CN114495065A - Target object identification method and device - Google Patents
Target object identification method and device Download PDFInfo
- Publication number
- CN114495065A CN114495065A CN202210104778.1A CN202210104778A CN114495065A CN 114495065 A CN114495065 A CN 114495065A CN 202210104778 A CN202210104778 A CN 202210104778A CN 114495065 A CN114495065 A CN 114495065A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- obstacle
- image
- target object
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/497—Means for monitoring or calibrating
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The specification discloses a method and a device for identifying a target object, and particularly discloses that a target image acquired by unmanned equipment for the target object and point cloud data scanned when the target image is acquired are acquired, then an imaging area of the image of the target object in the target image is predicted, point cloud data of an obstacle between the target object and the unmanned equipment is identified from the point cloud data, an imaging area of the image of the obstacle in the target image is determined based on the identified point cloud data, and then the target object is identified through the target image after the target object is determined not to be shielded by the obstacle in the target image according to the imaging area of the target object and the imaging area of the obstacle. Therefore, the accuracy of the target object identified by the unmanned equipment is improved, and the driving safety of the unmanned equipment is further guaranteed.
Description
Technical Field
The specification relates to the technical field of unmanned driving, in particular to a method and a device for identifying a target object.
Background
When the unmanned equipment realizes automatic driving, the signal lamp sensing module is required to be utilized to identify each signal lamp positioned on the traveling path of the unmanned equipment, and the unmanned equipment runs according to the identified result.
When the signal lamp on the advancing path of the unmanned equipment is sensed, the unmanned equipment needs to collect images containing the signal lamp, based on a preset high-precision map, according to positioning data of the unmanned equipment, when the signal lamp is shot by a camera mounted on the unmanned equipment in a pose mode when the collected image is used for predicting the position of the signal lamp in the image, and then an image area needing to be subjected to traffic light sensing detection is marked out of the collected image according to the position. And finally, the unmanned equipment identifies the color of the signal lamp of the image area, determines the color of the collected signal lamp, and controls the unmanned equipment to run according to the determined color of the signal lamp.
When discerning the signal lamp based on above-mentioned mode, if the signal lamp in unmanned aerial vehicle equipment the place ahead is sheltered from, this signal lamp perception module continues to carry out the signal lamp detection in the image to the camera collection, still probably monitors the signal lamp from it, like this, when unmanned aerial vehicle equipment carries out autopilot according to the testing result, will have very big safety risk.
Disclosure of Invention
The present specification provides a method and an apparatus for object recognition, which partially solve the above problems in the prior art.
The technical scheme adopted by the specification is as follows:
the present specification provides a method of object identification, comprising:
acquiring a target image acquired by unmanned equipment aiming at a target object and point cloud data scanned when the target image is acquired;
predicting an imaging area of the image of the target object in the target image as a target object image area, and recognizing point cloud data of an obstacle located between the target object and the unmanned aerial vehicle from the point cloud data as obstacle point cloud data, and determining the imaging area of the image of the obstacle in the target image as an obstacle image area based on the obstacle point cloud data;
judging whether the target object is blocked by the obstacle in the target image or not according to the target object image area and the obstacle image area;
and if the target object is not blocked, identifying the target object through the target image.
Optionally, predicting an imaging region of the image of the target object in the target image comprises:
determining a relative pose between the target object and the unmanned equipment as a first relative pose according to the spatial position of the target object recorded in a preset high-precision map and the pose when the unmanned equipment acquires the target image;
determining a relative pose between a camera and the target object when the target image is acquired as a second relative pose according to the first relative pose and a relative pose between the camera arranged on the unmanned equipment and the unmanned equipment;
and predicting an imaging area of the image of the target object in the target image according to the second relative pose.
Optionally, identifying point cloud data of an obstacle located between the target object and the unmanned device from the point cloud data specifically includes:
determining a relative pose between the target object and the unmanned equipment as a third relative pose according to the spatial position of the target object recorded in a preset high-precision map and the pose when the unmanned equipment scans the point cloud data;
determining a relative pose between the laser radar and the target object when the point cloud data is scanned according to the first relative pose and a relative pose between the laser radar arranged on the unmanned equipment and the unmanned equipment, and taking the relative pose as a fourth relative pose;
predicting the area occupied by the target object in the point cloud data according to the fourth relative pose to serve as a target object point cloud area;
determining an area capable of covering the target object point cloud area from the point cloud data as an area of interest corresponding to the target object point cloud area;
identifying point cloud data of an obstacle located between the target object and the unmanned device from the point cloud data contained within the region of interest.
Optionally, determining, from the area scanned by the laser radar, an area that can cover the target object point cloud area as an area of interest corresponding to the target object point cloud area, specifically including:
determining the center of the target object point cloud area, and expanding the target object point cloud area by taking the center as an area center according to a preset offset value;
and taking the expanded target point cloud area as an interesting area corresponding to the target point cloud area.
Optionally, identifying point cloud data of an obstacle located between the target object and the unmanned aerial vehicle from the point cloud data, as obstacle point cloud data, specifically includes:
for each point cloud point, determining the distance between the point cloud point and the unmanned equipment when the unmanned equipment collects the point cloud point as a first distance corresponding to the point cloud point, and determining the distance between the target object and the unmanned equipment when the unmanned equipment collects the point cloud point as a second distance corresponding to the point cloud point;
and determining the cloud points of the obstacle points according to the cloud points of which the first distance is smaller than the second distance, and determining the cloud data of the obstacle points according to the determined cloud points of the obstacle points.
Optionally, determining the cloud point of the obstacle point according to the cloud point of the point where the first distance is smaller than the second distance, specifically including:
aiming at each point cloud point, if the first distance corresponding to the point cloud point is smaller than the second distance corresponding to the point cloud point, taking the point cloud point as a candidate point cloud point;
and recognizing the point cloud point of the obstacle between the target object and the unmanned equipment from the candidate point cloud points to determine the obstacle point cloud data by taking the condition that the distance between two adjacent point cloud points contained in the point cloud data belonging to the same obstacle is smaller than a set distance as a constraint condition.
Optionally, determining an imaging area of the image of the obstacle in the target image based on the obstacle point cloud data, as an obstacle image area, specifically including:
determining the coordinates of corresponding pixel points of the point cloud points in the target image as the image coordinates corresponding to the point cloud points aiming at each point cloud point in the obstacle point cloud data;
and connecting pixel points corresponding to cloud points of each point in the obstacle point cloud data in the target image based on the relative positions of the cloud points of each point in the obstacle point cloud data and the image coordinates corresponding to the cloud points of each point in the obstacle point cloud data so as to determine the obstacle image area.
Optionally, the method further comprises:
and if the target object is determined to be shielded by the obstacle in the target image and does not pass through the target image, identifying the target object.
Optionally, the target comprises: a signal lamp.
The present specification provides an apparatus for object recognition, comprising:
the data acquisition module is used for acquiring a target image acquired by the unmanned equipment aiming at a target object and point cloud data scanned when the target image is acquired;
an image area determination module for predicting an imaging area of the image of the target object in the target image as a target object image area, identifying point cloud data of an obstacle located between the target object and the unmanned aerial vehicle from the point cloud data as obstacle point cloud data, and determining the imaging area of the image of the obstacle in the target image as an obstacle image area based on the obstacle point cloud data;
the judging module is used for judging whether the target object is shielded by the obstacle in the target image according to the target object image area and the obstacle image area;
and the identification module is used for identifying the target object through the target image if the target object is not blocked.
The present specification provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the above-described method of object identification.
The present specification provides an unmanned device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the above-described method of object identification when executing the program.
The technical scheme adopted by the specification can achieve the following beneficial effects:
in the method for identifying the target object, firstly, a target image collected by the unmanned equipment for the target object and point cloud data scanned when the target image is collected are obtained, then, an imaging area of the image of the object in the object image is predicted as an object image area, meanwhile, point cloud data of an obstacle located between the target object and the unmanned aerial vehicle is identified from the scanned point cloud data as obstacle point cloud data, and determining an imaging area of the image of the obstacle in the target image as an obstacle image area based on the identified obstacle point cloud data, and finally, and judging whether the target object is blocked by the obstacle in the target image or not according to the target object image area and the obstacle image area, and if not, identifying the target object through the target image.
According to the method, after the unmanned device acquires the target object image, whether the image contains the target object to be recognized is judged according to the predicted target object image area and the predicted obstacle image area, and the image is recognized only when the image contains the target object to be recognized is determined, so that the target object is recognized only on the image where the target object is not shielded by the obstacle, the target object recognized from the image can be guaranteed, namely the target object required to be perceived when the unmanned device runs, the accuracy of the target object recognized by the unmanned device is improved, and the running safety of the unmanned device is further guaranteed.
Drawings
The accompanying drawings, which are included to provide a further understanding of the specification and are incorporated in and constitute a part of this specification, illustrate embodiments of the specification and together with the description serve to explain the specification and not to limit the specification in a non-limiting sense. In the drawings:
FIG. 1 is a schematic flow chart of object recognition in the present specification;
FIGS. 2A-2B are schematic diagrams of a point cloud area of a signal lamp as identified herein;
FIG. 2C is a schematic illustration of the determination of the image area of an obstacle in this description;
FIG. 3 is a detailed flow chart of a method for identifying signal lights and controlling the drone according to the identified signal lights as provided herein;
FIG. 4 is a schematic view of an object recognition apparatus provided herein;
fig. 5 is a schematic diagram of an electronic device corresponding to fig. 1 provided in the present specification.
Detailed Description
To make the objects, technical solutions and advantages of the present specification clearer and more complete, the technical solutions of the present specification will be described in detail and completely with reference to the specific embodiments of the present specification and the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present disclosure, and not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present specification without any creative effort belong to the protection scope of the present specification.
The control scheme of the unmanned aerial device provided in this specification will be described in detail below with reference to embodiments.
Fig. 1 is a schematic flow chart of a method for identifying a target object in this specification, which specifically includes the following steps:
step S100, acquiring a target image acquired by the unmanned equipment aiming at a target object and point cloud data scanned when the target image is acquired.
In an actual scene, the unmanned equipment acquires surrounding image information through a camera in real time in an automatic driving process, meanwhile, scans and acquires surrounding point cloud data through a laser radar, then performs data processing on the acquired image information and the obtained point cloud data to obtain an environment sensing result, and controls the unmanned equipment to run according to the environment sensing result. When the unmanned equipment passes through a road area provided with a specified target object, determining the distance between the target object on a planned driving path and the unmanned equipment based on the spatial position of each target object recorded in a preset high-precision map, and when the distance is determined to be smaller than the set distance, determining that an image aiming at the target object needs to be acquired to identify the target object, so as to control the unmanned equipment to pass through the road area according to the obtained identification result.
When an obstacle exists between the unmanned device and the target object, the obstacle may block the target object in an image actually acquired by the unmanned device, at this time, once a color similar to the target object exists on the surface of the obstacle located between the target object and the unmanned device in the acquired image or an image of the target object in another direction reflected by the surface of the obstacle exists, if the unmanned device still directly performs target object identification on the acquired image, it is identified that the target object is not a target object that the unmanned device needs to identify in an actual scene (i.e., the target object needs to be used for indicating the unmanned device to pass through a road area where the target object exists when the unmanned device passes through the road area where the target object exists), which may have a great potential safety hazard.
To solve the problem, in this specification, when the unmanned device senses the target object according to the captured image, it is determined whether an image of an obstacle exists in an imaging area of the target object in the image with respect to the captured image, and if it is determined that the image is not blocked, it is determined that a target object detection result of the image is reliable, otherwise, it is determined that the target object detection result of the image is not reliable, and it is not possible to control the unmanned device to travel according to the target object detection result of the image
In specific implementation, when the unmanned equipment determines that the target object needs to be perceived, the unmanned equipment acquires a target image acquired by the unmanned equipment for the target object and point cloud data scanned when the target image is acquired.
The target image may be acquired for each target object when the unmanned device determines that the distance between the target object and the unmanned device is smaller than a set distance based on the spatial position of each target object recorded in a preset high-precision map. The point cloud data scanned when the target image is acquired is the point cloud data scanned by the laser radar in the nearest acquisition period away from the target image acquisition time.
The execution subject of the control method of the unmanned device provided in the present specification may be the unmanned device mentioned above, or may be a server providing service support for the unmanned device. The present specification will explain in detail a control method of the unmanned aerial vehicle provided in the present specification, with only the unmanned aerial vehicle as an execution subject.
The above-described unmanned facility may be a facility capable of realizing automatic driving, such as an unmanned vehicle, a robot, and an automatic distribution facility. Based on this, the unmanned device related to this specification can be used for carrying out the delivery task in the delivery field, for example, use unmanned device to carry out the business scene of deliveries such as express delivery, logistics, takeaway.
In the present description, the object to be recognized by the unmanned aerial vehicle may be a signal light installed on a road, but the object may also be other objects such as a road sign, a landmark building, and the like.
Step S102, predicting an imaging area of the image of the target object in the target image to be used as a target object imaging area, identifying point cloud data of an obstacle between the target object and the unmanned equipment from the point cloud data to be used as obstacle point cloud data, and determining the imaging area of the image of the obstacle in the target image to be used as an obstacle image area based on the obstacle point cloud data.
In this step, the unmanned device respectively predicts an imaging area (i.e., a signal lamp imaging area) of a signal lamp (i.e., a target object) which the unmanned device needs to sense in the acquired target image and an imaging area (i.e., an obstacle image area) of an obstacle which is actually detected by the laser radar and is positioned between the signal lamp and the unmanned device in the acquired target image, and further determines whether the signal lamp in the target image is blocked by the obstacle according to whether the predicted overlapping area exists between the imaging area of the signal lamp and the obstacle image area. The following explains how to determine the signal lamp imaging area and the obstacle imaging area, respectively.
Specifically, when the unmanned device determines the signal lamp imaging area, the relative pose between the signal lamp and the unmanned device is determined as a first relative pose according to the spatial position of the signal lamp recorded in a preset high-precision map and the pose when the unmanned device acquires the target image, then the relative pose between the camera and the signal lamp when the target image is acquired is determined as a second relative pose according to the first relative pose and the relative pose between the camera arranged on the unmanned device and the unmanned device, and then the imaging area of the signal lamp image in the target image is predicted as the signal lamp image area according to the second relative pose.
The method comprises the steps that Positioning information of the unmanned equipment can be obtained in real time through a Global Positioning System (GPS) in the driving process of the unmanned equipment, meanwhile, attitude information obtained through real-time Measurement of an Inertial Measurement Unit (IMU) arranged on the unmanned equipment is obtained, then when signal lamp identification needs to be carried out on an acquired image, the Positioning information of the unmanned equipment and the attitude information of the unmanned equipment when the unmanned equipment acquires the target image can be determined based on the acquired target image, and the attitude when the unmanned equipment acquires the target image is obtained. In addition, the relative pose between the camera provided on the unmanned aerial vehicle and the unmanned aerial vehicle can be calibrated in advance and stored on the unmanned aerial vehicle. In addition, the spatial position of the signal lamp recorded in the high-precision map may be a geographic coordinate of the signal lamp in a preset world coordinate system, such as longitude, latitude and altitude of each position on the outline of the signal lamp.
In addition, when the unmanned equipment determines an obstacle image area, the point cloud data of an obstacle between the signal lamp and the unmanned equipment is identified from the point cloud data actually scanned by the laser radar and is used as obstacle point cloud data, and then an imaging area of the image of the obstacle in the target image is determined based on the obstacle point cloud data and is used as the obstacle image area.
In specific implementation, the unmanned equipment firstly determines the relative pose between the laser radar which scans the point cloud data and is arranged on the unmanned equipment and the signal lamp, then predicts the area occupied by the signal lamp in the point cloud data based on the determined relative pose, and then determines the area needing to be subjected to obstacle identification based on the area occupied by the signal lamp in the point cloud data so as to identify the obstacle point cloud data.
Specifically, when the unmanned aerial vehicle determines the relative pose between the laser radar and the signal lamp, the relative pose between the signal lamp and the unmanned aerial vehicle is determined as a third relative pose according to the spatial position of the signal lamp recorded in a preset high-precision map and the pose when the unmanned aerial vehicle scans the point cloud data, and then the relative pose between the laser radar and the signal lamp when the point cloud data is scanned is determined as a fourth relative pose according to the third relative pose and the relative pose between the laser radar and the unmanned aerial vehicle arranged on the unmanned aerial vehicle.
The pose of the unmanned device when the point cloud data is scanned can be determined according to the monitored positioning information of the unmanned device and the pose information obtained by real-time measurement of an IMU (inertial measurement unit) arranged on the unmanned device.
Then, the unmanned aerial vehicle predicts an area occupied by the signal lamp in the point cloud data when the signal lamp is scanned by the laser radar as a point cloud area of the signal lamp according to the determined fourth relative pose (i.e., the relative pose between the laser radar and the signal lamp when the point cloud data is scanned).
In practical application, the predicted point cloud area of the signal lamp is a transverse scanning angle and a longitudinal scanning angle related to point cloud data of the signal lamp when the laser radar scans the signal lamp under the assumption that no obstacle exists between the unmanned device and the signal lamp, and specifically refer to fig. 2A-2B.
After the signal lamp point cloud area is determined, the area which can cover the determined signal lamp point cloud area is determined in the point cloud data actually scanned by the laser radar of the unmanned equipment and is used as the interesting area corresponding to the signal lamp point cloud area, and obstacle identification is carried out on the point cloud points contained in the interesting area so as to identify the point cloud data of the obstacle between the signal lamp and the unmanned equipment from the point cloud data contained in the interesting area.
When the unmanned equipment determines the region of interest corresponding to the signal lamp point cloud region, the center of the signal lamp point cloud region can be determined first, then the signal lamp point cloud region is expanded by taking the center as the region center according to a preset offset value (for example, the transverse scanning angle and the longitudinal scanning angle are respectively expanded by 2 degrees), and then the expanded signal lamp point cloud region is used as the region of interest corresponding to the signal lamp point cloud region. Of course, the region of interest corresponding to the signal lamp point cloud region may also be determined in other manners as long as the signal lamp point cloud region can be covered.
Then, the unmanned device identifies obstacle point cloud data from the point cloud data contained in the region of interest. Specifically, the unmanned aerial vehicle may determine, for each point cloud point located in the area of interest, a distance between the point cloud point and the unmanned aerial vehicle when the unmanned aerial vehicle acquires the point cloud point, as a first distance corresponding to the point cloud point, and determine, at the same time, a distance between the signal lamp and the unmanned aerial vehicle when the unmanned aerial vehicle acquires the point cloud point, as a second distance corresponding to the point cloud point, and then determine, according to the point cloud point where the first distance is smaller than the second distance, obstacle point cloud data of an obstacle located between the signal lamp and the unmanned aerial vehicle.
When the unmanned equipment determines the obstacle point cloud data, whether a first distance corresponding to the point cloud point is smaller than a second distance corresponding to the point cloud point is judged for each point cloud point, and then the point cloud point is used as a candidate point cloud point when the first distance corresponding to the point cloud point is determined to be smaller than the second distance corresponding to the point cloud point. And then, for each candidate point cloud point, determining other candidate point cloud points adjacent to the candidate point cloud point, determining the distance between the other candidate point cloud points and the candidate point cloud point as a third distance corresponding to the candidate point cloud point, and when determining that the third distance between at least one other candidate point cloud point and the candidate point cloud point is smaller than a set distance, determining the candidate point cloud point as an obstacle point cloud point between the unmanned equipment and the signal lamp.
Therefore, the obstacle point cloud points between the unmanned equipment and the signal lamp in the region of interest can be identified, and obstacle point cloud data can be obtained. Therefore, the distance between two adjacent point cloud points contained in the point cloud data belonging to the same obstacle is smaller than the set distance to serve as the constraint condition, the obstacle point cloud points between the signal lamp and the unmanned equipment are identified from the candidate point cloud points, and therefore tiny particles (such as dust particles) between the signal lamp and the unmanned equipment in the air can be prevented from being used as the obstacle, the image which is not shielded by the signal lamp is discarded, and therefore a large amount of effective image data can be lost, and the perception capability of the unmanned equipment on the signal lamp is influenced.
In the above, when the first distance corresponding to the point cloud point is smaller than the second distance corresponding to the point cloud point, it is indicated that the point cloud point is a point cloud point located between the unmanned device and the signal lamp and possibly blocking an obstacle of the signal lamp. In actual scene, the signal lamp often can set up the higher place in road top, like this, when the barrier in unmanned aerial vehicle the place ahead is closer from the signal lamp, because can appear and have certain difference in height between barrier and the signal lamp on the road certainly, therefore the probability that this barrier can shelter from the signal lamp is less, when the barrier in unmanned aerial vehicle the place ahead is closer from unmanned aerial vehicle, then this barrier is more easily sheltered from the field of vision that sets up the camera at unmanned aerial vehicle, under this kind of condition, the condition that the signal lamp was sheltered from appears more easily. Therefore, point cloud data located in the visual field range are selected according to the visual field range of a camera arranged on the unmanned equipment, then point cloud points with the distance between the point cloud data and the unmanned equipment smaller than the set distance are determined from the point cloud data and serve as point cloud data needing obstacle identification, and the point cloud points of obstacles located between the unmanned equipment and a signal lamp are identified from the determined point cloud data. Therefore, obstacle identification can be carried out on the point cloud data near the unmanned equipment, and the service execution efficiency is improved.
After the obstacle point cloud data is obtained, the unmanned equipment can draw an area where the image of the obstacle is located in the target image based on the point cloud points of the obstacle contained in the obstacle point cloud data.
Specifically, the unmanned equipment determines the coordinates of pixel points of the point cloud points in the target image according to the cloud points of each point in the obstacle point cloud data, and the coordinates are used as image coordinates corresponding to the point cloud points, and then the pixel points corresponding to the cloud points of each point in the obstacle point cloud data are connected in the target image based on the relative positions of the cloud points of each point in the obstacle point cloud data and the image coordinates corresponding to the cloud points of each point in the obstacle point cloud data, so that the obstacle image area is determined.
When determining the image coordinates corresponding to the point cloud points of each obstacle, the unmanned equipment acquires and stores the relative pose between a camera arranged on the unmanned equipment and the unmanned equipment, and the relative pose between a laser radar arranged on the unmanned equipment and the unmanned equipment in advance, determines the relative pose between the camera and the laser radar, and then performs coordinate transformation on the point cloud points of the obstacles according to the relative pose between the camera and the laser radar and the internal parameters of the camera to obtain the coordinates of pixel points corresponding to the point cloud points of each obstacle in the image, namely the image coordinates corresponding to the point cloud points. Therefore, according to the obtained image coordinates corresponding to the cloud points of each point, the corresponding pixel points can be found from the target image, and the found pixel points are connected according to the relative positions of the cloud points of each point in the obstacle point cloud data, so that the imaging area of the image of the obstacle in the target image can be obtained.
How to determine the imaging area of the image of the obstacle in the target image will be described below with reference to the drawings.
Referring to fig. 2C, the black square in the graph is a pixel point corresponding to a point cloud point of an obstacle determined from an area of interest corresponding to a signal lamp in the graph, when an imaging area of the obstacle is drawn in the target image, a connection is performed according to the pixel point corresponding to the edge point cloud point of the obstacle, so as to obtain a convex polygon, and an area covered by the convex polygon is an imaging area of the image of the obstacle in the target image.
And step S104, judging whether the target object is blocked in the target image according to the signal lamp image area and the obstacle image area.
And S106, if the target object is not blocked, identifying the target object through the target image.
In specific implementation, after the unmanned equipment determines the signal lamp image area and the obstacle image area, the area range of the signal lamp image area and the area range of the obstacle image area are compared, and whether the signal lamp in the target image is shielded or not is judged according to the obtained comparison result. And if the signal lamp in the target image is not shielded by the obstacle, performing signal lamp identification on the target image, and controlling the unmanned equipment according to the signal lamp states (such as a red lamp state, a green lamp state and the like) of the signal lamp corresponding to the lane where the unmanned equipment is located in the obtained identification result. And if the signal lamp in the target image is blocked by the barrier, the signal lamp identification is not carried out on the target image.
When the area range of the signal lamp image area is compared with the area range of the obstacle image area, it can be determined that the signal lamp in the target image is shielded by the obstacle when the area range of the obstacle image area and the area range of the signal lamp image area are overlapped, otherwise, it is determined that the signal lamp in the target image is not shielded by the obstacle.
In actual service, when the blocked part of the signal lamp image area of the target image is smaller, the signal lamp state of the signal lamp can be identified by identifying the signal lamp aiming at the unblocked signal lamp image area. At this time, if the signal light recognition is not performed on the target image, a large amount of valid data will be lost.
Therefore, when the unmanned device judges whether the signal lamp in the target image is shielded by the obstacle, the size of the overlapped image area overlapped between the area range of the obstacle image area and the area range of the signal lamp image area can be determined, then the ratio of the area size of the overlapped image area to the signal lamp image area is determined, when the ratio is larger than a set threshold value, the signal lamp in the target image is determined to be shielded by the obstacle, otherwise, the signal lamp in the target image is determined not to be shielded by the obstacle.
Through the above-mentioned step, unmanned aerial vehicle is after the signal lamp image of gathering, signal lamp image area and obstacle image area according to predicting earlier, judge whether to include the signal lamp that needs the discernment in this image, and just carry out signal lamp discernment to the image when including the signal lamp that needs the discernment in confirming this image, therefore, only carry out signal lamp discernment to the image that the signal lamp is not sheltered from by the obstacle, can guarantee the signal lamp of discerning from in the image, the signal lamp of required perception when promptly unmanned aerial vehicle traveles, the accuracy of the signal lamp that unmanned aerial vehicle perception arrived has been promoted, unmanned aerial vehicle's the safety of traveling has been ensured.
In addition, in this specification, the unmanned aerial vehicle may further perform signal lamp recognition on each acquired frame of image, and when it is determined that a signal lamp exists in the current image according to the obtained recognition result, by using the scheme in this specification, it is determined whether the signal lamp recognized in the current image is a signal lamp that the unmanned aerial vehicle needs to perceive, if so, the unmanned aerial vehicle is controlled according to the obtained recognition result, and if not, the recognition result for the current image is discarded.
A detailed flowchart of a method for identifying a signal lamp, and controlling an unmanned aerial device according to the identified signal lamp, provided in the present specification, in the case of identifying a signal lamp, will be given below, with specific reference to fig. 3.
Step S300, the unmanned equipment acquires a target image acquired by the unmanned equipment aiming at the signal lamp and point cloud data scanned when the target image is acquired.
In step S302, the unmanned aerial vehicle predicts an imaging area of the image of the signal lamp in the target image as a signal lamp image area.
And step S304, the unmanned equipment predicts the area occupied by the signal lamp in the point cloud data scanned by the laser radar as the point cloud area of the signal lamp.
And S306, determining the center of the signal lamp point cloud area by the unmanned equipment, expanding the signal lamp point cloud area by taking the center as the area center according to a preset offset value, and taking the expanded signal lamp point cloud area as an interesting area corresponding to the signal lamp point cloud area.
Step S308, the unmanned equipment determines the distance between the point cloud point and the unmanned equipment as a first distance corresponding to the point cloud point for each point cloud point, and simultaneously determines the distance between the signal lamp and the unmanned equipment when the unmanned equipment collects the point cloud point as a second distance corresponding to the point cloud point.
Step S310, the unmanned aerial vehicle uses the point cloud points with the first distance smaller than the second distance as candidate point cloud points, and recognizes the cloud data of the point located at the obstacle from the candidate point cloud points by using a constraint condition that the distance between two adjacent point cloud points included in the point cloud data belonging to the same obstacle is smaller than a set distance.
Step S312, the unmanned aerial vehicle determines, for each cloud point in the obstacle point cloud data, coordinates of a pixel point corresponding to the cloud point in the target image, as image coordinates corresponding to the cloud point, and connects the pixel points corresponding to the cloud points in the obstacle point cloud data in the target image based on a relative position between the cloud points of each point in the obstacle point cloud data and the image coordinates corresponding to the cloud points of each point in the obstacle point cloud data, so as to determine an obstacle image area.
Step S314, the unmanned device determines whether the signal lamp is shielded by the obstacle in the target image according to the signal lamp image area and the obstacle image area, if not, step S316 is executed, otherwise, step S318 is executed.
And step S316, the unmanned equipment identifies the signal lamp of the target image and controls the unmanned equipment according to the obtained identification result.
Step S318, the unmanned device discards the target image and does not perform signal light recognition on the target image.
The imaging area of the signal lamp in the target image is predicted in step S302, the imaging area of an obstacle between the signal lamp and the unmanned equipment in the target image is predicted in steps S304-S312, the two processes are not in sequence, and the two processes can be executed simultaneously.
Based on the same idea, the present specification further provides a control device of the unmanned aerial vehicle, as shown in fig. 4.
Fig. 4 is a schematic view of an apparatus for identifying an object provided in the present specification, which specifically includes:
the data acquisition module 400 is used for acquiring a target image acquired by the unmanned equipment for a target object and point cloud data scanned when the target image is acquired;
an image area determination module 401 configured to predict an imaging area of the image of the target object in the target image as a target object image area, identify point cloud data of an obstacle located between the target object and the unmanned aerial vehicle from the point cloud data as obstacle point cloud data, and determine the imaging area of the image of the obstacle in the target image as an obstacle image area based on the obstacle point cloud data;
a determining module 402, configured to determine whether the target object is blocked by the obstacle in the target image according to the target object image area and the obstacle image area;
and an identifying module 403, configured to identify the target object through the target image if the target object is not blocked.
Optionally, the image area determining module 401 is specifically configured to determine, according to a spatial position of the target object recorded in a preset high-precision map and a pose of the unmanned aerial vehicle when the unmanned aerial vehicle acquires the target image, a relative pose between the target object and the unmanned aerial vehicle as a first relative pose; determining a relative pose between a camera and the target object when the target image is acquired as a second relative pose according to the first relative pose and a relative pose between the camera arranged on the unmanned equipment and the unmanned equipment; and predicting an imaging area of the image of the target object in the target image according to the second relative pose.
Optionally, the image area determining module 401 is specifically configured to determine, according to a spatial position of the target object recorded in a preset high-precision map and a pose when the unmanned aerial vehicle scans the point cloud data, a relative pose between the target object and the unmanned aerial vehicle as a third relative pose; determining a relative pose between the laser radar and the target object when the point cloud data is scanned according to the third relative pose and a relative pose between the laser radar arranged on the unmanned equipment and the unmanned equipment, and taking the relative pose as a fourth relative pose; predicting the area occupied by the target object in the point cloud data according to the fourth relative pose to serve as a target object point cloud area; determining an area capable of covering the target object point cloud area from the point cloud data as an area of interest corresponding to the target object point cloud area; identifying point cloud data of an obstacle located between the target object and the unmanned device from the point cloud data contained within the region of interest.
Optionally, the image area determining module 401 is specifically configured to determine a center of the target object point cloud area, and expand the target object point cloud area according to a preset offset value by taking the center as an area center; and taking the expanded target point cloud area as an interesting area corresponding to the target point cloud area.
Optionally, the image area determining module 401 is specifically configured to determine, for each point cloud point, a distance between the point cloud point and the unmanned aerial vehicle when the unmanned aerial vehicle acquires the point cloud point as a first distance corresponding to the point cloud point, and determine a distance between the target object and the unmanned aerial vehicle when the unmanned aerial vehicle acquires the point cloud point as a second distance corresponding to the point cloud point; and determining the cloud points of the obstacle points according to the cloud points of which the first distance is less than the second distance, and determining the cloud data of the obstacle points according to the determined cloud points of the obstacle points.
Optionally, the image area determining module 401 is specifically configured to, for each point cloud point, if it is determined that a first distance corresponding to the point cloud point is smaller than a second distance corresponding to the point cloud point, use the point cloud point as a candidate point cloud point; and recognizing an obstacle point cloud point between the target object and the unmanned equipment from each candidate point cloud point by taking the constraint condition that the distance between two adjacent point cloud points contained in the point cloud data belonging to the same obstacle is smaller than a set distance so as to determine the obstacle point cloud data.
Optionally, the image area determining module 401 is specifically configured to determine, for each point cloud point in the obstacle point cloud data, a coordinate of a pixel point corresponding to the point cloud point in the target image as an image coordinate corresponding to the point cloud point; and connecting pixel points corresponding to cloud points of each point in the obstacle point cloud data in the target image based on the relative positions of the cloud points of each point in the obstacle point cloud data and the image coordinates corresponding to the cloud points of each point in the obstacle point cloud data so as to determine the obstacle image area.
Optionally, the identifying module 403 is further configured to identify the target object if it is determined that the target object is blocked by the obstacle in the target image and does not pass through the target image.
Optionally, the target comprises: a signal lamp.
The present specification also provides a computer readable storage medium having stored thereon a computer program operable to perform the method of object identification provided in figure 1 above.
This description also provides a schematic block diagram of the drone shown in figure 5. As shown in fig. 5, the drone includes, at the hardware level, a processor, an internal bus, a network interface, a memory, and a non-volatile storage, although it may include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs the computer program to implement the method for identifying the target object described in fig. 1. Of course, besides the software implementation, this specification does not exclude other implementations, such as logic devices or combination of software and hardware, and so on, that is, the execution subject of the following processing flow is not limited to each logic unit, and may be hardware or logic devices.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), vhal (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functions of the various elements may be implemented in the same one or more software and/or hardware implementations of the present description.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.
This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only an example of the present disclosure, and is not intended to limit the present disclosure. Various modifications and alterations to this description will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification should be included in the scope of the claims of the present specification.
Claims (12)
1. A method of object identification, comprising:
acquiring a target image acquired by unmanned equipment aiming at a target object and point cloud data scanned when the target image is acquired;
predicting an imaging area of the image of the target object in the target image as a target object image area, and recognizing point cloud data of an obstacle located between the target object and the unmanned aerial vehicle from the point cloud data as obstacle point cloud data, and determining the imaging area of the image of the obstacle in the target image as an obstacle image area based on the obstacle point cloud data;
judging whether the target object is shielded by the obstacle in the target image or not according to the target object image area and the obstacle image area;
and if the target object is not blocked, identifying the target object through the target image.
2. The method of claim 1, wherein predicting an imaged region of the image of the target object in the target image comprises:
determining a relative pose between the target object and the unmanned equipment as a first relative pose according to the spatial position of the target object recorded in a preset high-precision map and the pose when the unmanned equipment acquires the target image;
determining a relative pose between a camera and the target object when the target image is acquired as a second relative pose according to the first relative pose and a relative pose between the camera arranged on the unmanned equipment and the unmanned equipment;
and predicting an imaging area of the image of the target object in the target image according to the second relative pose.
3. The method of claim 1, wherein identifying point cloud data for an obstacle located between the target object and the drone from the point cloud data comprises:
determining a relative pose between the target object and the unmanned equipment as a third relative pose according to the spatial position of the target object recorded in a preset high-precision map and the pose when the unmanned equipment scans the point cloud data;
determining a relative pose between the laser radar and the target object when the point cloud data is scanned according to the third relative pose and a relative pose between the laser radar arranged on the unmanned equipment and the unmanned equipment, and taking the relative pose as a fourth relative pose;
predicting the area occupied by the target object in the point cloud data according to the fourth relative pose to serve as a target object point cloud area;
determining an area capable of covering the target object point cloud area from the point cloud data as an area of interest corresponding to the target object point cloud area;
identifying point cloud data of an obstacle located between the target object and the unmanned device from the point cloud data contained within the region of interest.
4. The method of claim 3, wherein determining an area capable of covering the target point cloud area from the areas scanned by the lidar as the area of interest corresponding to the target point cloud area comprises:
determining the center of the target object point cloud area, and expanding the target object point cloud area by taking the center as an area center according to a preset offset value;
and taking the expanded target point cloud area as an interesting area corresponding to the target point cloud area.
5. The method of claim 1, wherein identifying point cloud data of an obstacle located between the target object and the unmanned aerial device from the point cloud data as obstacle point cloud data comprises:
for each point cloud point, determining the distance between the point cloud point and the unmanned equipment when the unmanned equipment collects the point cloud point as a first distance corresponding to the point cloud point, and determining the distance between the target object and the unmanned equipment when the unmanned equipment collects the point cloud point as a second distance corresponding to the point cloud point;
and determining the cloud points of the obstacle points according to the cloud points of which the first distance is smaller than the second distance, and determining the cloud data of the obstacle points according to the determined cloud points of the obstacle points.
6. The method of claim 5, wherein determining the obstacle point cloud point from point cloud points for which the first distance is less than the second distance comprises:
aiming at each point cloud point, if the first distance corresponding to the point cloud point is smaller than the second distance corresponding to the point cloud point, taking the point cloud point as a candidate point cloud point;
and recognizing an obstacle point cloud point between the target object and the unmanned equipment from each candidate point cloud point by taking the constraint condition that the distance between two adjacent point cloud points contained in the point cloud data belonging to the same obstacle is smaller than a set distance.
7. The method of claim 1, wherein determining an imaging area of the image of the obstacle in the target image as an obstacle image area based on the obstacle point cloud data comprises:
determining the coordinates of corresponding pixel points of the point cloud points in the target image as the image coordinates corresponding to the point cloud points aiming at each point cloud point in the obstacle point cloud data;
and connecting pixel points corresponding to cloud points of each point in the obstacle point cloud data in the target image based on the relative positions of the cloud points of each point in the obstacle point cloud data and the image coordinates corresponding to the cloud points of each point in the obstacle point cloud data so as to determine the obstacle image area.
8. The method of claim 1, wherein the method further comprises:
and if the target object is determined to be shielded by the obstacle in the target image and does not pass through the target image, identifying the target object.
9. The method of any one of claims 1 to 8, wherein the target comprises: a signal lamp.
10. An apparatus for object recognition, comprising:
the data acquisition module is used for acquiring a target image acquired by the unmanned equipment aiming at a target object and point cloud data scanned when the target image is acquired;
an image area determination module for predicting an imaging area of the image of the target object in the target image as a target object image area, identifying point cloud data of an obstacle located between the target object and the unmanned aerial vehicle from the point cloud data as obstacle point cloud data, and determining the imaging area of the image of the obstacle in the target image as an obstacle image area based on the obstacle point cloud data;
the judging module is used for judging whether the target object is shielded by the obstacle in the target image according to the target object image area and the obstacle image area;
and the identification module is used for identifying the target object through the target image if the target object is not blocked.
11. A computer-readable storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method of any of the preceding claims 1 to 9.
12. An unmanned aerial vehicle comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the method of any of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210104778.1A CN114495065A (en) | 2022-01-28 | 2022-01-28 | Target object identification method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210104778.1A CN114495065A (en) | 2022-01-28 | 2022-01-28 | Target object identification method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114495065A true CN114495065A (en) | 2022-05-13 |
Family
ID=81476530
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210104778.1A Pending CN114495065A (en) | 2022-01-28 | 2022-01-28 | Target object identification method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114495065A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116755111A (en) * | 2023-08-24 | 2023-09-15 | 深圳市镭神智能系统有限公司 | Method and device for identifying obstacle of mine car, computer equipment and storage medium |
-
2022
- 2022-01-28 CN CN202210104778.1A patent/CN114495065A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116755111A (en) * | 2023-08-24 | 2023-09-15 | 深圳市镭神智能系统有限公司 | Method and device for identifying obstacle of mine car, computer equipment and storage medium |
CN116755111B (en) * | 2023-08-24 | 2023-11-03 | 深圳市镭神智能系统有限公司 | Method and device for identifying obstacle of mine car, computer equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111619560B (en) | Vehicle control method and device | |
CN111874006B (en) | Route planning processing method and device | |
CN111695546B (en) | Traffic signal lamp identification method and device for unmanned vehicle | |
CN109767637B (en) | Method and device for identifying and processing countdown signal lamp | |
CN112001456B (en) | Vehicle positioning method and device, storage medium and electronic equipment | |
US11371851B2 (en) | Method and system for determining landmarks in an environment of a vehicle | |
US11891091B2 (en) | Semantic occupancy grid management in ADAS/autonomous driving | |
CN111311709A (en) | Method and device for generating high-precision map | |
CN111873989B (en) | Vehicle control method and device | |
CN112712558A (en) | Positioning method and device of unmanned equipment | |
CN112327864A (en) | Control method and control device of unmanned equipment | |
CN112766241B (en) | Target object identification method and device | |
CN114295119A (en) | Map construction method and device | |
CN111797722A (en) | Method and device for drawing lane line | |
CN114495065A (en) | Target object identification method and device | |
CN112990099B (en) | Method and device for detecting lane line | |
CN114332808A (en) | Method and device for predicting steering intention | |
CN111426299B (en) | Method and device for ranging based on depth of field of target object | |
CN112712009A (en) | Method and device for detecting obstacle | |
CN115496930A (en) | Image processing method and device, storage medium and electronic equipment | |
CN114440902A (en) | Method and device for constructing elevation map | |
CN113096427B (en) | Information display method and device | |
CN114590250A (en) | Control method and device for unmanned equipment | |
CN112329547A (en) | Data processing method and device | |
CN112955928A (en) | Information processing method and information processing 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 |