JPH06242205A - Apparatus for tracking image - Google Patents

Abstract

PURPOSE:To obtain an image tracking apparatus which can determine a suitable threshold value irrespective of the change of the luminance of a target and has the remarkably improved efficiency to detect the target. CONSTITUTION:A measuring window read circuit 13 reads out the luminance data of a designated position within the screen when receiving a measuring window controlling data 14. The luminance data is supplied as a background luminance data 15 to a microcomputer 9. The microcomputer 9 determines a threshold value from the background luminance data 15, and feeds the value as a threshold value data 10 to a binarizing circuit 3. Therefore, the threshold value can be calculated on the basis of the luminance of the background in the periphery of the target. Even when the maximum luminance of a tracking target is lowered to a value smaller than a predetermined constant value in comparison with the case one field before, the target part can be detected stably as '1' area.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention prevents unstable tracking of the threshold value due to minute changes in the brightness of a target during flight object guidance or firearm control image tracking, and enables stable detection of the tracking target. The present invention relates to a possible image tracking device.

[0002]

2. Description of the Related Art FIG. 9 is a diagram showing an example of a conventional image tracking device, wherein 1 is an image pickup device for picking up an image of the outside world, 2 is a digital video signal output from the image pickup device 1, and 3 is a multi-valued signal. A binarization circuit that converts a digital video signal into a binary image,
4 is the binary image data in which the target area is "1" and the background area is "0", and 5 is the connected closed area that measures the geometrical characteristics of the connected closed area consisting of pixels with the value "1" in the binary image. Area feature measuring circuit, 6 is feature quantity data for each connected area, 7 is a feature vector buffer for recording feature quantity data of each area, 8 is a feature vector of each area, 9 is a microcomputer, 10 is threshold data , 11 is an interface circuit with an external device, 12 is tracking error data, and 20 is an input / output interface.

In the figure, an external scene is imaged by an image pickup device 1 and output as a digital video signal 2 in the scanning order of the image pickup device 1. The digital video signal 2 usually indicates scanning boundaries and is composed of a synchronization signal and luminance data of each pixel. The microcomputer 9 supplies the preset threshold value as threshold value data 10 to the binarization circuit 3. The binarization circuit 3 compares the luminance data of the digital video signal 2 with the threshold value data 10, and if the luminance is higher than the threshold value, the value is "1", and if not, it is "0". It is binarized and output as binary image data 4. The connected closed area feature measuring circuit 5 is based on the brightness data and the binary image data 4 of the digital video signal 2 described above.
Various feature amounts are measured for each connected “1” region in the binary image, and the feature amount data 6 is output. The feature amount data 6 includes the area of each region, the barycentric coordinate, the luminance, the circumscribed rectangular coordinate, and the like. These measured feature amount data 6 are feature vectors of a structure that is integrated into one for each connected closed region,
It is recorded in the feature vector buffer 7. The steps up to this point are performed for each field as needed. Next, when there are a plurality of areas in the field of view, the microcomputer 9 determines threshold value data from the maximum brightness in the screen one field before and the standard deviation of the sensor noise in order to determine which area to track. Decide on 10. The threshold TH is
If Gmax is the maximum brightness in the screen, σ is the standard deviation of the image pickup device noise, and α is a constant given experimentally, it can be determined as in “Equation 1”.

[0004]

[Equation 1]

The microcomputer 9 reads out the feature vector 8 of each region in the current field from the feature vector buffer 7, determines the region with the highest brightness as the tracking target, and starts the tracking operation. Then, after starting the tracking operation, the threshold value data 9 is determined from the maximum luminance of the area determined to be the tracking target one field before and the standard deviation of the sensor noise. The threshold value TH is determined, for example, as in “Equation 2”, where Tmax is the maximum luminance of the tracking target, σ is the standard deviation of the image pickup device noise, and α is an experimentally given constant.

[0006]

[Equation 2]

The microcomputer 9 compares the feature vector of the area determined to be the tracking target one field before with the feature vector of each area in the current field, and selects the area having the closest feature vector. The barycentric coordinates of the area are transferred as tracking error data 12 to the external device through the interface circuit 11.

[0008]

In the conventional apparatus, the maximum brightness of the area determined to be the tracking target is measured,
The value obtained by subtracting a certain value from the maximum brightness was set as the threshold value. The maximum brightness of the target is stable when the target area is large to some extent, but when the target is very small close to the detection limit of the imaging device, the brightness changes greatly due to the aperture ratio of the detection element of the imaging device. There are cases. Therefore,
As shown in FIG. 10, when the maximum luminance of the tracking target falls below a certain value as compared with the one field before, the threshold value exceeds the maximum luminance of the target, and the target portion is set as the “1” area. There was a problem that it could not be detected stably.

The present invention has been made to solve the above-mentioned problems, and a measurement window is set in the screen of a digital video signal from an image pickup device, and when the target area is small, the brightness around the target is adjusted. By measuring and determining the threshold value for binarization, the object is to stably detect the target area regardless of the change in the target brightness.

[0010]

An image tracking device according to the present invention is provided with a means for determining a threshold value by a measurement window reading circuit regardless of a change in target brightness.

The image tracking device according to the present invention is
When the background brightness is unstable, the measurement window read circuit and the brightness change calculation circuit wait for the background brightness to stabilize, and a means for determining the threshold value is provided.

Further, the image tracking device according to the present invention includes a measurement window reading circuit and a background luminance difference calculating circuit.
Even if a part of the target area covers the measurement window, a means for determining the threshold value is provided excluding the constituent pixels.

[0013]

The image tracking device according to the present invention can extract only the target regardless of the change in the brightness of the target and the change in the brightness of the background, and the target detection capability can be greatly improved.

[0014]

【Example】

Example 1. FIG. 1 shows a configuration of one embodiment of the present invention, in which a measurement window read circuit 13 sets a measurement window in one screen of a digital video signal 2 and measures luminance data in the measurement window. Reference numeral 14 is measurement control data, and 15 is measured background luminance data. In the figure, 1 to 12 are equivalent to those in FIG. 9 described in the prior art.

In the configuration shown in the figure, the microcomputer 9 sets a measurement window as shown in FIG. 2 on the basis of the position of the center of gravity of the area which has been determined to be the tracking target one field before after starting the tracking operation. The control data 14 is supplied to the measurement window read circuit 13. 2 may be determined as in, for example, “Equation 3”, where S is the area of the tracking target and α is a constant given experimentally.

[0016]

[Equation 3]

The measurement window read circuit 13 receives the measurement window control data 14, reads the luminance data at the position designated on the screen, and supplies it as the background luminance data 15 to the microcomputer 9. The configuration of the measurement window read circuit 13 may be, for example, a field memory that stores brightness data for one field, and may read only necessary addresses from the microcomputer 9. The microcomputer 9 determines a threshold value from the background luminance data 15 and supplies it as the threshold value data 10 to the binarization circuit 3. For the threshold value TH, W (i) is measured window constituent pixel (i
= 1. ． If N is the luminance data, σ is the standard deviation of the image pickup device noise, and α is a constant given experimentally, it can be determined as in “Equation 4”.

[0018]

[Equation 4]

The operation after the binarization circuit 3 is similar to that of the conventional device.

As described above, according to the above configuration, the threshold value can be calculated based on the brightness of the background around the target, and as shown in FIG. 3, the maximum brightness of the tracking target is a fixed value compared with the one field before. Even if the value drops below the value, the target part is set to "1".
It can be detected stably as a region.

By the way, in the above embodiment, when the background brightness around the target is stable, an appropriate threshold value can be set, but the background brightness is significantly increased, for example, at the moment when the background changes from cloud to sky. When the value changes to, the threshold value TH is set to a high value as shown in FIG. 4, the target portion cannot be detected as the “1” area, and the application is difficult with the above configuration.

Example 2. FIG. 5 shows a configuration of a second embodiment of the present invention for solving the above-mentioned problems, in which 16 is a luminance change calculation circuit and 17 is a background luminance change rate. 5, 1 to 12 and 13 to 15 are the same as those shown in FIG.

In the configuration shown in the figure, the brightness change calculation circuit 16
Receives the background luminance data 15 from the measurement window reading circuit 13 and calculates the background luminance change rate 17. If the period for calculating the rate of change is N fields, the background luminance data for N fields is stored and the rate of change is calculated. Assuming that the background luminance of the i-th field (i = 1..N) is L (i), the background luminance change rate Lt is calculated by "Equation 5".

[0024]

[Equation 5]

The microcomputer 9 compares the background luminance change rate 17 with a preset determination reference value, and when the determination reference value is exceeded, the background luminance data 15 is regarded as invalid, and the background luminance change rate 17 is determined by the determination reference value. Wait until it is within the value. Then, the threshold value is calculated based on the background luminance data 15 when the value is within the judgment reference value.

As described above, according to the above configuration, even when the background luminance changes, the rate of change can be calculated, and by waiting for the background luminance to stabilize, the target portion is stably detected as the "1" area. be able to.

By the way, in the second embodiment, when the aspect ratio of the target is close to 1: 1 as in the case where the target is sufficiently far away, the entire target area exists inside the measurement window composing pixel, so that it is stable. Although the threshold value for detecting the target area can be set, when the target has a large aspect ratio as in the case where the targets are close to each other, a part of the target area is measured window constituent pixels as shown in FIG. As a result, the threshold value TH is set higher, the target portion cannot be detected as the "1" region, and the application is difficult with the above configuration.

Example 3. FIG. 7 shows a configuration of a third embodiment of the present invention for solving the above problems, in which 18 is target luminance data, 19 is a background luminance difference calculation circuit, and 20 is background luminance difference data. .

In the configuration shown in the figure, the measurement window read circuit 1
As shown in FIG. 8, the pixel 3 has measurement window constituent pixels near the target center of gravity and supplies the target luminance data 18 in addition to the background data 15 to the background luminance difference calculation circuit 19. The background brightness difference calculation circuit 19 receives the background brightness data 15 and the target brightness data 18, calculates the brightness difference between the target and the background, and outputs the background brightness difference data 20. The microcomputer 9 compares the background luminance difference data 20 with the preset judgment reference value, and from the background luminance data 15 of all the measurement window constituent pixels,
The threshold value is calculated after excluding the luminance data of the constituent pixels that do not meet the determination reference value.

As described above, according to the above configuration, even when a part of the target region overlaps the measurement window constituent pixel such as when the target aspect ratio is large, the threshold value is calculated excluding the constituent pixel. Therefore, the target portion can be stably detected as the “1” region.

[0031]

As described above, according to the present invention, the measurement window read circuit can determine an appropriate threshold value irrespective of changes in the target brightness, and stabilizes the target portion as the "1" region. Can be detected.

Further, when the background brightness is unstable, the measurement window read circuit and the brightness change calculation circuit allow the threshold value to be determined after waiting until the background brightness stabilizes.
The target portion can be stably detected as the "1" region.

Further, even if a part of the target area covers the measurement window, the threshold value can be determined by excluding the constituent pixels by the measurement window read circuit and the background luminance difference calculation circuit. It can be stably detected as the "1" region.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram showing a first embodiment of the present invention.

FIG. 2 is a diagram showing an arrangement of measurement windows according to the present invention.

FIG. 3 is a diagram showing a relationship between a threshold and a target brightness according to the image tracking device of the present invention.

FIG. 4 is a diagram showing a relationship between a threshold and a target luminance when the background luminance changes.

FIG. 5 is an overall configuration diagram showing a second embodiment of the present invention.

FIG. 6 is a diagram showing the relationship between the arrangement of measurement windows according to the present invention and a target having a large aspect ratio.

FIG. 7 is an overall configuration diagram showing a third embodiment of the present invention.

FIG. 8 is a diagram showing an arrangement of measurement windows according to the present invention.

FIG. 9 is an overall configuration diagram showing an example of a conventional image tracking device.

FIG. 10 is a diagram showing a relationship between a threshold value and a target brightness according to a conventional image tracking device.

[Explanation of symbols]

 1 Imaging Device 3 Binarization Circuit 5 Connected Closed Region Feature Measurement Circuit 7 Feature Value Vector Buffer 9 Microcomputer 11 Interface Circuit 13 Measurement Window Read Circuit 16 Brightness Change Calculation Circuit 19 Background Brightness Difference Calculation Circuit 20 Input / Output Interface

Claims (3)

[Claims] 1. An image pickup apparatus for picking up an image of the outside world, a measurement window read circuit for setting a measurement window in one screen of a digital video signal output from the image pickup apparatus, and measuring luminance data in the measurement window, and a digital video. A binarization circuit that converts a signal into a binary image by a threshold value, and a connected closed region feature measurement circuit that measures a feature amount for each connected closed region from the output binary image of this binarization circuit, A feature vector buffer that stores a feature amount for each area, a microcomputer that determines the threshold value of the binarization circuit based on the output of the measurement window read circuit, and controls the entire device, and an external device. And an interface circuit for interfacing with the image tracking device. 2. The image tracking according to claim 1, further comprising a luminance change calculation circuit that receives the output of the measurement window read circuit, calculates the luminance change of the background, and supplies the calculation result to the microcomputer. apparatus. 3. A background luminance difference calculation circuit that receives the output of the measurement window read circuit, calculates the luminance difference between the target and the background, and supplies the calculation result to the microcomputer. Image tracking device.