KR101813790B1 - Apparatus and Method for multi-sensor information fusion based on feature information - Google Patents

Abstract

The present invention relates to a technique of tracking a target using feature information of a target, and more particularly, to a technique of tracking a target using characteristic information of a target, such as a target size, And more particularly, to a method and apparatus for identifying a target and a defacement object and tracking the target in a situation in which there is a deaf object that gives a similar or stronger signal to the target around the target by using the feature information.

Description

TECHNICAL FIELD [0001] The present invention relates to a multi-sensor information fusion apparatus,

The present invention relates to a technique of tracking a target using feature information of a target, and more particularly, to a technique of tracking a target using characteristic information of a target, such as a target size, And more particularly, to a method and apparatus for identifying a target and a defacement object and tracking the target in a situation in which there is a deaf object that gives a similar or stronger signal to the target around the target by using the feature information.

Traps and / or fighters operate various forms of dexterity such as chaff to deceive active microwave sensors and flare to deceive infrared sensors to enhance their survivability from threats. In addition, it also manages a compound device that operates chaff and flare at the same time.

Because the dexterity is operated with a very close and larger signal intensity to the target, the multi-sensor information fusion method using only the position information (distance, angle) of the target can not effectively distinguish the target from the dexterity.

1. Korean Registered Patent No. 10-1051716 (July 19, 2011)

1. Nah, JH, "Image minutia information detection method for position estimation" Chosun University 2013 2. Han, Doohee, et al., "Improvement of Induction Performance by Multi-sensor Information Fusion in Clutter Environment" Control ㅇ Robot ㅇ Systems Journal Vol 16 No.2 (Feb. 2010) pp.177-187

The present invention has been proposed in order to solve the problems according to the above background art, and it is an object of the present invention to provide a multi-sensor information fusion system capable of tracking a target in a stable manner by identifying a target and a defacement even in the case where various types of de- And an object of the present invention is to provide an apparatus and a method.

The present invention provides a feature-based multi-sensor information fusion device capable of reliably tracking a target by identifying a target and a defacement even in the case where various types of dexterous objects are operated around the target.

The multi-sensor information fusion apparatus includes:

A plurality of sensing units for generating target position information having positional information on a target and target characteristic information having a characteristic point for the target; And

The target position information and the target feature information, selects the measurement values closest to the target only through the target position information and the target feature information within the target effective area set in advance among the target position information and the target feature information, And a data processor for calculating a target predicted center point in the next time zone using the information fusion filter to update the information fusion filter.

In this case, the target location information may be at least one of a distance, an angle, and a signal to noise ratio (SNR) of the target.

The target feature information may be at least one of a size, a shape, and a brightness of the target.

The data processor may further include: a target identifying unit that identifies a target using the target location information and the target feature information; And applying a weight to the measurements with a difference according to the distance from the one central predictor and using a measure having the largest weight calculated as a result of the application, A target tracking unit for tracking a target through a tracking window control for updating a filter and estimating a target prediction center point of a next time zone through an updated information fusion filter to change an operation mode for the plurality of sensing units and a size of the target effective area, And a sensor information fusion filter unit.

The target identifying unit may convert the target position information and the target feature information into a navigation coordinate system to determine whether the target exists within the target effective area.

The plurality of sensing units may include a first sensor and a second sensor, wherein the first sensor is an infrared ray / visible ray sensor, the second sensor is a microwave sensor, and the position information of the infrared ray / And the feature information of the infrared / visible light sensor is a width, a length, and a brightness of a target, the position information of the micro sensor is a distance and an angle, and the feature information of the micro sensor is a length and height of a target .

The information fusion may be performed in order of the target position information and the target feature information, which are input to the data processor, in a time stamp order.

On the other hand, another embodiment of the present invention provides a method of generating target position information, the method comprising: generating target position information having position information for a plurality of sensing unit targets and target feature information having characteristic points for the target; The data processor receiving the target position information and the target feature information; Selecting only the target location information and the target feature information in the target valid area that is set in advance among the target location information and the target feature information received by the data processor; Selecting the measurements closest to the target using only the target location information and the target feature information that are within the predetermined target effective area; And a step of the data processor updating the information fusion filter using the selected measurement values.

According to the present invention, a method of using target feature information is added to a conventional target position-based target tracking method to use not only target position information of each sensor but also feature information such as size, intensity, and / A target can be stably traced by identifying the target and the deceit in situations where various forms of dexterity are operated near the target.

Another advantage of the present invention is that the target tracking performance can be improved through such stable tracking.

1 is a configuration diagram of a feature-based multi-sensor information fusion device according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating a multi-sensor information fusion process according to another embodiment of the present invention.
FIG. 3 is a conceptual diagram for explaining multi-sensor information fusion time alignment according to another embodiment of the present invention.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

Like reference numerals are used for similar elements in describing each drawing.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. The term "and / or" includes any combination of a plurality of related listed items or any of a plurality of related listed items.

Unless otherwise defined, 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 this invention belongs.

Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Should not.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a feature-based multi-sensor information fusion apparatus and method according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a configuration diagram of a feature-based multi-sensor information fusion device according to an embodiment of the present invention. Referring to FIG. 1, a feature-based multiple sensor information fusion apparatus 100 for multi-sensor information fusion processing includes first and second sensing units (not shown) for measuring target position information, 110-1 and 110-2, a data processor 120 for receiving the target position information and the target feature information, and performing information fusion. Here, the target position information may be a distance, an angle, and a signal to noise ratio (SNR) of the target. Also, the target feature information may be the size, shape, brightness, etc. of the target.

The first and second sensing units 110-1 and 110-2 detect the target (not shown) and generate target feature information having the target position information having the position for the target (not shown) and the target feature points .

In addition, the first sensing unit 110-1 includes a first sensor 111-1 for detecting the position and / or the characteristic points of the target, a first signal processor (for example, a first sensor 111-1 for processing the detected position and / 112-1). The first sensor 111-1 may be an infrared sensor, a visible light sensor, an illuminance sensor, an angle sensor, a temperature sensor, a microper sensor, or the like, in order to generate such positions and minutiae.

The first signal processor 112-1 performs a function of converting analog measurement information measured through the first sensor 111-1 into digital measurement information. For this purpose, a DSP (Digital Signal Processor), a memory, and the like may be configured in the first signal processor 112-1. The memory may be a memory provided in the first signal processor 112-1 or may be a separate memory. Accordingly, a flash memory (SSD), a hard disk drive, a flash memory, an electrically erasable programmable read-only memory (EEPROM), a static random access memory (SRAM), a ferro- Volatile memory such as a magnetic random access memory (MRAM) and / or a volatile memory such as a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), or a double date rate (SDRAM) .

Similarly to the first sensing unit 110-1, the second sensing unit 110-2 includes a second sensor 111-2 and a second signal processor 112-2. The second sensor 111-2 may be an infrared sensor, a visible light sensor, an illuminance sensor, an angle sensor, a temperature sensor, a microper sensor, or the like, in order to generate position and characteristic points. The second signal processor 112-2 performs a function of converting analog measurement information generated by the second sensor 111-2 into digital measurement information.

Although two first and second sensing units 110-1 and 110-2 are illustrated in FIG. 1 for the sake of understanding, the present invention is not limited thereto and may include two or more sensing units.

The data processor 120 receives the target position information and / or the target feature information generated by the first and second sensing units 110-1 and 110-2 and performs information fusion. A diagram showing this information fusion is shown in FIG. 2 will be described later.

With continued reference to Figure 1, the data processor 120 uses the target location information and / or the target feature information to identify the target and the target location information and the target feature information close to the identified target using the target location information and / A plurality of measured values in a predetermined target validation gate having a predetermined boundary with respect to the one central predicted value are divided by a distance from the center predicted value, A multi-sensor information fusion filter unit 122 for performing target tracking through an operation mode and a tracking window control for the plurality of sensing units 110-1 and 110-2 by updating the information fusion filter using the largest weight in the week And the like. Here, the operation mode becomes an operation mode of the first and second sensors 111-1 and 111-2, and is changed by the multiple sensor information fusion filter unit 1220. [

FIG. 2 is a flowchart illustrating a multi-sensor information fusion process according to another embodiment of the present invention. Referring to FIG. 2, target feature information such as a distance, an angle, and a target size, which are target position information on a target, are input from the second sensor 111-2 in accordance with the movement of the target (not shown) (Step S210, S220, S221, S230, S250).

In addition, the first sensor 111-1 receives the target feature information such as the angle of the target position information, the size of the target, and the brightness of the target, and uses the output of the INS (Inertial Navigation System) (Step S210, S220, S221, S240, S260).

The target position information and the target feature information, which are the sensor input information closest to the target, are determined by determining whether the target according to each of the converted target position information and the target feature information exists in the validation gate and whether the target feature information is a specific target. The information fusion filter is updated to perform target tracking (steps S251, S261, S270, S281, S291, S292).

Further, if there are m measurement values in the effective area of each sensor (111-1, 111-2 in FIG. 1), a weight is calculated for each measurement value (steps S251, S261).

The closest measure to the predicted position and feature of the target being tracked has the largest weight. The information fusion filter is updated using the measurement value having the largest weight to track the target (steps S270 and S281). That is, one central predicted value is calculated using the target position information and the target feature information close to the identified target.

Thereafter, a plurality of measured values in a predetermined target validation gate, which is a constant boundary based on the one central predicted value, are weighted by a distance from a center predicted value (center point of the target effective station) Is applied. After applying the weight, the information fusion filter is updated using the measurement having the largest weight.

Thereafter, the target prediction center point of the next time zone is estimated through the updated information fusion filter so that the first and second sensors 111-1 and 111-2 perform detection and / or tracking (steps S291, S292, S230, S240 ). In other words, since the defacement object is very close to the target and operates with a larger signal intensity, the largest weight is used to separate the dexterity object from the target, and the target prediction center point is again calculated for detection / tracking, The operating modes of the sensing units 110-1 and 110-2 and the size of the target effective area are changed. In other words, it performs tracking window control.

The location information of the microwave sensor is distance and angle, and the characteristic information is the size of the target (length, height). The position information of the infrared / visible light sensor is an angle, and the characteristic information is the width (i.e., length), the height (i.e., height) size of the target, and the brightness (average, maximum, minimum, variance) of the target.

FIG. 3 is a conceptual diagram for explaining multi-sensor information fusion time alignment according to another embodiment of the present invention. Referring to FIG. 3, since the acquisition time of target information between sensors is different, each sensor obtains the instantaneous time to acquire the target information, and attaches a time stamp to the sensor information (i.e., the target position information and the target feature information) . And processes the sensor information in chronological order by arranging each sensor information inputted to the data processor 120 in time stamp order.

100: Multi-sensor information fusion device
110-1: first sensing unit
111-1: first sensor 112-1: first signal processor
110-2: second sensing unit
111-2: second sensor 112-2: second signal processor
120: Data processor
121: target identification unit
122: multiple sensor information fusion filter unit

Claims (8)

A plurality of sensing units for generating target position information having positional information on a target and target characteristic information having a characteristic point for the target; And
Receiving the target position information and the target feature information, selecting the measurement values for the target only through the target position information and the target feature information within the target effective area set in advance among the target position information and the target feature information, And a data processor for calculating a target prediction center point of the next time zone using the information fusion filter to update the information fusion filter,
The data processor includes:
A target identifying unit for identifying the target using the target position information and the target feature information; And
Calculating a central predicted value of one of the measured values, applying a weight to the measured values with a difference according to a distance from the one central predicted value, and using a measured value having the largest weight calculated as an application result, For performing a target tracking through a tracking window control for updating an operation mode for the plurality of sensing units and a size of the target effective area by estimating a target prediction center point of the next time zone through an updated information fusion filter, And an information fusion filter unit.
The method according to claim 1,
Wherein the target location information is at least one of a distance, an angle, and a SNR (Signal To Noise Ratio) of the target.
The method according to claim 1,
Wherein the target feature information is at least one of size, shape, and brightness of the target.
delete The method according to claim 1,
Wherein the target identifying unit converts the target position information and the target feature information into a navigation coordinate system to determine whether the target exists within the target effective area. The method of claim 3,
Wherein the plurality of sensing units comprise a first sensor and a second sensor, wherein the first sensor is an infrared / visible light sensor, the second sensor is a microwave sensor, the position information of the infrared / visible light sensor is an angle, Wherein the feature information of the infrared / visible light sensor is a width, a length, and a brightness of a target, the position information of the micro sensor is distance and angle, and the feature information of the micro sensor is a length and height of the target Based multi-sensor information fusion device.
The method according to claim 1,
Wherein the information fusion is processed in a chronological order by arranging the target position information and the target feature information inputted to the data processor in a time stamp order.
Generating target feature information having target location information having location information for a plurality of sensing unit targets and feature points for the target;
The data processor receiving the target position information and the target feature information;
Selecting only the target location information and the target feature information in the target valid area that is set in advance among the target location information and the target feature information received by the data processor;
Selecting the measurements for the target using only the target location information and the target feature information within the predetermined target effective area; And
And the data processor updating the information fusion filter using the selected measurements,
Wherein the step of selecting only the target location information and the target feature information comprises:
And identifying the target using the target location information and the target feature information,
Wherein the updating comprises:
Wherein the multiple sensor information fusion filter unit calculates a center predicted value of one of the measurements and applies a weight to the measurements with a difference according to the distance from the one center predicted value, And a tracking window control unit that estimates a target prediction center point of the next time zone by using the updated information fusion filter to change the operation mode of the plurality of sensing units and the size of the target effective area, And performing tracking based on the feature information.

Images