JP4364834B2 - Automatic exposure control device - Google Patents

Description

  The present invention relates to an automatic exposure control device for optical equipment such as a digital video camera and a digital still camera.

  Conventionally, methods such as average photometry, center-weighted photometry, spot photometry, and multi-pattern photometry are known as automatic exposure determination methods for exposure control devices. In either case, a plurality of evaluation areas are set in advance in the area of the captured image, and the exposure is determined using the average luminance information of each evaluation area. FIG. 1 shows an example of the conventional automatic exposure determination method. In this example, the evaluation area of a captured image is divided into 10 × 10 100 small areas. A plurality of evaluation areas are set for the entire photographed image and the like, and the minimum unit is referred to as a small area here. An evaluation area formed by combining a plurality of small areas is referred to as a middle area. Areas 1 to 6 surrounded by a thick line in FIG.

  For example, in center-weighted metering, exposure control is performed using metering data from the middle area of area 3 and area 4. In center-weighted metering, exposure control is generally performed by weighting the area 3 existing in the center of the two areas 3 and 4 with weight.

  In spot photometry, exposure control is performed using photometric data from an area that is the same as or narrower than the area 3 existing in the center.

  By the way, it is desirable to place the middle region at the center of the image area under shooting conditions with a small luminance difference, but under shooting conditions with a large luminance difference, a correction process called high luminance processing is required. Become.

  For example, as shown in FIG. 2, when photographing a landscape such as a sky at the top and a mountain or the ground at the bottom, the sky with much higher brightness than the brightness of the mountain or ground at the bottom is almost from the center of the screen. The shooting condition is located above. In such a case, since the high-luminance sky is located at the center of the screen, when center-weighted metering or spot metering is performed, the brightness of the sky greatly affects, resulting in a dark image as a whole.

  In automatic exposure control, exposure control such as shutter time and aperture is performed so that the average brightness at the time of shooting is approximately at the center of the shooting brightness range, so some correction is necessary to obtain a bright image suitable for mountains and the ground. Necessary.

  Here, a correction method has been proposed in which a small area having high luminance equal to or higher than a certain threshold (referred to as a high luminance limit threshold) is replaced with a high luminance limit value. Such a correction method is called high luminance processing.

  For example, in the technique disclosed in Japanese Patent Laid-Open No. 5-122600, which is similar to the present invention, a luminance signal having a preset clip level or higher is converted into a luminance signal having a clip level as an upper limit threshold value. Two types of clip levels are set in advance, and are selected based on photometric information.

  Further, according to the technique disclosed in Japanese Patent Laid-Open No. 7-298131 similar to the present invention, a high luminance signal exceeding a preset high luminance removal threshold is removed from the luminance signal (data), and the high luminance signal is removed. Exposure control is performed based on the remaining luminance signal.

  However, this type of conventional exposure control device still has the following problems.

 For example, the brightness histogram obtained when the upper part is empty, the lower part is a mountain, and the ground is as shown in FIG. Here, the horizontal axis is a partition number that is partitioned, for example, by 1 eV from the lower luminance to the higher luminance. The vertical axis represents the number of small areas having luminance existing in the section. The section number “4” corresponding to the sensitivity center of the CCD is the distribution center of the luminance histogram.

  In FIG. 3, the high luminance part (sky) is almost located at the center of the luminance histogram distribution. When this is removed or clipped by high luminance processing, the luminance histogram shown in FIG. 4 is obtained. In FIG. 4, the low luminance part (mountain or ground) is located almost at the center of the distribution of the luminance histogram. On the contrary, when the upper part is a forest, a mountain or the like and the lower part is a field, the distribution of the luminance histogram as shown in FIG. 5 is obtained.

  In FIG. 5, the high luminance part (field) is located almost at the center of the distribution of the luminance histogram. When this is corrected by adjusting the exposure to the low luminance part by high luminance processing, as shown in FIG. 6, the low luminance part (forest, mountain, etc.) becomes the center of the distribution of the luminance histogram, and the field is so bright. It becomes an image of whiteout.

Therefore, an automatic exposure control device that can select a high-intensity removal process that is optimal for removing the influence of a high-intensity captured image portion by obtaining a high-frequency section using a luminance histogram according to the situation. It has been proposed (see Patent Document 1).
JP 2002-372733 A

  However, the one disclosed in Japanese Patent Application Laid-Open No. 2002-372733 is able to select an optimum high-intensity removal process for removing the influence of a high-intensity photographed image part depending on the situation, but a certain high Because it was configured to determine whether or not each small area is high luminance based on the luminance removal threshold or the high luminance restriction threshold, it was possible to approach the optimum high luminance removal process one step. Even so, there are still aspects that are not sufficient from the viewpoint of obtaining appropriate automatic exposure.

  The present invention has been made in view of such a situation, and a photographed image is divided into a plurality of rough middle areas composed of a set of a plurality of small areas, and on the photographed image based on the average luminance of the middle area. The threshold value for determining whether the brightness of the small area in the middle area is high or not is changed for each middle area, and the high brightness processing is performed for each small area. By adopting the configuration, an object is to provide an automatic exposure control apparatus that can perform fine high-intensity processing and can further contribute to obtaining an appropriate automatic exposure.

The automatic exposure control device according to claim 1,
Luminance value calculation means for dividing a captured image into a plurality of small areas based on image data from the image sensor and calculating a luminance value for each small area;
Histogram creation means for creating a brightness histogram by dividing the brightness value for each small area into sections;
A high-brightness region extraction means for extracting a high-brightness region from each of the small regions by a high-brightness extraction threshold determined according to the distribution state of the luminance histogram;
The high luminance removal threshold as a reference value for determining that the luminance of the small area extracted by the high luminance area extracting means should be high luminance and the luminance of the small area extracted by the high luminance area extracting means are used for exposure evaluation. A high brightness limit threshold value as a reference value for determining whether or not to replace the high brightness limit value to be used, and when the brightness for each small area is greater than the high brightness removal threshold value, the brightness is used for exposure evaluation. The luminance of the remaining small area excluded from the target for use is used as luminance information for exposure evaluation, and when the luminance of each small area is smaller than the high luminance removal threshold and larger than the high luminance limitation threshold, High luminance processing means for replacing luminance with the high luminance limit value to obtain luminance information for exposure evaluation;
Comprising calculating means for calculating an exposure control value based on the luminance information of each small area obtained based on the processing of the high luminance processing means,
The photographed image is divided into a plurality of middle areas composed of a set of a plurality of small areas, an average luminance value is calculated based on the luminance of each small area existing in each middle area, and the average for each middle area Determination means for determining the high luminance removal threshold value, the high luminance limitation threshold value, and the high luminance limitation value for each middle region according to a middle region luminance pattern obtained by performing multi-value processing based on the luminance value. Have
The high brightness processing means performs an evaluation process based on the high brightness removal threshold, the high brightness limit threshold, and the high brightness limit value determined for each of the middle regions, and the calculation means is based on the processing result. An exposure control value is calculated.

  The automatic exposure control apparatus according to claim 2 is characterized in that the relationship between the thresholds satisfies the following magnitude relationship.

The automatic exposure control apparatus according to claim 3, wherein the multi-value conversion process is a binarization process based on a comparison between the average brightness and a predetermined threshold value. Features.

  The automatic exposure control apparatus according to a fourth aspect is characterized in that the predetermined threshold value for binarization is the high luminance extraction threshold value.

  The automatic exposure control apparatus according to claim 5, wherein the middle area luminance pattern is arranged from top to bottom on the captured image, and the average luminance of the middle area luminance pattern is higher than the lower middle area. When the pattern exhibits a high pattern, the high-intensity removal threshold value of the middle region existing on the pattern is set lower than the high-intensity removal threshold value used for the remaining middle region.

  The automatic exposure control apparatus according to claim 6, wherein the middle region luminance pattern is arranged from top to bottom on the captured image, and the average luminance of the middle region luminance pattern is higher than the lower middle region. When a high pattern is exhibited, the high brightness removal threshold value of the middle region existing on the pattern is set to the same value as the high brightness extraction threshold value.

The automatic exposure control device according to claim 7,
Luminance value calculation means for dividing a captured image into a plurality of small areas based on image data from the image sensor and calculating a luminance value for each small area;
Histogram creation means for creating a brightness histogram by dividing the brightness value for each small area into sections;
A high-brightness region extraction means for extracting a high-brightness region from each of the small regions by a high-brightness extraction threshold determined according to the distribution state of the luminance histogram;
Having a high luminance removal threshold value as a reference value for determining that the luminance of the small region extracted by the high luminance region extracting means should be high luminance, and the luminance for each small region is higher than the high luminance removal threshold value High luminance processing means for excluding the luminance from the use target for exposure evaluation and setting the luminance of the remaining small area as luminance information for exposure evaluation when large,
In an automatic exposure control apparatus having a calculation means for calculating an exposure control value based on luminance information of each small area obtained based on the processing of the high luminance processing means,
The photographed image is divided into a plurality of middle areas composed of a set of a plurality of small areas, an average luminance value is calculated based on the luminance of each small area existing in each middle area, and the average for each middle area In accordance with the middle area luminance pattern obtained by performing the multi-value processing based on the luminance value, the determining means for determining the high luminance removal threshold for each middle area,
The high-intensity processing unit performs an evaluation process based on the high-intensity removal threshold value determined for each middle region, and the calculation unit calculates an exposure control value based on the processing result.

The automatic exposure control device according to claim 8,
Luminance value calculation means for dividing a captured image into a plurality of small areas based on image data from the image sensor and calculating a luminance value for each small area;
Histogram creation means for creating a brightness histogram by dividing the brightness value for each small area into sections;
A high-brightness region extraction means for extracting a high-brightness region from each of the small regions by a high-brightness extraction threshold determined according to the distribution state of the luminance histogram;
Having a high brightness limit threshold value as a reference value for determining whether or not to replace the brightness of the small area extracted by the high brightness area extracting means with a high brightness limit value used for exposure evaluation; High luminance processing means for replacing the luminance with the high luminance limitation value when the luminance is greater than the high luminance limitation threshold, and obtaining luminance information for exposure evaluation;
In an automatic exposure control apparatus having a calculation means for calculating an exposure control value based on luminance information of each small area obtained based on the processing of the high luminance processing means,
The photographed image is divided into a plurality of middle areas composed of a set of a plurality of small areas, an average luminance value is calculated based on the luminance of each small area existing in each middle area, and the average for each middle area In accordance with the middle region luminance pattern obtained by performing multi-value processing based on the luminance value, the high luminance limit threshold and the high luminance limit value are respectively determined for each middle region,
The high brightness processing means executes an evaluation process based on the high brightness limit threshold and the high brightness limit value determined for each of the middle regions, and the calculation means calculates an exposure control value based on the processing result. It is characterized by that.

  As described above, the automatic exposure control apparatus according to the present invention divides a photographed image into a plurality of rough middle areas composed of a set of a plurality of small areas, and takes a photographed image based on the average brightness of the middle area. Judging the brightness distribution above, changing the threshold for determining whether the brightness of the small area in the middle area is high for each middle area, and performing high brightness processing for each small area Since the configuration to be used is adopted, fine high-intensity processing can be performed, and an effect can be obtained that can further contribute to obtaining appropriate automatic exposure.

  Hereinafter, an embodiment in which an automatic exposure control apparatus according to the present invention is applied to a digital camera will be described with reference to the drawings.

  7 to 9 are external views showing an example of a digital camera as an imaging apparatus to which the automatic exposure control apparatus according to the present invention is applied. FIG. 10 is a block circuit diagram of the digital camera shown in FIGS.

  7 to 10, reference numeral 7 denotes a lens barrel unit. The lens barrel unit 7 includes a zoom optical system 71, a focus optical system 72, a diaphragm unit 73, and a mechanical shutter unit 74.

  The zoom optical system 71 is roughly composed of a zoom lens 71a for capturing an optical image of a subject and a zoom drive motor 71b for driving the zoom lens 71a. The focus optical system 72 is roughly composed of a focus lens 72a and a focus drive motor 72b for driving the focus lens 72a.

  The diaphragm unit 73 is roughly composed of a diaphragm 73a and a diaphragm motor 73b that drives the diaphragm 73a. The mechanical shutter unit 74 includes a mechanical shutter 74a and a mechanical shutter motor 74b that drives the mechanical shutter 74a. The lens barrel unit 7 has a motor driver 75 that drives these motors.

  The motor driver 75 is driven and controlled by a drive command from a CPU block 1043 in the digital still camera processor 104, which will be described later, based on an input from the remote control light receiving unit 6 and an operation input of each operation switch SW1 to SW13 of the operation unit key unit. Is done. In addition, since the function of each operation switch SW1-SW13 is well-known, only a name is attached | subjected to drawing and the detailed description is abbreviate | omitted.

  The ROM 108 stores a control program and control parameters described by codes readable by the CPU block 1043. When the power SW 13 of the digital camera is turned on, the control program is loaded into the main memory (not shown), and the CPU block 1043 controls the operation of each part of the apparatus according to the control program, and the data necessary for the control, etc. Are temporarily stored in the RAM 107 and a local SRAM 1044 in the digital still camera processor 104 described later. By using a rewritable flash ROM for the ROM 108, the control program and control parameters can be changed, and the function can be easily upgraded.

  The CCD 101 is a solid-state image sensor for photoelectric conversion of an optical image. The F / E (front end) -IC 102 includes a CDS 1021, an AGC 1022, an A / D 1023, and a TG 1024. The CDS 1021 performs correlated double sampling for image noise removal. The AGC 1022 performs gain adjustment. The A / D 1023 performs digital signal conversion. The TG 1024 is supplied with a vertical synchronizing signal (hereinafter referred to as VD) and a horizontal synchronizing signal (hereinafter referred to as HD) from the CCD1 signal processing block 1041 and is controlled by the CPU block 1043, the CCD 101 and the F / E-IC 102. A drive timing signal is generated.

  The digital still camera processor 104 includes a CCD1 signal processing block 1041, a CCD2 signal processing block 1042, a CPU block 1043, a local SRAM 1044, a USB block 1045, a serial block 1046, a JPEG CODEC block 1047, a RESIZE block 1048, a TV signal display block 1049, and a memory. A card controller block 10410 is included.

  The CCD1 signal processing block 1041 performs white balance setting and gamma setting on the output data of the F / E-IC 102 from the CCD 101, and supplies a VD signal and an HD signal. The CCD2 signal processing block 1042 performs conversion into luminance data and color difference data by filtering processing. The CPU block 1043 controls the operation of each part of the device. The local SRAM 1044 temporarily stores data necessary for the above-described control. The USB block 1045 performs USB communication with an external device such as a personal computer. The serial block 1046 performs serial communication with an external device such as a personal computer. The JPEG CODEC block 1047 performs JPEG compression / decompression. The RESIZE block 1048 enlarges / reduces the size of the image data by interpolation processing. The TV signal display block 1049 converts the image data into a video signal for display on an external display device such as a liquid crystal monitor or a TV. A memory card controller block 10410 controls a memory card that records captured image data.

  The SDRAM 103 temporarily stores image data when the digital still camera processor 104 performs various processes on the image data. The image data to be stored is, for example, “RAW-RGB image data” in which the CCD 1 signal processing block 1041 takes in the white balance setting and gamma setting from the CCD 101 via the F / E-IC 102. “YUV image data” in which luminance data / color difference data conversion has been performed in the CCD2 signal processing block 1042, “JPEG image data” compressed in JPEG by the JPEG CODEC block 1047, and the like.

  The memory card throttle 121 is a throttle for mounting a removable memory card. The built-in memory 120 is a memory for storing captured image data even when no memory card is attached to the memory card throttle 121 described above. The LCD driver 117 is a drive circuit that drives the LCD monitor 10 described later, and also has a function of converting the video signal output from the TV signal display block 1049 into a signal for displaying on the LCD monitor 10.

  The LCD monitor 10 is used for monitoring the state of a subject before photographing, confirming a photographed image, displaying image data recorded in a memory card or the built-in memory 120 described above, and the like. The video AMP 118 is used for impedance conversion of the video signal output from the TV signal display block 1049 by 75Ω. The video jack 119 is used to connect to an external display device such as a TV. The USB connector 122 is used for USB connection with an external device such as a personal computer. The serial driver circuit 1231 is used for voltage conversion of the output signal of the serial block 1046 described above in order to perform serial communication with an external device such as a personal computer, and the RS-232C connector 1232 is serially connected to an external device such as a personal computer. Used to do.

  The SUB-CPU 109 is a CPU in which ROM and RAM are built in one chip, and outputs the operation signals SW1 to SW13 of the operation key unit and the output signal of the remote control light receiving unit 6 to the CPU block 1043 as user operation information. In addition, a signal indicating the state of the camera output from the CPU block 1043 is converted into a control signal for the sub LCD 1, AF LED 8, strobe LED 9, and buzzer 113, which will be described later, and output.

  The sub LCD 1 is a display unit that displays, for example, the number of shootable images, and the LCD driver 111 is a drive circuit that drives the sub LCD 1 based on an output signal of the SUB-CPU 109. The AF LED 8 is used to display a focus state at the time of shooting, and the strobe LED 9 is used to indicate a strobe charging state. Note that the AF LED 8 and the strobe LED 9 may be used for other purposes such as when a memory card is being accessed. The operation switches SW1 to SW13 of the operation key unit constitute a part of the key circuit operated by the user, and the remote control light receiving unit 6 is a signal receiving unit of the remote control transmitter operated by the user.

  The audio recording unit 115 is roughly composed of a microphone 1153 to which a user inputs an audio signal, a microphone AMP 1152 for amplifying the input audio signal, and an audio recording circuit 1151 for recording the amplified audio signal. The audio reproduction unit 116 includes an audio reproduction circuit 1161 that converts the recorded audio signal into a signal that can be output from the speaker, an audio AMP 1162 that amplifies the converted audio signal, and drives the speaker, and a speaker 1163 that outputs the audio signal. It is abbreviated.

  The CCD1 signal processing block 1041 functions as a luminance calculation means, and performs luminance integration processing for each small evaluation area. This is performed by converting RGB signal data included in the image signal of each evaluation small area into luminance data (brightness value) and integrating the luminance value for each evaluation small area.

  The integrated luminance value for each small evaluation area is processed by a control program (ROM: 108) as a histogram generating means, and a luminance histogram based on the luminance distribution is generated by the control program. Based on the brightness histogram and the exposure conditions at the time of shooting, each threshold value for high brightness and a high brightness processing method are determined.

  Each threshold value of high luminance here is determined based on the distribution of the high frequency sections of the luminance histogram. The high brightness processing thus determined is executed, and an exposure control value for exposure control is obtained, and the exposure control is performed by controlling the F / E-IC 102 and the aperture motor 73b based on the exposure control value. .

  Hereinafter, the high luminance processing will be described. FIG. 11 is a flowchart of high luminance processing.

  First, the CCD signal processing block (exposure control unit) 1041 creates a luminance histogram in consideration of the difference in exposure conditions up to the previous time (S.801). Here, the luminance histogram is created from the luminance information (exposure evaluation information) of each small region obtained by dividing the captured image region G into 100 of 10 × 10 as shown in FIG. Note that the exposure evaluation information is a general term for information used for determining exposure in addition to luminance information which is a luminance signal of each small region.

  In the example shown in FIG. 12, since the number of small regions used in the histogram is only 100, the width per section of the horizontal axis (corresponding to the luminance) of the luminance histogram cannot be made too small. This is because if the width per section is narrowed, the total number of sections increases, and the number (frequency number) in which the luminance of each small area enters per section decreases. Therefore, considering that the sensitivity of the CCD 101 is around ± 3 ev (Expousure Value), the width per section is set to 1.0 ev. Alternatively, the width per section may be 0.5 ev, and the adjacent sections may be 1.0 ev. The number of sections is set to seven sections in consideration of corrections such as exposure correction. Note that it is known that the unit of the section is set to the ev unit, so that it is less affected by the high luminance image portion.

  When the exposure evaluation information of each small region is output as a linear value, the threshold value for each section when creating the histogram is converted from the ev value to the linear value, and the histogram is created. For example, when the linear data at the target exposure (luminance as the central section of the luminance histogram) is 1000, the histogram sections (shown by the center value in Table 1) shown in Table 1 below are set.

図１３はこのようにして得られた輝度ヒストグラムの一例である。ＣＣＤ信号処理ブロック（露出制御部の演算手段）１０４１は輝度ヒストグラムの作成後、輝度ヒストグラムの平均値（頻度平均値＝輝度平均値）を求める（Ｓ.８０２）。頻度平均値（輝度平均値）の計算式は以下の通りである。
頻度平均値（輝度平均値）＝（区画番号×各区画毎の頻度）の総和／小領域の総個数
なお、ここでは、小領域の総個数は１００であり、その図１３において、符号Ｚ１は頻度平均値（輝度平均値）を示す。

FIG. 13 is an example of the luminance histogram obtained in this way. The CCD signal processing block (calculation means of the exposure control unit) 1041 obtains an average value (frequency average value = luminance average value) of the luminance histogram after creating the luminance histogram (S.802). The calculation formula of the frequency average value (luminance average value) is as follows.
Frequency average value (luminance average value) = (sum of partition number × frequency for each partition) / total number of small regions Here, the total number of small regions is 100, and in FIG. The frequency average value (luminance average value) is shown.

  Next, the CCD signal processing block (exposure control unit) 1041 determines a frequency threshold value from the created luminance histogram (S.803). In FIG. 13, reference numeral Z2 indicates the frequency threshold.

  Incidentally, when the luminance is concentrated, as shown in FIG. 14A, the frequency of the luminance histogram exceeds the frequency threshold Z2, but when the luminance on the screen is dispersed, for example, As shown in FIG. 14B, the frequency of the luminance histogram is divided into two, and the phenomenon occurs that the frequency threshold Z2 is not exceeded.

  Therefore, it is conceivable to prepare two frequency threshold values, a high frequency threshold value Z2H and a low frequency threshold value Z2L, as shown in FIG. 15. However, as shown in FIG. If the frequency histogram is fixed to the high frequency threshold Z2H in the luminance histogram obtained from the above, a section with a frequency higher than the frequency threshold Z2H cannot be obtained. In such a case, it is desirable to set a low frequency threshold Z2L.

  On the other hand, as shown in FIG. 15B, when the luminance histogram obtained from the photographed image is concentrated so that the luminance distribution is concentrated, if the frequency threshold Z2L is fixed, a section that is not so high is also selected as a high frequency. In such a case, it is desirable to set a high frequency threshold value Z2H.

That is, if the frequency threshold value Z2 is fixed, it is not desirable anyway. Therefore, here, the CCD signal processing block (exposure control unit) 1041 sets the minimum frequency threshold value Z2MIN as an empirical value, and this minimum The number N of partitions having a frequency equal to or higher than the frequency threshold Z2MIN is counted, and the dynamic frequency threshold ZT is determined from the reciprocal thereof. That is, an empirically obtained value P is set, and the dynamic frequency threshold ZT is obtained by the following equation.
ZT = P / N
When such processing is performed, N is increased when the luminance distribution of the captured image is substantially uniform, so that the frequency threshold ZT is decreased, and it is possible to prevent a high-frequency section higher than the frequency threshold ZT from being present. In addition, when the luminance distribution is locally concentrated, N is small, so the frequency threshold ZT is large, so that a partition with a frequency that is not high can be removed, and a higher-frequency partition can be extracted (S.804). .

  Thus, by determining the dynamic frequency threshold value ZT from the distribution state of the luminance histogram, it is possible to stably extract a high luminance small region.

  Therefore, when the frequency threshold is fixed to a certain value, the exposure histogram changes, or the angle of view changes, the luminance histogram distribution is changed, and the section where the ratio of the high luminance small area is larger than the frequency threshold Z2 changes. The problem of instability is resolved, the frequency threshold value can be determined more finely, and a small area with high brightness can be extracted stably according to the distribution state of the brightness histogram.

  Next, for each section of the luminance histogram, a section having a frequency greater than the frequency threshold Z2 is obtained (S.805). Hereinafter, the section obtained in this way is referred to as a high-frequency section. In FIG. 13, section number 2 and section number 6 are high-frequency sections.

  Among the obtained high-frequency sections, a section having the largest section number (referred to as the highest-frequency section) and a section having the smallest section number (referred to as the minimum-frequency section) are obtained. In FIG. 13, section number 2 corresponds to the minimum high-frequency section, and section number 6 corresponds to the maximum high-frequency section. When only one section has a frequency greater than the frequency threshold Z2 (ZT), the maximum high frequency section and the minimum high frequency section are naturally the same.

  Next, the CCD1 signal processing block 1041 compares whether or not the number of the minimum high-frequency section is larger than a certain section number (S.806). For example, a certain partition number is “4”. The CCD1 signal processing block 1041 resets the minimum high-frequency section when the minimum high-frequency section number is larger than the fixed section number (S.807).

  First, the frequency is sequentially added from the low-luminance side division of the luminance histogram toward the high-luminance side division, and the addition value is compared with the frequency threshold value Z2 (ZT) for each addition. The section at the time when the added value exceeds the frequency threshold value Z2 is newly set as the minimum high-frequency section.

  This resetting process improves that it is difficult to obtain sections having a frequency higher than the frequency threshold Z2 (ZT) when the luminance of a landscape image or the like is dispersed.

  Next, the CCD1 signal processing block 1041 compares whether or not the difference between the section number of the maximum frequent section and the section number of the least frequent section is a certain value or more (S.808). If the difference is greater than or equal to a certain value, the high-brightness extraction threshold Z3 is set when the luminance difference is large (S.809). That is, a value obtained by adding a constant value to the evaluation value (luminance value) corresponding to the minimum high frequency section is set as the high luminance extraction threshold. If the difference is equal to or smaller than a certain value, the high luminance extraction threshold value Z3 'when the luminance difference is small is set (S.810). That is, a value obtained by adding a constant value to the frequency average value (average luminance value) is set as the high luminance extraction threshold value Z3 '. Here, a value obtained by adding a constant value to the evaluation value corresponding to the minimum high-frequency section is shown as a high-frequency extraction threshold Z3 in FIG.

  Subsequently, multi-value conversion processing described below is executed on the nine middle regions shown in FIG. 12 (S. 811). In FIG. 12, for convenience of explanation, symbols Mi (i = 1 to 9: M1 to M9) are attached to the nine middle regions in order from the upper left to the lower right.

  FIG. 16 is a flowchart of the multivalue processing.

First, the CCD1 signal processing block 1041 calculates an average luminance value of nine middle regions (S.701). Here, the histogram frequency average value of the small areas in each of the middle areas M1 to M9 is set as the average luminance value. That is, the average luminance value of each of the middle areas M1 to M9 is obtained by a value obtained by dividing the total luminance of the small areas for which the average luminance value is calculated by the number of small areas in the middle area. For example, if the target middle area is the reference numeral M1 in FIG. 12, the average luminance value of the middle area M1 is enlarged as shown in FIG. 17, and the small areas m1 _{1 to} m1 ₉ in the middle area M1 are shown. Is a value obtained by the total number of small regions of 9.

Further, the average luminance value in the region M5 that is a value obtained by the total number 16 of the small regions the sum of the brightness of the small region m5 ₁ ~m5 ₁₆ of the middle region M5 (S.702).

  Next, a comparison is made between each average luminance value of each obtained middle region Mi (i = 1 to 9) and a high luminance extraction threshold value Z3 obtained from the luminance histogram (S.703), and each average luminance value is calculated. Is larger than the high luminance extraction threshold, the luminance value of the middle region is set to “1” (S.704), and when it is smaller, it is binarized as “0” (S.705). Thereby, the binarized middle region pattern value is created.

Here, as shown in FIG. 18, when the upper left middle region M1 is associated with the least significant bit and the lower right middle region M9 is associated with the most significant bit, the middle region pattern value is
When all binarized middle areas are 0: 000000000000 (decimal value: 0)
When all binarized middle areas are 1: 111111111 (decimal value: 511)
Corresponding to

  Further, for example, as shown in FIG. 18, when the lower left middle region has a low luminance and the upper right middle region has a high luminance distribution, the middle region pattern value is 00100111 (decimal value: 39). As shown in FIG. 19, when the luminance of the middle region on the left side and the middle region on the upper right side is low and the luminance of the middle region on the right side is high, the middle region pattern value is 100100000 (decimal number). Value: 288). This is repeated for Mi (i = 1 to 9) (S.706), and the middle region pattern value is determined (S.707).

  Next, the threshold level value Q of each of the middle areas M1 to M9 is set according to the middle area pattern value determined in this way (S.708). Here, the threshold level value Q has three stages of 0-2. By using the threshold level values Q0 to Q2, a middle region dedicated threshold described later is switched for each middle region. The smaller the threshold level value Q, the higher the removal rate of the small area with high luminance.

  For example, in the case of a bright subject at the top, “1” appears frequently as the pattern value of the upper middle region, and the pattern value of the lower middle region becomes the middle region pattern value where “0” appears frequently. In such a case, the threshold level value Q of the upper middle region is set to a small value, and the threshold level value Q of the lower middle region is set to a large value.

  Further, for example, in the case of a subject whose upper part is dark, it becomes a middle area pattern value in which “0” appears frequently as the pattern value of the upper middle area. In such a case, the threshold level value Q in the upper middle region is not set to a small value, but is set to a threshold level value Q that is approximately equal to the threshold level value Q in the lower middle region.

  For example, when the middle area pattern value is “000100111”, the threshold levels Q of the middle areas M1 to M3 are set to “0”, and the threshold levels Q of the middle areas M4 and M6 to M9 are set to “1”. For example, when the middle area pattern value is “100100000”, the threshold levels Q of the middle areas M1 to M4 and M6 to M9 are set to “1”.

  Further, here, the threshold level value Q of the central middle region M5 is set to a threshold level value Q higher than the threshold level values Q of the other middle regions M1 to M4 and M6 to M9. It becomes possible to improve the reproducibility of the high luminance part. Here, the threshold level Q of the middle region M5 is both “2”.

  Further, for example, when the middle area pattern values are “000000000000” and “111111111”, the threshold levels Q of the middle areas M1 to M9 are all set to “1”.

  This threshold level Q is assigned to each of the middle regions M1 to M9 according to 512 middle region pattern values.

  By performing such processing, the threshold level value Q can be set finely in accordance with the luminance pattern of the subject, and as a result, a fine, high-luminance small region can be removed.

  In this embodiment, since the average luminance value of each of the middle regions M1 to M9 is binarized by comparing with the high luminance extraction thresholds Z3 and Z3 ′, the number of luminance pattern values (512 in this embodiment) ) And the calculation time is shortened. Further, since the threshold values for the binarization determination are set to the high luminance extraction threshold values Z3 and Z3 ', high luminance extraction is possible for a rough region such as the middle region.

  In this embodiment, binarization processing is performed. However, if multi-value processing such as ternarization and quaternarization is performed, finer exposure control can be performed. In this embodiment, the threshold values for the binarization determination are set to the high luminance extraction threshold values Z3 and Z3 ′. However, the average luminance value of the entire small area may be used as the threshold value for the binarization determination. Good.

  Next, the high luminance removal process will be described. Here, the high luminance removal processing refers to processing including replacement.

  FIG. 20 is a flowchart of the high luminance removal process.

  The CCD1 signal processing block 1041 executes high-intensity extraction for each small area, and performs high-intensity removal processing on the extracted high-intensity small area.

  First, a middle area as a processing target is acquired (S.601). Next, a high brightness removal threshold value, a high brightness limit threshold value, and a high brightness limit value are set as the middle area dedicated threshold values for each middle area (S.602).

  In accordance with the threshold level Q obtained by the multi-value processing, the middle region dedicated threshold values of the middle regions M1 to M9 are switched.

  When the threshold level Q is “0”, the high luminance removal threshold = the high luminance restriction threshold = the high luminance extraction threshold is set, and all the extracted high luminance small regions are to be removed. Ultimately, not all high-luminance small areas are removed due to the limitation on the number of high-luminance removal areas.

  As a result, the influence of the sky part can be reduced for a luminance pattern such as a photographed image whose upper part corresponds to the sky part.

When the threshold level Q is “1”,
High luminance removal threshold = High luminance extraction threshold + 1.5 Ev
High luminance limit threshold = High luminance extraction threshold + 0.5 Ev
High luminance limit value (Yclip) = high luminance limit threshold value.

When the threshold level Q is “2”,
High luminance removal threshold = High luminance extraction threshold + 2.5 Ev
High luminance limit threshold = High luminance extraction threshold + 1.5 Ev
High luminance limit value (Yclip) = high luminance limit threshold value.

  We decided to use the high-brightness extraction threshold value obtained from the luminance histogram as the threshold for each middle region as a reference because it can follow the change in luminance and the effect of hunting that the image display for monitoring becomes brighter or darker. This is to reduce it.

  In addition, depending on the brightness value of the small area, it was decided to switch between removal, replacement with a high brightness limit value, or use as it is, such as a small area corresponding to strange high brightness such as a point light source. In addition, the brightness of the small area corresponding to the small area such as the sky part in the backlight state is removed using the high brightness removal threshold, and the high brightness of the normal level is replaced with the high brightness limit threshold without removing it. By doing this, it is possible to suppress the shift to the high luminance side and reduce the overexposure phenomenon.

  As described above, when the threshold level Q is changed according to the middle region pattern, it is possible to set fine appropriate high luminance extraction threshold values and high luminance limit threshold values for each middle region according to the luminance pattern of the subject. .

  As described above, the CCD1 signal processing block 1041 divides the photographed image G into a plurality of middle regions made up of a set of a plurality of small regions, and calculates an average luminance value based on the luminance of each small region existing in each middle region. And a high luminance removal threshold, a high luminance limit threshold, and a high luminance limit for each middle region according to the middle region luminance pattern obtained by performing multilevel processing based on the average luminance value for each middle region. It functions as a determining means for determining each value.

  Subsequently, the CCD1 signal processing block 1041 counts the small areas and integrates the luminance values for the small areas in each middle area.

  First, the brightness value of the small area to be processed is acquired (S.603), and it is determined whether or not the brightness value is larger than the high brightness extraction threshold value determined from the distribution state of the brightness histogram (S.604).

  When the luminance value of the small region to be processed in the middle region to be processed is smaller than the high luminance extraction threshold, the number Nn of normal regions is counted as the normal luminance region (S.605). Further, the integrated luminance value Yn of the luminance values in the normal area is obtained (S.606).

  If the luminance value of the small region to be processed in the middle region of the processing target is larger than the high luminance extraction threshold, it is next determined whether or not the luminance value of the small region is larger than the high luminance removal threshold ( S.607).

  If the luminance value to be processed is larger than the high luminance removal threshold, the small region is determined as a high luminance removal region, and the number of high luminance removal regions Nh3 is counted (S.608). Next, the integrated luminance value Yh3 of the high luminance removal area is obtained (S.609).

  On the other hand, when the brightness of the small area is smaller than the high brightness removal threshold, it is determined whether or not the brightness value of the small area is larger than the high brightness limit threshold (S.610).

  If the brightness value of the small area is larger than the high brightness limit threshold, it is determined as a high brightness limit area and the number of high brightness limit areas Nh2 is counted (S.611). Next, the integrated luminance value Yh2 of the high luminance limited area is obtained (S.612).

  If the brightness value of the small area is smaller than the high brightness limit threshold, the number of high brightness areas Nh1 is counted as a high brightness area (S.613). Next, the luminance integrated value Yh1 of the high luminance region is obtained (S.614).

  After the count and integration of the corresponding small area are completed, it is determined whether the count and integration of all the small areas in the target middle area has been completed (S.615). The small area is acquired, and the comparison process, the count process, and the integration process are repeated until all the small areas in the middle area are completed.

  The CCD1 signal processing block 1041 calculates the number of effective areas (Ne) when the counting and integration in the middle area is completed (S.616). Subsequently, an effective luminance integrated value (Ye) is calculated (S.617).

  Here, the number Ne of effective areas is obtained by subtracting the number Nh3 of high luminance removal areas, which is the number of removal areas to be removed, from the number of all small areas (Nall = 100).

  The effective luminance integrated value Ye is a luminance integrated value obtained by clipping the luminance integrated value Yn of the normal region, the luminance integrated value Yh1 of the high luminance region, and the luminance of the high luminance limiting region with the high luminance limiting value (= number of high luminance limiting regions). (Nh2) × high luminance limit value (Yclip)).

Then, each middle region luminance value (Yi: i = 1 to 9) is calculated (S.618). This middle region luminance value is obtained by dividing the effective luminance integrated value Ye by the number Ne of effective regions. This is repeated for all the middle regions M1 to M9 (S.619).
When the luminance value calculation for all the middle regions is completed, final luminance value calculation (Y) is performed (S.620). This final luminance value Y is obtained by dividing the sum of the luminance values (Y1 to Y9) of each middle region by the number of middle regions (Narea = 9).

  The CCD1 signal processing block 1041 executes an evaluation process based on the final luminance value Y and calculates an exposure control value.

  The processing according to the present invention can be performed by a program. The program including this processing is stored in an external storage medium such as a memory card, and the processing according to the present invention is performed by rewriting the camera program as necessary. It may be.

  In the embodiment of the present invention, the setting of the small evaluation area and the evaluation middle area has been described as the nine division mode shown in FIG. 12, but the present invention is not limited to this.

  In the embodiment of the present invention, the brightness histogram used for exposure control is described as having a width per section of 1 ev and the number of sections as shown in FIG. However, the present invention is not limited to this.

  Furthermore, in the embodiment of the present invention, the captured image is described as being stored in the memory card. However, the present invention is not limited to the memory card, and for example, a video tape may be used as a storage medium. At this time, the memory card storing the control program is set in the video camera separately from the storage medium storing the image.

  In the embodiment of the present invention, the exposure control device of the optical device has been described as the exposure control device of a digital still camera. However, the present invention is not limited to this as long as it is an optical device using an image sensor. It can also be applied to a scope using a camera or a CCD.

It is explanatory drawing which shows an example of the division area of the conventional picked-up image. It is a figure which shows an example of a to-be-photographed object typically. It is a figure which shows the brightness | luminance of the picked-up image obtained when image | photographing the to-be-photographed object shown in FIG. 2 with a brightness | luminance histogram. It is a figure which shows the brightness | luminance histogram obtained when performing a high brightness | luminance process about the sky part of the picked-up image which has a brightness | luminance histogram shown in FIG. It is a figure which shows the brightness | luminance of the picked-up image obtained when image | photographing the scenery of a mountain on the upper part and a field on the lower part with a luminance histogram. It is a figure which shows the brightness | luminance histogram obtained when a high-intensity process is performed about the field of the picked-up image which has a brightness | luminance histogram shown in FIG. It is a front view which shows an example of the digital camera to which the automatic exposure control apparatus concerning this invention was applied. It is a rear view of the digital camera shown in FIG. FIG. 8 is a top view of the digital camera shown in FIG. 7. FIG. 8 is a block circuit diagram of the digital camera shown in FIG. 7. It is a high-intensity processing flow figure concerning this invention. It is explanatory drawing which shows an example of the division | segmentation of the small area | region and middle area | region of the picked-up image concerning this invention. It is explanatory drawing which shows an example of the brightness | luminance histogram obtained by the automatic exposure control apparatus concerning this invention. It is explanatory drawing which illustrates typically the malfunction by the luminance concentration of a luminance histogram, dispersion | distribution, Comprising: (a) is explanatory drawing which shows the malfunction when luminance concentrates, (b) is a case where luminance is disperse | distributing It is explanatory drawing which shows this malfunction. It is explanatory drawing which shows an example of the solution method of the malfunction by preparing two frequency threshold values and luminance concentration and dispersion | distribution, Comprising: (a) is explanatory drawing which shows typically the case where luminance distribution is disperse | distributed, (b) These are explanatory drawings schematically showing a case where the luminance distribution is concentrated. It is a multi-value process flow diagram according to the present invention. It is explanatory drawing which expands and shows a small area | region, in order to demonstrate an example of the calculation method of the average luminance concerning this invention. It is explanatory drawing which shows the correspondence of the middle area | region and bit concerning this invention. It is a figure for demonstrating an example of the middle region pattern value corresponding to the brightness | luminance pattern of the to-be-photographed object concerning this invention. It is a high-intensity removal process flowchart concerning this invention.

Explanation of symbols

101 ... CCD (imaging device)
108 ... ROM (histogram creation means)
1041... CCD1 signal processing block (high luminance processing means, determination means)

Claims (8)

Luminance value calculation means for dividing a captured image into a plurality of small areas based on image data from the image sensor and calculating a luminance value for each small area;
Histogram creation means for creating a brightness histogram by dividing the brightness value for each small area into sections;
A high-brightness region extraction means for extracting a high-brightness region from each of the small regions by a high-brightness extraction threshold determined according to the distribution state of the luminance histogram;
The high luminance removal threshold as a reference value for determining that the luminance of the small area extracted by the high luminance area extracting means should be high luminance and the luminance of the small area extracted by the high luminance area extracting means are used for exposure evaluation. A high brightness limit threshold value as a reference value for determining whether or not to replace the high brightness limit value to be used, and when the brightness for each small area is greater than the high brightness removal threshold value, the brightness is used for exposure evaluation. The luminance of the remaining small area excluded from the target for use is used as luminance information for exposure evaluation, and when the luminance of each small area is smaller than the high luminance removal threshold and larger than the high luminance limitation threshold, High luminance processing means for replacing luminance with the high luminance limit value to obtain luminance information for exposure evaluation;
In an automatic exposure control apparatus having a calculation means for calculating an exposure control value based on luminance information of each small area obtained based on the processing of the high luminance processing means,
The photographed image is divided into a plurality of middle areas composed of a set of a plurality of small areas, an average luminance value is calculated based on the luminance of each small area existing in each middle area, and the average for each middle area Determination means for determining the high luminance removal threshold value, the high luminance limitation threshold value, and the high luminance limitation value for each middle region according to a middle region luminance pattern obtained by performing multi-value processing based on the luminance value. Have
The high brightness processing means performs an evaluation process based on the high brightness removal threshold, the high brightness limit threshold, and the high brightness limit value determined for each of the middle regions, and the calculation means is based on the processing result. An automatic exposure control device that calculates an exposure control value. 2. The automatic exposure control apparatus according to claim 1, wherein the relationship between the threshold values satisfies the following magnitude relationship.
High luminance extraction threshold ≦ High luminance limit threshold ≦ High luminance removal threshold   The automatic exposure control apparatus according to claim 1, wherein the multi-value processing is a binarization processing based on a comparison between the average luminance and a predetermined threshold value.   The automatic exposure control apparatus according to claim 3, wherein the predetermined threshold value for binarization is the high luminance extraction threshold value.   When the middle area luminance pattern is arranged from top to bottom on the photographed image and the average luminance of the middle area luminance pattern is higher in the upper area than the lower middle area, 5. The automatic exposure control according to claim 1, wherein a high-intensity removal threshold value of an existing middle region is set lower than a high-intensity removal threshold value used for the remaining middle region. apparatus.   When the middle area luminance pattern is arranged from top to bottom on the photographed image and the average luminance of the middle area luminance pattern is higher in the upper area than the lower middle area, 5. The automatic exposure control apparatus according to claim 1, wherein a high-luminance removal threshold value of an existing middle region is set to the same value as the high-luminance extraction threshold value. Luminance value calculation means for dividing a captured image into a plurality of small areas based on image data from the image sensor and calculating a luminance value for each small area;
Histogram creation means for creating a brightness histogram by dividing the brightness value for each small area into sections;
A high-brightness region extraction means for extracting a high-brightness region from each of the small regions by a high-brightness extraction threshold determined according to the distribution state of the luminance histogram;
Having a high luminance removal threshold value as a reference value for determining that the luminance of the small region extracted by the high luminance region extracting means should be high luminance, and the luminance for each small region is higher than the high luminance removal threshold value High luminance processing means for excluding the luminance from the use target for exposure evaluation and setting the luminance of the remaining small area as luminance information for exposure evaluation when large,
In an automatic exposure control apparatus having a calculation means for calculating an exposure control value based on luminance information of each small area obtained based on the processing of the high luminance processing means,
The photographed image is divided into a plurality of middle areas composed of a set of a plurality of small areas, an average luminance value is calculated based on the luminance of each small area existing in each middle area, and the average for each middle area In accordance with the middle area luminance pattern obtained by performing the multi-value processing based on the luminance value, the determining means for determining the high luminance removal threshold for each middle area,
The high brightness processing means performs an evaluation process based on the high brightness removal threshold value determined for each of the middle regions, and the calculation means calculates an exposure control value based on the processing result. Exposure control device. Luminance value calculation means for dividing a captured image into a plurality of small areas based on image data from the image sensor and calculating a luminance value for each small area;
Histogram creation means for creating a brightness histogram by dividing the brightness value for each small area into sections;
A high-brightness region extraction means for extracting a high-brightness region from each of the small regions by a high-brightness extraction threshold determined according to the distribution state of the luminance histogram;
Having a high brightness limit threshold value as a reference value for determining whether or not to replace the brightness of the small area extracted by the high brightness area extracting means with a high brightness limit value used for exposure evaluation; High luminance processing means for replacing the luminance with the high luminance limitation value when the luminance is greater than the high luminance limitation threshold, and obtaining luminance information for exposure evaluation;
In an automatic exposure control apparatus having a calculation means for calculating an exposure control value based on luminance information of each small area obtained based on the processing of the high luminance processing means,
The photographed image is divided into a plurality of middle areas composed of a set of a plurality of small areas, an average luminance value is calculated based on the luminance of each small area existing in each middle area, and the average for each middle area In accordance with the middle region luminance pattern obtained by performing multi-value processing based on the luminance value, the high luminance limit threshold and the high luminance limit value are respectively determined for each middle region,
The high brightness processing means executes an evaluation process based on the high brightness limit threshold and the high brightness limit value determined for each of the middle regions, and the calculation means calculates an exposure control value based on the processing result. An automatic exposure control device characterized by that.

Images