JPS583483A - Detector for mobile body - Google Patents

Abstract

PURPOSE:To cope with a travelling body which moves in a complicated high speed, by processing a real and an imaginary part of a space filter function from a memory and a video signal from a television camera. CONSTITUTION:A space filter function of a complex number is stored in response to each picture element of a television camera 1 by a memory 3. A control circuit 2 reads out a real part and an imaginary part of a space filter function in response to each picture element from the memory 3 in synchronizing with the scanning of the television camera 1 respectively. Each component of the read out space filter function and the video signal from the television camera 1 are respectively multiplied at a multiplication circuit 5 individually, integrated at an integration circuit 6, then inputted to a CPU4 via an A/D conversion circuit 7. The CPU4 calculates required travelling body information from the result of integration.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a moving object detection device, more specifically, a moving object detection device, which detects a moving object by capturing an image of an object with a television camera and processing a video signal output from the television camera. The present invention relates to a moving object detection device that measures the speed and determines the shape of a moving object.

One type of moving object detection device is one that utilizes the spatial fill effect. A spatial filter is basically constructed at the image plane of an objective lens that transmits infrared or visible light or a reflector that reflects it, and suppresses the influence of radiation from the background other than the object included in the field of view. It can be used effectively for speed measurement. A spatial filter is constructed by a reticle placed on the image plane of an optical system, or a detector made up of a large number of photoelectric conversion elements arranged parallel to each other. There are also electro-optic reticles that are controlled to move in one direction with alternating opaque and transparent parts.However, in such reticles or photodetector plays, their configuration is Since the spacing of the array of elements is fixed, the spatial frequency is fixed to a value defined by the spacing, and there is a problem that only a specific frequency component of the object can be detected, resulting in a lack of versatility.

Another object of the present invention is to provide a highly versatile moving object detection device that can select any spatial frequency component. It is an object of the present invention to provide a moving object detection device that can detect a direction and can handle moving objects that make complex movements and high-speed moving objects.

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

In this invention, a television camera is used as a detector for detecting an object. The image pattern of the object projected onto the television camera through the lens system is expressed as a function f(x,
y). If the spatial frequency function of this video pattern is F(U, V), it is expressed by the following equation.

F (U, V) ・dx-dy ... (1) Here, S
indicates the area to be integrated, which is the area of the image projected onto the television camera or a smaller area. exp
(-j (U-x+V・y)) is called an intercharacter filter function. U and V are spatial frequency components (angular velocities) in the XIY direction, respectively, and if λX and ^y are wavelengths in the XIY direction, they are expressed by the following equations.

U22π/λX...(2)■=
2π/λy (3) Also, the spatial frequency 0 and the direction θ of its vector with respect to the X-axis direction are expressed by the following equation.

Φ=, S possible 7...(4) θ= ja
n' (5) Expressing equation (1) using trigonometric functions, the following equation is obtained.

F(U,V) − ・(cos(U−x+V′y)−jsin(U−x+V
-y))・dx-dy...(6) This function F(U, V) is later! Since this is the output of a spatial filter constructed isometrically by an electric circuit, this is called a spatial filter output. This space? , λ − il? When an object within the field of view of the television camera is moving, D (u, v) changes sinusoidally with a frequency proportional to the speed of movement of the object.

Since F(U, V) is a complex number, it can be expressed as a vector on a complex plane as shown in FIG. Here...(7)...(8) R 9°゛°(9) Obtain this phase angle α for each scan of one screen, and calculate its change △
By calculating α, the moving speed W of the object can be obtained using the following equation.

Here, τ is the scanning time for one screen, and is approximately 16.7 ms in the case of one field. The direction of movement of the object is
This can be determined by the polarity (positive or negative) of Δα.

If the wavelength of the spatial frequency (the pitch of the spatial filter) and the period at which the P1 output F (U, V) changes are T, then the speed W can also be obtained from the following equation.

Kini P/T...aυHowever, P21/f is "lower 7.

Since the spatial frequency 0 is defined by the values of U and V, the spatial frequency 0 can be changed by changing these U and V. This is equivalent to changing the interval between the components of the reticle and light receiving element array that constitute the above-mentioned spatial filter. In this invention, as described below, since the values of U and ■ can be arbitrarily set, any spatial frequency component of the object can be obtained, and the invention is known for its versatility.

FIG. 2 is a block diagram showing the configuration of the moving object detection device. The output signals of each of these blocks are shown in FIG. FIG. 8a shows each signal of one field, FIG. 8b shows an enlarged signal of one horizontal scan, and FIG. 8c shows a further enlarged signal of one pixel. A television camera (1) images an object and outputs a video signal (5) thereof, and is controlled by a control circuit (2). The video signal (A) is the function f(x, y
). The video signal (A) has its low frequency components and DC components cut off by a high band pass filter OQ.
This results in a video signal (A1) containing only high frequency components. As shown by the broken line in FIG. 5, the video signal (A) contains many low frequency components and DC components. This mainly depends on the brightness of the background of the object. The video signal (A1) after passing through the filter QO is shown by a solid line in FIG. It is preferable that the filter OO cuts components having a frequency lower than the spatial frequency in the y direction to be extracted. It is also effective to cut frequency components below the scanning frequency (approximately 60 Hz) of the television camera (1). Filter OQ
Specifically, as shown in Figure 4, the capacitor (T)
It consists of a differentiator circuit consisting of a resistor Q and an amplifier Q. .

The control circuit (2) outputs a vertical synchronization signal and a horizontal synchronization signal to control the video camera (1), and also outputs a vertical address signal (B) and a horizontal address signal (C) in synchronization with these synchronization signals. Output. The control circuit (2) also generates a reset signal (-) and an area designation signal (E), which will be described later.

Storage device (hereinafter simply referred to as RAM) (3) is No. (7)
And the value of (U-x+v-y) (radian) of equation (8) is stored. (U-x+v-y) can be expressed as variables X and y if the values of U and V are determined in advance.
Uniquely determined by The RAM (3) has storage locations equal to the number of pixels of the television camera (1), as shown in Figure 6, and each storage location has a number of bits that can represent the number Nx - Ny. It has information storage capacity. The number of pixels (number of memory locations) in the horizontal scanning direction (X direction) is N
x, and the number of pixels in the vertical scanning direction (y direction) is Ny. R
Each memory of AM (3) contains (U −x+V−y
) values are stored. Each memory location of RAM' (3) is controlled by an address signal (B) output from the control circuit (2).
) (C), and the value of (U-x+v-y) is read out in synchronization with the scanning of camera (1).

ROM (8) and (9) are (IJ'x+v-y)
Depending on the value of cos(U-x+V-y) (hereinafter simply (
cos)), and -5in (U-x+v-y
) (hereinafter simply expressed as (sin)). (COS) or (sin) takes positive and negative values. In ROM (8) or (9) (
cos) or (Si,) is represented by 8 bits, for example, the most significant digit (MSB) represents a positive or negative sign, and the other 7 bits represent (cos) or [sin]. It represents the absolute value.

A more detailed configuration of ROM (8) (9) is shown in FIG.
The address specifying where is stored is decoded and
The result of this decoding is (cos
) or (sin) is read. The read data is the signal (G) and wholesale.

The multiplier circuit (5) contains the video signal of the television camera (1)) and the data read from the ROM (8) (9) [c
Signal (G) (H) indicating the value of os] or (sin)
is input and fcx.

y)·[cos] or f(x, y)(sin) is calculated.

An example of this multiplication circuit (5) is shown in FIG.

The video signal (Al) is an analog signal, and (cos) and (sin) are digital signals. Multiplication circuit (5)
It has a FiDA conversion function and converts the multiplication result into an analog signal (
1) Output as (J). The multiplier circuit (5) receives the video signal (Al) as input, connects to a ladder-like resistance circuit 0η having eight output terminals, each output terminal of this resistance circuit 0η, and outputs each of the signals (G) (H). Analog changeover switches 翰 to 翰 controlled by bits, a first adding circuit consisting of an inverting amplifier (31) and a resistor, which adds the outputs of the terminals (a) of these switches, and this first adding circuit. The output of the second adding circuit is composed of an inverting amplifier (32) and a resistor (2), which adds the output of the adding circuit and the output of the terminal (b) side of the switch. The output signal is (1) (J). The signal (G) (goods) has a heavy current (2's "#") according to each digit. The resistor circuit 0υ outputs the video signal (A1). It is divided according to these weights.If the current flowing into the switch (1) is 1, a current of 2'I flows into the switch (goods).Signal (G)
Only the MSB of (H) is inverted by the NOT circuit (2), and the switch (B) is controlled by this inverted bit. When a certain bit of the signals (G) and (H) is 0, the corresponding analog switches except switch @ are connected to the (b) side, and when it is 1, they are switched to the (a) side. The state in which the switches (E) to (A) are shown in FIG. 8 is when the signal (G) G() is ooooooooo. In this case, a resistor () is connected between each resistor and the output terminal of the amplifier C (+1) and the inverting input terminal of the amplifier C (+1) so that the signal (I) (J) becomes 0. The value of the signal (I) (t or (J)) corresponding to the value of the input signal (G) (or Q()) is determined.
) is as follows. MS here
When B is 0, (COS) and (s in) are positive, and when B is l, they are negative.

(G) (1')0 1 1
1 1 1 1 1 → CAI) X 12
7/128? 1 0000 0 00 1 −+ (AI) X
1/128ooooooooo →
01 1 1 1 1 1.1 1 → (A
t) x (”/128)/,
/ 10000001 → (Al) x C! :1m?
/1゜8)10000000 7) (A1)x
(-1) The integration circuit (6) integrates the output signal (I) or (J) of the multiplication circuit (5). Integrator circuit (
A specific example of 6) is shown in FIG. The integrating circuit (6) includes an operational amplifier 11, its input resistance O2, a capacitor connected between the inverting input terminal and the output terminal of the amplifier I11, and a capacitor connected between the resistor O and the inverting input terminal. It consists of a switch @ chrysanthemum connected to the capacitor, and a switch (9) for shorting the capacitor. A reset signal Q)) is output at the start of scanning one screen (one field or one frame), and this reset signal (DJ
The switch (9) is turned on. switch (4
When υ is turned on, the charge on the capacitor 0'3 is increased by 8, so that the integrating circuit (6) is reset. .

The scanning period for one screen includes a blanking period in which there is no video signal. In addition, as shown in FIG. 3c, RAM (3
) for reading (U-x+v-y) from (
TDI), (cos) from ROM(8)(9)
, (sin) of the readout time delay (TD2) and the multiplier circuit (5) delay (TD8). The area designation signal (E) eliminates such retrace period and delay time and determines the integration period so that only valid input signals are integrated, and the switch (B) is turned on only during this integration period. Then, signals (I) and (J) are input. The output camera (L) of the integrating circuit (6) is the (7) (L), respectively.
8) represents -Q and R in the formula. This output signal 4 (L
) is converted into a digital signal by the AD conversion circuit (7),
Read by CPU (4).

As mentioned above, the values of (COS) and (sin) are digital quantities, and considering economy, it is appropriate to represent them with about 8 bits. The larger the capacity of each memory location in ROM (8) (9), the more accurately (cos) and (s in) can be represented. However, the ROM becomes expensive and the multiplication circuit (5) becomes complex. The video signal (5) is an analog signal as described above, and contains many low frequency components and DC components. Therefore, the signal (
5) and the signal (F) depends on the low frequency component and DC component of the signal distribution, and since the signals (G) and (H) are approximate values expressed by 8-bit information,
If so, the signals (1) (U) and (K) (L) tend to become inaccurate. -However, since the low frequency components and DC components of the video signal (8) are removed by the high bandpass filter 01), the multiplication result (1)
U) and the integration result (K) (L) will be accurate.

The CPU (Central Processing Unit) (4) performs processing such as writing (U-x×y) into the RAM (3) and calculating the moving speed of the object based on the data read from the space conversion circuit (7). Do the following. CPU (4) is preferably a microprocessor.

FIG. 10 shows the field of view 1511 of the television camera (1),
The lens in the TV camera (1) and the RAM (3) corresponding to the image connected to the TV camera (1) i
It shows.

As mentioned above, RAM (3) contains the television camera (
There are a number (NxXNy) of storage locations equal to the number of pixels in 1).

The length of the visual field υ corresponding to the horizontal length of -1 pixel is MK+The length of the visual field @υ corresponding to the vertical length of 1 pixel is expressed as MY. The horizontal and vertical lengths of the field of view (c) I) are Mx-Nx and My-Ny, respectively.

Figure 11 shows the (
This shows the write processing of U-x+V'y). This process is performed during the initial process of the CPU (4). An address counter that specifies the X-direction and X-direction addresses of each memory location in RAM (3) is Cx,
It is represented by Cy, and its contents are represented by (Cx) and (Cy), respectively. In addition, the above-mentioned sky frequency components U, V
The value obtained by dividing .pi. by 2.pi. represents the bare number of the spatial filter electrically constructed with the circuit of FIG. It is assumed that U and V are set in advance. Further, the above-mentioned magnifications Mx and My are also set in advance.

Referring to FIG. 11, each address counter Cx.

Cy is set to 0 (step Iυ), and the contents of these counters (Cx) (Cy) are given a multiplier M.
The values obtained by multiplying x and My, respectively, are placed as X and Y (step 183).

These x, y, above-mentioned U, V, Nx, Ny and Mx
.

Using UV My, (Nx 0Mx −X+Ny 1My・
Y) is calculated and this result is set as 2 (step -
). The Z obtained in Nut7' 141 is written to the memory location in RAM 3) addressed by the address counters CX, Cy (step I6+il). counter C
The contents of x are incremented by 1 (step -), and it is checked whether the new (Cx) exceeds (Nx-1) (step 171). If the step is ready and the answer is No, the process returns to step -, and the above-described process is repeated using a new (Cx). If step η is YES, one horizontal scan has been completed, so the contents of the counter cy are incremented by 1 (step embroidery), and this new (Cy) becomes (Ny-1).
is exceeded (step II). If NO in step -, the process is returned to the control of the step, (Cx) is set to 0, and the processes from step to step are repeated.

If YES in step -, writing of (U-x+v-y) to all memory locations in RAM (3) is completed.

FIG. 12 shows the moving speed W of the object by the CPU (4).
It shows the calculation process. This process is performed by the AD conversion circuit (
5) is started by an interrupt input from 7) and is executed for each interrupt. As mentioned above, the scanning period of one screen is set to τC1 (for example, 16.7 ms). It is assumed that the pitch P of the spatial filter has been calculated in advance for the signal (L) read from the slope conversion circuit (7) for each side filter. Furthermore, let α1 be the phase angle calculated during the previous screen scan.

Referring to FIG. 12, from the oil conversion circuit (7) to the AD conversion point fc 7' -fi (AD)K, (AD)L
are read and loaded into registers Ra and Rb in CPU (4), respectively (step ff11-). next,
Using the data in each register Ra, Rh, the phase angle α is calculated according to equation (9) (step (72). From this phase angle α and the previous phase angle α1, the change Δα is calculated. (step 9), and the current phase angle is stored as α1 for the next calculation (step σ41). Then, the velocity W is calculated by the calculation of the equation Q1) (step 4). As mentioned above, the moving direction of the object can be detected by checking the polarity of Δα.

In the above embodiment, (U-x+v-y) is stored in RAM (3).
However, two such Oboro units are provided, and (cos) and [5in) are stored in the ROM (8
) (9) may be omitted. Also, the 7th
) and (8) are performed by the multiplication circuit (5) and the integration circuit (6), but the video signal (A1)
It is also possible to perform these calculations on the CPU (4) after AD converting the data and importing the data into the CPU (4).

As described in detail above, according to the present invention, the pitches U and V can be set arbitrarily, so that the present invention is much more versatile than a moving object detection device using a conventional spatial filter. Also, the scanning time of a TV camera is usually about 16.7ms for one field + about 88.8ms for 17 frames.
Therefore, at least the moving speed and direction of the moving body can be detected from the scanning information of one screen, and high-speed processing is possible. Therefore, it can immediately respond to moving objects that make complex movements or move at high speeds. Furthermore, it is also possible to determine the shape of the object by analyzing the obtained spatial filter output.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram showing vectors of spatial filter output;
FIG. 2 is a block diagram showing an embodiment of the present invention, FIG. 8 is a time chart showing output signals of each block in FIG. 2, and FIG. 4 is a circuit diagram showing an example of a high band pass filter. Figure 5 is a graph showing the frequency components of the video signal, Figure 6 is a diagram showing the configuration of the ROM, Figure 7 is a graph showing the configuration of the ROM,
Fig. 8 is a circuit diagram showing an example of a multiplication circuit, Fig. 9 is a circuit diagram showing an example of an integrating circuit, Fig. 10 is a diagram showing the relationship between the field of view of a television camera and RAM, and Fig. 11 is a circuit diagram showing an example of an integrating circuit. FIG. 12 is a flowchart showing the procedure for writing processing. FIG. 12 is a flowchart showing the procedure for measuring moving speed.゛(1)...Television camera, (2)...Control circuit,
(3)...RAM. (4)...CPU, (5)...Multiplication circuit, (6)...
...Integrator circuit, (8) (9)...ROM. Patent Applicant: Tateishi Electric Co., Ltd. Figure 1 Figure 5 JXIJ Number

Claims (1)

[Claims] A memory that stores a spatial filter function of complex numbers corresponding to each pixel of a TV camera, a real number component and an imaginary number of the spatial filter function corresponding to each pixel from the memory in synchronization with the scanning of the TV camera. means for reading out each component; multiplication means for separately multiplying each component of the read spatial filter function by a video signal from a television camera; A moving object detection device comprising an integrating means for integrating, and an arithmetic means for calculating required moving object information from the results of these integrations. (2) the calculation means calculates the phase angle from the integration result of the real component and the imaginary component for one screen of the television camera 2;
The moving object detection device according to claim 1, wherein the speed and direction of movement of the moving object are determined from the change in the phase angle.