JPS6033353B2 - image tracking device - Google Patents

Description

[Detailed description of the invention] The present invention relates to an improvement of an image tracking device.

As a means of extracting a specific target from an image, it is necessary to use information such as the shape, size, brightness, color, etc. of the specific image, but this invention uses the brightness between the background and the target to extract the target. It is characterized by the fact that it extracts the following information.

When an image in which there is a difference in brightness between the target and the background is viewed as an electrical signal (hereinafter referred to as a video signal), this appears as a difference in voltage level.

Therefore, in order to separate the target and the background, which have different brightness, by setting a threshold between the target voltage level and the background voltage level, the target can be extracted using a voltage comparator. . However, in order to set the threshold voltage in this way, it is necessary to know the target and background voltage levels. However, the target and background voltage levels may change from time to time, making it difficult to constantly detect each. In contrast, conventional methods of setting thresholds Once the operator specifies the target, the hardware detects the initial target voltage level. Next, fixed threshold voltages are set above and below this detected voltage level to perform target extraction. From then on, the method is to always detect the voltage level of the target at the center of the tracked target. Although this method allows target extraction even when the target voltage level changes, it is a fixed threshold, and although the target voltage level can be detected, the background voltage level cannot be detected, so it is not optimal. There was a problem in that it was not possible to set a precise threshold voltage. Another problem is that when the target is a small target, it is necessary to constantly detect the target lining, so if the tracking point temporarily deviates from the target due to disturbance etc., tracking is not possible. To solve the former problem, methods such as manually varying the threshold have been considered, but from the viewpoint of operability, it has been difficult to perform tracking corresponding to various targets. Therefore, the key point of the present invention is to detect the background voltage level as well as the target voltage level, which will be described in detail below.

It is necessary to specify whether the target is bright or dark relative to the background, but if the target is darker than the background (
(lower voltage level), first find the average value AVE{1} and MIN value VM... of the video signal in the processing area, set the middle value between them, and set the value VTH{1' as the value of the video signal below it. Find the average value AVE■. And further VM
...and AVE'2- is the threshold voltage VTH
[2] This is a method in which a video signal with a voltage level lower than this is targeted and a video signal with a voltage level higher than this is separated as a background. This is illustrated in FIGS. 1 to 4.

1 in Figure 1 indicates the processing area, in which the background 2 and target 3
Indicates the case where there is.

4 is one of the scanning lines passing through the target, and the video signal of this scanning line is shown in FIG.

In FIG. 2, 5 is the video signal, 7 is the average value AVEm of the video signal within the processing area, and 8 is the MIN value VMI.
N, 9 is the threshold between AVE{1} and VMIN, the value V
TH, average value AV of the video signal below the threshold value
E[2}, 11 is the threshold VTH■. As can be seen from this figure, when the target size is small compared to the background size in a simple background, AVE'1 indicates approximately the background voltage level, and VM... indicates the target voltage level. (If the target is brighter than the background, VMx indicates the voltage level of the target.) Therefore, the target and background can be separated using the intermediate value AVE■ as a threshold, but there is a problem. In other words, AVE{1) corresponds to the background voltage level under the above conditions, but if we consider the case where the ratio of the area of the target to the background and the background are not uniform, AVE(1' corresponds to the background voltage level. An example of this is shown in Figures 3 and 4. Figure 3 is a diagram with another background 6 added to Figure 1, and the video signal of the scanning line containing the target is shown in Figure 4. In the figure, 7 is AVE'1}, 8 is VM, small 9 is VTH [11, 10 is AVE[2', 1 1 is V
As shown in the figure, AVE'1
} is affected by the background of 6, and VTH'1) cannot be said to be an appropriate threshold for separating the target and background.

The threshold VTH■ used in this invention reduces the influence of this background of 6 by introducing the average value AVE{21. Below is the threshold VTHm, VT
The formula for determining H■ is shown below. {i} When the target is brighter than the background, VTH■=vMX12AVE■. ．． ．．
[1}VTHu)=vMAX+2AVE{1}
．． ．． ．． [2''o} If the target is darker than the background VTH (2) = vM, N+2AVE {2}
．． ．． ．． [31VTH[1,=v side N+2AVE'
11. ．． ．． [4' Next, the configuration of the present invention will be explained with reference to FIGS. 5 and 6.

FIG. 5 shows an example of a configuration diagram of an image tracking device, in which a target extraction circuit 1 is extracted from a video signal obtained from an imaging device 12.
3, the target is extracted, and the position detection circuit 14 detects its position.

One of these position signals is input to the mechanism driving section of the green camouflage layer 12 and fed back so that the target is always at the center of the visual field.
The other one is input to the processing area setting circuit 15 and changes the position of the processing area within the visual field according to the position signal. A mixer circuit 16 mixes the image from the imaging device 12 and the marker indicating the processing area.

Next, the receiver 17 displays this signal as an image, and performs initial target selection and setting, and visual confirmation of the approaching situation.
FIG. 6 shows an example of the configuration of the target extraction circuit 13 shown in FIG. 5, and is a feature of the present invention.

19 is an AD converter for quantizing the analog video signal from the imaging device 12, and 20 is an image memory circuit for the area designated by the processing area setting circuit, which is an image memory for the quantized video signal.

21 is an average value calculation circuit which takes this signal as input and calculates AVE'1}, AVE■, 22 is an Mx, M, N value detection circuit which detects VM^x, VM, N, and 23 performs these arithmetic operations. conduction threshold VTHm, V. This is an ALU circuit that calculates H■.

24 is a voltage comparator that compares the quantized video signal in the processing area with the threshold VTH (23) determined by the ALU circuit 23, thereby making it possible to extract the target signal.

As described above, according to the present invention, when extracting a target using the brightness and darkness between the background and the target, the average value of the video signal within the processing area set within the field of view is calculated, and this is The intermediate value between this first average value and the minimum value (or maximum value) is set as the threshold, and the average value of the signals below this is determined again and this is used as the second average value. By setting the average value and setting the threshold between the second average value and the minimum value (or maximum value), it is possible to improve the tracking performance for the target in automatic tracking, and therefore the load on the operator can be improved. can be dramatically reduced.

[Brief explanation of the drawing]

1 to 4 are diagrams for explaining the threshold voltage on a video signal according to the present invention, FIG. 5 is a diagram showing a configuration example of an image tracking device according to the present invention, and FIG. 1 is a diagram showing a configuration example of a target extraction circuit, in which 1 is a processing area;
2 is the background, 3 is the target, 4 is the scanning line, 5 is the video signal of the 4th scanning line, 6 is another background, 7 is AVE (1', 8 is VMI
N, 9 is VTHm, 10 is AVE {2-, 1 1 is V river ■, 12 is an imaging device, 13 is a target extraction circuit, 14 is a position detection circuit, 15 is a processing area setting circuit, 16 is a mixer circuit, 17 is a receiver, 18 is a mechanical drive unit, 19 is an AD converter, 20 is an image memory circuit, 21 is an average value calculation circuit, 22 is an M^x, skin N value detection circuit, 23 is an ALU circuit, and 24 is a voltage comparison circuit. It is a vessel. Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6

Claims (1)

[Claims] 1. A specific image within the visual field is extracted using a video signal obtained from an imaging device such as a TV camera or an infrared imaging device, and a position signal of the specific image with respect to the center of the visual field is output, and the position of the specific image is extracted with respect to the center of the visual field. In an image tracking device that has a servo mechanism that controls the signal so that it is always at the center of the field of view, when the target is darker than the background and is extracted as a lower level signal than the background, the signal within the processing area set within the field of view. The average value of the video signal is determined, and this is set as the first average value.The intermediate value between this first average value and the minimum value is set as the threshold, and the average value is determined again for signals below that threshold, and this is set as the first average value. 1. An image tracking device characterized in that the image tracking device is configured to automatically extract a target using an average value of 2 and an intermediate value between the second average value and the minimum value as a threshold. 2 A specific image within the field of view is extracted using a video signal obtained from an imaging device such as a TV camera or an infrared imaging device, and a position signal of the specific image relative to the center of the field of view is output, and the position signal is always at the center of the field of view. In an image tracking device that has a servo mechanism that controls the target to is determined, this is set as the first average value, and this first
The intermediate value between the average value and the maximum value is set as the threshold, and the average value of the signals above that value is determined again and this is set as the second average value.The intermediate value between the second average value and the maximum value is set as the threshold, and the target value is set. An image tracking device characterized by automatically extracting.