JP3439511B2 - Infrared imaging device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an infrared imaging device,
In particular, the present invention relates to an infrared image pickup device using an infrared charge transfer device (hereinafter sometimes abbreviated as IRCCD) as an image pickup device.

In recent years, the development of infrared image pickup devices to which the IR CCD is applied has been actively developed. However, this IR CCD has problems such as non-linearization of the output level and pixel saturation with respect to a strong infrared ray incident, and measures to prevent this are taken. It has been demanded.

[0003]

2. Description of the Related Art FIG. 6 shows a conventional infrared image pickup device. 1 is a detector including an IR CCD, 2 is an A / D converter for converting an analog output signal of the detector 1 into a digital signal, and 3 is a detector. It is a correction circuit for correcting the characteristic variation of each element of the digital output signal of the A / D converter 2.

Reference numeral 8 is a driver for amplifying and supplying the set charge injection rate signal and integration time signal to the detector 1.

FIG. 7 shows a configuration example (for one pixel) of the input circuit section of the infrared image pickup apparatus shown in FIG. 6, in which the output of the photodiode PD is the charge injection rate which constitutes the charge accumulation control section TG. The charge is accumulated in the charge accumulation unit SG via the setting gate G1 and the integration time setting gate G2 connected in series to the gate G1, and the accumulated charges are sequentially transferred and output to the outside.

Then, the set amounts of the gates G1 and G2 are manually set in advance to set the charge injection amount and the integration time, respectively.

The operation of the charge storage controller TG in the conventional infrared image pickup device will be described with reference to FIG.
In FIG. 7A, characteristic curves P1 to P3 show the voltage-current characteristics of the photodiode PD shown in FIG.
This shows that the amount of light incident on the IR CCD increases as the characteristic curve P1 goes to P3.

Characteristic curves 1 to 3 show voltage-current characteristics when the charge injection rate set in the gate G1 shown in FIGS. 6 and 7 is changed. This charge injection rate characteristic is I
The input circuit characteristics of the RCCD are shown, and characteristic curves P1 to
The intersection of P3 and the characteristic curve-corresponds to the charge injection amount.

The characteristic curve shown in FIG. 8 (a) shows a normal incident light amount, but when a strong infrared ray is incident, characteristic curves P1'-P3 'as shown in FIG. 8 (b). Becomes

In the conventional infrared image pickup device against such a strong incidence of infrared rays, the manual setting of the integration time in the integration time setting gate G2 shown in FIG. 7 is shortened as compared with FIG. 8A. An attempt is made to prevent the output saturation state of the pixel caused by the limitation of the charge storage capacity of the IRCCD.

[0011]

In the case of the normal incident light quantity shown in FIG. 8 (a), the charge injection quantity is in the state shown by the characteristic curve or in (however, the stability is more preferable). That is, the charge injection amount is manually set at the time when the characteristic curves P1 to P3 start rising. This is because if the charge injection amount or is selected, the output data of the conventional example shown in FIG. 6 is in a linearized state and the largest output range can be obtained.

However, the problem here is that I
In the case where the RCCD input circuit characteristic is the charge injection amount characteristic curve in FIG. 9A, the output shortages are x and x ′ are “0”, whereas in FIG. Regarding the quantity characteristic curve ', it becomes large as y and y', respectively.

It is considered that this is caused by a voltage drop due to the substrate resistance of the photodiode PD in the charge generation / injection process.

As described above, it is impossible to prevent the phenomenon that the charge injection amount is changed and the charge injection efficiency is decreased with the increase of the incident light amount, and thus the nonlinearity of the output level is increased. there were.

Therefore, according to the present invention, an infrared imaging in which an imaging output signal of a detector having an infrared charge transfer device as an imaging device is converted into a digital signal by an A / D converter and a variation between pixels is corrected by a correction circuit to output. In the device, the non-linearity which cannot be solved only by controlling the charge integration time is corrected and I
The purpose is to make the output signal of the RCCD proportional to the incident light.

[0016]

In the infrared imaging device according to the present invention (1) for achieving the above object, as shown in principle in FIG. A frame memory 4 for sequentially holding one frame, and a circuit 5 for judging a level based on a threshold value based on the data for one frame to determine whether or not the infrared ray incident amount is saturated.
And the pre-stored data for setting the charge injection rate so that the input / output characteristics of the infrared charge transfer element are linear are read out based on the output of the determination circuit 5 and the D / A converter 6
And a charge injection rate setting data storage section 8 to be given to the detector 1 via the driver 7.

In the present invention, the pre-stored data for setting the charge accumulation time for preventing the saturation of the infrared charge transfer device due to the incidence of strong infrared rays is read out based on the output of the judgment circuit 5 and the driver is read. An integration time setting data storage unit 9 provided to the detector 1 via 7 can be further provided.

Further, in the present invention, corresponding to the output data of the integration time setting data storage unit 9, prestored data for correcting the characteristic variation among the infrared charge transfer elements is read out and the correction circuit 3 is read. It is also possible to further include a correction data storage unit 10 for giving to.

The data reading from the charge injection rate setting data storage unit 8 may be controlled by the integration time setting data storage unit 9 instead of the saturation determination circuit 5.

Further, in the present invention (part 2), FIG.
As shown in principle in FIG. 3, the image pickup output signal of the detector 1 having the infrared charge transfer element as the image pickup element is converted into a digital signal by the A / D converter 2 and the correction data stored in the correction data storage section 10 ′ is stored in advance. In the infrared imaging device that corrects the variation between pixels by the correction circuit 3 with the use data and outputs
A frame memory 4'which sequentially holds one frame of data output from the / D converter 2 and whether or not the infrared incident amount is saturated by taking the difference between the one frame data and the current frame data Based on the output of the determination circuit 5, a saturation determination circuit 5 ′ for determining whether or not the charge injection rate is set so that the input / output characteristics of the infrared charge transfer element are linear is generated and D / It is provided with a charge injection rate setting data generation unit 8 ′ which is given to the detector 1 via the A converter 6 and the driver 7.

In the present invention, the decision circuit 5'is
Comparing the subtracter 11 that takes the difference between the data for one frame and the current frame data, the memory 12 that stores at least two threshold values, the output of the subtractor 11 and each threshold value of the memory 12 Circuit 1 for determining the infrared incident level
3 and an up / down counter 14 that counts up, counts down, or does not count according to the determination result.

Further, according to the present invention, the charge injection rate setting data generator 8'receives a frame synchronization signal and alternately generates "1" and "0", and an output of the flip-flop 15. It can also be configured with an adder 16 that adds the output of the determination circuit 5 '.

[0023]

(1) The present invention (No. 1): FIG. 1 In FIG. 1, the analog signal output from the detector 1 is converted into a digital signal in the A / D converter 2, and then the correction circuit 3 interpixels. Is corrected to form output data, but one frame of this output data is held in the memory 4.

Then, the data for one frame is compared with a threshold value in the saturation determination circuit 5 to determine the level, and output whether or not the incident amount of infrared rays is saturated.

The result of this determination is sent to the storage unit 8, and the storage unit 8 changes the setting data of the charge injection rate depending on whether or not the incident amount of infrared rays is saturated, and this data is changed by the D / A converter 6. After being converted into an analog signal, a charge injection amount is given to the detector 1 via the driver 7.

This will be described with reference to the characteristic curve of FIG. 8 described above. When the saturation determination circuit 5 shown in FIG. 1 determines that the pixels of the data held in the frame memory 4 are saturated as a whole. It is considered that the state shown in FIG. 8A has shifted to the state shown in FIG. 8B. Therefore, the saturation determination circuit 5 changes the charge injection amount with respect to the charge injection rate setting data storage unit 8. There must be.

In this case, if the IRCCD input circuit characteristic is the characteristic curve at the time of the normal incident light quantity as shown in FIG. 9A, the output shortages x and x'can be eliminated by changing the characteristic curve. You can do that, so you don't need to change from curve to curve.

However, when it is necessary to change the intensity of the incident light as shown in FIG.
Even if it moves from the characteristic curve'to ', the output becomes insufficient by the amount of z and z'shown in the figure.
Need to change to.

Therefore, from the saturation determination circuit 5 to the storage unit 8, the input / output characteristic of the IR CCD becomes linear, that is, the characteristic of the charge injection amount at the time when the characteristic curves P1 'to P3' just rise. Data for setting the charge injection rate so as to form a curve 'is read and given to the detector 1 via the D / A converter 6 and the driver 7.

In this way, it becomes possible to restore the nonlinearity of the output to the linearity due to the change in the photodiode characteristics caused by the input infrared ray amount.

In the above description, only the charge injection rate is changed based on the saturation of the pixel, but the integration time data preset in the storage unit 9 is changed based on the determination result of the saturation determination circuit 5, and this is also changed. Based on this, the detector 1 can be controlled to control the charge integration time in a shorter direction.

Further, the correction data used in the correction circuit 3 may be changed in accordance with the change of the integration time data set in the storage section. In this case, the correction stored in the storage section 10 is corrected. By selecting the use data corresponding to the integration time setting data, it is possible to perform correction with less variation between pixels.

(2) Present Invention ( No. 2): FIG. 2 In the present invention, the output data corrected by the correction circuit 3 is not used, but the data with variations between pixels before correction is used for the previous frame. And the current frame are given to the saturation judgment circuit 5 ′.

In the present invention, the charge injection rate setting data is changed by one frame by using the saturation judgment in the same saturation judgment circuit as that shown in FIG. 1 provided at regular intervals or separately. Next, the change saturation judgment circuit 5'of the image data before and after the change of the charge injection rate setting data is obtained.

In the saturation judgment circuit 5 ', the level of the current frame data is higher than that of the previous frame data, that is, the output shortage y, y', z, z as shown in FIG. When'is obtained as a difference, it can be determined that the infrared incident amount is saturated when this difference exceeds the threshold value, and the charge injection amount setting data generation unit 8'is determined based on this determination result. Control.

Therefore, the charge injection rate setting data generator 8 '
Generates data for setting the charge injection rate so that the input / output characteristic of the IRCCD becomes linear and supplies it to the detector 1 via the D / A converter 6 and the driver 7.

In this way, the output shortage y, shown in FIG.
It is possible to eliminate y ′, z, and z ′, and it is possible to linearize the nonlinearity of the output due to the change in the photodiode characteristics, as in the case of FIG.

[0038]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS As an embodiment of the infrared imaging device according to the present invention shown in FIG. 1, the charge injection rate setting data storage section 8 and the correction data storage section 10 are 64 Kbits and 2 bits respectively.
A ROM having an M-bit capacity is used, and the frame memory 4
For this, a RAM having a capacity of 4 Mbits is used. Also, an example of the number of parallel bits between each part is shown in parentheses.

Further, FIG. 3 shows an embodiment of the saturation judgment circuit shown in FIG. 1, and 51 is a ROM storing a threshold value.
The ROM 51 has a 17-bit output, and 12 bits of the ROM 51 are sent to the comparator 52, are compared with the input data (12 bits), and the comparison result is accumulated in the counter 53.

Cumulative result of this counter 53 (5 bits)
Is compared with a threshold value for 5 bits of the 17-bit output of the ROM 51, and the comparison result is accumulated in the up / down counter 55.

This will be explained by taking an example. First, in the comparator 52, the level of the input data, that is, the output data (12 bits) from the frame memory 4 is the ROM 5.
The number of pixels having a threshold value set by 12 bits from 1 to, for example, "3000" or more is output and accumulated in the counter 53.

The accumulated value is also stored in the ROM 51.
Threshold value set by the output 5 bits of, for example, "10
The comparator 54 compares whether or not it is equal to or greater than "0".

The output of the counter 53 is "100".
When it is determined by the comparator 54 that the above is the case, the up / down counter 55 is counted up by "1" to obtain the output data. Then, this operation is repeated for one frame.

Similarly, the number of pieces of data equal to or less than the threshold value set by the ROM 51 is detected, and if the number is equal to or larger than a predetermined value, the count is counted down by 1 count.

That is, the highest level of one frame is the threshold "1".
When it is 500 or less, the up / down counter 55 is down-counted when the number is 100 or more, and the integration time setting data, the charge injection rate setting data, and the correction data correspond to this. To change.

In the case of the charge injection rate setting data, the characteristic curve shown in FIG. 8 is shifted from "or" to "," and the integration time setting data is operated in the direction of shortening by one step. Then, the data is switched to the correction data due to the reduction of the integration time by one step. In this case, the integration time setting data can be adjusted in 8 steps.

When the number of pixels having a threshold level of "3000" or more in one frame is "100" or less and the maximum level of pixels is the threshold value "1500" or more, the integration time setting data and the charge injection rate setting are normally set. The data and the correction data may be left as they are.

FIG. 4 shows an embodiment of the input level judgment circuit 13 in the saturation judgment circuit 5'shown in FIG.
In this embodiment, the output data D1 of the subtractor 11 is commonly input and the threshold value D2 read from the ROM 12 is input.
Two comparators 131 and 132 which respectively input D3
And counters 133 and 134 that count the outputs of these comparators 131 and 132, and these counters 13
It is composed of two comparators 135 and 136 which compare the outputs of the circuits 3,134 and the other thresholds D4 and D5 of the ROM 12 and send the outputs to the up / down counter 14.

In the operation of this input level judging device, first, in the comparator 131, the output D of the subtractor 11 is outputted.
The counter 133 counts up only when 1 exceeds the threshold value D2 of the ROM 12, and the counter 133
The comparator 135 outputs a logic level "1" when the output exceeds the other threshold D4 of the ROM 12, and otherwise outputs "0".

Similarly, in the comparator 132, the counter 134 counts up only when the output data D1 from the subtractor 11 is smaller than the threshold value D3 set by the ROM 12, and the cumulative value of the counter 134 is RO.
The output of comparator 136 will be a logical "1" only if it is greater than another threshold D5 set by M12, otherwise a logical "0" output will be generated and provided to up / down counter 14.

The operation of this input level determination circuit is similar to that of the embodiment of the saturation determination circuit shown in FIG.
The threshold value D2 corresponds to the threshold value “3000” in FIG. 3, and the threshold value D3 corresponds to the threshold value “1500”. Further, it corresponds to the threshold value D4 = D5 = threshold value “100”.

Therefore, the values of x, x ', y, y', z, z'in FIG.
If the data D1> D2 is determined by the thresholds D2 to D5 set to 12, the process proceeds in the direction of the charge injection rate curve →→ in FIG. 8A, and conversely the data D1.
<D3, the process proceeds to the curve →→ or “→” → ”in FIG. Furthermore, the data D3 <
When it is determined that D1 <D2, no change is made.

FIG. 5 is a diagram for explaining the operation of the correction circuit shown in FIGS. 1 and 2. Now, the input / output characteristics of the pixel i output from the A / D converter 2 are obtained by observation. The straight line A is obtained by correcting the two data IL and IH.

Further, the average data obtained from one frame of the above data IL and IH is ML and MH as shown in the figure.
Then, a straight line B is obtained by these two two-point linear approximations.
Is obtained.

It is the role of the correction circuit 3 to make the straight line A coincide with the straight line B.

Therefore, first, the slopes of the straight lines A and B are obtained by the following equation.

[0057]

[Equation 1]

Then, the gain (ratio) between the slope of the straight line A and the slope of the straight line B is calculated by the following equation.

[0059]

[Equation 2]

When the gain thus obtained is applied to the straight line A, a straight line C is obtained.

If an offset amount of (ML-IL) is added to this straight line C, it will coincide with the straight line B.

Now, when the output of the pixel i is K ₀ , the inclination is corrected as described above to obtain the following equation.

[0063]

[Equation 3]

When the offset amount is corrected for this, the following equation is obtained.

[0065]

[Equation 4]

When this equation is modified, it becomes the following equation.

[0067]

[Equation 5]

As a result, K ₀ can be corrected to K ₂ .

Here, the correction can be performed by knowing the gain and the offset.

Therefore, the correction data storage unit 10 'stores the causes and offsets of all the pixels, and the correction circuit 3 calculates based on this data.

The correction circuit 3 and the correction data 1
Although 0'is known in the related art, data for correcting this correction data is changed in advance according to the output data of the integration time setting data storage unit 9 (or the determination result of the saturation determination circuit 5). Then, the correction data storage unit 10 shown in FIG. 1 can be obtained.

[0072]

As described above, according to the infrared image pickup device of the present invention, one frame of data from the detector is sequentially held, and whether the infrared incident amount is saturated from this one frame of data. Whether or not the level is judged based on the threshold value, and the charge injection rate is set in advance so that the input / output characteristic of the IRCCD is linear is given to the read detector based on the above level judgment result. Moreover, since the level determination result is configured to be changed also in the data for setting the charge integration time and the data for the correction circuit, it is possible to obtain the effect of canceling the change in the photodiode characteristics, and to obtain the effect with respect to the input level. The linearity of the output level can be improved.

Further, due to such improvements, it is possible to expect improvement in processing performance such as correction of pixel variation, target identification and tracking.

[Brief description of drawings]

FIG. 1 is a block diagram showing a principle configuration (No. 1) of an infrared imaging device according to the present invention.

FIG. 2 is a block diagram showing a basic configuration (No. 2) of the infrared imaging device according to the present invention.

FIG. 3 is a block diagram showing an embodiment of a saturation determination circuit shown in FIG.

FIG. 4 is a block diagram showing an embodiment of the input level determination circuit shown in FIG.

FIG. 5 is a graph diagram for explaining the operation of the correction circuit.

FIG. 6 is a block diagram showing a conventional infrared imaging device.

FIG. 7 is a block diagram showing a configuration example (for one pixel) of an input circuit unit of the infrared imaging device.

FIG. 8 is a characteristic graph of an infrared charge transfer device.

[Explanation of symbols]

1 detector 2 A / D converter 3 Correction circuit 4,4 'frame memory 5,5 'saturation determination circuit 6 D / A converter 7 driver 8 Charge injection rate setting data storage section 9 Integration time setting data storage 10, 10 'Correction data storage unit 11 Subtractor 12 ROM 13 Input level judgment circuit 14 up / down counter 15 D-flip-flop 16 adder In the drawings, the same reference numerals indicate the same or corresponding parts.

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Hiroki Shimozenma 1015 Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture, Fujitsu Limited (72) Inventor, Masaki Kamata 1015, Kamitadanaka, Nakahara-ku, Kawasaki, Kanagawa Prefecture, Fujitsu Limited (56) References Japanese Unexamined Patent Publication No. 5-328222 (JP, A) Japanese Unexamined Patent Publication No. 4-156189 (JP, A) Japanese Unexamined Patent Publication No. 4-355576 (JP, A) Kaihei 4-72779 (JP, U) (58) Fields investigated (Int.Cl. ⁷ , DB name) H04N 5/30-5/335 G01J 1/00-1/60, 11/00

Claims (7)

(57) [Claims] 1. An A / D converter (2) converts an imaging output signal of a detector (1) having an infrared charge transfer device as an imaging device into a digital signal, and a correction circuit (3) corrects variations between pixels. In the infrared imaging device that outputs the data, the frame memory (4) that sequentially holds one frame of the data output from the correction circuit (3) and whether the infrared incident amount is saturated from the one frame of data. A circuit (5) for judging the level based on a threshold value, and pre-stored data for setting the charge injection rate so that the input / output characteristics of the infrared charge transfer element are linear are stored in the judgment circuit (5). An infrared ray, comprising: a charge injection rate setting data storage unit (8) to be supplied to the detector (1) through a read D / A converter (6) and a driver (7) based on an output. Imaging device. 2. A driver (7) which reads pre-stored data for setting a charge storage time for preventing saturation of the infrared charge transfer device due to the incidence of strong infrared rays based on the output of the judgment circuit (5). ) Through the detector (1)
The infrared imaging device according to claim 1, further comprising an integration time setting data storage section (9) provided to the. 3. The correction circuit (3) for reading the pre-stored data for correcting the characteristic variation among the infrared charge transfer elements corresponding to the output data of the integration time setting data storage section (9). 3. The infrared image pickup device according to claim 2 , further comprising a correction data storage unit (10) provided to (1). 4. The charge injection rate setting data storage section (8)
But claim 2 or 3 characterized in that reading the data stored beforehand corresponding to the output data of the saturation determining circuit the integrating time setting data storage unit instead of the output of the (5) (9)
Infrared imaging apparatus according to. 5. An imaging output signal of a detector (1) having an infrared charge transfer device as an imaging device is converted into a digital signal by an A / D converter (2) and stored in a correction data storage unit (10 ′) in advance. In an infrared imaging device that corrects variations between pixels in a correction circuit (3) using the corrected data and outputs the frame, a frame that sequentially holds one frame of data output from the A / D converter (2) A memory (4 '), a saturation determination circuit (5') for determining the level of whether or not the infrared ray incident amount is saturated by taking the difference between the data for one frame and the current frame data; Data for setting the charge injection rate so that the input / output characteristic of the transfer element becomes linear is generated based on the output of the judgment circuit (5), and is generated via the D / A converter (6) and the driver (7). And give it to the detector (1) An infrared image pickup device comprising: a charge injection rate setting data generation unit (8 '); 6. The subtraction circuit (5) for determining the difference between the one frame data and the current frame data, the determination circuit (5 ′), a memory (12) storing at least two threshold values, The output of the subtractor (11) and the memory (1
A circuit (13) for judging the infrared incident level by comparing each threshold value of 2), and counting up by the judgment result,
The infrared image pickup device according to claim 5, wherein the infrared image pickup device comprises a down-counting or non-counting up / down counter (14). 7. The charge injection rate setting data generator (8 ')
Is a flip-flop (15) that alternately generates "1" and "0" in response to a frame synchronization signal, and an addition that adds the output of the flip-flop (15) and the output of the determination circuit (5 '). The infrared image pickup device according to claim 5, wherein the infrared image pickup device is configured with a device (16).