JP3518550B2 - Image reading 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 image reading apparatus for reading an image on a color film or the like to obtain image data.

[0002]

2. Description of the Related Art An image reading apparatus such as an image scanner or a film scanner reads an image printed on paper or the like to obtain image data, or reads an image on a color film to obtain image data. be able to.

Therefore, in this type of apparatus, generally, a target image is picked up by a line sensor at an appropriate scanning speed, the picked-up image signal is temporarily stored in a frame memory, and the stored signal is predetermined. The target video signal of, for example, the NTSC system is read out at the synchronous speed.

[0004]

According to the present invention, in such a device, 1. The optimum reading condition can be set for the image to be read. 2. Quantization error is not noticeable in the displayed image even after gradation correction. 3. The circuit scale is small and the cost is low. Is to realize.

[0005]

In the first invention,
An image reading apparatus that obtains image data of the image to be read by scanning the image to be read twice, that is, a prescan and a main scan, and a photoelectric conversion element that photoelectrically converts incident light from the image to be read, and the photoelectric conversion element. Control means for controlling the level of the output signal of the conversion element, gradation correction means for gradation correction of the output signal of the photoelectric conversion element, and conversion of the output signal gradation-corrected by the gradation correction means into a digital signal. The gradation correction means is configured to be able to change its correction characteristic, sets the correction characteristic to a linear input / output characteristic at the time of the prescan, and sets the correction characteristic to a linear input / output characteristic at the time of the main scan. The control means obtains during the prescan so that the level of the output signal of the photoelectric conversion means matches the dynamic range of the A / D converter. The based on the A / D converter output signal in which the image reading apparatus adapted to control the level of the output signal of the photoelectric conversion means.

[0006]

By the first scanning, the conditions for loading the image pickup signal S11 of the image to be read 1 into the memory 15R are set, and the second scanning is performed according to the set conditions, and the image pickup signal S11 is stored in the memory. The image data is officially taken into 15R and the target image data is output.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In FIG. 1, reference numeral 1 denotes an image to be read,
In this example, a 35 mm positive color film (reversal film) is used for simplicity. Further, reference numeral 20 denotes a microcomputer (CPU system) which controls the operation or processing of each part of this apparatus.

Then, the light from the light source 2 is applied to the film 1 through the color separation filter 3, and the image of the film 1 is formed on the line sensor 11 by the lens 4.

In this case, the film 1 is adapted to move at a predetermined speed in the sub-scanning direction (horizontal direction) by the drive motor 21 during image pickup. Further, as the light source 2, a three-wavelength fluorescent lamp or a halogen lamp can be used. Furthermore, the color separation filter 3
A filter section that respectively transmits the three primary color lights, that is,
It has a filter section R for transmitting red light, a filter section G for transmitting green light, and a filter section B for transmitting blue light, and is rotated by a drive motor 23 so that the filter sections RB are light sources. 2 and the film 1 are arranged in order on the optical path. The rotations of the motors 21 and 23 are controlled by the microcomputer 20.

In this example, the line sensor 11 is a CCD line sensor having a VOFD (vertical overflow drain) structure, and the charge sweep-out pulse P20 controls the charge accumulation time in each pixel. Is. Therefore, the pulse P20 controls the shutter time (shutter speed) of the line sensor 11 to control the signal level of the output signal S11 of the line sensor 11.

Further, the line sensor 11 is provided at the image forming position of the lens 4 so that the scanning direction thereof is the main scanning direction (vertical direction), and the microcomputer 20 reads out the main scanning clock and the charge sweep pulse P20. Is supplied.

Further, 12 is a gamma correction amplifier, 13 is A
/ D converters, 14R to 14B are frame memories. In this case, the gamma correction amplifier 12 is, for example, as shown in FIG.
.About.C have three kinds of correction characteristics .gamma.1 to .gamma.3. That is, as shown in FIG. 2A, the first correction characteristic γ1 has a smaller gain near the rising edge than the theoretical characteristic of γ = 0.45 and exhibits a knee characteristic from a relatively low input level. Used when Image 1 is a positive film.

As shown in FIG. 2B, the second correction characteristic γ 2 also has a small gain near the rising edge as compared with the theoretical characteristic of γ = 0.45, but after the rising edge, the gain is made larger than the theoretical characteristic. ing. This correction characteristic γ2 is mainly used when the image to be read 1 is a negative film.

Further, the third correction characteristic γ3 is set to γ3 = 1, that is, a linear input / output characteristic, as shown in FIG. 2C. Then, these gamma correction characteristics γ 1 ~
γ3 is selected by a control signal from the microcomputer 20. Note that such a gamma correction amplifier 12 is commercialized as a one-chip IC.

The A / D converter 13 is, for example, 8
The number of quantized bits is 1 bit / sample, and the dynamic range of the A / D converter 13 is, for example, 10 of the level of the image signal S11 that is not gamma-corrected.
Set to be equal to 0 %.

Further, 15R to 15B are lookup tables, 16R to 16B are D / A converters, and 17 is N.
3 illustrates a TSC encoder. In this case, the look-up tables 15R to 15B correspond to the device to which the final video signal is supplied, that is, when the video signal is supplied to the CRT display and to the printer according to the instruction of the microcomputer 20. It is for correcting the gradation and hue of the video signal in correspondence with the time when the video signal is supplied.

18 is a digital interface,
Reference numeral 19 represents an image processing circuit. The image processing circuit 19
The frame memory 14R based on the instruction from the microcomputer 20
14B write / read control, arithmetic processing for a video signal (imaging signal), and the like are performed.

When a video signal of the image on the film 1 is to be obtained, prescan is performed according to an instruction from the microcomputer 20. According to the result of, set the conditions for the following. Perform a main scan. Are executed in order.

That is, the motor 20 is operated by the microcomputer 20.
Is controlled, and the red color filter portion R of the color separation filter 3 is positioned on the optical path between the light source 2 and the film 1. Subsequently, the motor 21 is controlled by the microcomputer 20 and the film 1 is transported in the sub-scanning direction at a predetermined speed for the target frame.

Further, the charge accumulation time of the line sensor 11 is set so that the signal level of the image pickup signal S11 output from the line sensor 11 is 1/2 of the signal level when a standard image is picked up. Is done (this is microcomputer 2
It is set by the charge sweep pulse P20 supplied from 0 to the line sensor 11).

Therefore, the line sensor 11 takes out the image pickup signal S11 of the red component of the target frame of the film 1 at a speed corresponding to the moving speed of the film 1 and the frequency of the read clock.

The red component signal S11 is supplied to the gamma correction amplifier 12, and at this time, the correction amplifier 12 is set to the correction characteristic γ3 (linear characteristic) by the microcomputer 20, and the signal S11 is A / A as it is. It is supplied to the D converter 13 and A / D converted into a digital red signal S3R, and this signal S3R is written in the frame memory 14R. The write address at this time is designated by the image processing circuit 19 and is a write address corresponding to the scanning position of the line sensor 11.

In this way, the target frame of the film 1 is imaged by the line sensor 11, and the red signal S3R for one frame is stored in the frame memory 14R.

Next, the motor 23 drives the color separation filter 3
The green filter part G is located on the optical path between the light source 2 and the film 1, and the film 21 is transferred by the motor 21 at a predetermined speed in the sub-scanning direction to the target frame. Furthermore, the accumulation time of the line sensor 11 and the correction characteristics of the gamma correction amplifier 12 are determined by the red signal S3R.
It is set in the same manner as in the case of.

Therefore, the image sensor signal S11 of the green component of the target frame of the film 1 is taken out from the line sensor 11, this signal S11 is A / D converted into the digital green signal S3G by the A / D converter 13, and this Signal S3G
Are written in the frame memory 14G. Thus, the green signal S3G for one frame of the target frame of the film 1
Are stored in the frame memory 14G.

Subsequently, the motor 23 positions the blue filter portion B of the color separation filter 3 on the optical path between the light source 2 and the film 1, and the motor 21 sets the film 1 to a predetermined target frame. It is transported in the sub-scanning direction at a speed of. Furthermore, the accumulation time of the line sensor 11 and the correction characteristics of the gamma correction amplifier 12 are determined by the red signal S
It is set as in the case of 3R.

[0027] Thus, from the line sensor 11, the imaging signal S11 of the blue component of the frame of interest of the film 1 is taken out, the signal S11 is A / D converted into a digital blue signal S3B by the A / D converter 13, the Signal S3B
Is written in the frame memory 14B. Thus, the blue signal S3B for one frame of the target frame of the film 1
Is stored in the frame memory 14B.

Thus, the target frame of the film 1 is
The color separation filter 3 and the line sensor 11 capture images in a frame sequential manner, and the digital color signals S3R to S3B are stored in the frame memories 14R to 14B, respectively. In addition,
As described above, no gamma correction is performed on any of the signals S3R to S3B.

As a result of the above, the prescan is completed, and the target film 1 is stored in the frame memories 14R to 14B.
That is, the color signals S3R to S3B of the frame are stored as data by prescan.

Next, the average value of the color signals S3R to S3B obtained by this prescan is obtained. That is, the image processing circuit 19, frame memory 14R~14B color signals S3R~S3B is read, Y AV = 0.30E3R + 0.59E3G + 0.11E3B E3R: Average signal S3R level E3G: average level of the signal S3G E3B: signal The average value YAV of one frame of the luminance signal represented by the average level of S3B is calculated.

In this case, since the A / D converter 13 quantizes 8 bits / 1 sample, its dynamic range is "0" to "255" and its median value is "128". is there.

[0032] At the time of pre-scanning, so that the average value YAV when imaging the standard specific image becomes 64, charge storage time of the line sensor 11 is set. Also, the dynamic range of the A / D converter 13 is 100 % of the image pickup signal S11 at the time of main scanning of a standard image.
Is equal to.

Furthermore, the main scan of this standard image
The average value of the brightness signal when the brightness is on is 128. Rye at this time
The charge accumulation time of the sensor 11 is 128 during prescan.
/ 64 times.

[0034] Therefore, when the film 1 of the pre-scanned <br/> was signals S3R～S3B, when the signal S3R～S3B at the time of prescanning 128 / YAV multiplies, a signal level of the 100% A / D The dynamic range of the converter 13 can be almost covered.

Therefore, during the main scan, the charge accumulation time of the line sensor 11 is 128 /
It is set to YAV times. Also, in this case, since the image to be read 1 is a positive color film, the gamma correction amplifier 1
2 is set to the correction characteristic γ1.

Then, after that, the main scan is performed,
The target frames of the film 1 are frame-sequentially imaged in the same manner as in the prescan, and the digital color signals S3R to S3B
Are stored in the frame memories 14R to 14B, respectively. However, in this case, unlike the pre-scan,
The levels of the signals S3R to S3B stored in the memories 14R to 14B have the maximum value "255" when the level is almost 100 %. Gamma correction is also performed.

Thus, when the main scan is performed and the signals S3R to S3B are stored in the memories 14R to 14B,
By repeatedly reading out the signals S3R to S3B of the memories 14R to 14B in a predetermined order, the digital video signals S4R to S4B having the target synchronous speed are taken out, and the signals S4R to S4B are supplied to the lookup tables 15R to 15B. Video signals S4R to S4B that have been corrected for hue and the like
It is said that Then, the signals S4R to S4B are supplied to the D / A converters 16R to 16B to generate the analog video signal S.
The signals S4R to S4B are D / A converted, and the signals S4R to S4B are taken out as output signals.

The D / A converted signals S4R to S4
4B is supplied to the NTSC encoder 17 to extract a color composite signal, or a luminance signal and a carrier color signal. Furthermore, look-up tables 15R to 15B
Signals S4R to S4B from the digital interface 1
It is taken out through 8 as digital image data.

In this case, the output signal S4R ~
The signal level of S4B is increased based on the average value of the signals S3R to S3B at the time of prescan, and the dynamic range of the A / D converter 13 is fully used, so that the quantization error can be reduced. .

In the above description, the main scan is performed using the average value of the color signals S3R to S3B obtained by the prescan, but the peak value of the signals S3R to S3B can also be used.

That is, after the prescan, the image processing circuit 19 causes the color signals S of the frame memories 14R to 14B to be obtained.
The peak values EPR, EPG, EPB of the signals S3R, S3G, S3B are respectively obtained by reading out 3R-S3B or using the color signals S3R-S3B at that time as they are at the time of prescan.

After that, the main scan is performed,
The target frames of the film 1 are imaged frame-sequentially in the same manner as during prescan, and in this case, the charge accumulation time of the line sensor 11 is 255 / EPR times that during prescan when reading the red component. Is set to 255 when reading the green component / EPG times the prescan, and 255 when reading the blue component when prescanning.
It is set to / EPB times.

In this way, the main scan is performed, the signals S3R to S3B are stored in the memories 14R to 14B, and thereafter, the analog video signals S4R to S4B and the like are taken out as described above.

Also in this example, the output signals S4R to S4B are increased in signal level on the basis of the peak values of the signals S3R to S3B at the time of pre-scan, and the dynamic range of the A / D converter 13 is made full. , The quantization error can be reduced.

Further, the main scan can be performed using the histogram of the signals S3R to S3B.

That is, after the prescan, the color signal S3R of the frame memory 14R is read by the image processing circuit 19, and the signal S3R is read as shown in FIG.
A histogram of the signal level E3R and its appearance frequency is obtained.

The appearance frequency of the signal S3R is 0.
A certain frequency CTH in the vicinity is set as a threshold, and the level at the point where the signal level is the minimum or higher than the threshold CTH is set as the level ELR and the level at the point where the signal level is the maximum is set as the level EHR.

A similar histogram and a similar level are obtained for the signals S3G and S3B.

Then, after that, the main scan is performed,
In the same way as during pre-scan, the target frames of film 1 are imaged frame-sequentially, and when reading the red component, the charge accumulation time of line sensor 11 is set to 255 / EHR times that during pre-scan. To be done.

Further, the signal S by this main scan
When 3R is written in the memory 14R, ECR = {E3R- (ELR.255 / EHR) ** γ} × 255 / {255- (ELR ・ 255 / EHR) ** γ} where ECR: signal after correction The S3R level γ = γ1 or γ2 ** indicates a power. The level correction indicated by is performed, and the signal S3R subjected to the level correction is written in the memory 14R.

Also for the signals S3G and S3B, the charge accumulation time of the line sensor 11 is similarly corrected, and the level correction similar to the above equation is performed, and the level-corrected signals S3G and S3B are also corrected. Are written in the memories 14G and 14B.

In this way, the main scan is performed, the signals S3R to S3B are stored in the memories 14R to 14B, and thereafter, the analog video signals S4R to S4B and the like are taken out as described above.

Also in this example, the signal levels of the output signals S4R to S4B are corrected based on the histogram of the signals S3R to S3B at the time of pre-scan, and the dynamic range of the A / D converter 13 is made full. Since it is used, the quantization error can be reduced.

Furthermore, according to this example, since the signal level having a very low appearance frequency is ignored by the threshold CTH, even if there is a sudden noise, it is not affected.

When the image to be read 1 is an orange-based negative film, the histogram thereof is as shown in FIG. 4, for example, and an orange-based peak is generated at a portion where the signal level is high. Therefore, in this case, ignoring this orange-based peak,
The correction may be performed in the same manner as the case.

In the above description, the color separation filter 3
Is a filter unit R that transmits red light, green light, and blue light
.About.B, but may have a filter portion that transmits cyan light, magenta light, and yellow light. Further, when a filter unit that transmits white light in addition to the three primary color lights is provided and the image to be read 1 is read in the monochrome mode, the use efficiency of light is improved by using this filter unit, so that high-speed reading is possible. it can.

Further, the reading density at the prescan can be roughened to shorten the time required for the prescan. In addition, the image obtained by the prescan,
It can also be used as an image for setting the trimming position. Further, the image processing circuit 19 can also be used for shading correction of the line sensor 11.

Further, a variable gain amplifier is provided between the line sensor 11 and the gamma correction amplifier 12, and the gain of this variable gain amplifier is changed in the same manner as the accumulation time of the line sensor 11 and the A / D converter 13 is changed. It is also possible to change the level of the supplied image pickup signal S11.

Furthermore, when a fluorescent lamp is used as the light source 2, a heater for keeping the lamp warm can be attached in order to prevent a decrease in brightness at low temperatures. Further, an infrared cut filter may be arranged on the optical path between the light source 2 and the line sensor 11 in order to prevent the influence of infrared rays from the light source 2 or the heater for keeping the temperature thereof.

Further, in the above description, the target frame of the film 1 is imaged in the frame-sequential manner by the color separation filter 3 and the line sensor 11, but it may be imaged in the line-sequential manner.

[0061]

According to the present invention, by performing the prescan, the dynamic range of the image pickup signal S11 at the time of the main scan is set so as to fully use the dynamic range of the A / D converter 13. A / D
An 8-bit converter 13 is sufficient, and the capacities of the memories 14R to 14B can be reduced. Therefore, the circuit scale can be reduced and
Can be low cost.

Further, since the gamma characteristic of the gamma correction amplifier 12 can be made linear during the pre-scan, it is possible to measure the physical quantity such as the average value, the peak value and the histogram, and the optimum reading for the main scan. Conditions can be set.

[Brief description of drawings]

FIG. 1 is a system diagram showing an example of the present invention.

FIG. 2 is an explanatory diagram of gamma characteristics.

FIG. 3 is a diagram showing an example of a histogram.

FIG. 4 is a diagram showing another example of a histogram.

[Explanation of symbols]

1 Read image 2 light sources 3 color separation filter 4 lenses 11 CCD line sensor 12 Gamma correction amplifier 13 A / D converter 14R-14B frame memory 15R-15B Look-up table 16R-16B D / A converter 17 NTSC converter 18 digital interface 19 Image processing circuit 20 microcomputer 21, 23 Drive motor

Claims (7)

(57) [Claims] 1. An image reading apparatus for scanning an image to be read twice, that is, a prescan and a main scan, to obtain image data of the image to be read, which photoelectrically converts incident light from the image to be read. An element, a control means for controlling the level of the output signal of the photoelectric conversion element, a gradation correction means for correcting the gradation of the output signal of the photoelectric conversion element, and an output signal gradation-corrected by the gradation correction means And an A / D converter for converting the digital signal into a digital signal, the gradation correction unit is configured to change its correction characteristic, and sets the correction characteristic to a linear input / output characteristic during the prescan. During the main scan, the control means sets the output signal of the photoelectric conversion means to match the dynamic range of the A / D converter. An image reading apparatus for controlling the level of the output signal of the photoelectric conversion means based on the output signal of the A / D converter obtained during rescanning. 2. An image reading device which obtains image data of the image to be read by scanning the image to be read twice, that is, pre-scanning and main scanning, wherein photoelectric conversion is performed on incident light from the image to be read. An element, control means for controlling the level of the output signal of the photoelectric conversion element, gamma correction means for gamma correction of the output signal of the photoelectric conversion element, and a digital signal for the output signal gamma-corrected by the gamma correction means. The gamma correction means is configured to be able to change the gamma correction coefficient of the correction characteristic, and the gamma correction coefficient is set to 1 during the pre-scan, and the main scan is performed. At times, the level of the output signal of the photoelectric conversion means is adjusted to match the dynamic range of the A / D converter. An image reading apparatus in which the control means controls the level of the output signal of the photoelectric conversion means based on the output signal of the A / D converter obtained during the prescan. 3. The image reading apparatus according to claim 1, wherein the photoelectric conversion element is of a charge storage type, and the control means controls the charge storage time of the photoelectric conversion element. An image reading device adapted to control the level of an output signal of the photoelectric conversion means. 4. The image reading apparatus according to claim 1 or 2, wherein the control means is based on an average value of output signals of the A / D converter obtained during the prescan during the main scan. An image reading device adapted to control the level of an output signal of the photoelectric conversion element. 5. The image reading apparatus according to claim 1 or 2, wherein the control means is based on a peak value of an output signal of the A / D converter obtained during the pre-scan during the main scan. An image reading device adapted to control the level of an output signal of the photoelectric conversion element. 6. The image reading apparatus according to claim 5, wherein the peak value of the output signal of the A / D converter when the frequency of appearance of the output signal of the A / D converter is equal to or higher than a predetermined value. An image reading device adapted to control the level of an output signal of a photoelectric conversion element. 7. The image reading device according to claim 6, wherein the minimum level and the maximum level in a range in which the appearance frequency of the output signal of the A / D converter obtained during the prescan is a predetermined value or more, and the correction An image reading apparatus adapted to correct the level of an output signal of an A / D converter during the main scan, based on a coefficient of characteristics.