Background technique
Automatic Pilot technology includes environment sensing, decision rule and vehicle control three parts.It can be used for intelligent driving ring
The hardware device of border perception has very much, mainly includes camera, laser radar, millimetre-wave radar, ultrasonic radar etc..Vision technique
It is chiefly used in the lane grade positioning of intelligent vehicle, identifies the barriers such as road geometry, the vehicles or pedestrians for detecting periphery, identification
Traffic lights and traffic sign etc..
Environment sensing is as the first link, the key position in intelligent driving vehicle Yu external environment information exchange,
Key is to make the sensing capability of the more preferable simulation human driver of intelligent driving vehicle, to understand the driving of itself and periphery
Situation.
The exploitation of Function for Automatic Pilot, which needs sensory perceptual system to provide main front, can travel region, and prior art uses
The scheme that vision+map combines, technical difficulty are higher.
Industry solves the problems, such as that this mainstream algorithm has using textural characteristics and extracts the travelable region of segmentation, also there is use
Divide road surface, vehicle, pavement etc. based on the parted pattern of deep learning, may also used to detection can exercise region.But
It is that the Generalization Capability that the textural characteristics based on conventional machines vision detect is poor, Various Complex scene can not be adapted to；Although the latter
Various targets in image can accurately be partitioned into, but it is computationally intensive, it is difficult to accomplish that speed is unable to reach demand in real time.
Summary of the invention
The object of the present invention is to provide a kind of structured roads of view-based access control model sensor technology can travel region detection side
Case proposes a kind of road drivable region detection method for being based purely on visual sensor, realizes that the vehicle of structured road can
The detection method of running region carries out path planning to be supplied to the decision system of vehicle.
To realize the above-mentioned technical purpose, the automatic driving vehicle that the present invention provides a kind of view-based access control model sensor can travel area
Area detecting method, step include: to input the image step of visual sensor, detect the travelable region step in described image
Suddenly, export can travel region coordinate points arrive Vehicle Decision Method layer step, in which: it is described input visual sensor image step in
Described image only pass through visual sensor obtain；In travelable region step in the detection described image further include: (1)
Image preprocessing: reduction and size adjusting are carried out to described image；(2) image after pretreatment image detection: is input to network
It feedovers to it, detects the boundary of barrier, one travelable area is returned out according to file for image after the pretreatment
Domain coordinate points, the travelable area coordinate point are the down contour point of the nearest barrier in distance shooting source, and can described in output
Running region coordinate points；(3) post processing of image: the travelable area coordinate point of the network is handled, is computed and restores
More accurate coordinate value is as a result, calculate the travelable region in former described image.
The present invention can only pass through the visual sensor system in automatic driving vehicle sensory perceptual system, structured road vehicle
It can travel region detection, can travel the statement in region and its output to lower layer, i.e. Vehicle Decision Method planning module, carry out vehicle row
Sail path planning.
As a further improvement, the reduction are as follows: it is big that described image from 1920*1208 size is cut to 1920*640
It is small, to ignore interference region；The size adjusting are as follows: 2 times of image down sampling after the reduction are arrived with the size of 960*320.
As a further improvement, it is described reduction be since the longitudinal coordinate 430 of described image to longitudinal coordinate 1070
Until, and retain 640 points between the two.
As a further improvement, the travelable area coordinate point is returned out according to every 16 files, with compression
The information content of output simultaneously promotes speed.
As a further improvement, it is feedovered in described image detection by Caffe engine.
As a further improvement, the travelable area coordinate point in described image detection is 60*20* three-dimensional
Spend matrix, in which: 60*20 be 960*320 image input size down-sampling 16 obtain again, the three dimensionality matrix: coordinate value, under
Sampling error value and obstacle classification attribute three parts.
As a further improvement, the down-sampling error amount is former coordinate value divided by the remainder after down-sampling multiple, with
Promote the accuracy of output result.
As a further improvement, the obstacle classification attribute includes: general barrier and curb, and the general barrier
Hinder object be denoted as 1 and curb be denoted as 0.
As a further improvement, the calculating in described image post-processing are as follows: adopt the coordinate value of output under
The result of sample error amount combines, and is calculated according to down-sampling multiple.
The present invention proposes a kind of road drivable region detection method for being based purely on visual sensor, realizes structuring road
The vehicle on road can travel the detection method in region, carry out path planning to be supplied to the decision system of vehicle.In the process of moving,
Road surface is most safe in the image that travelable region detection module can help automatic driving vehicle analysis to capture based on camera
Running region, and can be vehicle routing plan by detection road edge, barrier, the non-travelable region such as pedestrian
Reference information is provided.
Specific embodiment
The present invention provides a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor, step packet
It includes: inputting the image step of visual sensor, detect the travelable region step in described image, export the seat that can travel region
Punctuate is to Vehicle Decision Method layer step, in which: the described image in the image step of the input visual sensor only passes through vision
Sensor obtains；In travelable region step in the detection described image further include: (1) image preprocessing: to the figure
As carrying out reduction and size adjusting；(2) image detection: being input to network for image after pretreatment and feedover to it, detection barrier
The boundary for hindering object returns out a travelable area coordinate point according to file for image after the pretreatment, described to can travel
Area coordinate point is the down contour point of the nearest barrier in distance shooting source, and exports the travelable area coordinate point；(3) figure
As post-processing: handle the travelable area coordinate point of the network, be computed restore more accurate coordinate value as a result,
Calculate the travelable region in former described image.
The present invention can only pass through the visual sensor system in automatic driving vehicle sensory perceptual system, structured road vehicle
It can travel region detection, can travel the statement in region and its output to lower layer, i.e. Vehicle Decision Method planning module, carry out vehicle row
Sail path planning.
As a further improvement, the reduction are as follows: it is big that described image from 1920*1208 size is cut to 1920*640
It is small, to ignore interference region；The size adjusting are as follows: 2 times of image down sampling after the reduction are arrived with the size of 960*320.
As a further improvement, it is described reduction be since the longitudinal coordinate 430 of described image to longitudinal coordinate 1070
Until, and retain 640 points between the two.
As a further improvement, the travelable area coordinate point is returned out according to every 16 files, with compression
The information content of output simultaneously promotes speed.
As a further improvement, it is feedovered in described image detection by Caffe engine.
As a further improvement, the travelable area coordinate point in described image detection is 60*20* three-dimensional
Spend matrix, in which: 60*20 be 960*320 image input size down-sampling 16 obtain again, the three dimensionality matrix: coordinate value, under
Sampling error value and obstacle classification attribute three parts.
As a further improvement, the down-sampling error amount is former coordinate value divided by the remainder after down-sampling multiple, with
Promote the accuracy of output result.
As a further improvement, the obstacle classification attribute includes: general barrier and curb, and the general barrier
Hinder object be denoted as 1 and curb be denoted as 0.
As a further improvement, the calculating in described image post-processing are as follows: adopt the coordinate value of output under
The result of sample error amount combines, and is calculated according to down-sampling multiple.
The present invention proposes a kind of road drivable region detection method for being based purely on visual sensor, realizes structuring road
The vehicle on road can travel the detection method in region, carry out path planning to be supplied to the decision system of vehicle.In the process of moving,
Road surface is most safe in the image that travelable region detection module can help automatic driving vehicle analysis to capture based on camera
Running region, and can be vehicle routing plan by detection road edge, barrier, the non-travelable region such as pedestrian
Reference information is provided.
The travelable region of vehicle includes the road surface, semi-structured road surface, non-structured road surface of structuring.Structure
The road surface of change is usually to have road edge line, and pavement structure is single, such as major urban arterial highway, high speed, national highway, provincial highway etc., this
The structure sheaf on road surface executes certain standard, and the color and material of surface layer are unified.Semi-structured road surface refers to general nonstandard
The road surface of standardization, top course are that color and material differ greatly, such as parking lot, square etc., and there are also some distributor roads.It is non-
The road surface of structuring does not have structure sheaf, natural road scene.The present invention temporarily pays close attention to structuring road surface.
The algorithm on boundary and coordinate points:
Based on the thought of " Stixel World " (cylindrical region), detection of obstacles and road surface segmentation problem are converted into one
A deep learning regression problem focuses on barrier/curb boundary, for each column of image, returns out a coordinate points,
This coordinate points is the down contour point of the nearest traveling-prohibited area (barrier/curb) in distance shooting source.This coordinate position is of equal value
In the farthest point for sailing region based on present frame safe and feasible.Also, for the method because required calculation amount is relatively fewer, speed is opposite
The method of deep learning segmentation faster, can accomplish in real time, the needs of meeting us.This module is using using deep learning
The mode that coordinate points return solves the problems, such as this.When actual vehicle is tested, it can travel region and show.It is to be noted that feasible
Sailing the boundary between region and barrier is present invention coordinate points of interest, in road image, traveling-prohibited area (barrier
Hinder object/curb) down contour point, as can travel the up contour point in region, therebetween shared a line circle.
Image of the present invention inputs preprocessing module first, is cut out ROI region, is adjusted to the input ruler of network model requirement
Entering after very little can travel region detection module, enters post-processing module after returning out the original prediction result of network, post-processes mould
Block can be calculated in original image using the original output information of primitive network, be can travel the coordinate points in region, be then output to downstream
Module.
Image preprocessing:
In order to further enhance speed, input picture is cut first, reduces the redundant computation of extraneous areas, from
1920*1208 input is cut to 1920*640, ignores excessively high region (sky, cloud) in image, then adjusts greatly to image
Small processing, 2 times of down-sampling to 960*320 of size；The another way for promoting speed is the information content of compression output: sampling
Every 16 column return the coordinate points of a file.Image preprocessing submodule cuts out input picture, size change over operations, will be defeated
Enter the input size that Image Adjusting to image detection submodule needs, the input size that can travel region detection module is 960*
320, it is cut out by original image to 1920*640 down-sampling and is obtained for twice.Original image is 1920*1208 size, because of Chinese herbaceous peony portion engine
The road surface that certain area can be blocked, the ordinate starting point selected when cutting out to 1920*640 size is 430, by test, indulges and sits
640 all road surfaces of point between mark 430 to 1070, it is maximum comprising information content, ignore the image at both ends.
Image detection:
Pretreated image is input to network, is feedovered by Caffe engine to picture, output result point three
Point: region ordinate point, down-sampling error amount and current point categorical attribute can be exercised.Down-sampling error amount can be understood as former seat
Scale value is recorded divided by the remainder after down-sampling multiple (N) for promoting output result accuracy.The final output knot of network
Fruit is the matrix of a 60*20*3 dimension.60 and 20 be that 960 and 320 images input size down-sampling 16 obtains again, is stored in
Prediction result of the current size lower network to three dimensional informations；3 dimensions respectively represent the location information of coordinate points, down-sampling
The difference information and the categorical attribute of 16 times of generation errors.
In view of the use of downstream module, this module is while returning out the coordinate points that can travel zone boundary also to every
The categorical attribute (curb, barrier) of a point judges, and represents the physical significance of each point.
Post processing of image:
Module is responsible for handling the original output of network, the result of the ordinate value of output and down-sampling error amount is combined,
Assuming that down-sampling times, more accurate coordinate value result is restored according to formula.
By the above method, detects that vehicle front can travel region, testing result is retouched with following data structure
It states.With x_position, the y_position of double type, the coordinate in figure is represented, it is most lower to record histogram used in figure
Hold the position of point at recurrence.The point_class parameter of integer is defined in another external feature, 0 represents curb, and 1 represents obstacle
The type of object, subsequent barrier expands, and the selectable value of this classification parameter will also increase therewith.
It should be appreciated that invention which is intended to be protected is not limited to non-limiting embodiments, it should be understood that non-limiting embodiment party
Case is illustrated as just example.The protection scope of the desired essence of the application is more embodied in independent claims offer
Range and its dependent claims.
Claims (9)
1. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor, step include: input vision
The image step of sensor detects the travelable region step in described image, exports the coordinate points that can travel region to vehicle
Decision-making level's step, it is characterised in that:
Described image in the image step of the input visual sensor is only obtained by visual sensor；
In travelable region step in the detection described image further include:
(1) reduction and size adjusting image preprocessing: are carried out to described image；
(2) image detection: being input to network for image after pretreatment and feedover to it, detects the boundary of barrier, for institute
It states after pretreatment image and returns out a travelable area coordinate point according to file, the travelable area coordinate point is that distance is clapped
The down contour point of the nearest barrier in source is taken the photograph, and exports the travelable area coordinate point；
(3) post processing of image: the travelable area coordinate point of the network is handled, is computed and restores more accurate seat
Scale value is as a result, calculate the travelable region in former described image.
2. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor as described in claim 1,
It is characterized in that:
The reduction are as follows: described image is cut to 1920*640 size from 1920*1208 size, to ignore interference region；
The size adjusting are as follows: 2 times of image down sampling after the reduction are arrived with the size of 960*320.
3. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor as claimed in claim 2,
It is characterized in that:
It is described reduction be since the longitudinal coordinate 430 of described image start until longitudinal coordinate 1070, and retain therebetween
640 points.
4. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor as claimed in claim 2,
It is characterized in that: returning out the travelable area coordinate point according to every 16 files, to compress the information content of output and mention
Lifting speed.
5. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor as described in claim 1,
It is characterized in that: being feedovered in described image detection by Caffe engine.
6. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor as described in claim 1,
Be characterized in that: the travelable area coordinate point in described image detection is 60*20* three dimensionality matrix, in which: 60*20
It is that 960*320 image input size down-sampling 16 obtains again, the three dimensionality matrix: coordinate value, down-sampling error amount and obstacle
Object categorical attribute three parts.
7. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor as claimed in claim 6,
Be characterized in that: the down-sampling error amount is former coordinate value divided by the remainder after down-sampling multiple, to promote the essence of output result
Exactness.
8. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor as claimed in claim 7,
Be characterized in that: the obstacle classification attribute includes: general barrier and curb, and the general barrier is denoted as 1 and curb
It is denoted as 0.
9. a kind of automatic driving vehicle drivable region detection method of view-based access control model sensor as claimed in claim 8,
It is characterized in that: the calculating in described image post-processing are as follows: by the result knot of the coordinate value of output and down-sampling error amount
It closes, and is calculated according to down-sampling multiple.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811511924.2A CN110084086A (en) | 2018-12-11 | 2018-12-11 | A kind of automatic driving vehicle drivable region detection method of view-based access control model sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811511924.2A CN110084086A (en) | 2018-12-11 | 2018-12-11 | A kind of automatic driving vehicle drivable region detection method of view-based access control model sensor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110084086A true CN110084086A (en) | 2019-08-02 |
Family
ID=67412947
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811511924.2A Pending CN110084086A (en) | 2018-12-11 | 2018-12-11 | A kind of automatic driving vehicle drivable region detection method of view-based access control model sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110084086A (en) |
Cited By (37)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111291650A (en) * | 2020-01-21 | 2020-06-16 | 北京百度网讯科技有限公司 | Method and device for automatic parking assistance |
| CN112172828A (en) * | 2020-10-14 | 2021-01-05 | 斑马网络技术有限公司 | Narrow road passage method, device, electronic device, and computer-readable storage medium |
| CN112419154A (en) * | 2020-11-26 | 2021-02-26 | 三一专用汽车有限责任公司 | Method, device, equipment and computer readable storage medium for detecting travelable area |
| CN112419776A (en) * | 2020-10-16 | 2021-02-26 | 爱驰汽车（上海）有限公司 | Autonomous parking method and device, automobile and computing equipment |
| CN113744518A (en) * | 2020-05-30 | 2021-12-03 | 华为技术有限公司 | Method and device for detecting vehicle travelable area |
| CN113838060A (en) * | 2020-06-23 | 2021-12-24 | 图森有限公司 | Perception systems for autonomous vehicles |
| CN114282597A (en) * | 2020-05-11 | 2022-04-05 | 华为技术有限公司 | Method and system for detecting vehicle travelable area and automatic driving vehicle adopting system |
| CN114359869A (en) * | 2021-12-31 | 2022-04-15 | 中国第一汽车股份有限公司 | Method and device for detecting boundary on vehicle driving area |
| CN114565904A (en) * | 2022-03-02 | 2022-05-31 | 北京百度网讯科技有限公司 | Method, apparatus, apparatus, medium and vehicle for determining a drivable area |
| CN114648744A (en) * | 2020-12-02 | 2022-06-21 | Aptiv技术有限公司 | Method for determining semantic collision-free space |
| CN114663397A (en) * | 2022-03-22 | 2022-06-24 | 小米汽车科技有限公司 | Method, device, equipment and storage medium for detecting travelable area |
| CN114821504A (en) * | 2022-04-29 | 2022-07-29 | 上海商汤临港智能科技有限公司 | Object detection, driving control method, device, electronic device, and storage medium |
| US11403069B2 (en) | 2017-07-24 | 2022-08-02 | Tesla, Inc. | Accelerated mathematical engine |
| US11409692B2 (en) | 2017-07-24 | 2022-08-09 | Tesla, Inc. | Vector computational unit |
| US11487288B2 (en) | 2017-03-23 | 2022-11-01 | Tesla, Inc. | Data synthesis for autonomous control systems |
| US11537811B2 (en) | 2018-12-04 | 2022-12-27 | Tesla, Inc. | Enhanced object detection for autonomous vehicles based on field view |
| US11562231B2 (en) | 2018-09-03 | 2023-01-24 | Tesla, Inc. | Neural networks for embedded devices |
| US11561791B2 (en) | 2018-02-01 | 2023-01-24 | Tesla, Inc. | Vector computational unit receiving data elements in parallel from a last row of a computational array |
| US11567514B2 (en) | 2019-02-11 | 2023-01-31 | Tesla, Inc. | Autonomous and user controlled vehicle summon to a target |
| CN115761662A (en) * | 2022-09-16 | 2023-03-07 | 浙江华锐捷技术有限公司 | Feasible region detection method and device |
| US11610117B2 (en) | 2018-12-27 | 2023-03-21 | Tesla, Inc. | System and method for adapting a neural network model on a hardware platform |
| US11636333B2 (en) | 2018-07-26 | 2023-04-25 | Tesla, Inc. | Optimizing neural network structures for embedded systems |
| CN116052122A (en) * | 2023-01-28 | 2023-05-02 | 广汽埃安新能源汽车股份有限公司 | Method and device for detecting drivable space, electronic equipment and storage medium |
| US11665108B2 (en) | 2018-10-25 | 2023-05-30 | Tesla, Inc. | QoS manager for system on a chip communications |
| US11681649B2 (en) | 2017-07-24 | 2023-06-20 | Tesla, Inc. | Computational array microprocessor system using non-consecutive data formatting |
| US11734562B2 (en) | 2018-06-20 | 2023-08-22 | Tesla, Inc. | Data pipeline and deep learning system for autonomous driving |
| US11748620B2 (en) | 2019-02-01 | 2023-09-05 | Tesla, Inc. | Generating ground truth for machine learning from time series elements |
| US11790664B2 (en) | 2019-02-19 | 2023-10-17 | Tesla, Inc. | Estimating object properties using visual image data |
| US11816585B2 (en) | 2018-12-03 | 2023-11-14 | Tesla, Inc. | Machine learning models operating at different frequencies for autonomous vehicles |
| US11841434B2 (en) | 2018-07-20 | 2023-12-12 | Tesla, Inc. | Annotation cross-labeling for autonomous control systems |
| US11893774B2 (en) | 2018-10-11 | 2024-02-06 | Tesla, Inc. | Systems and methods for training machine models with augmented data |
| US11893393B2 (en) | 2017-07-24 | 2024-02-06 | Tesla, Inc. | Computational array microprocessor system with hardware arbiter managing memory requests |
| CN118038409A (en) * | 2024-02-27 | 2024-05-14 | 重庆赛力斯凤凰智创科技有限公司 | Vehicle drivable area detection method, device, electronic equipment and storage medium |
| US12014553B2 (en) | 2019-02-01 | 2024-06-18 | Tesla, Inc. | Predicting three-dimensional features for autonomous driving |
| US12307350B2 (en) | 2018-01-04 | 2025-05-20 | Tesla, Inc. | Systems and methods for hardware-based pooling |
| US12462575B2 (en) | 2021-08-19 | 2025-11-04 | Tesla, Inc. | Vision-based machine learning model for autonomous driving with adjustable virtual camera |
| US12522243B2 (en) | 2021-08-19 | 2026-01-13 | Tesla, Inc. | Vision-based system training with simulated content |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485233A (en) * | 2016-10-21 | 2017-03-08 | 深圳地平线机器人科技有限公司 | Drivable region detection method, device and electronic equipment |
| CN108470147A (en) * | 2017-02-23 | 2018-08-31 | 福特全球技术公司 | It is detected using monocular-camera and the free space of deep learning |
-
2018
- 2018-12-11 CN CN201811511924.2A patent/CN110084086A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485233A (en) * | 2016-10-21 | 2017-03-08 | 深圳地平线机器人科技有限公司 | Drivable region detection method, device and electronic equipment |
| CN108470147A (en) * | 2017-02-23 | 2018-08-31 | 福特全球技术公司 | It is detected using monocular-camera and the free space of deep learning |
Non-Patent Citations (2)
| Title |
|---|
| DAN LEVI等: "StixelNet: A Deep Convolutional Network for Obstacle Detection and Road Segmentation", 《BRITISH MACHINE VISION CONFERENCE 2015》 * |
| MARIUS CORDTS等: "The Stixel World: A medium-level representation of traffic scenes", 《IMAGE AND VISION COMPUTING》 * |
Cited By (57)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12020476B2 (en) | 2017-03-23 | 2024-06-25 | Tesla, Inc. | Data synthesis for autonomous control systems |
| US11487288B2 (en) | 2017-03-23 | 2022-11-01 | Tesla, Inc. | Data synthesis for autonomous control systems |
| US11403069B2 (en) | 2017-07-24 | 2022-08-02 | Tesla, Inc. | Accelerated mathematical engine |
| US12536131B2 (en) | 2017-07-24 | 2026-01-27 | Tesla, Inc. | Vector computational unit |
| US12216610B2 (en) | 2017-07-24 | 2025-02-04 | Tesla, Inc. | Computational array microprocessor system using non-consecutive data formatting |
| US11681649B2 (en) | 2017-07-24 | 2023-06-20 | Tesla, Inc. | Computational array microprocessor system using non-consecutive data formatting |
| US12086097B2 (en) | 2017-07-24 | 2024-09-10 | Tesla, Inc. | Vector computational unit |
| US11893393B2 (en) | 2017-07-24 | 2024-02-06 | Tesla, Inc. | Computational array microprocessor system with hardware arbiter managing memory requests |
| US11409692B2 (en) | 2017-07-24 | 2022-08-09 | Tesla, Inc. | Vector computational unit |
| US12307350B2 (en) | 2018-01-04 | 2025-05-20 | Tesla, Inc. | Systems and methods for hardware-based pooling |
| US11797304B2 (en) | 2018-02-01 | 2023-10-24 | Tesla, Inc. | Instruction set architecture for a vector computational unit |
| US12455739B2 (en) | 2018-02-01 | 2025-10-28 | Tesla, Inc. | Instruction set architecture for a vector computational unit |
| US11561791B2 (en) | 2018-02-01 | 2023-01-24 | Tesla, Inc. | Vector computational unit receiving data elements in parallel from a last row of a computational array |
| US11734562B2 (en) | 2018-06-20 | 2023-08-22 | Tesla, Inc. | Data pipeline and deep learning system for autonomous driving |
| US11841434B2 (en) | 2018-07-20 | 2023-12-12 | Tesla, Inc. | Annotation cross-labeling for autonomous control systems |
| US12079723B2 (en) | 2018-07-26 | 2024-09-03 | Tesla, Inc. | Optimizing neural network structures for embedded systems |
| US11636333B2 (en) | 2018-07-26 | 2023-04-25 | Tesla, Inc. | Optimizing neural network structures for embedded systems |
| US11983630B2 (en) | 2018-09-03 | 2024-05-14 | Tesla, Inc. | Neural networks for embedded devices |
| US11562231B2 (en) | 2018-09-03 | 2023-01-24 | Tesla, Inc. | Neural networks for embedded devices |
| US12346816B2 (en) | 2018-09-03 | 2025-07-01 | Tesla, Inc. | Neural networks for embedded devices |
| US11893774B2 (en) | 2018-10-11 | 2024-02-06 | Tesla, Inc. | Systems and methods for training machine models with augmented data |
| US11665108B2 (en) | 2018-10-25 | 2023-05-30 | Tesla, Inc. | QoS manager for system on a chip communications |
| US11816585B2 (en) | 2018-12-03 | 2023-11-14 | Tesla, Inc. | Machine learning models operating at different frequencies for autonomous vehicles |
| US12367405B2 (en) | 2018-12-03 | 2025-07-22 | Tesla, Inc. | Machine learning models operating at different frequencies for autonomous vehicles |
| US11908171B2 (en) | 2018-12-04 | 2024-02-20 | Tesla, Inc. | Enhanced object detection for autonomous vehicles based on field view |
| US11537811B2 (en) | 2018-12-04 | 2022-12-27 | Tesla, Inc. | Enhanced object detection for autonomous vehicles based on field view |
| US12198396B2 (en) | 2018-12-04 | 2025-01-14 | Tesla, Inc. | Enhanced object detection for autonomous vehicles based on field view |
| US12136030B2 (en) | 2018-12-27 | 2024-11-05 | Tesla, Inc. | System and method for adapting a neural network model on a hardware platform |
| US11610117B2 (en) | 2018-12-27 | 2023-03-21 | Tesla, Inc. | System and method for adapting a neural network model on a hardware platform |
| US12014553B2 (en) | 2019-02-01 | 2024-06-18 | Tesla, Inc. | Predicting three-dimensional features for autonomous driving |
| US12223428B2 (en) | 2019-02-01 | 2025-02-11 | Tesla, Inc. | Generating ground truth for machine learning from time series elements |
| US11748620B2 (en) | 2019-02-01 | 2023-09-05 | Tesla, Inc. | Generating ground truth for machine learning from time series elements |
| US12164310B2 (en) | 2019-02-11 | 2024-12-10 | Tesla, Inc. | Autonomous and user controlled vehicle summon to a target |
| US11567514B2 (en) | 2019-02-11 | 2023-01-31 | Tesla, Inc. | Autonomous and user controlled vehicle summon to a target |
| US11790664B2 (en) | 2019-02-19 | 2023-10-17 | Tesla, Inc. | Estimating object properties using visual image data |
| US12236689B2 (en) | 2019-02-19 | 2025-02-25 | Tesla, Inc. | Estimating object properties using visual image data |
| CN111291650A (en) * | 2020-01-21 | 2020-06-16 | 北京百度网讯科技有限公司 | Method and device for automatic parking assistance |
| CN111291650B (en) * | 2020-01-21 | 2023-06-20 | 北京百度网讯科技有限公司 | Automatic parking assisting method and device |
| US12340593B2 (en) | 2020-05-11 | 2025-06-24 | Shenzhen Yinwang Intelligent Technologies Co., Ltd. | Vehicle drivable area detection method, system, and autonomous vehicle using the system |
| CN114282597A (en) * | 2020-05-11 | 2022-04-05 | 华为技术有限公司 | Method and system for detecting vehicle travelable area and automatic driving vehicle adopting system |
| CN113744518B (en) * | 2020-05-30 | 2023-04-18 | 华为技术有限公司 | Method and device for detecting vehicle travelable area |
| CN113744518A (en) * | 2020-05-30 | 2021-12-03 | 华为技术有限公司 | Method and device for detecting vehicle travelable area |
| CN113838060A (en) * | 2020-06-23 | 2021-12-24 | 图森有限公司 | Perception systems for autonomous vehicles |
| CN112172828A (en) * | 2020-10-14 | 2021-01-05 | 斑马网络技术有限公司 | Narrow road passage method, device, electronic device, and computer-readable storage medium |
| CN112419776A (en) * | 2020-10-16 | 2021-02-26 | 爱驰汽车（上海）有限公司 | Autonomous parking method and device, automobile and computing equipment |
| CN112419154A (en) * | 2020-11-26 | 2021-02-26 | 三一专用汽车有限责任公司 | Method, device, equipment and computer readable storage medium for detecting travelable area |
| CN114648744A (en) * | 2020-12-02 | 2022-06-21 | Aptiv技术有限公司 | Method for determining semantic collision-free space |
| US12522243B2 (en) | 2021-08-19 | 2026-01-13 | Tesla, Inc. | Vision-based system training with simulated content |
| US12462575B2 (en) | 2021-08-19 | 2025-11-04 | Tesla, Inc. | Vision-based machine learning model for autonomous driving with adjustable virtual camera |
| CN114359869B (en) * | 2021-12-31 | 2025-09-09 | 中国第一汽车股份有限公司 | Method and device for detecting boundary on vehicle driving area |
| CN114359869A (en) * | 2021-12-31 | 2022-04-15 | 中国第一汽车股份有限公司 | Method and device for detecting boundary on vehicle driving area |
| CN114565904A (en) * | 2022-03-02 | 2022-05-31 | 北京百度网讯科技有限公司 | Method, apparatus, apparatus, medium and vehicle for determining a drivable area |
| CN114663397A (en) * | 2022-03-22 | 2022-06-24 | 小米汽车科技有限公司 | Method, device, equipment and storage medium for detecting travelable area |
| CN114821504A (en) * | 2022-04-29 | 2022-07-29 | 上海商汤临港智能科技有限公司 | Object detection, driving control method, device, electronic device, and storage medium |
| CN115761662A (en) * | 2022-09-16 | 2023-03-07 | 浙江华锐捷技术有限公司 | Feasible region detection method and device |
| CN116052122A (en) * | 2023-01-28 | 2023-05-02 | 广汽埃安新能源汽车股份有限公司 | Method and device for detecting drivable space, electronic equipment and storage medium |
| CN118038409A (en) * | 2024-02-27 | 2024-05-14 | 重庆赛力斯凤凰智创科技有限公司 | Vehicle drivable area detection method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110084086A (en) | A kind of automatic driving vehicle drivable region detection method of view-based access control model sensor | |
| Tümen et al. | Intersections and crosswalk detection using deep learning and image processing techniques | |
| Song et al. | Real-time obstacles detection and status classification for collision warning in a vehicle active safety system | |
| EP2574958B1 (en) | Road-terrain detection method and system for driver assistance systems | |
| Song et al. | Lane detection and classification for forward collision warning system based on stereo vision | |
| McCall et al. | Video-based lane estimation and tracking for driver assistance: survey, system, and evaluation | |
| US10137890B2 (en) | Occluded obstacle classification for vehicles | |
| CN110531376A (en) | Detection of obstacles and tracking for harbour automatic driving vehicle | |
| Huang et al. | On-board vision system for lane recognition and front-vehicle detection to enhance driver's awareness | |
| Wu et al. | Applying a functional neurofuzzy network to real-time lane detection and front-vehicle distance measurement | |
| CN106199558A (en) | Barrier method for quick | |
| WO2018133851A1 (en) | Point cloud data processing method and apparatus, and computer storage medium | |
| DE102019118999A1 (en) | LIDAR-BASED OBJECT DETECTION AND CLASSIFICATION | |
| CN111027430A (en) | Traffic scene complexity calculation method for intelligent evaluation of unmanned vehicles | |
| Tumen et al. | Recognition of road type and quality for advanced driver assistance systems with deep learning | |
| DE102021132082A1 (en) | END-TO-END SYSTEM TRAINING USING UNITED IMAGES | |
| CN114119896B (en) | Driving path planning method | |
| CN115294545A (en) | Complex road surface lane identification method and chip based on deep learning | |
| Kühnl et al. | Visual ego-vehicle lane assignment using spatial ray features | |
| Labayrade et al. | Robust and fast stereovision based obstacles detection for driving safety assistance | |
| Aron et al. | Current approaches in traffic lane detection: a minireview | |
| Cerri et al. | Computer vision at the hyundai autonomous challenge | |
| DE102023109040A1 (en) | UNIFIED FRAMEWORK AND TOOLS FOR ANNOTATION OF LANE BOUNDARIES | |
| US12460941B2 (en) | Map generation apparatus | |
| Singh et al. | Lane estimation using improved Hough transform for isolated and metro highway |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190802 |