KR102535590B1 - Object detection apparatus and method based on sensor fusion - Google Patents

Abstract

The present invention relates to an object detection technology, and an object detection device based on sensor fusion according to an embodiment of the present invention includes a first detection device generating first detection information by detecting an object within a first monitoring area and a second monitoring area. and a second detection device that detects an object and generates second detection information, and a fusion detection device that generates fusion detection information based on the first detection information and the second detection information.

Description

Object detection apparatus and method based on sensor fusion

The present invention relates to object detection technology, and more particularly, to a sensor fusion-based object detection device and method for detecting an object around a vehicle.

Recently, many studies have been conducted for the development and commercialization of self-driving cars as well as advanced driver assisted systems (ADAS), which are driver assistance systems.

Driving assistance devices such as ADAS detect objects around the vehicle even when the driver is not aware of it, recognize the situation, determine the risk of an accident, and provide a warning to the driver when the risk exists or automatically brakes when the risk increases. Alternatively, it performs a function of preventing a collision accident by controlling the steering system.

In order to provide these functions, the driving assistance device recognizes the surrounding situation using various sensors such as a camera, radar, and lidar, and performs judgment and control using a high-performance processor.

The radar device has excellent performance for detecting objects regardless of the external environment such as weather, but it is impossible to identify objects, and the camera can identify objects, but its performance is poor in poor environments (eg, raining conditions). In addition, lidar has excellent performance, but there is a problem that it is not suitable for commercialization because it is expensive.

Accordingly, a convergence type detection device in which a camera, radar, and lidar are converged is widely used in autonomous vehicles.

However, since most of the current converged detection devices give priority to the camera device, there is a problem in that a serious error may occur in a situation where the camera cannot detect the device.

The present invention has been made to solve the problems of the prior art as described above, and an object of the present invention is to generate a reliability index of detection information detected by different types of sensors, and based on the reliability index, to an object It is an object of the present invention to provide a sensor fusion-based object detection device and detection method capable of improving reliability by performing detection on the object.

The object of the present invention is not limited to the object mentioned above, and other objects not mentioned will be clearly understood by those skilled in the art from the description below.

An object detection device based on sensor fusion according to an embodiment of the present invention for achieving the object to be solved as described above includes: a first detection device generating first detection information by detecting an object within a first monitoring area; a second detection device generating second detection information by detecting an object within the second monitoring area; and fusion generating fusion detection information based on the first detection information, a first reliability index for an object in the first detection information, the second detection information, and a second reliability index for an object in the second detection information. a detecting device, wherein the first detecting device tracks the location of an object included in the first detection information, derives the number of traces of the object included in the first detection information, and determines the first reliability index. , and the second detection device performs position tracking on the object included in the second detection information, derives the number of tracking objects included in the second detection information, and generates the second reliability index. and if the first detection device succeeds in detecting the object within the first surveillance region, increases the tracking number associated with the first reliability index, and if it fails to detect the object within the first surveillance region, the first reliability index and The number of traces is set while decreasing the number of associated traces, and when the second detection device succeeds in detecting an object within the second monitoring area, the number of traces associated with the second reliability index is increased, and within the second surveillance area If object detection fails, the number of traces associated with the second reliability index is decreased and the number of traces is set.

The first detection device generates the first reliability index for the object in the first detection information and provides it to the convergence detection device, and the second detection device generates the second reliability index for the object in the second detection information. An index is generated and provided to the fusion detection device.

The first detection device is, for example, an image-based detection device, and the second detection device is, for example, a radar-based detection device or a laser radar-based detection device.

delete

The convergence detection device generates an integrated reliability index based on the first and second reliability indexes, and generates the convergence detection information for the detected object when the integrated reliability index is greater than or equal to a threshold value.
A sensor fusion-based object detection device according to another embodiment of the present invention for achieving the object to be solved as described above includes a first detection device generating first detection information by detecting an object within a first monitoring area; and a second detection device generating second detection information by detecting an object within a second monitoring area and a fusion detection device generating fusion detection information based on the first detection information and the second detection information.
A sensor fusion-based object detection device according to another embodiment of the present invention for achieving the object to be solved as described above is a plurality of detection devices for generating detection information by detecting an object within a preset monitoring area, and and a fusion detection device generating fusion detection information based on the detection information generated by each of the plurality of detection devices and a reliability index for an object in the detection information.
A sensor fusion-based object detection method according to an embodiment of the present invention for achieving the object to be solved as described above is to detect an object within a first monitoring area and to obtain first detection information and information about the first detection information. generating a first reliability index, generating second detection information and a second reliability index for the second detection information by detecting an object within a second surveillance area, and generating the first detection information and the second detection information; generating fusion detection information based on the detection information, a first reliability index, and a second reliability index, wherein generating a first reliability index for an object in the first detection information performing location tracking on an object included in the first detection information and generating the first reliability index by deriving the tracking number of the object included in the first detection information; Deriving the number of traces for the step of generating the first reliability index increases the number of traces associated with the first reliability index if the object in the first surveillance area is successfully detected, and detects the object in the first surveillance area. fails, deriving the number of traces while decreasing the number of traces associated with the first reliability index, and generating a second confidence index for an object in the second detection information. performing location tracking on an object included in the second detection information and generating the second reliability index by deriving a tracking number for the object included in the second detection information; In the step of deriving the tracking number for and generating the second reliability index, if the object within the second surveillance area is successfully detected, the tracking number associated with the second reliability index is increased, and the detection of the object within the second surveillance area is successful. If it fails, deriving the number of traces while decreasing the number of traces associated with the second reliability index.

According to an embodiment of the present invention as described above, a sensor fusion-based object detection technology for detecting an object around a vehicle is proposed.

Since the sensor fusion-based object detection technology proposed in the present invention generates a reliability index of detection information detected by different types of sensors and performs object detection based on the reliability index, the reliability of object detection is improved. can improve

1 is a diagram showing an example of an object detection device based on sensor fusion according to a preferred embodiment of the present invention.
2 is a diagram showing an example of a first detection device of a sensor fusion-based object detection device according to a preferred embodiment of the present invention.
3 is a diagram showing an example of a second detection device of a sensor fusion-based object detection device according to a preferred embodiment of the present invention.
4 is a configuration diagram illustrating an example of a fusion detection device of a sensor fusion-based object detection device according to a preferred embodiment of the present invention.
5 is a diagram showing an example of detection information generated by a sensor fusion-based object detection apparatus according to a preferred embodiment of the present invention.
6 is a diagram showing an example of integrated information generated by the sensor fusion-based object detection apparatus according to a preferred embodiment of the present invention.
7 is a diagram for explaining a fusion-based object detection method using a sensor fusion-based object detection apparatus according to a preferred embodiment of the present invention.

For the embodiments of the present invention disclosed in the text, specific structural or functional descriptions are only exemplified for the purpose of describing the embodiments of the present invention, and the embodiments of the present invention can be implemented in various forms, and It should not be construed as limited to the described embodiments.

Since the present invention can have various changes and various forms, specific embodiments will be illustrated in the drawings and described in detail in the text. However, it should be understood that this is not intended to limit the present invention to the specific disclosed form, and includes all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention.

It is understood that when an element is referred to as being “connected” or “connected” to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when a component is referred to as “directly connected” or “directly connected” to another component, it should be understood that no other component exists in the middle. Other expressions describing the relationship between components, such as "between" and "directly between" or "adjacent to" and "directly adjacent to", etc., should be interpreted similarly.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, terms such as “comprise” or “have” are intended to indicate that the disclosed feature, number, step, operation, component, part, or combination thereof exists, but that one or more other features or numbers, It should be understood that the presence or addition of steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

Meanwhile, when a certain embodiment can be implemented differently, functions or operations specified in a specific block may occur in a different order from the order specified in the flowchart. For example, two successive blocks may actually be performed substantially concurrently, or the blocks may be performed backwards depending on the function or operation involved.

Hereinafter, a sensor fusion-based object detection device and detection method to be proposed in the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram showing an example of an object detection device based on sensor fusion according to a preferred embodiment of the present invention, and FIG. 2 shows an example of a first detection device of the object detection device based on sensor fusion according to a preferred embodiment of the present invention. FIG. 3 is a diagram showing an example of a second detection device of a sensor fusion-based object detection device according to a preferred embodiment of the present invention.

Referring to FIG. 1 , a sensor fusion-based object detection device 1 according to a preferred embodiment of the present invention generates a reliability index for detection information obtained from at least two or more detection devices, and converts the detection information and the reliability index to each other. Based on this, it is implemented to generate and provide fusion detection information.

To this end, the sensor fusion-based object detection device 1 includes at least two or more detection devices 10 and 20 and a fusion detection device 30, and may further include a control device 40.

For convenience of description, hereinafter, an example in which the sensor fusion-based object detection device 1 includes two detection devices, that is, the first detection device 10 and the second detection device 20 will be described.

The first detection device 10 is a component for detecting an object within a region (first surveillance region, S1) preset as a surveillance region for object detection, and may be, for example, an image-based detection device.

The second detection device 20 is a component for detecting an object in an area preset as a monitoring area (second monitoring area, S2) for object detection, for example, a radar-based detection device or lidar. (lidar) based detection device.

The first surveillance region S1 and the second surveillance region S2 overlap each other to form a common surveillance region C.

And, it is assumed that the first and second detection devices 10 and 20 are temporally and spatially synchronized.

The first detection device 10 detects an object in a captured image of the first monitoring area S1 to generate and provide first detection information.

For example, the first detection device 10 may generate first detection information including one or more pieces of information such as location information, type information, and location tracking information related to the detected object.

In addition to this, the first detection device 10 may additionally provide a first reliability index indicating how reliable the first detection information is.

That is, the first detection device 10 may provide first detection information or generate and provide first detection information and a first reliability index.

To this end, as shown in FIG. 2, the first detection device 10 includes an image sensor 11, a first detection module 12, a first location tracking module 13, and a first processing unit 14. can include

The image sensor 11 obtains an image of the first monitoring area S1 and may be selected from various types of image sensors known in the art.

The first detection module 12 detects an object by performing image processing on an image acquired by the image sensor 11 .

The first detection module 12 may detect an object by performing image processing on an image, and may generate location information and type information related to the detected object.

The first location tracking module 13 may generate location tracking information by performing location tracking on an object by applying a location tracking algorithm to the object detected by the first detection module 12 .

The first processing unit 14 generates and provides first detection information by synthesizing information from the first detection module 12 and information from the first location tracking module 13 .

In addition, the first processing unit 14 may generate and provide a first reliability index indicating the level of reliability of the first detection information.

For example, the first processing unit 14 performs location tracking on the object included in the first detection information, derives the tracking number of the object included in the first detection information, and derives the derived tracking number (tracking_cnt1). ) may be generated as the first reliability index.

At this time, the first processing unit 14 may increase the tracking number (tracking_cnt1) by 1 if object detection is successful, and may set the tracking number (tracking_cnt1) by 1 if object detection fails.

For example, the first processing unit 14 may generate the first detection information in the format shown in FIG. 5, but the format for the first detection information shown in FIG. 5 is for the understanding of the present invention. As an example, it is natural that various changes can be made.

Referring to FIG. 5 , the first processing unit 14 may generate first detection information based on location information for each detected object.

In addition, the first processing unit 14 may record the reliability index of the detected object in the first detection information.

The second detection device 20 generates and provides second detection information by detecting an object based on a signal received by reflecting a laser signal or a radio wave signal output from the second monitoring area 21 by an object. .

For example, the second detection device 20 may generate second detection information including one or more pieces of information such as location information, reflection intensity information, type information, location tracking information, etc. of the detected object.

In addition to this, the second detection device 20 may additionally provide a second reliability index indicating how reliable the second detection information is.

That is, the second detection device 20 may provide second detection information or generate and provide second detection information and a second reliability index.

To this end, as shown in FIG. 3, the second detection device 20 includes a signal transmission/reception unit 21, a second detection module 22, a second location tracking module 23, and a second processing unit 24. can include

The signal transmission/reception unit 21 transmits a signal (laser signal or radio wave signal) to the first surveillance area S2 and receives a signal reflected back by an object ('reflected signal').

The second detection module 22 detects an object by performing signal processing on the reflected signal received by the signal transceiver 21 .

The second detection module 22 may detect an object by performing signal processing on the reflected signal and generate location information, type information, reflection intensity information, etc. related to the detected object.

The second location tracking module 23 may generate location tracking information by performing location tracking on an object by applying a location tracking algorithm to the object detected by the second detection module 22 .

The second processing unit 24 generates and provides second detection information by synthesizing information from the second detection module 22 and information from the second location tracking module 23 .

In addition, the second processing unit 24 may generate and provide a second reliability index indicating how reliable the second detection information is.

For example, the second processing unit 24 performs location tracking on the object included in the second detection information, derives the tracking number of the object included in the second detection information, and derives the derived tracking number (tracking_cnt2 ) may be generated as the second reliability index.

At this time, the second processing unit 24 may increase the tracking number (tracking_cnt2) by 1 if object detection is successful, and may set the tracking number (tracking_cnt2) by 1 if object detection fails.

For example, the second processing unit 24 may generate the second detection information in the format shown in FIG. 5, but the format for the second detection information shown in FIG. 5 is for the understanding of the present invention. As an example, it is natural that various changes can be made.

Referring to FIG. 5 , the second processing unit 24 may generate second detection information based on location information for each detected object.

In addition, the second processing unit 24 may record the reliability index of the detected object in the second detection information.
The convergence detection device 30 is based on the first detection information and the first reliability index from the first detection device 10 and the second detection information and the second reliability index from the second detection device 20, convergence Detection information can be generated and provided.

Alternatively, the fusion detection device 30 generates and provides fusion detection information based on the first detection information from the first detection device 10 and the second detection information from the second detection device 20. .

The convergence detection device 30 generates a first reliability index using the first detection information, generates a second reliability index using the second detection information, and generates the first detection information, the first reliability index, and the second reliability index. Convergence detection information may be generated and provided based on the detection information and the second reliability index.

delete

A detailed configuration of the fusion detection device 30 will be described later with reference to FIG. 4 .

The control device 40 determines the risk based on the fusion detection information from the fusion detection device 30, and performs a control operation (ex, alarm, brake control, steering control, etc.) according to the determination result.

4 is a configuration diagram illustrating an example of a fusion detection device of a sensor fusion-based object detection device according to a preferred embodiment of the present invention.

The fusion detection device 400 shown in FIG. 4 may be applied to the fusion detection device 30 of FIG. 1 , wherein the fusion detection device 400 includes the first detection information from the first detection device 10 and the first detection information. Based on the first reliability index and the second detection information from the second detection device 20 and the second reliability index, a final decision on object detection may be made, and fusion detection information may be generated.
Alternatively, the fusion detection device 400 generates a first reliability index and a second reliability index based on the first detection information from the first detection device 10 and the second detection information from the second detection device 20. And, based on the first and second detection information and the first and second reliability indices, a final decision on object detection may be made, and convergence detection information may be generated.

delete

In this case, the convergence detection device 400 may generate an integrated reliability index based on the first reliability index and the second reliability index, and may use the integrated reliability index for final determination of object detection.

Specifically, the fusion detection device 400 may include a receiving unit 410, a reliability index generating unit 420, an integrating unit 430, a detecting unit 440, a tracking unit 450, and a control unit 460. .

The receiver 410 receives first detection information from the first detection device 10 and second detection information from the second detection device 20 .

In addition, the receiving unit 410 may additionally receive a first reliability index from the first detection device 10 and a second reliability index from the second detection device 20 .

The reliability index generator 420 generates a first reliability index for the first detection information based on the first detection information from the receiver 410, and generates a first reliability index based on the second detection information from the receiver 410. 2 A second reliability index for detection information may be generated.

For example, the reliability index generation unit 420 performs location tracking of objects included in the first and second detection information, and determines the number of tracking objects included in the first detection information (tracking_cnt1) and the second detection information. The number of tracking objects included in the detection information (tracking_cnt2) may be derived, and the derived information (tracking_cnt2, tracking_cnt2) may be used as first and second reliability indices.

In this case, the reliability index generation unit 420 may increase the tracking number (tracking_cnt) by 1 if object detection is successful, and may set the tracking number (tracking_cnt) by 1 if object detection fails.

As another example, the reliability index generation unit 420 may identify a detection area using the first and second detection information, set an index area, and calculate a reliability index for the index area.

At this time, the index area may be set based on the object's mobility information (eg, speed, direction, etc.).

In addition, in the case of detection information generated by an image-based detection device, the reliability index can be calculated based on the histogram distribution of the exponential area, and in the case of deep learning-based detection information, the detection confirmation information provided by the deep learning part can be calculated based on

And, in the case of detection information generated by a lidar-based detection device, the reliability index may be set to a variance value of a distance in an index region or a size of a reflection coefficient in an index region.

The integrator 430 generates integrated information based on the first detection information, the first reliability index, the second detection information, and the second reliability index, which are input information.

Depending on the embodiment, the integration unit 430 may generate integrated information based on the first detection information, the first reliability index, the second detection information, and the second reliability index from the reception unit 410 .

Depending on the embodiment, the integration unit 430 may generate integrated information based on the first and second detection information from the receiver 410 and the first and second reliability indexes from the reliability index setting unit 420. can

The integration unit 430 may generate integrated information in a format as shown in FIG. 6 , for example.

However, it is natural that the format of integrated information shown in FIG. 6 can be variously changed as an example for understanding the present invention.

Referring to FIG. 6 , the integrator 430 may generate integrated information on the detected object based on location information for each object included in the first and second detection information.

Also, the integrator 430 may define and record a type of an object included in integrated information.

For example, the integration unit 430 defines 'Type A' as an object included in both the first and second detection information, and the object included in the first detection information but not included in the second detection information. is defined as 'Type B' for , and 'Type C' is defined for objects not included in the first detection information but included in the second detection information, but not included in the first and second detection information, but previously detected. An object included in information and managed in integrated information may be implemented to be defined as 'Type D'.

In addition, the integrator 430 may record reliability indices for each detected object in integrated information.

That is, the integrator 430 may generate integrated information including location information, type information, and confidence indices of currently detected and previously detected objects.

The detection unit 440 makes a final decision on object detection based on the integrated information from the integrating unit 430 and generates fusion detection information for the detected object.

Accordingly, the fusion detection information includes information about the object detected by the first detection device 10 and information about the object detected by the second detection device 20 .

In addition, the detection unit 440 may generate an integrated reliability index based on the first reliability index and the second reliability index, and make a final determination on object detection based on the integrated reliability index to generate convergence detection information.

For example, the detector 440 may calculate an integrated reliability index according to Equation 1 or Equation 2 below.

[Equation 1]

Integrated Reliability Index = 1st Reliability Index + 2nd Reliability Index

[Equation 2]

Integrated Reliability Index = MAX{First Reliability Index, Second Reliability Index}

That is, the detection unit 440 may calculate the sum of the first reliability index and the second reliability index as the integrated reliability index, or may calculate the higher reliability index of the first reliability index and the second reliability index as the integrated reliability index.

As another example, the detection unit 440 may calculate an integrated reliability index according to Equation 3 below.

[Equation 3]

Integrated Confidence Index =

Here, tracking_cnt1 is the number of tracking objects in the first detection information, which is the difference between the number of successful detection and the number of detection failures, and tracking_cnt2 is the number of tracking objects in the second detection information, which is the difference between the number of successful detection and the number of detection failures, MAX_CNT is the maximum threshold of tracking_cnt1 and tracking_cnt2.

When the detection unit 440 makes a final decision on object detection based on the integrated reliability index calculated according to Equation 1 or Equation 2, if the integrated reliability index is greater than or equal to a predetermined threshold value (ex, 3), the object is detected. can be judged to have been

And, when the detection unit 440 makes a final decision on object detection based on the integrated reliability index calculated according to Equation 3, if the integrated reliability index is greater than or equal to a predetermined threshold value (ex, 0.5), it means that the object has been detected. can judge

The tracking unit 450 may generate tracking information for each object included in the fusion detection information based on the fusion detection information from the detection unit 440 .

The control unit 460 performs overall control of the integration detection unit 400, and receives, integrates, and outputs fusion detection information from the detection unit 440 and estimation information from the tracking unit 450.

In the above, the configuration and functions of each component of the sensor fusion-based object detection device according to a preferred embodiment of the present invention have been reviewed. Hereinafter, a sensor fusion-based object detection method according to a preferred embodiment of the present invention will be described.

7 is a diagram for explaining a fusion-based object detection method using a sensor fusion-based object detection apparatus according to a preferred embodiment of the present invention.

The step-by-step operation shown in FIG. 7 can be performed by the sensor fusion-based object detection device 1 described with reference to FIGS. 1 to 6, and the first detection device 10 detects an object within the first monitoring area S1 is detected to generate first detection information (S700), and the second detection device 20 detects an object within the second monitoring area S2 to generate second detection information (S710).

Of course, if the sensor fusion-based object detection device 1 of FIG. 1 includes a larger number of detection devices, the detection information generated in step S710 may increase.

The first and second detection information may include location information, type information, and location tracking information related to the detected object.

In addition, the first detection device 10 generates a first reliability index for the object in the first detection information, and the second detection device 20 generates a second reliability index for the object in the second detection information. can do.

For example, the first detection device 10 may be an image-based detection device, and the second detection device 20 may be a radar-based detection device or a laser radar-based detection device.

After the step S710, the fusion detection device 30 generates fusion detection information based on the first detection information and the second detection information (S720).

After the step S720, the control device 40 determines the risk based on the convergence detection information, and performs a control operation according to the determination result (S730).

Even though all components constituting the embodiments of the present invention described above are described as being combined or operated as one, the present invention is not necessarily limited to these embodiments. That is, within the scope of the object of the present invention, all of the components may be selectively combined with one or more to operate. In addition, although all of the components may be implemented as a single independent piece of hardware, some or all of the components are selectively combined to perform some or all of the functions combined in one or a plurality of pieces of hardware. It may be implemented as a computer program having. In addition, such a computer program can implement an embodiment of the present invention by being stored in a computer readable medium such as a USB memory, a CD disk, a flash memory, etc. and read and executed by a computer. A recording medium of a computer program may include a magnetic recording medium, an optical recording medium, a carrier wave medium, and the like.

On the other hand, although the sensor fusion-based object detection device and detection method of the present invention have been described according to embodiments, the scope of the present invention is not limited to specific embodiments, and will be apparent to those skilled in the art in relation to the present invention. Various alternatives, modifications, and changes can be implemented within the scope.

Therefore, the embodiments described in the present invention and the accompanying drawings are not intended to limit the technical idea of the present invention, but to explain, and the scope of the technical idea of the present invention is not limited by these embodiments and the accompanying drawings. . The protection scope of the present invention should be construed according to the claims, and all technical ideas within the equivalent range should be construed as being included in the scope of the present invention.

1: sensor fusion based object detection device
10: first detection device 11: sensor
12: first detection module 13: first location tracking module
14: first processing unit 20: second detection device
21: signal transmission and reception unit 22: second detection module
23: second location tracking module 24: second processing unit
30, 400: fusion detection device 40: control device
410: receiving unit 420: reliability index generating unit
430: integration unit 440: detection unit
450: tracking unit 460: control unit

Claims (10)

a first detection device generating first detection information by detecting an object within the first surveillance area;
a second detection device generating second detection information by detecting an object within the second monitoring area; and
A fusion detection device for generating fusion detection information based on the first detection information, a first reliability index for the first detection information, the second detection information, and a second reliability index for the second detection information. do,
The first detection device performs position tracking on an object included in the first detection information, derives the number of tracking objects included in the first detection information, and generates it as the first reliability index;
The second detection device performs location tracking on an object included in the second detection information, derives the number of tracking objects included in the second detection information, and generates it as the second reliability index;
If the first detection device succeeds in detecting an object within the first surveillance area, it increases the number of traces associated with the first reliability index, and if it fails to detect an object within the first surveillance region, the number of traces associated with the first reliability index increases. Set the number of traces while decreasing the number,
If the second detection device succeeds in detecting an object within the second surveillance area, it increases the number of traces associated with the second reliability index, and if it fails to detect an object within the second surveillance area, the number of traces associated with the second reliability index increases. A sensor fusion-based object detection device that sets the tracking number while decreasing the number. According to claim 1,
A sensor fusion-based object detection device further comprising a control device that determines a risk based on the fusion detection information from the fusion detection device and performs a control operation according to a result of the determination. According to claim 1,
The first detection device is an image-based detection device, and the second detection device corresponds to any one of a radar-based detection device and a laser radar-based detection device. delete According to claim 1,
wherein the fusion detection device generates the fusion detection information for the detected object when an integrated reliability index generated based on the first reliability index and the second reliability index is equal to or greater than a predetermined threshold value. generating first detection information and a first reliability index for the first detection information by detecting an object within a first monitoring area;
generating second detection information and a second reliability index for the second detection information by detecting an object within a second monitoring area; and
Generating fusion detection information based on the generated first detection information, first reliability index, second detection information, and second reliability index,
Generating a first reliability index for an object in the first detection information may include tracking the location of an object included in the first detection information and deriving the number of tracking objects included in the first detection information. And generating the first reliability index by doing so,
Deriving the number of traces of the object included in the first detection information and generating the first reliability index increases the number of traces associated with the first reliability index when the object within the first monitoring area is successfully detected. , deriving the number of traces while decreasing the number of traces associated with the first reliability index when the object detection in the first surveillance area fails,
Generating a second reliability index for an object in the second detection information may include tracking the location of an object included in the second detection information and deriving the number of tracking objects included in the second detection information. And generating the second reliability index by doing so,
In the step of generating the second reliability index by deriving the number of traces of the object included in the second detection information, the number of traces associated with the second reliability index is increased if the object within the second surveillance area is successfully detected. , deriving the number of traces while decreasing the number of traces associated with the second reliability index when the detection of an object within the second surveillance area fails.
Sensor fusion based object detection method. According to claim 6,
The sensor fusion-based object detection method further comprising determining danger based on the generated fusion detection information and performing a control operation based on the determined danger. delete delete According to claim 6,
Generating fusion detection information based on the generated first detection information, first reliability index, second detection information, and second reliability index
generating an integrated reliability index based on the generated first and second reliability scores; and
and generating the fusion detection information for the detected object when the generated integrated reliability index is greater than or equal to a predetermined threshold value.

Images