JP2004242015A - Histogram calculation method, histogram calculation circuit, and imaging apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a histogram calculation method for requiring a small computing load with a high histogram detection capability, and provide a histogram calculation circuit and an imaging apparatus employing a signal processing circuit. <P>SOLUTION: The histogram calculation circuit comprises: a first histogram generation means for generating a first histogram representing a signal strength distribution of a first digital signal; and a second generation means for generating a second histogram representing a signal strength distribution of a second digital signal corresponding to the first digital signal. The first and second histograms are generated with different gradation detection accuracy respectively. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

【００３６】
ここで、式（１），（２），（３）において、パラメータｐは、加算されたデジタル信号全体に対するゲインであり、これによって加算後のデータに対するダイナミックレンジの制御を行う。この値が小さいほど、ダイナミックレンジは広く、逆に大きいほど、ダイナミックレンジは狭くなる。具体的には、コントラストの高いシーン（真夏の晴天など）では０．８に、曇りや日陰では０．８６、室内蛍光灯下では０．９といったようにシーンに応じてこの値を変化させてやることにより、８ｂｉｔの階調値を有効に使用することが可能となる。
【００３７】
パラメータｔｈは、高感度デジタル信号Ｈ_ｂに対し低感度デジタル信号Ｌ_ｂをどの程度混ぜ合わせるかの比率を表すものであり、高感度デジタル信号Ｈ_ｂと低感度デジタル信号Ｌ_ｂの信号強度に応じて予め決定することができる。実際の合成処理においては、高感度デジタル信号Ｈ_ｂと、パラメータｔｈ， ｐと、上記式（２）及び（３）に従い決定されるｈ＿ｇａｉｎ及びｌ＿ｇａｉｎを算出し、高感度デジタル信号Ｈ_ｂ及び低感度デジタル信号Ｌ_ｂに、それぞれｈ＿ｇａｉｎ及びｌ＿ｇａｉｎを乗算して、各信号を加算することにより、合成デジタル信号Ｓを生成する。一例として、ｔｈを固定とし、ｐをシーンに応じて可変とすることで処理を実行することが可能である。
【００３８】
ＹＣ処理部５６は、所定の同時化処理を行い、合成信号ＳのＲ信号、Ｇ信号、Ｂ信号から輝度信号Ｙとクロマ信号Ｃｒ，Ｃｂとを作成する回路である。また、ＹＣ処理部５６は、生成した輝度信号Ｙとクロマ信号Ｃｒ，Ｃｂに対し、ノイズ低減等の各処理を施し、非圧縮状態の画像データを生成する。生成された画像データは、圧縮される場合には圧縮処理部６０で圧縮された後に、また非圧縮のままの場合には、直接画像記録再生部１００に送られ、記録媒体１０５に記録される。
【００３９】
以下、ヒストグラム検出部５１について詳述する。
【００４０】
高感度信号積算部１５１は、高感度デジタル信号Ｈ_ａをそれぞれＲ，Ｇ，Ｂの色毎に積算し、色毎のヒストグラム及び累積頻度を作成するものである。ここで、高感度信号積算部１５１は、１２ビット階調の高感度デジタル信号Ｈ_ａを６ビットシフトし、６ビット換算のデータ（６４階調）としてヒストグラムを求める。図４は、高感度デジタル信号Ｈ_ａのＧについて作成されたヒストグラムの例を示す。
【００４１】
信号レベル検出部１５２は、低感度デジタル信号Ｌ_ａの各色Ｒ，Ｇ，Ｂ信号レベルの特性を算出し、後段に設けられた低感度信号積算部１５３に出力するものである。ここで、信号レベル検出部１５２が検出する信号レベルの特性とは、信号のレベルの大まかな分布状態、一例として、１２ビット階調で表されている各色Ｒ，Ｇ，Ｂの信号が、ハイライト側にどの程度信号成分を含んでいるかを検出する。
【００４２】
具体的に、信号レベル検出部１５２は、１２ビット階調で表現された各色信号が、（１）階調レベル２０４８〜４０９５の成分（１２ビット目が１）、（２）階調レベル１０２４〜２０４７の成分、（３）階調レベル５１２〜１０２３の成分、そして５１１以下の成分をそれぞれいくつ含むかをカウントする。以下、具体的な算出方法を図５のフローチャートを基に説明する。
【００４３】
まず、信号レベル検出部１５２は、低感度デジタル信号Ｌ_ａを受領すると（ステップＳ１）、受け取った信号の各色毎の各信号成分の１２ビット目が１であるかどうかを確認する（ステップＳ２）。ステップＳ２において、１２ビット目が１である場合には、その成分は、２０４８〜４０９５の内の何れかの大きさを持つ（最大で４０９５）としてカウントを行い（ステップＳ３）、０である場合には、１１ビット目が１であるかどうかを確認する（ステップＳ４）。
【００４４】
ステップＳ４において、１１ビット目が１である場合には、その成分は、１０２４〜２０４７の内の何れかの大きさを持つ（最大で２０４７）としてカウントを行い（ステップＳ５）、０である場合には、１０ビット目が１であるかどうかを確認する（ステップＳ６）。
【００４５】
ステップＳ６において、１０ビット目が１である場合には、その成分は、５１２〜１０２３の内の何れかの大きさを持つ（最大で１０２３）としてカウントを行い（ステップＳ７）、０である場合には、その成分は、５１１以下の何れかの大きさを持つ（最大で５１１）としてカウントを行う（ステップＳ８）。１成分についてのカウント終了後、次の成分があるかどうかを確認し、次成分が存在するのであれば、ステップＳ２に戻り、次成分について同様のカウントを行う（ステップＳ９）。次成分が無ければ、カウントを終了する。
【００４６】
信号レベル検出部１５２は、カウントが終了すると、各成分毎のカウント数を確認し、信号レベルを決定する（ステップＳ１０、ステップＳ１１）。まず、階調レベル２０４８〜４０９５の成分についてのカウント数が予め定められた所定値以上であるかどうかを確認し、所定値以上であれば信号レベル１であるとする。ここで、所定値より小さければ、１０２４〜２０４７の成分が予め定められた所定値以上であるかどうかを確認し、所定値以上であれば信号レベル２であるとする。さらにここで、所定値より小さければ、５１２〜１０２３の成分が予め定められた所定値以上であるかどうかを確認し、所定値以上であれば信号レベル３であるとし、所定値より小さければ信号レベル４であるとする。
【００４７】
低感度信号積算部１５３は、信号レベル検出部１５２によって確認された信号レベルを基に、低感度デジタル信号Ｌ_ａをそれぞれＲ，Ｇ，Ｂの色毎に積算し、色毎のヒストグラム及び累積頻度を作成するものである。ここで、信号レベルに応じてとは、１２ビット階調の高感度デジタル信号Ｈ_ａをレベルに応じてシフトするビット数を変更することを意味する。
【００４８】
図６は、信号レベルと、ヒストグラム作成時の階調分割の関係を示す図である。低感度信号積算部１５３では、信号レベル１の場合には、１２ビット階調の信号を６ビットシフトし、信号成分を積算することにより、１２ビットのダイナミックレンジ幅分を６４階調で表現する。一方、信号レベル２の場合には、１２ビット階調の信号を５ビットシフトしかつ１２ビット目を無視して、信号成分を積算することにより、１１ビットのダイナミックレンジ幅分を６４階調で表現する。同様に、信号レベル３の場合には、１２ビット階調の信号を４ビットシフトし、１２ビット目及び１１ビット目を無視することにより、信号成分を積算し、１０ビットのダイナミックレンジ幅分を６４階調で表現する。最後に、信号レベル４の場合には、１２ビット階調の信号を３ビットシフトし、１０〜１２ビット目を無視して信号成分を積算することにより、９ビットのダイナミックレンジ幅分を６４階調で表現する。
【００４９】
なお、以上の説明では、低感度信号の信号レベルが、２０４７以上であるかどうか（信号レベル１かどうか）についても確認を行ったが、低感度信号の信号レベルが信号レベル１に実効的な数値を持たない場合には、信号レベル１であるかどうかの判断を省略し、低感度信号が信号レベル２〜４の何れに属するかを判定した後に、ヒストグラムを作成するように構成してもよい。一般に、ＷハニカムＣＣＤの場合には、低感度信号の信号レベルは、対応する高感度信号の信号レベルの１／４程度であるため、信号レベル１についての確認は不要である。
【００５０】
また、以上の説明では、各信号レベルに該当する成分の数が所定値以上であるかどうかで、ヒストグラム作成範囲を変更したが、特にこれに限られることはなく、一度１２ビット階調の信号を６ビットシフトして、ヒストグラムを作成し、作成されたヒストグラムから最適な信号レベルを判断し、ヒストグラムを再作成するように構成してもよい。
【００５１】
なお、ヒストグラム検出部５１によって検出された、高感度信号Ｈ_ａ及び低感度信号Ｌ_ａの各ヒストグラムは、ゲイン調整部５３，階調補正部５４，合成処理部５５等におけるゲイン、補正係数、ｈ＿ｇａｉｎ及びｌ＿ｇａｉｎ等を決定するのに用いられる。
【００５２】
以上、本実施形態によれば、低感度信号Ｌ_ａについては、その信号特性に応じてヒストグラムが作成される。一般に低感度信号Ｌ_ａは、高感度信号Ｈ_ａに比べてダイナミックレンジ幅が狭く、１２ビット目、１１ビット目等に実効的な信号が含まれないような場合も多々考えられる。したがって、信号の特性に応じて、検出するダイナミックレンジを変更することによって、１２ビットのダイナミックレンジ幅よりも狭いダイナミックレンジ幅について６４階調で表現することにより、１２ビットのダイナミックレンジ幅をそのまま６４階調表現していた場合と比べて、低感度信号の信号特性をよりよく表現するヒストグラムを作成することが可能である。
【００５３】
また、本実施形態によれば、検出するダイナミックレンジを変更することにより検出精度の向上を図るため、特別な回路の構成や記憶装置の追加を必要としない。したがって、ＬＳＩなどハードかする際のコストアップを抑えつつ、検出精度の向上を図ることができる。
【００５４】
なお、本実施形態においてはＷハニカムＣＣＤと呼ばれる撮像素子を備えた撮像装置として説明を行ったが、これに限らず、高感度信号及び低感度信号を取得可能な各種撮像装置に上記信号処理回路を適用しても構わない。
【００５５】
【発明の効果】
本発明によれば、ヒストグラムの階調分解能を落とす際に、検出信号の特性に応じて、検出するダイナミックレンジを変更することによって、回路構成の簡素化を図り、ＬＳＩなどハード化する際のコストアップを抑えつつ、ヒストグラムの検出精度の向上を図ることができる。
【図面の簡単な説明】
【図１】本実施形態のデジタルカメラを示すブロック図である。
【図２】固体撮像素子の部分拡大平面図である。
【図３】信号処理部における処理の詳細を示すブロック図である。
【図４】高感度デジタル信号Ｈ_ａのＧについて作成されたヒストグラムの例を示す。
【図５】ヒストグラム算出処理を示すフローチャートである。
【図６】信号レベルと、ヒストグラム作成時の階調分割の関係を示す図である。
【符号の説明】
１ デジタルカメラ
１０ 光学系
２０ 撮像部
２１ 固体撮像素子
２２ 受光素子
２４ 垂直転送路
２６ 水平転送路
２８ 出力部
３０ Ａ／Ｄ撮像処理部
４０ 駆動制御部
５０ 信号処理部
５１ ヒストグラム検出部
５２ 最適化合成処理回路
５３ ゲイン調整部
５４ 階調補正部
５５ 合成処理部
５６ ＹＣ処理部
６０ 圧縮処理部
７０ 表示部
８０ 制御部
９０ メモリ
１００ 画像記録再生部
１０５ 記録媒体[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a signal processing method, a signal processing circuit, and an imaging device, and more particularly, to an imaging device such as a digital camera or a digital video camera equipped with an imaging device capable of outputting a high-sensitivity signal and a low-sensitivity signal, and a signal used therefor The present invention relates to a processing circuit and a signal processing method.
[0002]
[Prior art]
In general, an image pickup apparatus such as a digital camera or a digital video camera has a specific dynamic range determined by a charge storage capacity of a built-in image pickup device. The dynamic range of this image sensor is narrower than the dynamic range of a silver salt negative film obtained by a camera using a silver salt film.
[0003]
Therefore, when shooting a scene in which a very bright (high brightness) subject and a relatively dark (low brightness) subject are mixed, such as when shooting indoors or outdoors, either a high brightness subject or a low brightness subject The exposure time is controlled so that one is at an appropriate level. Under such control, if the low-brightness subject is set to the proper level, the high-brightness part will be overexposed, and if the high-brightness subject is set to the proper level, the low-brightness part will be blacked out. The case where it cannot be obtained is also assumed.
[0004]
In order to solve such problems, two types of image signals or video signals of high sensitivity signals and low sensitivity signals having different dynamic ranges are acquired for one scene, and the two types of image signals are optimized. An imaging apparatus that obtains a signal having a wide dynamic range by performing synthesis processing has been developed (see Patent Document 1).
[0005]
Here, if uniform optimization processing is performed in any scene, the image quality may be deteriorated depending on the imaging scene. Therefore, a signal having a wide dynamic range as described in Patent Document 1 is used. It is desirable to perform adaptive optimization processing according to the scene.
[0006]
In order to perform this adaptive control, it is necessary to design the gradation based on information such as the contrast (highlight) of the subject. In general, in order to obtain the contrast (highlight) information of the subject, it is conceivable to use a histogram of the captured image (see Patent Document 2). When creating a histogram here, it is desirable to provide a gradation resolution of the histogram equivalent to the resolution of the image to be detected.
[0007]
[Patent Document 1]
JP 2001-94999 A
[Patent Document 2]
JP2001-8104
[0008]
[Problems to be solved by the invention]
However, when a histogram having a gradation resolution equivalent to the resolution of the detected image is created for each of the two types of image signals, the capacity is 2 compared to an apparatus that captures only one type of image signal. It is necessary to provide a double storage device. For example, when the detection data of one signal has 12-bit gradation, the gradation resolution of the histogram for one signal is also 12-bit gradation (= 4096 gradation), and for two signals, it is twice the 4096 gradation. It is necessary to provide an area for storing data for 8192 gradations. In addition, when the number of data in the histogram is large, the amount of calculation at the time of comparison determination increases, and a heavy load is applied to the arithmetic circuit.
[0009]
In order to solve this, it is conceivable to reduce the resolution of the histogram. However, if the resolution of the histogram is lowered, the detection accuracy of the histogram is lowered, image optimization is not performed properly, and high-quality image data There is also a problem that cannot be obtained.
[0010]
The present invention has been made in view of the above-described problems, and provides a histogram detection method, a histogram detection circuit, and an image pickup apparatus using the signal processing circuit with a low calculation load while having a high histogram detection capability. The purpose is to do.
[0011]
[Means for Solving the Problems]
The above object of the present invention is achieved by the following method.
(1) creating a first histogram showing a signal intensity distribution of the first digital signal;
Creating a second histogram showing a signal intensity distribution of a second digital signal corresponding to the first digital signal,
The histogram calculation method, wherein the first histogram and the second histogram are created with different gradation detection accuracy.
(2) a signal characteristic detection step of detecting a signal characteristic of the second digital signal;
The histogram detection method according to (1), wherein the second histogram is created in accordance with the signal characteristic detected by the signal characteristic detection unit.
(3) First histogram creating means for creating a first histogram showing the signal intensity distribution of the first digital signal, and a second showing the signal intensity distribution of the second digital signal corresponding to the first digital signal. Second histogram creating means for creating a histogram of
The histogram calculation circuit, wherein the first histogram and the second histogram are created with different gradation detection accuracy.
(4) comprising signal characteristic detecting means for detecting the signal characteristic of the second digital signal;
The histogram detection circuit according to (3), wherein the second histogram creation means creates the second histogram according to the signal characteristic detected by the signal characteristic detection unit.
(5) an imaging device having a plurality of light receiving elements each composed of a first light receiving region and a second light receiving region each having different sensitivities;
A first analog signal composed of a plurality of output signals output from each of the first light receiving elements, and a second analog signal composed of a plurality of output signals output from each of the second light receiving elements. An A / D conversion circuit for A / D converting each analog signal to generate a first digital signal and a second digital signal;
A signal processing circuit that performs a plurality of processes on the first digital signal and the second digital signal to generate image data;
The signal processing circuit creates a first histogram creating means for creating a first histogram showing a signal intensity distribution of the first digital signal, and a second histogram showing a signal intensity distribution of the second digital signal. A histogram detection circuit for generating the first histogram and the second histogram with different gradation detection accuracy, respectively.
The image processing apparatus, wherein the signal processing circuit optimizes and synthesizes the first digital signal and the second digital signal according to the first histogram and the second histogram.
(6) The histogram detection circuit includes signal characteristic detection means for detecting a signal characteristic of the second digital signal,
The imaging apparatus according to (5), wherein the second histogram creating unit creates the second histogram according to the signal characteristic detected by the signal characteristic detection unit.
[0012]
According to the above, the gradation detection accuracy of the histogram is different between the first histogram and the second histogram. The gradation detection accuracy of the second histogram is determined according to the signal characteristics of the second digital signal. Therefore, when reducing the gradation resolution of the histogram, the dynamic range to be detected is changed according to the characteristics of the detection signal, thereby simplifying the circuit configuration and suppressing an increase in the cost of hardware such as an LSI. The detection accuracy can be improved.
[0013]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a digital camera which is an example of an imaging apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings.
[0014]
FIG. 1 is a block diagram showing a digital camera 1 of the present embodiment. The digital camera 1 includes an optical system 10, an imaging unit 20, an A / D imaging processing unit 30, a drive control unit 40, a signal processing circuit 50, an image data recording unit 60, a display unit 70, and a control unit. 80, a memory 90, and an image recording / reproducing unit 100.
[0015]
The optical system 10 includes a single focus lens configured by combining a single lens or a plurality of lenses to which light from the outside is incident, a lens such as a zoom lens, an aperture for limiting the amount of light from the outside, and the like. It is a light collecting member that focuses incident light from the outside onto the image pickup device of the image pickup unit 20 at the subsequent stage. In the optical system 10, in accordance with the drive signal from the drive control unit 40, focus adjustment according to the distance between the subject and the camera, adjustment of the angle of view by zoom operation, and the like are performed.
[0016]
The imaging unit 20 is arranged at the subsequent stage of the optical system 10 and photoelectrically converts light incident in the digital camera 1 via the optical system 10 for each incident position and extracts it as an electrical signal. In the present embodiment, the imaging unit 20 includes a solid-state imaging device 21 called a W honeycomb CCD, and can acquire two types of signals having different sensitivities by one imaging.
[0017]
FIG. 2 is a partially enlarged plan view of the solid-state image sensor 21. The solid-state imaging device 21 is a so-called honeycomb-structured solid-state imaging device, and is disposed on the surface of a semiconductor substrate in a row direction (direction indicated by an arrow X) and a column direction (direction indicated by an arrow Y) perpendicular thereto. Light receiving element 22 (only a part is numbered in the figure), a vertical transfer unit 24, a horizontal transfer unit 26, and an output unit 28. Among the plurality of light receiving elements 22, the odd-numbered ones are shifted in the column direction by about ½ of the column direction pitch between the adjacent light receiving elements with respect to the even-numbered ones. Are arranged so as to be shifted in the row direction by about ½ of the row direction pitch between the light receiving elements adjacent to the light receiving elements in the even-numbered rows. In FIG. 2, the light receiving elements of 5 rows and 8 columns are shown, but actually, more light receiving elements are provided.
[0018]
The light receiving element 22 generates and accumulates signal charges corresponding to the amount of incident light, and is, for example, a photodiode. Each of the light receiving elements 22 has a main region m having a relatively large light receiving area and a sub region s having a relatively small light receiving area. The main region m and the sub region s generate and accumulate signal charges corresponding to light having a predetermined spectral sensitivity. The sub-region s has a light receiving sensitivity lower than that of the main region m, and the amount of accumulated signal charges is smaller than that of the main region m.
[0019]
Above each light receiving element 22, any one of red (indicated by R), green (indicated by G), or blue (indicated by B) color filter (not shown) is provided. . Here, the main region m and the sub region s in the same light receiving element 22 are provided with the same color filter. The light receiving regions m and s of each light receiving element 22 generate and accumulate signal charges corresponding to the light of each color.
[0020]
The vertical transfer unit 24 reads signal charges from the respective light receiving elements 22 and transfers them in the column direction. The vertical transfer units 24 are provided on the side of the light receiving elements 22 corresponding to the columns. The vertical transfer unit 24 reads the signal charges in the main region m and the signal charges in the sub region s independently in the column direction and transfers them to the horizontal transfer unit 26. The horizontal transfer unit 26 transfers the signal charges from the plurality of vertical transfer units 24 and transfers the transferred signal charges in the row direction. The output unit 400 outputs a voltage signal corresponding to the signal charge amount transferred from the horizontal transfer unit 26. Hereinafter, a plurality of voltage signals generated based on a plurality of charge signals from a plurality of main regions m are collectively generated based on a high sensitivity signal H and a plurality of charge signals from a plurality of subregions s. The plurality of voltage signals are referred to as low sensitivity signals L. Both the high sensitivity signal H and the low sensitivity signal L are sent to the A / D imaging processing unit 30.
[0021]
The A / D imaging processing unit 30 performs predetermined analog signal processing such as bias adjustment and gain adjustment of the high sensitivity signal H and the low sensitivity signal L, and then performs A / D conversion to digitize each signal. is there. Here, high-sensitivity signal H and the low-sensitivity signal L is converted into a digital signal by the independently 12-bit gradation, high-sensitivity digital signal H _a and the low-sensitivity digital signal H _b is generated. High-sensitivity digital signal H _a and the low-sensitivity digital signal H _b generated is sent to the signal processing unit 50.
[0022]
The drive control unit 40 drives the lens of the optical system 10, the transfer timing signal for transferring the charge signal generated by the imaging unit 20, and the driving timing of the A / D imaging processing unit 30 based on an instruction from the control unit 80. Etc. to control each part. The focus control and the control at the time of imaging are determined by the drive control unit 40 based on an instruction from the control unit, and each signal generated after the imaging is sequentially sent to the signal processing unit 50.
[0023]
The signal processing unit 50 performs an optimization process of a high-sensitivity digital signal H _a and the low-sensitivity digital signal H _b, synthesized by the one optimized sensitivity digital signal H _a and the low-sensitivity digital signal H _b Image data is generated. The signal processing unit 50 is connected to the bus 110, and the generated image data is recorded on the recording medium 105 connected to the image recording / playback unit 100 via the compression processing unit 60 or the bus 110. The signal processing unit 50 will be described in detail later.
[0024]
The compression processing unit 60 is connected to the bus line 110 and performs compression processing according to a predetermined image format such as TIFF or JPEG on the image data generated by the signal processing unit 50 to change the data amount of the image data. Let Whether or not to perform data compression depends on user settings. When data compression processing is not performed, image data may be directly written from the signal processing unit 50 to the recording medium 105.
[0025]
The display unit 70 is connected to the bus line 110 and has a display made of liquid crystal or the like. The display unit 70 displays on the display a subject image acquired before imaging and image data after imaging, and displays settings of the digital camera and the like to the user.
[0026]
The control unit 80 includes a CPU that performs high-speed calculation, performs various calculations based on an imaging instruction from a shutter button (not shown), and performs overall control of the entire imaging operation of the entire imaging operation of each unit of the digital camera 1. Further, at the time of imaging, it functions as a part of specific control such as control of auto focus, automatic exposure control, auto white balance, etc. and calculation assistance of each processing unit as necessary.
[0027]
The memory 90 is a built-in recording unit that stores various setting values of the digital camera and temporarily records the calculation results calculated by each unit. Each part of the digital camera exchanges data through a memory if necessary.
[0028]
The image recording / playback unit 100 can be connected to an external recording medium 105 such as smart media or xD-Picture Card, and records and saves the created image data in the external recording medium 105. The image recording / playback unit 100 is configured to be able to display data on the display of the display unit 70 by reading data from the external recording medium 105 and outputting the data to the display unit 70 during playback of the image.
[0029]
FIG. 3 is a block diagram showing details of processing in the signal processing unit 50. The signal processing unit 50 includes a histogram detection unit 51, a gain adjustment unit 53, a gradation correction unit 54, a synthesis processing unit 55, and a YC processing unit 56. Both high-sensitivity digital signal H _a and the low-sensitivity digital signal H _b inputted to the signal processing unit 50 is sent to the histogram detection unit 51 and the gain adjusting unit 53.
[0030]
Histogram detector 51, respectively a high-sensitivity digital signal H _a and low-sensitivity digital signal L _a by multiplying R, G, for each color of B, and calculates the Color histogram and Color of average integration value. The histogram detection unit 51 includes a high sensitivity signal integration unit 151, a signal level detection unit 152, and a low sensitivity signal integration unit 153.
[0031]
The gain adjustment unit 53 is a multiplication circuit that performs gain adjustment of a digital signal such as white balance adjustment. The white balance adjustment is a process of correcting the color balance according to the color (color temperature) of the light source so that a white subject is imaged as white. In general, when the white balance is not adjusted, due to the difference in color temperature, an image obtained under a fluorescent lamp tends to be greenish, and an image obtained under an incandescent lamp tends to be reddish.
[0032]
Gain adjusting unit 53, in particular white balance adjustment, R obtained by the signal integration unit 51 for the high-sensitivity digital signal _{H a,} G, the integrated value _R sum of B _signal, G _{sum, B sum} based on , G _sum and other signals R _sum , B _sum R _sum / G _sum , B _sum / G _sum are calculated. Then, the color level between RGB is adjusted by multiplying the signals R and B by the gain correction coefficient for each color calculated according to the obtained ratio. Here, the gain correction coefficients obtained from the high-sensitivity digital signal H _a, the high-sensitivity digital signal H _a R, B signals and low-sensitivity digital signal L _a of R, each digital signal H _a fit over the B _signal, L to adjust the white balance of _a.
[0033]
Tone correction unit 54, gradation correction 8bit from 12bit independently using a predetermined coefficient the gradation of the high-sensitivity digital signal after the gain adjustment H _a and low-sensitivity digital signal L _a. For this gradation correction, a high sensitivity digital signal gradation correction LUT and a low sensitivity digital signal gradation correction LUT in a matrix format are used. LUT for gradation correction, the high-sensitivity digital signal H _a and low-sensitivity digital signal L _a is input, and outputs the data according to the value of the signal, performs tone conversion, high sensitivity after tone correction The digital signal _Hb and the low sensitivity digital signal _Lb are output. These gradation correction LUTs may be stored in advance in the memory area of the control unit 80 or the memory 90, or may be adaptively determined by calculations performed based on user settings, digital signal histograms, and the like. It may be required.
[0034]
Synthesis processing unit 55, a high-sensitivity digital signal H _b and the low-sensitivity digital signal L _b of the gradation-corrected synthesized for each R, G, B, R signal, G signal, consisting of B signal composite digital signal S Is generated. Here, as an example, the synthesis is performed using the following equation.
[0035]
[Expression 1]

[0036]
Here, in Expressions (1), (2), and (3), the parameter p is a gain for the entire added digital signal, and thereby controls the dynamic range for the added data. The smaller the value, the wider the dynamic range, and vice versa. Specifically, this value is changed according to the scene, such as 0.8 for high-contrast scenes (midsummer sunny weather, etc.), 0.86 for cloudy or shaded, 0.9 for indoor fluorescent lights, etc. By doing so, it is possible to effectively use the 8-bit gradation value.
[0037]
Parameter th is representative of the high-sensitivity digital signal H _b to one of the ratios extent mix the low-sensitivity digital signal L _b, according to the signal strength of the high-sensitivity digital signal H _b and the low-sensitivity digital signal L _b Can be determined in advance. In actual synthesis processing, the high-sensitivity digital signal _Hb , the parameters th and p, and h_gain and l_gain determined according to the above equations (2) and (3) are calculated, and the high-sensitivity digital signal _Hb and low-sensitivity are calculated. into a digital signal _{L b,} respectively multiplied by h_gain and L_gain, by adding the respective signals to generate a composite digital signal S. As an example, it is possible to execute processing by fixing th and changing p according to the scene.
[0038]
The YC processing unit 56 is a circuit that performs a predetermined synchronization process and creates a luminance signal Y and chroma signals Cr and Cb from the R signal, G signal, and B signal of the synthesized signal S. In addition, the YC processing unit 56 performs various processes such as noise reduction on the generated luminance signal Y and chroma signals Cr and Cb to generate uncompressed image data. The generated image data is compressed by the compression processing unit 60 when compressed, or directly sent to the image recording / reproducing unit 100 and recorded on the recording medium 105 when not compressed. .
[0039]
Hereinafter, the histogram detection unit 51 will be described in detail.
[0040]
High-sensitivity signal integration section 151 is for the high-sensitivity digital signal H _a integrating R, G, for each color of B, respectively, to create a histogram and the cumulative frequency of each color. Here, the high sensitivity signal accumulator 151, 12 a high-sensitivity digital signal H _a bit gradation shifted by 6 bits, a histogram as data of 6 bit conversion (64 gradations). Figure 4 shows an example of a histogram created for G sensitive digital signal H _a.
[0041]
Signal level detecting unit 152 is for calculating the color R of the low-sensitivity digital signal L _a, G, the characteristics of the B signal level, and outputs the low-sensitivity signal integration section 153 provided on the subsequent stage. Here, the characteristic of the signal level detected by the signal level detection unit 152 is a rough distribution state of the signal level, for example, signals of each color R, G, B represented by 12-bit gradation are high. It detects how much signal components are included on the write side.
[0042]
Specifically, the signal level detection unit 152 indicates that each color signal expressed in 12-bit gradation is (1) a component of gradation levels 2048 to 4095 (the 12th bit is 1), and (2) gradation level 1024 to The number of components including 2047, (3) components of gradation levels 512 to 1023, and 511 or less are counted. Hereinafter, a specific calculation method will be described based on the flowchart of FIG.
[0043]
First, the signal level detector 152, upon receiving the low-sensitivity digital signal L _a (step S1), 12 bit of the signal components of each color of the received signal to determine whether it is 1 (step S2) . If the 12th bit is 1 in step S2, the component is counted as having any size from 2048 to 4095 (maximum 4095) (step S3). Is checked whether the 11th bit is 1 (step S4).
[0044]
If the 11th bit is 1 in step S4, the component is counted as having any one of 1024 to 2047 (maximum 2047) (step S5) and is 0. Is checked whether the 10th bit is 1 (step S6).
[0045]
In step S6, when the 10th bit is 1, the component is counted as having any one of 512 to 1023 (maximum 1023) (step S7), and when it is 0 The component is counted as having any size of 511 or less (maximum 511) (step S8). After the counting for one component is completed, it is confirmed whether or not there is a next component. If there is a next component, the process returns to step S2 and the same counting is performed for the next component (step S9). If there is no next component, the counting ends.
[0046]
When the count ends, the signal level detection unit 152 checks the count number for each component and determines the signal level (steps S10 and S11). First, it is confirmed whether or not the count number for the components of the gradation levels 2048 to 4095 is equal to or greater than a predetermined value, and if it is equal to or greater than the predetermined value, it is assumed that the signal level is 1. Here, if it is smaller than the predetermined value, it is confirmed whether or not the components of 1024 to 2047 are not less than a predetermined value, and if it is not less than the predetermined value, it is assumed that the signal level is 2. Further, if it is smaller than the predetermined value, it is checked whether or not the components 512 to 1023 are equal to or higher than a predetermined value. If it is equal to or higher than the predetermined value, the signal level 3 is assumed. Assume that it is level 4.
[0047]
Low-sensitivity signal integration section 153, based on the signal level checked by the signal level detector 152, a low-sensitivity digital signal L _a by multiplying R, G, for each color of B, respectively, histograms and cumulative frequency for each color Is to create. Here, according to the signal level, means changing the number of bits to be shifted in accordance with high-sensitivity digital signal H _a 12-bit gradation level.
[0048]
FIG. 6 is a diagram showing the relationship between the signal level and the gradation division at the time of creating the histogram. In the case of the signal level 1, the low-sensitivity signal integrating unit 153 shifts the 12-bit gradation signal by 6 bits and integrates the signal component to express the 12-bit dynamic range width with 64 gradations. . On the other hand, in the case of signal level 2, the 12-bit gradation signal is shifted by 5 bits and the 12th bit is ignored, and the signal components are integrated, so that the 11-bit dynamic range width can be reduced to 64 gradations. Express. Similarly, in the case of the signal level 3, the signal component is integrated by shifting the 12-bit gradation signal by 4 bits and ignoring the 12th and 11th bits, and the 10-bit dynamic range width is obtained. Expressed in 64 gradations. Finally, in the case of the signal level 4, the 12-bit gradation signal is shifted by 3 bits, and the signal components are integrated ignoring the 10th to 12th bits, so that the 9-bit dynamic range width is increased to the 64th floor. Express in key.
[0049]
In the above description, whether or not the signal level of the low-sensitivity signal is 2047 or more (whether or not the signal level is 1) is also confirmed. However, the signal level of the low-sensitivity signal is effective for the signal level 1. In the case of not having a numerical value, the determination of whether or not the signal level is 1 may be omitted, and the histogram may be created after determining which of the signal levels 2 to 4 the low sensitivity signal belongs to. Good. In general, in the case of the W honeycomb CCD, the signal level of the low sensitivity signal is about ¼ of the signal level of the corresponding high sensitivity signal, and therefore confirmation of the signal level 1 is unnecessary.
[0050]
In the above description, the histogram creation range is changed depending on whether or not the number of components corresponding to each signal level is greater than or equal to a predetermined value. However, the present invention is not limited to this. May be shifted by 6 bits to create a histogram, determine an optimum signal level from the created histogram, and recreate the histogram.
[0051]
Incidentally, detected by the histogram detector 51, the histogram of the high-sensitivity signal _{H a} and low-sensitivity signals _{L a,} the gain adjustment section 53, the gradation correction section 54, a gain in the synthesis processing section 55 or the like, the correction coefficient, H_gain And l_gain etc. are used.
[0052]
As described above, according to the present embodiment, for the low-sensitivity signal L _a, the histogram is created in response to the signal characteristics. Generally low-sensitivity signal L _a has a narrow dynamic range width as compared with the high-sensitivity signal H _a, 12 bit, contemplated many cases that do not contain an effective signal to the 11 bit or the like. Therefore, by changing the dynamic range to be detected in accordance with the characteristics of the signal, the dynamic range width narrower than the 12-bit dynamic range width is expressed in 64 gradations, so that the 12-bit dynamic range width is directly 64. It is possible to create a histogram that better expresses the signal characteristics of the low-sensitivity signal as compared to the case where gradation is expressed.
[0053]
In addition, according to the present embodiment, since the detection accuracy is improved by changing the dynamic range to be detected, no special circuit configuration or storage device is required. Therefore, it is possible to improve detection accuracy while suppressing an increase in cost when hardware such as LSI is used.
[0054]
Although the present embodiment has been described as an image pickup apparatus including an image pickup device called a W honeycomb CCD, the present invention is not limited thereto, and the signal processing circuit described above can be applied to various image pickup apparatuses capable of acquiring a high sensitivity signal and a low sensitivity signal. May be applied.
[0055]
【The invention's effect】
According to the present invention, when reducing the gradation resolution of a histogram, the dynamic range to be detected is changed according to the characteristics of the detection signal, thereby simplifying the circuit configuration and reducing the cost of hardware such as an LSI. It is possible to improve the detection accuracy of the histogram while suppressing the increase.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a digital camera according to an embodiment.
FIG. 2 is a partially enlarged plan view of a solid-state image sensor.
FIG. 3 is a block diagram showing details of processing in a signal processing unit.
Figure 4 shows an example of a histogram created for G sensitive digital signal H _a.
FIG. 5 is a flowchart showing a histogram calculation process.
FIG. 6 is a diagram illustrating a relationship between a signal level and gradation division when creating a histogram.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Digital camera 10 Optical system 20 Image pick-up part 21 Solid-state image pick-up element 22 Light receiving element 24 Vertical transfer path 26 Horizontal transfer path 28 Output part 30 A / D image pick-up process part 40 Drive control part 50 Signal process part 51 Histogram detection part 52 Optimization synthesis | combination Processing circuit 53 Gain adjustment unit 54 Gradation correction unit 55 Composition processing unit 56 YC processing unit 60 Compression processing unit 70 Display unit 80 Control unit 90 Memory 100 Image recording / reproducing unit 105 Recording medium

Claims (6)

Creating a first histogram showing a signal intensity distribution of the first digital signal;
Creating a second histogram showing a signal intensity distribution of a second digital signal corresponding to the first digital signal;
The histogram calculation method, wherein the first histogram and the second histogram are created with different gradation detection accuracy. A signal characteristic detection step of detecting a signal characteristic of the second digital signal;
The histogram detection method according to claim 1, wherein the second histogram is created according to the signal characteristic detected by the signal characteristic detection unit. First histogram creating means for creating a first histogram showing the signal intensity distribution of the first digital signal, and a second histogram showing the signal intensity distribution of the second digital signal corresponding to the first digital signal Second histogram creating means for creating,
The histogram calculation circuit, wherein the first histogram and the second histogram are created with different gradation detection accuracy. Comprising signal characteristic detecting means for detecting a signal characteristic of the second digital signal;
4. The histogram detection circuit according to claim 3, wherein the second histogram creation means creates the second histogram according to the signal characteristic detected by the signal characteristic detection unit. An imaging device having a plurality of light receiving elements each composed of a first light receiving region and a second light receiving region each having a different sensitivity;
A first analog signal composed of a plurality of output signals output from each of the first light receiving elements, and a second analog signal composed of a plurality of output signals output from each of the second light receiving elements. An A / D conversion circuit for A / D converting each analog signal to generate a first digital signal and a second digital signal;
A signal processing circuit that performs a plurality of processes on the first digital signal and the second digital signal to generate image data;
The signal processing circuit creates a first histogram creating means for creating a first histogram showing a signal intensity distribution of the first digital signal, and a second histogram showing a signal intensity distribution of the second digital signal. A histogram detection circuit for generating the first histogram and the second histogram with different gradation detection accuracy, respectively.
The image processing apparatus, wherein the signal processing circuit optimizes and synthesizes the first digital signal and the second digital signal according to the first histogram and the second histogram. The histogram detection circuit includes signal characteristic detection means for detecting a signal characteristic of the second digital signal,
The imaging apparatus according to claim 5, wherein the second histogram creating unit creates the second histogram according to the signal characteristic detected by the signal characteristic detection unit.

Images