KR101704635B1 - Method and apparatus for detecting target using radar and image raster data - Google Patents

Abstract

The present specification relates to a method and an apparatus for detecting a target using radar and image raster data. An embodiment of the present specification includes: a step in which a location where the target is predicted to be present during the next measurement is estimated by the use of radar information on the target measured by a radar sensor; a step in which a change in raster data is measured by comparison being performed on the raster data on previous and current image information in the image information measured by an image sensor and similarity is checked between the estimated and predicted target location and the measured raster data change; a step in which a raster region of interest with respect to the raster data is selected based on the measured radar information; and a step in which target information is detected in accordance with the selected raster region of interest.

Description

TECHNICAL FIELD [0001] The present invention relates to a method and an apparatus for detecting a target using radar and image raster data,

The present invention relates to a method and apparatus for detecting a target using radar and image raster data, and more particularly, to a method and an apparatus for detecting a target by combining a radar information measured through a radar sensor and image information measured through an image sensor, And more particularly, to a method and apparatus for detecting a target using radar and image raster data.

Intelligent safety automobiles and unmanned autonomous driving technology assist the driver in recognizing the road environment when the driver can not accurately recognize the road environment due to driver's negligence, negligence, and visual limit, So that the vehicle can be moved without any operation of the vehicle.

In order to support safe driving of the vehicle, a technique of recognizing the distance from the preceding vehicle and grasping the road situation outside the driver's sight has been introduced. Especially, distance sensors such as laser radar and ultrasonic sensor have been introduced to detect obstacles, and camera sensors have been introduced to recognize objects.

In other words, a vehicle equipped with various sensors such as a long distance measuring sensor (radar) mounted on the front of the vehicle, a distance measuring sensor for a near distance mounted on the side / rear of the vehicle, and a camera mounted on the side mirror, Technology has been introduced.

On the other hand, the method of extracting outline points from the distance measurement data generated by the radar sensor has been difficult in actual implementation. As a result, the size and cost of the radar sensor are increased in order to accurately extract the contour points. Due to the problem of computation speed, real - time performance can not be guaranteed.

Also, in a real environment, such a radar data has a 1: 1 correspondence with the image raster data.

Korean Patent Laid-Open Publication No. 10-2012-0072131 (published on July 3, 2012)

Embodiments of the present invention are directed to a method for detecting a target using radar and image raster data, which can accurately detect a target by selecting a raster candidate group by fusing radar information measured through a radar sensor and image information measured through an image sensor And apparatus.

Embodiments of the present invention are directed to radar and image rasters that can correct distance information of radar information based on velocity information and estimate the expected position of the target using the corrected radar information, And a method and an apparatus for detecting a target using data.

Embodiments of the present invention are directed to a method and apparatus for detecting targets using radar and image raster data that can improve the speed of operation of image information by selecting a raster candidate region for raster data based on the measured radar information Method, and apparatus.

According to a first aspect of the present invention, there is provided a method for estimating an expected position of a target in a next measurement using radar information on a target measured through a radar sensor; A change in raster data is measured by comparing raster data of previous and current image information in the image information measured through the image sensor, and the similarity between the estimated position of the estimated target and the change in the measured raster data Checking; Selecting a raster candidate region for raster data based on the measured radar information; And detecting the target information in accordance with the selected raster candidate group.

In estimating the expected position of the target, the distance information may be corrected using the velocity information from the radar information about the target to estimate the predicted position of the target in the next measurement.

Wherein the step of verifying the similarity of the changes in the measured raster data comprises comparing the raster data of the previous and current image information to measure the change of the position and size of the raster, Can be confirmed.

The method may further include correcting the raster data based on the measured radar information after confirming the similarity of the change of the measured raster data.

Wherein the step of selecting the raster candidate group includes adjusting the priority of the selected raster candidate group using the radar information and the image recognition result, and the step of detecting the target information may include the step of, based on the priority of the adjusted raster candidate group, Can be detected.

According to a second aspect of the present invention, there is provided an information processing apparatus comprising: a vehicle sensor unit having a radar sensor and an image sensor to measure radar information and image information; A position estimating unit estimating a predicted position of the target in the next measurement using the measured radar information about the target; A raster data processing unit for measuring a change in raster data by comparing raster data of previous and current image information in the measured image information and confirming a similarity between the estimated position of the estimated target and the measured raster data; A raster candidate group selection unit for selecting a raster candidate region for the raster data based on the measured radar information; And a target detection unit for detecting target information according to the selected raster candidate group.

The position estimating unit may estimate the predicted position of the target in the next measurement by correcting the distance information using the velocity information from the radar information about the target.

The raster data processing unit compares the raster data of the previous and current image information to measure the change of the position and the size of the raster and confirm the similarity between the estimated position of the target and the change of the position and size of the measured raster have.

The raster data processing unit may correct the raster data based on the measured radar information after confirming the similarity of the change of the measured raster data.

The raster candidate group selection unit may adjust the priority of the selected raster candidate group using the radar information and the image recognition result, and the target detection unit may detect the target information according to the priority of the adjusted raster candidate group.

Embodiments of the present invention can accurately detect a target by selecting a raster candidate group by fusing radar information measured through a radar sensor and image information measured through an image sensor.

Embodiments of the present invention can improve the accuracy of measurement data by correcting the distance information of the radar information based on the speed information and estimating the predicted position of the target using the corrected radar information.

Embodiments of the present invention can improve the operation speed of image information by selecting a raster candidate region for raster data based on the measured radar information.

Embodiments of the present disclosure can improve the matching accuracy between radar information and raster data of image information.

Embodiments of the present disclosure can improve the detection performance of the target by adjusting the priority of the selected raster candidate group.

1 is a configuration diagram of a target detection apparatus using radar and image raster data according to an embodiment of the present invention.
2 is an explanatory view of a target detection process using radar information according to an embodiment of the present invention.
3 is a flowchart of a method of detecting a target using radar and image raster data according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Will be described in detail with reference to the portions necessary for understanding the operation and operation according to the present specification. In describing the embodiments of the present invention, description of technical contents which are well known in the technical field to which the present invention belongs and which are not directly related to the present specification will be omitted. This is for the sake of clarity without omitting the unnecessary explanation and without giving the gist of the present invention.

In describing the components of the present specification, the same reference numerals may be given to components having the same name, and the same reference numerals may be given to different drawings. However, even in such a case, it does not mean that the corresponding component has different functions according to the embodiment, or does not mean that it has the same function in different embodiments, and the function of each component is different from that of the corresponding embodiment Based on the description of each component in FIG.

1 is a configuration diagram of a target detection apparatus using radar and image raster data according to an embodiment of the present invention.

1, a target detecting apparatus 100 using radar and image raster data according to an embodiment of the present invention includes a vehicle sensor unit 110, a position estimating unit 120, a raster data processing unit 130, A raster candidate group selection unit 140 and a target detection unit 150. [

The specific configuration and operation of each component of the target detection apparatus 100 using the radar and the image raster data of FIG. 1 will be described below.

First, the vehicle sensor unit 110 includes a radar sensor 111 and an image sensor 112 installed in the vehicle. The vehicle sensor unit 110 includes a radar sensor 111 and an image sensor 112 to measure radar information and image information. Here, the radar sensor 111 is a device for detecting the position of an object by using reflected waves generated by the radiation of radio waves when transmitting and receiving are performed in the same place. The reflected wave is captured by using the reflection of the radio wave by colliding with the target to find out the existence of the target. A Doppler effect may be used or the frequency of the transmission radio wave may be changed with time or a pulse wave may be used as a transmission wave to prevent the transmitted radio wave from overlapping with the received radio wave to be difficult to distinguish. Also, by using the rotating device, the antenna can be moved to the left and right to find not only the distance and the direction but also the altitude of the target. By arranging the antenna up and down, it is possible to search and track in the horizontal and vertical directions.

The position estimating unit 120 estimates the expected position of the target in the next measurement using the radar information about the target measured through the radar sensor 111 of the vehicle sensor unit 110. [ Here, the target may be a moving obstacle such as another vehicle or a pedestrian in front of the vehicle, a road sign existing in a fixed form, or the like.

Here, the position estimating unit 120 corrects the distance information using the velocity information among the radar information about the target. Subsequently, the position estimating unit 120 estimates the predicted position of the target at the next measurement using the corrected distance information. That is, the position estimating unit 120 corrects the distance information using the speed information to improve the accuracy of the radar, and estimates the predicted position of the target at the next measurement.

The raster data processing unit 130 compares the raster data of the previous and current image information in the image information measured by the image sensor 112 of the vehicle sensor unit 110 to measure the change of the raster data.

Then, the raster data processing unit 130 confirms the estimated position of the target estimated by the position estimating unit 120 and the similarity of the change of the measured raster data. Specifically, the raster data processing unit 130 calculates the position change and size change of the raster in the image information, compares the calculated position and size change of the raster with the radar information, finds the similarity between the raster data of the image information and the radar information can confirm.

Here, the raster data processing unit 130 compares the raster data of the previous and current image information to measure the position and size change of the raster. Then, the raster data processing unit 130 can confirm the similarity between the change in the position and size of the measured raster and the predicted position of the target estimated in the position estimating unit 120. [

Then, the raster data processing unit 130 can check the similarity of the change of the measured raster data, and then correct the raster data based on the measured radar information.

The raster candidate group selection unit 140 selects a raster candidate region for the raster data based on the radar information measured through the radar sensor 111 of the vehicle sensor unit 110. The radar information and the image information are mutually complemented and fused, and the raster candidate group selection unit 140 selects a raster data candidate group in order to improve the computation speed.

Then, the raster candidate group selection unit 140 adjusts the priority of the selected raster candidate group using the radar information and the image recognition result. That is, the raster candidate group selection unit 140 adjusts the priority of the raster candidate group by fusing the radar information and the image recognition result. Here, the target detecting unit 150 can detect the target quickly according to the priority of the adjusted raster candidate group, thereby detecting the target quickly.

The target detection unit 150 detects the target information according to the raster candidate group selected by the raster candidate group selection unit 140. [ At this time, the target detecting unit 150 can detect the target information according to the priority order of the raster candidate group adjusted by the raster candidate group selecting unit 140. [

2 is an explanatory view of a target detection process using radar information according to an embodiment of the present invention.

As shown in FIG. 2, the vehicle 210 is moving at an actual speed.

The first target 220 and the second target 230 located around the vehicle 210 each have an actual speed. The first target 220 is measured in the same direction as the speed of the vehicle 210 and different speed data. The second target 230 measures velocity data in which the velocity and direction of the vehicle 210 are different from each other.

Here, the vehicle 210 measures the angle and velocity data together with the relative distance data between the first target 220 and the second target 230 using the radar sensor 111 provided.

Since the velocity of each of the vehicle 210, the first target 220 and the second target 230 is a velocity vector, an estimated velocity vector is calculated using the current position and angle information of the vehicle 210 as the first target 220, And the second target 230, respectively.

The vehicle 210 then detects the position of the first target 220 and the second target 230 and then uses the velocity data to estimate the first target 220 and the second target 230 at the next measurement The position can be estimated.

3 is a flowchart of a method of detecting a target using radar and image raster data according to an embodiment of the present invention.

The vehicle sensor unit 110 measures radar information including distance, angle, and speed information using the radar sensor 111 (S302). Here, the vehicle sensor unit 110 can simultaneously measure the angle data and the velocity data together with the distance data through the radar sensor 111. [

The position estimating unit 120 estimates a velocity vector using velocity information from the radar information measured by the vehicle sensor unit 110 (S304). Since the velocity information is a vector between the radar and the target, the position estimator 120 estimates the estimated velocity vector using the current position and the angle information.

The position estimating unit 120 corrects the distance information using the estimated velocity vector, and estimates the expected position of the target at the next measurement using the corrected distance information (S306). Here, the position estimating unit 120 may first estimate the position of the target at the next measurement using the velocity information after detecting the position of the target. Subsequently, the position estimating unit 120 corrects the predicted position at the next measurement using the estimated velocity vector. Here, the velocity vector is estimated as the reference velocity of the actual road frame (World Frame), not the reference velocity of the radar frame.

When the vehicle sensor unit 110 measures an image of the target using the image sensor 112, the raster data processing unit 130 receives the image information measured by the vehicle sensor unit 110 (S308).

Then, the raster data processing unit 130 extracts raster data of the previous and current image information from the image information measured by the image sensor 112 of the vehicle sensor unit 110 (S310).

Next, the raster data processing unit 130 compares raster data of the extracted previous and current image information to measure changes in raster data (S312). The raster data processing unit 130 may measure the change in the position of the center of the raster and the change in the size of the raster by comparing the raster data of the previous image information with the raster data of the current image in the image information.

Then, the raster data processing unit 130 confirms the similarity between the predicted position of the target estimated by the position estimating unit 120 and the measured raster data, and corrects the raster data based on the measured radar information ( S314). That is, the raster data processing unit 130 can confirm the similarity of the raster change with respect to the estimated future position information in the radar, that is, the expected position of the target, and correct the raster data based on the measured radar information.

Thereafter, the raster candidate group selection unit 140 selects a raster candidate region (Region of Interest) based on the radar information measured through the radar sensor 111 of the vehicle sensor unit 110 (S316). The raster candidate group selection unit 140 selects raster data candidates based on the radar information for the purpose of improving the operation speed of the video signal processing.

In addition to selecting the raster candidate group, the raster candidate group selection unit 140 performs image recognition on the target based on the image information input in step S308 (S318).

The raster candidate group selection unit 140 sets the priority of the raster candidate group selected in step S316 using the radar information and the image recognition result (S320). The raster candidate group selection unit 140 can adjust the priority of the raster data candidate group using the intensity of the radar signal and the result of image recognition (for example, nearest vehicles, pedestrians, etc.).

Then, the target detection unit 150 detects the target information according to the raster candidate group selected by the raster candidate group selection unit 140 (S322). At this time, the target detecting unit 150 can detect the target information according to the priority order of the raster candidate group adjusted by the raster candidate group selecting unit 140. [

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or essential characteristics thereof. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

100: Target detection device
110: vehicle sensor unit
111: Radar sensor
112: Image sensor
120:
130: Raster data processor
140: Raster candidate group selection
150: Target detection unit

Claims (10)

Estimating a predicted position of the target in the next measurement using the radar information on the target measured through the radar sensor;
A change in raster data is measured by comparing raster data of previous and current image information in the image information measured through the image sensor, and the similarity between the estimated position of the estimated target and the change in the measured raster data Checking;
Selecting a raster candidate region for raster data based on the measured radar information; And
Detecting target information according to the selected raster candidate group
A method of detecting a target using radar and image raster data. The method according to claim 1,
The step of estimating the expected position of the target
A method of detecting a target using radar and image raster data for correcting distance information using velocity information among radar information on the target and estimating a predicted position of the target in a next measurement. The method according to claim 1,
The step of ascertaining the similarity of the changes of the measured raster data
Comparing the raster data of the previous and current image information to measure raster position and magnitude variation and determining a similarity between the estimated position of the estimated raster and the estimated position of the raster, Target Detection Method Using. The method according to claim 1,
After confirming the similarity of the change of the measured raster data, correcting the raster data based on the measured radar information
And a target detection method using the radar and image raster data. The method according to claim 1,
The step of selecting the raster candidate group
Adjusting a priority of the selected raster candidate group using the radar information and the image recognition result,
Wherein the step of detecting the target information comprises detecting radar and image raster data for detecting target information according to the priority of the adjusted raster candidate group. A vehicle sensor unit having a radar sensor and an image sensor to measure radar information and image information;
A position estimating unit estimating a predicted position of the target in the next measurement using the measured radar information about the target;
A raster data processing unit for measuring a change in raster data by comparing raster data of previous and current image information in the measured image information and confirming a similarity between the estimated position of the estimated target and the measured raster data;
A raster candidate group selection unit for selecting a raster candidate region for the raster data based on the measured radar information; And
A target detection unit for detecting target information according to the selected raster candidate group,
And a target detection device using image raster data. The method according to claim 6,
The position estimating unit
And estimating a predicted position of the target in the next measurement by correcting the distance information using the velocity information from the radar information on the target. The method according to claim 6,
The raster data processing unit
Comparing the raster data of the previous and current image information to measure raster position and magnitude variation and determining a similarity between the estimated position of the estimated raster and the estimated position of the raster, . The method according to claim 6,
The raster data processing unit
And a radar and image raster data for correcting the raster data based on the measured radar information after confirming the similarity of the change of the measured raster data. The method according to claim 6,
The raster candidate group selection unit
Adjusting a priority of the selected raster candidate group using the radar information and the image recognition result,
Wherein the target detection unit uses the radar and the image raster data to detect the target information according to the priority of the adjusted raster candidate group.

Images