JP2001256500A - System and method for judging target - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a target judging system with high performance by performing a processing at higher speed and improving processing precision even when multiple similar objects (object candidates) exist and also an environmental change is intense. SOLUTION: A target judging system for judging a target object by using a picture processing system is presented. This system is provided with a picture input system 20 for inputting picture data comprising the target object and simultaneously performing a plurality kind of picture processing systems 200 and 201 in parallel. An object judgement processing system 1 calculates respective featured values corresponding to the object candidates obtained by the picture processings of the picture processing systems and calculates an evaluation value and a synthetic evaluation value based on the featured values. A synthetic judgement part 30 judges and processes the object candidate with the largest value as the final target object from the synthetic evaluation value of the object candidates.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a target judging system and a judging method for judging a target object required for monitoring and following a target object by using an image processing system.

[0002]

2. Description of the Related Art Conventionally, for example, in a missile target determination system and a monitoring system, a target determination system for specifying a target object is provided in order to follow and monitor a target object.

[0003] The target determination system inputs input image data obtained from an imaging device such as a camera, extracts target candidates from the image data, and finally executes a processing process for specifying a target target. . The system is
Using an image processing system, an object candidate extraction process based on luminance and contrast detection by a still image processing method, and an object candidate extraction based on detection of a moving region by a moving image processing (moving vector processing) method Execute the process.
In short, the target determination system extracts the feature amount of the target object candidate by each image processing method, and executes the final target object determination process based on the extracted feature amount.

[0004]

A target determination system using a conventional image processing system is relatively easy if the characteristics of the object are unified and can be determined by comparing the measured values. From the target object.

[0005] However, particularly when the target object exists in the field, there is a high possibility that many similar objects (object candidates) exist. For this reason, performing image processing for each object candidate increases the amount of processing, and as a result, delays the determination processing. Also, when the type of image processing is simply increased, the processing amount increases, and the processing is similarly delayed. Further, the processing accuracy is not always improved.

[0006] On the other hand, in many cases, the environment changes drastically in the field, and when unified image processing is applied, it may be impossible to detect a target candidate temporarily due to a change in image conditions, an increase in noise, and the like. High in nature. As a result of this,
In the continuous determination process, there is a high possibility that the process may be stopped or a situation in which a sufficient feature amount cannot be extracted for the target candidate.

Accordingly, an object of the present invention is to improve the processing speed and improve the processing accuracy even when a large number of similar objects (object candidates) are present and the environment changes drastically. A determination system and a determination method are provided.

[0008]

SUMMARY OF THE INVENTION The present invention relates to a target determination system for determining a target object using an image processing system. This is a system that performs integrated processing of feature amounts of target object candidates obtained by the method. Specifically, the present invention provides image input means for inputting image data including a target object and processing the input image using a plurality of image processing methods, and an object obtained by image processing of each image processing method. Calculating means for calculating each feature amount corresponding to the object candidate, and obtaining a comprehensive evaluation value of the object candidate based on each feature amount corresponding to each image processing method, and finally determining the target object And a determination processing means for performing the determination.

With such a configuration, for example, a plurality of types of processing methods such as a still image processing method and a moving image processing (moving vector processing) method are executed simultaneously and in parallel, and an object candidate is determined by each processing method. Can be obtained. Therefore, it is possible to capture characteristics of the object candidate from various angles, and it is possible to secure a sufficient characteristic amount even in a situation where some processing results become invalid due to an environmental change or the like. In addition, if a plurality of types of processing methods are executed simultaneously and in parallel, even if there are a large number of target object candidates, the determination processing speed does not decrease as a result. Furthermore, the accuracy of the target object is finally improved by obtaining the comprehensive evaluation value of the object candidate by processing the feature values of the object candidate obtained by each processing method in a so-called integrated manner. Can be.

[0010]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a target determination system 1 using a digital image processing system.
FIG. 3 is a block diagram showing the configuration of FIG. The system 1 is roughly divided into an image input system 20 and an object determination processing system 10.
And a comprehensive determination unit 30.

The image input system 20 has, for example, a still image processing unit 200 and a moving image processing unit (moving vector processing unit) 201 using a plurality of image processing methods. Image input system 20
Performs different image processing on an input image simultaneously and in parallel, and executes respective image processing data (SD), (M
D). Here, the input image is image data obtained by imaging an area including the target object by an imaging device (not shown) such as a camera.

The still image processing unit 200 has a two-dimensional C
FAR (two dimensional Constant False Alarm Rate)
A processing unit that uses a processing method or the like, detects a region having high luminance and contrast from an input image and outputs the detected region as image processing data (SD). Also, the moving image processing unit 201
Executes a movement vector process from an input image, detects a movement area, and outputs it as image processing data (MD).

Here, the area detected by each image processing is defined as a segment. Each segment belongs to any of the target object, similar object candidates, and their background. Each image processing data (SD),
(MD) includes segment feature values (Ua), which will be described later.
(Ub).

The object determination processing system 10 has a segment collation processing unit 100, a discriminant function value calculation unit 101, and a gradient correlation processing unit 102. The segment matching processing unit 100 receives the image processing data (SD) and (MD) output from the image input system 20, extracts and outputs the feature amounts (Ua) and (Ub) for each image processing method. I do. Furthermore, the segment matching processing unit 100 has a function of executing a segment matching process based on position information and the like, and determining whether or not each segment is the same object.

The discriminant function value calculator 101 calculates the feature quantities (Ua), (Ua) given by the segment collation processor 100.
b) and a predetermined feature amount weighting coefficient (ωa, ωb) are used to calculate a discriminant function value (evaluation value (V)) of a plurality of object candidates (segments). Here, the feature weight coefficient (ω
a, ωb) are preset values so that the target object can be calculated most accurately. Slope correlation processing unit 1
02 is a process of integrating evaluation values of a plurality of frames while discounting them by a gradient correlation process, and has a function of obtaining a total evaluation value (CV) of a plurality of object candidates in consideration of a time-series element.

The comprehensive judgment section 30 compares the comprehensive evaluation values (CV) of a plurality of candidate objects, and finally, for example, a segment of the comprehensive evaluation value (CV) showing a relatively maximum value is finally determined as a target object. It has a function to determine that there is.

(Procedure of Target Judgment Processing) Hereinafter, with reference to FIG.
With reference to the flowcharts of FIGS. 2 to 5 and FIG. 6,
A processing procedure of the system according to the embodiment will be described.

First, an input image such as that shown in FIG. 2A is input to an image input system 20 and processed by an image pickup device such as a camera. As described above, the image input system 20 executes different image processing on an input image simultaneously and in parallel, and executes respective image processing data (SD), (M
D) is generated (step S1). Here, as input images, the object candidates (T1) and (T2) and the background (A,
B) is assumed.

The segment collation processing unit 100 processes the image processing data (SD), (M
D), the feature amounts (Ua), (U
b) is extracted and output (step S2). Specifically, as shown in FIG. 2 (B), for the image shown in FIG. 2 (A), object candidates (T1), ( T2) and the feature amount (Ua) corresponding to the background (A, B) are extracted. Further, as shown in FIG. 3B, the image processing data (M
From (D), feature quantities (Ub) corresponding to the object candidates (T1) and (T2) and the background (A, B) are extracted.

Further, the segment matching processing unit 100
Perform segment matching based on location information, etc.
It is determined whether or not each segment is the same object (step S3). Here, by the determination process, it is determined that the segments that are the object candidates (T1) and (T2) and the background (A and B) are not the same object.

As shown in FIG. 4, the discriminant function value calculation unit 101 includes the feature amounts (Ua) and (Ub) given by the segment matching processing unit 100 and a predetermined feature amount weighting coefficient (ω).
a, ωb), a plurality of object candidates (T1, T
Calculate the evaluation value (V1, V2) of 2) (Step S)
4). Here, each segment of the background (A, B) is
On the basis of the evaluation value by the calculation unit 101, it is excluded from the target object. Further, the inclination correlation processing unit 102 evaluates the evaluation values (V1, V2) of the plurality of object candidates (T1, T2).
From the total evaluation value (CV1, CV1,
2) is obtained (step S5).

The comprehensive judgment unit 30 compares the comprehensive evaluation values (CV1 and CV2) of the plurality of object candidates, and finally determines a segment of the overall evaluation value showing a relatively maximum value as a target object. It is determined that there is (step S6). Here, as shown in FIG. 5, a relatively large overall evaluation value (C
The object candidate (T1) indicating V1) is finally determined as the target object (50).

As described above, according to this embodiment, a plurality of different image processing methods are simultaneously and concurrently executed on an input image, and a plurality of feature amounts of a candidate object are obtained by each processing method. Can be. Therefore, even in the case of the same object candidate, it is possible to capture features from multiple angles. As a result, even in a situation where some processing results become invalid due to an environmental change or the like, a sufficient feature amount can be secured.
Also, by executing a plurality of types of processing methods simultaneously and in parallel, it is possible to increase the processing speed even when a large number of object candidates exist. Therefore, as a result, the speed of the determination processing of the target object does not decrease. Furthermore, the feature amount of the object candidate obtained by each processing method is
In other words, since the target object is determined based on the comprehensive evaluation value by performing the integrated processing, the determination accuracy can be improved.

[0025]

As described in detail above, according to the present invention, even when a large number of similar objects (object candidates) are present and the environment changes drastically, the processing speed is improved and the processing accuracy is improved. Accordingly, a high-performance target determination system is provided. Therefore, if the system of the present invention is applied to, for example, a missile system that tracks a target object existing outdoors or a monitoring system that monitors an object existing outdoors where the environment changes rapidly, the target object can be specified without fail. It is very useful because it can.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a main part of a target determination system according to an embodiment of the present invention.

FIG. 2 is an exemplary view for explaining processing of a segment matching processing unit relating to the embodiment;

FIG. 3 is an exemplary view for explaining processing of a segment matching processing unit relating to the embodiment;

FIG. 4 is an exemplary view for explaining processing of a discrimination function value calculation unit related to the embodiment;

FIG. 5 is an exemplary view for explaining processing of a comprehensive judgment unit related to the embodiment;

FIG. 6 is an exemplary flowchart for explaining the procedure of a determination process according to the embodiment;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Target judgment system 10 ... Object judgment processing system 20 ... Image input system 30 ... General judgment part 100 ... Segment collation processing part 101 ... Discriminant function value calculation part 102 ... Slope correlation processing part 200 ... Still image processing part 201 ... Video Image processing unit (motion vector processing unit)

Claims (6)

[Claims] 1. A target determination system for determining a target object using an image processing system, comprising: inputting image data including a target object and processing the input image by a plurality of image processing methods. An input unit, a calculating unit that calculates each feature amount corresponding to the candidate object obtained by the image processing of each of the image processing methods, and an object candidate based on each feature amount corresponding to each of the image processing methods. And a determination processing means for determining a final target object by obtaining an overall evaluation value of (i). 2. The image input unit according to claim 1, wherein each of the first image processing method for performing still image processing and the second image processing method for performing moving image processing is simultaneously and concurrently executed. 2. The target determination system according to claim 1, wherein each image processing result including an extraction result of the object candidate is generated. 3. The segment matching processing means for executing a matching process for determining whether or not each object candidate obtained by the image processing of each of the image processing methods is the same object. The target determination system according to claim 1, wherein the target determination system includes: 4. A discriminant function value calculating means for calculating an evaluation value of an object candidate that has been collated as the same object by the segment matching processing means using a predetermined feature amount weighting coefficient. 4. The method according to claim 3, wherein
The target determination system described in the above. 5. The determination processing means compares the comprehensive evaluation values of a plurality of object candidates, and determines an object candidate corresponding to an overall evaluation value indicating a relatively maximum value as a final target object. The target determination system according to claim 1, wherein it is determined that there is a target. 6. A determination method applied to a target determination system that determines a target object using an image processing system, wherein image data including a target object is input, and the input is performed by a plurality of types of image processing methods. Processing an image; calculating each feature amount corresponding to the candidate object obtained by the image processing of each image processing method; each candidate object obtained by the image processing of each image processing method Performing a matching process to determine whether the object is the same object, and the evaluation value of the target candidate that is matched as the same object in the matching processing step, a predetermined feature amount weighting coefficient Calculating using the evaluation value; calculating the overall evaluation value of the target candidate by the inclination correlation processing of the evaluation value; and comparing the overall evaluation values of the plurality of object candidates with each other. A step of determining that an object candidate corresponding to the comprehensive evaluation value indicating the maximum value is the final target object.