JPH0523560B2 - - Google Patents

Description

[Detailed description of the invention]

(Technical Field) The present invention relates to a colorimetric sensor device with a wide dynamic range. (Prior Art) Conventionally, it is known to obtain color temperature information using the output ratio of an R sensor that detects an R (red) color component and a B sensor that detects a B (blue) color component. However, in a colorimetric sensor device that obtains color temperature information in this way, if we try to perform processing such as logarithmic compression on the output of each color sensor using two signal processing circuits, not only will the circuit configuration become complicated, but Differences in logarithmic compression characteristics and temperature characteristics between the two circuits cause large discrepancies in color temperature information. Also, in order to accurately process the output of the color sensor mentioned above over a wide level range, it is possible to convert it into a digital signal and perform calculation processing, but in that case, the amount of light that can be handled will be greatly affected by the performance of the A/D converter. There were often broad restrictions. (Objective) It is an object of the present invention to provide a colorimetric sensor device that can overcome the drawbacks of the prior art. Another object of the present invention is to provide a colorimetric sensor device having a wide dynamic range. Another object of the present invention is to provide a colorimetric sensor device that can obtain accurate color temperature information with a simple configuration. (Examples) The present invention will be described below based on Examples. FIG. 1 is a diagram showing the principle of a colorimetric sensor device according to the present invention. 1 is a decoding section which detects a colorimetric sensor device according to the signal input to the select input terminal 2.
Control what is output from the CMS. Reference numeral 3 denotes a switch section which, under the control of the decoding section 1, decides whether to send the photocurrent from the colorimetric sensor 4 to the log amplifier section 5 or not. When not being sent, it is shot to remove excess electron-hole pairs.
Note that the colorimetric sensor 4 includes an R sensor that detects red color and a B sensor that detects blue color. The log amplifier unit 5 logarithmically compresses the photocurrent of the colorimetric sensor 4 and converts it into voltage. 6 is the final amplifier section, which inputs the output of the log amplifier section 5 to the next stage A/D.
The signal is amplified to match the input level of the converter 11. 7 is an I _REF control section that controls the magnitude of the reference current (by using an external resistor).
(adjusted by _R0 ), the offset of the amplifier section 6, etc. are canceled. Reference numeral 8 denotes a current shunting section which generates a current used to cancel temperature fluctuations in the voltage in the log amplifier section 5. 9 is an offset section as a signal source and offset adjustment means so that the R signal and B signal can be used within the specified input signal range of the A/D converter.
Forms a current to provide an offset voltage to Vout. Reference numeral 10 denotes a mode set input terminal for controlling the offset value of the offset section 9, into which a mode set signal from the system controller 13 is input. A select signal from the system controller 13 is also input to the select input terminal 2. 11 is an A/D converter for converting the output of the amplifier section 6 into a digital signal, and 12 is a memory for storing the output of this converter. Reference numeral 14 denotes an arithmetic circuit as an arithmetic means for forming color temperature information using the output of this memory. 15
is an image sensor that converts a subject image into an electrical signal. Reference numerals 16 and 17 are gain control amplifiers that adjust the mutual gains of the R (red) signal, G (green) signal, and B (blue) signals that are output from the image sensor 15; in this embodiment, the R signal and B signals, respectively. Further, the amplifiers 16 and 17 are connected to the arithmetic circuit 14, respectively.
The gain is controlled by the output of , and the level of each color signal is adjusted according to the color temperature of the subject. A process encoder 18 forms a standard television signal such as an NTSC signal based on the color signal. Further, LM is a level detection circuit serving as a level detection means, which discriminates in which level range the level of the output Vout of the final amplifier section 6 falls, and sends this discrimination signal to the system controller 13.
Enter. Next, the operation will be explained. Colorimetric sensor 4 consisting of R sensor and B sensor
The photocurrent that has flowed into the log amplifier section is selectively sent to or shunted by the decoder section and the switch section. The photocurrent sent to the log amplifier section is logarithmically compressed into voltage, amplified by the final amplifier,
Output. The above flow is called signal acquisition mode. Further, under the control of the decoding section, no current is caused to flow from the sensor to the log amplifier, but a constant current is caused to flow from the current shunting section to the log amplifier, and the output is performed in the same manner. This measures the ambient temperature, so this is set as the temperature measurement mode. Further, by controlling the mode terminal 10 by the system controller 13, the voltage of the final amplifier section is varied by several volts, thereby changing the relationship between the reference voltage of the A/D converter and the brightness of the light source. In this way, the outputs of the R sensor, B sensor, and current shunt section 8 are A/D converted in time series via the log amplifier section 5 and the final amplifier section 6 according to the select signal of the system controller 13, and then the memory 12 is memorized. As a result, the outputs of the R sensor, B sensor, and current shunting section are stored in the memory 12. The calculation circuit 14 calculates the output ratio of the R sensor and the B sensor excluding the temperature coefficient based on these stored values. The details are disclosed in Japanese Patent Application No. 58-203403 (Japanese Unexamined Patent Publication No. 60-93929) filed by the applicant, so the explanation will be limited here. Here, the output ratio of the R sensor and the B sensor corresponds to the color temperature on a one-to-one basis. FIG. 2 is a diagram showing an example of the configuration of the colorimetric sensor device CMS, in which the same reference numerals as in FIG. 1 indicate the same elements. 19 is an R sensor as a color sensor, and 20 is a B sensor as a color sensor. Also, 21
is an operational amplifier. In this embodiment, each input terminal A, B of the select input terminal 2 is outputted from the sequence controller 13.
When is (0, 0), the output of the B sensor is output from the output terminal Vout of the colorimetric sensor device via the log amplifier section 5 and the final amplifier section 6. Fork, (A, B)=
When (0, 1), the output of the R sensor is processed in the same way and is output from Vout, and when (A, B) = (1, 0), a constant current i is sent from the current shunt section 8 to the log amplifier section 5,
The signal is processed through the final amplifier section 6 and output from Vout. Also, when (A, B) = (1, 1), the current shunt section 8
A constant current of 16i ₁ is output from this, which is processed in the same way and output from Vout. Since the outputs of each sensor and constant current source are processed in a time-division manner in this way, the circuit configuration is simplified. Furthermore, in this embodiment, multiple stages of offset adjustment are performed in the final amplifier section 6, so that the dynamic range can be greatly expanded. That is, when the final amplifier section is configured as shown in Figure 3, if the potential at point A is considered as the sum of the signal component Vs and the reference voltage V _REF (Vs + V _REF ), Is = Vs / R ₁ , so the output Vout is It is expressed as Vout=Vs(R ₁ +R ₂ )/R ₁ +V _REF . Therefore, in the configuration shown in Figure 3, when the A/D converter 11 performs, for example, LSB 13 mV, 8-bit A/D conversion, the input level range is 0 to 3.328 V, so considering the saturation of the amplifier, the Vout The effective level range is about 0.3 to 3V. On the other hand, Vs is about 17~ when the brightness of light is doubled.
If the increase is 18mV and the amplifier gain is 14.6,
Vout will increase by 0.26V. That is, within the input level range of the A/D converter.
In order to keep Vout within the range, only about 10 stops of brightness can be tolerated. However, this results in an extremely narrow dynamic range as a colorimetric sensor device. As a practical matter, it is not practical unless it can process light in the range of Ev=0 to 23. Therefore, in the present invention, when Vout is outside the range of 0.3 to 3V, which is the allowable input of the A/D converter,
0.3 to 3V by applying an offset voltage to Vout
It is configured to return within That is, when configured as shown in Figures 2 and 4, Is = (Vs + V _REF - V _REF ) / R ₁ = Vs / R ₁ On the other hand, Is can be expressed as Is = Is' - I ₀ , so Is' = Vs It can be expressed as /R ₁ +I ₀ . Therefore, it can be expressed as Vout=Vs+V _REF +Vs/R ₁・R ₂ +I ₀・R ₂ =R ₁ +R ₂ /R ₁・Vs+V _REF +I ₀・R ₂ . Therefore, by changing the magnitude of I ₀ , Vout
It is also possible to adjust the offset of For example, as shown in Figure 5, when I ₀ · R ₂ = 0,
Although it can only handle light intensity from Ev5 to Ev10, I ₀・R ₂ =
When the voltage is 1.5V, it is possible to handle a light amount of Ev2.5 to Ev7.5, and when I ₀ · R ₂ = 3V, it is possible to handle a light amount of Ev0 to Ev5. That is, by adding the offset _I0・_R2 , Ev0~
Becomes able to handle light intensity up to Ev10. In addition, in order to provide such an offset, a current mirror circuit is installed in the offset section as shown in the second figure.
CM1 to CM3 are provided, and the respective outputs I ₁ , I ₂ , I ₃ are
The ratio of I ₁ :I ₂ :I ₃ is 4:2:1. Here, I ₁ to _{I 3} each correspond to a reference signal. Further, the relationship is set such that I ₂ =Ic with respect to constant current Ic. Also, when a mode signal of 1 or 0 is input to each terminal C, D, E of the mode terminal 10, I ₁ to I ₃ are combined, and the value of I ₀ as an output value is changed to 0, 0.5Ic,
It can be changed to Ic, 1.5Ic, 2Ic, ......, 3.5Ic. Here, in the colorimetric sensor device of the present invention, the signals of the R sensor and the B sensor may be read as Vs, and the ratio thereof may be determined. In that case, it is necessary that the two types of signals are both in the same mode. This is because otherwise, the above-mentioned error will have an effect. That is, in the case where redundancy is not provided to the offset as shown in Fig. 6A, if the two types of signals are R and B, then
It is fine if R and B are in the same mode, as in 22 in Figure 6A, but in a certain mode, as in 23 in the same figure, R is in but B is in the same mode. mode, and in the lower mode,
Naturally, B is included, but if R is in the mode above it, calculating the ratio of R and B will result in a completely incorrect value. Therefore, in this embodiment, this drawback is prevented by providing an overlapping portion between each mode as shown in FIG. 6B. However, the difference between R and B is set to be smaller than the overlapping portion thereof. In this way, for example, even if the mode says that R is on but B is lower than its lower limit, both will be on when the mode is set. Incidentally, the relationship between the input of each terminal (C, D, E) and I ₀ is determined as shown in the table below.

[Table] For specific mode settings, follow the steps below to set the system controller 13 and level detection circuit LM.
This is done automatically. First, set the mode to the reference mode and check whether R is within that range. If R is greater than the upper limit (-5V), jump to mode. If it is less than the lower limit (-9V), jump to mode. If it becomes even weaker, compare its upper and lower limits with R and move in mode or mode. In this way, find the mode in which R is included. Once a mode in which R is found is found, check whether B is included. Here, if B is also in that mode, input a code signal from the system controller to the mode terminal 10 to give an offset corresponding to that mode, give a predetermined offset, and in this state, the R sensor The output, the B sensor output, and the output of the current shunt section are each sequentially processed using a select signal, A/D converted, and then stored in the memory 12, and then calculated on each other to form color temperature information. Here, if the current values of the outputs _I1 to _I3 of the current mirror circuits CM1 to CM3 have an error, the offset voltage also includes the error and cannot cover the target light brightness range. It is assumed that the difference between R and B is 0.5 EV steps. At this time, two problems occur. The first is the lower limit (mode), so that the target light brightness range can be covered.
The point is that a sufficient margin must be maintained. That is, it is assumed that there is an error of ±10% in the currents I ₁ to _{I 3} . When Ic = 100 μA, I ₁ = 200 μA → 220 to 180 μA I ₂ = 100 μA → 110 to 90 μA I ₃ = 50 μA → 55 to 45 μA. If the maximum error occurs in the mode, 350μA becomes 385μA or 315μA. In this case, the problem is 315μA, but the error is 35μA and R ₂
The lower limit EV of the mode is set in consideration of the applied error voltage.
You can solve this problem by finding the value. (If the error voltage is 1V, it corresponds to one EV step, so if the lower limit of the mode is set to EV1, it can be guaranteed up to EV2.) The second problem occurs in the following cases. Suppose we now know that the mode is set to R. However, if B is smaller than the lower limit of the mode, then in this case we jump to the mode. And in the mode, by creating an overlap,
Both R and B should be included. Now, at this time, if the mode becomes too high by 0.5V due to an error and the mode becomes low by 0.5V, the overlap will disappear. In other words, even if the error is subtracted, if the overlapping portion is not larger than the difference between R and B, a problem arises in that R and B are not covered in the same mode. In order to deal with this problem, the present embodiment does the following. That is, when calculating the difference (pitch) between each mode, the worst case is as follows.

[Table] In other words, the maximum error between each mode is, and
The current between and is 5μA each, but between and is 15μA each, and between and is 35μA. Moreover, such errors are unevenly distributed. In other words, the constant current for offset setting is
If I ₁ to _{I 3} are set at equal intervals, the maximum error between the modes will vary, and there is a possibility that a very large maximum error will occur. Therefore, the overlap between each mode must be increased accordingly, and as a result, the configuration and control of the constant current source for forming I ₀ becomes complicated. Therefore, in this embodiment, in order to reduce this error, the output of each reference signal source is set so that the above-mentioned error is averaged. This can be explained theoretically as follows. That is, in order to minimize the conversion error to the adjacent mode, the ratio of the output current of each signal source is set to 1:2:
Instead of 4:8, it is 1+α:2+β:4+γ:... The specific way to determine α, β, and γ is as follows. I ₁ = x ₁ , I ₂ = x ₂ , I ₃ = x ₃ , x ₁ = 4x, x ₂ = 2x,
x ₃ =x If x is considered to have an error of ±y%, the currents containing errors are x′ ₁ =4x×(1±y), x′ ₂ =2x×(1±y), x′ ₃
=x×(1±y), and the difference from the next mode is at maximum x×(1+y)=x+xy (mode, between) 2x×(1+y)−x×(1−y)=x+3xy
(mode, interval) 3x×(1+y)−{2x×(1+y)}=x+xy
(mode, between) 4x (1 + y) - 3x (1 - y) = x + 7xy
(Mode, between) The red line above is the error, and the mode,
The time is maximum. So to compensate for this shortcoming
Set x ₁ , x ₂ , x ₃ as below. X ₁ = 4x−Z, X ₂ = 2x+(Z−Z′), X ₃ = x+
As Z′, X ₁ +X ₂ +X ₃ = 7x, in the mode,
Make sure that the value that takes into account the error approaches the specified value. Now, assuming an error of ±y%, X' ₁ = (4x - Z) x (1 ± y), X' ₂ = (2x + (Z -
Z′))×(1±y), X′ ₃ =(x+Z′)×(1±y)
The difference from the next mode is (x+Z') x (1+y) = x+xy+Z'+Zy
(mode, interval) (2x+Z−Z′)×(1+y)−(x+Z′)×(1−
y)=x+3xy+Z−2Z′+Zy (mode, interval) (4x−Z)×(1+y)−{(2x+Z−Z′)+(x
+
Z')}+(1-y)=x+7xy-2Z
(mode, interval) Z when all these errors are equal,
All you have to do is find Z', so xy+Z'+Z'y=3xy+Z-2Z'+Zy=7xy-2Z ∴2xy-3Z'+Z+Zy-Z'y=0 4xy-3Z+2Z'-Zy=0 Z(1+y)-Z' (3+y)+2xy=0 -(3+y)Z+2Z'+4xy=0 ...(1) -(3+y)Z/2×(3+y)+Z×(1+y) +4xy×3+y/2+2xy=0 −(3+y) ^2/2・Z+(1+y) ・Z+2xy×(1+3+y)=0 Z×(−1−y+(3+y) ² /2=2xy(4+y) Z((3+y) ² −2−2y)/2)=2xy(4+y) Z (y ² + 4y + 7/2) = 2xy (4 + y) Z = 4xy × (4 + y) / y ² + 4y + 7 ... (2) Put this into formula (1) - (3 + y) × 4xy × (4 + y) / y ² +4
y+7+2Z′+4xy=0 2Z′=(3+y)(4+y)・4xy−4xy
・(y ² +4y+7)/y ² +4y+7 ∴Z'=2xy・(3y+5)/y ² +4y+7 ...(3) Therefore, for example, when x=50, y=0.1, Z=82/1.41=11.06 Z'= 7.152 X ₁ = 200 _{− 11.06} = 188.94 X ₂ = 100 + (11.06 − 7.152) _{= 103.908} Even if there is an error, it falls within the range of I ₁ = 189 ± 10% = 207.9 ~ 170.1 I ₂ = 104 ± 10% = 114.4 ~ 93.6 I ₃ = 57 ± 10% = 62.7 ~ 51.3, and the difference between each mode is the worst Even if

[Table] Therefore, the maximum error is 13.1 between modes and between modes, which is greatly reduced. Furthermore, according to this embodiment, the outputs of a plurality of color sensors are sequentially inputted to a common log amplifier section and a final amplifier section to obtain colorimetric information, so the log amplifier etc. are separately inputted to each color sensor. Compared to the case where
Since there is no variation in the characteristics of each system, not only accurate color temperature information can be obtained, but also the configuration can be simplified by making it common. Further, by providing an offset, a relatively inexpensive A/D converter with a narrow dynamic range can be used, and digital processing for forming highly accurate color temperature information is possible with a simple configuration. (Effects) As described above, according to the present invention, by giving an offset to the output of the color sensor, the output of each color sensor is kept within a predetermined level range. This makes it possible to convert into a digital signal by using the method, making it possible to obtain highly accurate color temperature information at low cost.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an example of an imaging device using a colorimetric sensor device according to the present invention, FIG. 2 is a diagram showing an example of the configuration of the colorimetric sensor device, and FIG. 3 is a diagram showing an example of the configuration of the final amplifier section. 4 is a diagram showing an example of the configuration of the final amplifier section according to the present invention, FIG. 5 is a diagram showing the principle of offset adjustment, FIG. 6A is a diagram showing an example of the offset adjustment method, and FIG. 1 is a diagram showing an example of an offset adjustment method according to the present invention. 9...offset unit as a signal source and offset adjustment means, CMS...colorimetric sensor device, _I1
~I ₃ ...Current as reference signal, 19,20...
R sensor, B sensor, LM as color sensors...
...Level detection circuit as level detection means, 14
...A calculation circuit as a calculation means.

Claims (1)

[Claims] 1. A photoelectric conversion means that converts incident light into an electrical signal; and a photoelectric conversion means that has a characteristic that varies at a predetermined ratio depending on the environmental temperature, and non-linearly converts the electrical signal obtained from the photoelectric conversion means. A plurality of substantially different constant current sources are supplied to the signal conversion means for obtaining a signal conversion output including a temperature fluctuation component corresponding to the above-mentioned environmental temperature fluctuation, and the signal conversion means having the above-mentioned characteristics. By processing the conversion output of each constant current source obtained,
a correction signal generating means for generating a correction signal having a temperature fluctuation component that fluctuates at approximately the same rate as the temperature fluctuation component; and an input range within a predetermined level range, and performs arithmetic processing on the signal conversion output and the correction signal. a correction means for removing the temperature fluctuation component in the signal conversion output by adjusting the signal conversion output and the correction signal at a constant level in accordance with the level of the electric signal so that the signal conversion output and the correction signal fall within the predetermined level range; A colorimetric sensor device comprising: level control means for performing offset control.