JP3643042B2 - Image processing apparatus, image processing method, and recording medium - Google Patents

Description

【００６９】
ただし、数１式中のＡは255/(255cγ)から計算される定数であり、この例ではＡ＝21.1となる。このようにして計算されたf(X1)，f(X2)・・・を再びプロットしたものが図９（ｂ）である。
【００７０】
次に、各画素についてエッジ強調処理を行う。ここで、エッジ強調処理は図８に示した空間フィルタを使うので、数２式に示す計算式が与えられる。
【００７１】
【数２】

【００７２】
数２式を用いてエッジ強調計算を行うと、f(X2)=133.2，f(X3)=98.8，f(X4)=202.0，f(X5)=167.6となる。このようにして計算されたf(X2)，f(X3)・・・を再びプロットしたものが図９（ｃ）である。
【００７３】
＜４−２ エッジ強調処理の後に階調変換を行った場合＞
次に、エッジ強調処理の後に階調変換を行う場合の処理結果を計算する。図１０（ａ）は、処理前のエッジ形状であり、横軸に画素位置、縦軸にデジタル値をとったものである。同様に、各画素位置X1，X2，X3，X4，X5，X6におけるデジタル値を、それぞれf(X1)，f(X2)，f(X3)，f(X4)，f(X5)，f(X6)とし、初期状態のf(X1)，f(X2)，f(X3)，f(X4)，f(X5)，f(X6)を、それぞれd1,d2,d3,d4,d5,d6(d1=d2=d3=60,d4=d5=d6=100)とする。
【００７４】
そして、各画素についてエッジ強調処理を行う。ここで、エッジ強調処理は図８に示した空間フィルタを使うので、数３式に示す計算式が与えられる。
【００７５】
【数３】

【００７６】
数２式を用いてエッジ強調計算を行うと、f(X2)=60，f(X3)=20，f(X4)=140，f(X5)=100となる。このようにして計算されたf(X2)，f(X3)・・・を再びプロットしたものが図１０（ｂ）である。
【００７７】
次に、階調変換処理を行う。ここで、変換特性はcγ＝0.45であるので、数４式に示した変換式を用いて、各画素について階調変換計算を行うと、f(X2)=133.2，f(X3)=81.2，f(X4)=195.0，f(X5)=167.6となる。ただし、数１式中のＡ＝255/(255cγ)＝21.1である。
【００７８】
【数４】

【００７９】
このようにして計算されたf(X2)，f(X3)・・・・・・を再びプロットしたものが図１０（ｃ）である。
【００８０】
＜４−３ 処理結果の比較＞
このように、階調変換処理とエッジ強調処理の処理順序を入れ換えて処理をおこなった場合には、図９（ｃ）、図１０（ｃ）に示すように、いずれの場合にもエッジ強調が行われているが、エッジ強調の度合が異なる。
【００８１】
つまり、エッジ立ち上がり部のコントラストを数５式で示す計算式を用いて求めると、階調変換処理を先に行った場合（上述４−１）には、Ｃｏ＝0.34、エッジ強調処理を先に行った場合（上述４−２）には、Ｃｏ＝0.41となり、Ｃｏの値が大きい程コントラストが高くなったことを示すので、エッジ強調処理を先に行った場合の方がエッジのコントラストが高くなったことがわかる。つまり、階調変換の後、エッジ強調を行ったものと、エッジ強調の後、階調変換を行ったものでは、エッジの立ち上がりの形が異なるため、処理順序によってエッジ強調度合を異ならせることができるのである。
【００８２】
【数５】

【００８３】
このような理由から、上述した実施の形態においては、「ポートレート」モードなど、エッジ強調を弱くしたい場合には、階調変換処理を先に行い、「マクロ」モードなど、エッジ強調を強くしたい場合には、エッジ強調処理を先に行うようにしているのである。
【００８４】
本実施の形態においては、階調変換の変換特性であるγ値（cγ）をcγ＜１としているので、上述した計算で示したように、エッジ強調処理を先に行うことで、エッジ強調度合を強くすることができるが、逆に、cγ＞１の関係にある場合には、エッジ強調度合を強くするためには、階調変換処理を先に行うような制御を行うようにすればよい。cγ＞１である場合の計算も同様な計算により求められるが、ここでは省略する。
【００８５】
このことは、階調変換の特性がcγ＜１であっても、cγ＞１であっても、階調変換とエッジ強調の処理順序を切り換えることで、エッジ強調度合を変化させられることを意味している。
【００８６】
以上、＜４−１＞および＜４−２＞の比較は、１次元の信号（エッジ形状）を例とした計算処理である。これを上述した実施の形態におけるデジタルカメラ１のＧデータ（２次元信号）に適用すれば、同様の効果が得られることとなる。つまり、Ｇデータに対する階調変換特性γ値（cγ）の値を１以外の値（たとえば0.45）とし、階調変換とエッジ強調の処理順序を切り換えることにより、Ｇデータの強調度合を変化させることができる。そして、図４で示したように、Ｒ、Ｂデータに対しても、強調度合の異なるＧデータの高周波成分が加算されることとなるので、出力画像の強調度合を変化させることができるのである。
【００８７】
本実施の形態においては、階調変換とエッジ強調の処理を切り換えることにより、エッジ強調の度合を変更することを特徴としているが、従来から行われていたエッジ強調の度合の切り換え方法と併用するようにしてもよい。
【００８８】
エッジ強調部４１におけるゲイン調整部４１３のゲイン値を複数種保有しておき、このゲイン値を切り換えながらエッジの強調度合を変化させる方法と、階調変換とエッジ強調の処理順序を切り換える方法を併用するのである。これにより、ゲイン値の種類に対して２倍の種類のエッジ強調度合を切り換えることが可能となる。
【００８９】
｛５．変形例｝
上述した、実施の形態の中で、デジタルカメラ１が備える各処理部（ＷＢ補正部３３、画素補間部３４、エッジ強調調整部３６など）は、ハードウェアによる処理で実現されてもよいし、ソフトウェアによる処理で実現されてもよい。
【００９０】
そして、各処理をソフトウェア処理により実現することで、本実施の形態にかかるデジタルカメラ１において実行したような処理を、一般のコンピュータにおいても実行させることが可能である。そして、このような処理を実行するプログラムを記録媒体として配布することが可能である。
【００９１】
図１１は、コンピュータ５の概観図である。コンピュータ５は、たとえば、パーソナルコンピュータであり、フロッピディスクやＣＤ−ＲＯＭ等の記録媒体５１を読み込むことが可能である。
【００９２】
そして、記録媒体５１に本実施の形態にかかるデジタルカメラ１と同様の画像処理を実行可能なプログラム６を記録しておくことにより、コンピュータ５は、プログラム６をハードディスクにインストールすることにより実行することが可能である。もしくは、記録媒体５１に記録されたプログラム６を直接実行するようにしてもよい。
【００９３】
図１２は、プログラム６の処理フローである。プログラム６は、まず、データの読み出し（ステップＳ２０１）を行う。たとえば、プログラム６は、所定のディレクトリに格納されている画像データの読み出しを行う。ここでの画像データは、ベイヤー配列のＣＣＤから出力されたＲＧＢデータであるものとする。
【００９４】
次に、プログラム６は、ＷＢ補正処理（ステップＳ２０２）、画素補間処理（ステップＳ２０３）を行う。
【００９５】
そして、図３で示した実施の形態においてエッジ強調調整部３６に相当する機能が実行されて処理順序が決定される（ステップＳ２０４）。ここでは、たとえば、画像ヘッダ情報に含まれている撮影モード情報や、他の定義ファイルなどから撮影モード情報を読み取るものとする。
【００９６】
そして、階調変換処理を先に行う場合（ステップＳ２０５でYes）には、階調変換処理（ステップＳ２０６）を行い、続いてエッジ強調処理（ステップＳ２０７）を行う。ただし、階調変換処理の変換特性はcγ＜１であり、前述した実施の形態において図６で示した場合と同様な関係で階調変換処理とエッジ強調処理の順序が切り換えられる。
【００９７】
階調変換処理を先に行わない場合（ステップＳ２０５でNo）には、エッジ強調処理（ステップＳ２０８）を行い、続いて階調変換処理（ステップＳ２０９）を行う。このように、いずれかの処理順序により、階調変換処理とエッジ強調処理とが施された画像データが出力されると、画像圧縮処理を行い（ステップＳ２１０）、ハードディスクや記録媒体等にデータの書込みを行う（ステップＳ２１１）。
【００９８】
【発明の効果】
以上説明したように、請求項１ないし請求項３に記載の発明では、階調変換処理とエッジ強調処理の処理順序を切り換える方法であり、エッジ強調処理の変換特性を複数保有することなく、エッジの強調度合を変化させることが可能である。
【０１０９】
請求項４記載の発明では、請求項１ないし請求項３に記載の方法を実行可能なプログラムを配布することにより、記録媒体読み取り可能なコンピュータにおいて、本発明の画像処理方法が実行可能となる。
【図面の簡単な説明】
【図１】本実施の形態にかかるデジタルカメラの概観斜視図である。
【図２】デジタルカメラの背面図である。
【図３】デジタルカメラの内部ブロック構成図である。
【図４】エッジ強調部の処理ブロック図である。
【図５】コアリング部およびゲイン調整部により制御される出力レベルを示す図である。
【図６】デジタルカメラが有する撮影モードと、エッジ強調の度合と、処理順序の関係を示す図である。
【図７】デジタルカメラにおけるエッジ強調調整処理を示すフローチャートである。
【図８】エッジ強調部における空間フィルタを示す図である。
【図９】階調変換処理を先に行った場合の、各画素の演算結果を示す図である。
【図１０】エッジ強調処理を先に行った場合の、各画素の演算結果を示す図である。
【図１１】コンピュータの概観図である。
【図１２】コンピュータ上で実行されるエッジ強調調整処理を示すフローチャートである。
【符号の説明】
１ デジタルカメラ
１６ 設定キー
３４ 画素補間部
３５ シーン判定部
３６ エッジ強調調整部
４０ 処理順序制御部
４１ エッジ強調部
４２ 階調変換部[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus and method for performing gradation conversion and edge enhancement.
[0002]
[Prior art]
In an image processing apparatus such as a digital camera, gradation conversion processing and edge enhancement processing are performed on input image data. The gradation conversion process is a conversion process according to the input / output characteristics of the monitor. The edge enhancement processing is processing for sharpening or enhancing an image, but it is desirable to perform appropriate edge enhancement depending on the characteristics of the image.
[0003]
Conventionally, in an image processing apparatus such as a digital camera, first, gradation conversion processing is performed on input image data, and then edge enhancement processing is performed on the image data subjected to gradation conversion. .
[0004]
A general edge enhancement process in the image processing apparatus will be described. Among the input RGB data, HPF (High Pass Filter) is applied to G data to extract high frequency components. A component below a predetermined level is removed (noise removal) from the high-frequency component extracted in the HPF. Then, gain adjustment is performed on the high-frequency component from which noise has been removed.
[0005]
The high frequency component of the G data whose gain is adjusted is added to the original G data, and the G data after the addition is output. Further, the high frequency component of the G data whose gain is adjusted is added to the B data and the R data, and the added B and R data are output.
[0006]
Edge enhancement processing is performed by such processing. Conventionally, in order to switch the degree of edge enhancement, a plurality of gain values, which are conversion characteristics of the gain adjusting unit, are held. Then, by switching a plurality of gain values, the conversion characteristics in the gain adjusting unit are switched, and the degree of edge is changed according to the characteristics of the image.
[0007]
[Problems to be solved by the invention]
However, in the method of switching the degree of edge enhancement by switching the gain, it is necessary to store a plurality of gain values in the memory, and there is a problem that extra memory capacity is used.
[0008]
Accordingly, an object of the present invention is to provide an image processing apparatus and the like that can switch the degree of edge enhancement even when a plurality of gain values are not held.
[0009]
[Means for Solving the Problems]
  In order to solve the above problems, the invention of claim 1A processing method for performing gradation conversion processing and edge enhancement processing on an image, wherein the gradation conversion processing and edge enhancement processing for the processing target image are performed according to a required level of the degree of edge enhancement for the processing target image. A processing order determination step for determining a processing order; and a step of sequentially executing a gradation conversion process and an edge enhancement process for the processing target image according to the order determined in the processing order determination step. To do.
[0020]
  Claim2The invention of claim1In the image processing method described in (1), the conversion characteristic of the gradation conversion process is expressed by a γ value, and the γ value is a value other than 1.
[0021]
  Claim3The invention of claim1Or claims2The image processing method according to claim 1, further comprising a scene determination step for determining a scene of the processing target image, wherein the processing order determination step is based on a determination result of the scene determination step. The order is determined.
[0022]
  Claim4The present invention is a computer-readable recording medium computer,1Or claims3A program for executing the image processing method according to any one of the above is recorded. Claim5According to the invention, a gradation conversion unit that performs gradation conversion of an image, an edge enhancement unit that performs edge enhancement of an image, and a scene determination that determines a scene of an image to be processed by the gradation conversion unit and the edge enhancement unit And means for switching the processing order of the image to be processed by the gradation conversion means and the edge enhancement means based on the determination result of the scene determination means.
[0023]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0024]
{1. Configuration of digital camera}
The image processing apparatus according to this embodiment will be described using a digital camera as an example. FIG. 1 is a front perspective view of the digital camera 1, and FIG. 2 is a rear view of the digital camera 1.
[0025]
A photographing lens 11, a finder window 12 and a flash 13 are provided in front of the digital camera 1, and a subject image input from the photographing lens 11 is formed on a CCD element 14.
[0026]
Further, on the upper surface of the digital camera 1, a release button 15 for giving a shutter instruction, a setting key 16a capable of performing various setting operations such as a shooting mode selection operation, and a liquid crystal panel 17 for displaying the state of the shooting mode and the like are provided. It has been.
[0027]
The side surface of the digital camera 1 is provided with a slot 18 serving as a recording medium insertion slot. By inserting a memory card 20 into the slot 18, it is possible to exchange image data captured by the digital camera 1. .
[0028]
Further, on the back surface 1B of the digital camera 1, a setting key 16b capable of performing various setting operations such as a shooting mode selection operation, a liquid crystal monitor 19 for displaying a preview of a captured image, various setting menus, and the like. And are provided. In the following description, the setting keys 16a and 16b will be collectively referred to as the setting key 16.
[0029]
Next, the internal configuration of the digital camera 1 will be described with reference to FIG. The CCD element 14 is an imaging element in which a primary color or complementary color filter is arranged for each pixel, and photoelectrically converts a light image of a subject into an RGB image signal and outputs it. The color filters of the CCD element 14 in the present embodiment are arranged in a Bayer array.
[0030]
The A / D converter 31 converts the electrical signal of each pixel output from the CCD element 14 into a digital signal. The black level correction unit 32 corrects the density part of the pixel signal to the reference black level. The white balance (WB) correction unit 33 performs level conversion of RGB pixel signals in order to correct a color balance shift due to the spectrum of the photographing light source. Then, the pixel interpolation unit 34 interpolates the missing color pixels and outputs RGB three-color electrical signals for each pixel.
[0031]
The edge enhancement adjustment unit 36 is an image processing unit that includes a gradation conversion unit 42 and an edge enhancement unit 41 and is a feature of the present invention.
[0032]
The gradation conversion unit 42 has a gradation conversion LUT for each color of RGB, and performs gradation conversion of RGB data. Assuming that the image is observed on the monitor, the reverse characteristic of the input / output characteristic of the monitor is corrected. Here, the gradation conversion LUT is characterized by a γ value that is a conversion characteristic.
[0033]
The edge enhancement unit 41 includes an HPF 411, a coring unit 412, and a gain adjustment unit 413, and performs edge enhancement. The configuration of the edge enhancement unit 41 is shown in FIG.
[0034]
Of the input RGB data, HPF 411 is applied to G data to extract high frequency components. The coring unit 412 removes a component below a predetermined level (noise removal) from the high-frequency component extracted by the HPF 411. Then, the gain adjustment unit 413 performs gain adjustment on the high frequency component output from the coring unit 412.
[0035]
The high frequency component of the G data whose gain is adjusted is added to the original G data, and the G data after the addition is output. Further, the high frequency component of the G data whose gain is adjusted is added to the B data and the R data, and the added B and R data are output. In this way, edge enhancement processing is performed using the high frequency component of G.
[0036]
FIG. 5 shows the processing characteristics of the coring unit 412 and the gain adjustment unit 413 in the present embodiment. In the figure, the horizontal axis is the level when the value output from the HPF 411 is input, and the vertical axis is the output level after the processing in the coring unit 411 and the gain adjustment unit 413 is performed.
[0037]
The coring unit 412 converts the value g of the G data output from the HPF 411 to 0 if it is −V1 <g <V1, and outputs the level as it is if it is any other value. The gain adjusting unit 413 performs conversion with a slope as shown in the figure according to the level output from the coring unit 412. That is, when the level output from the HPF 411 is from -V1 to V1, 0 is output, and when the level is other than that, it is converted by a characteristic (gain value) having a slope as shown in the figure. The level is output.
[0038]
Reference is again made to FIG. The processing order control unit 40 determines the processing order of the gradation conversion unit 42 and the edge enhancement unit 41 based on the determination result of the scene determination unit 35 described later, and the processing order of the gradation conversion unit 42 and the edge enhancement unit 41 is determined. Take control.
[0039]
The recorded image generation unit 37 generates a compressed image for recording, adds image information, and records the image on the memory card 20.
[0040]
{2. Scene determination and processing order control}
Next, the scene determination process in the scene determination unit 35 and the control process in the processing order control unit 40 will be described.
[0041]
The scene determination unit 35 determines an image scene from the shooting mode information, distance information (subject magnification), luminance distribution information, light source detection information, contour detection information, and the like, and outputs the result to the processing order control unit 40.
[0042]
Among these, first, a determination method from the shooting mode information will be described. The determination from the shooting mode information is based on the information when the shooting mode is explicitly selected by an operator operation using the setting key 16 described above.
[0043]
FIG. 6 shows an example of a scene type. In the present embodiment, six scenes (photographing modes) such as “portrait”, “landscape”, “night view”, “slow sync”, “macro”, and “character mode” can be selected.
[0044]
When the operator performs a predetermined operation using the setting key 16, six scenes are displayed on the liquid crystal monitor 19 in a selectable manner. Then, by performing a predetermined operation using the setting key 16, the operator selects one scene.
[0045]
As described above, the scene determination unit 35 can determine the scene by explicitly setting the shooting mode by the operator. Furthermore, in the present embodiment, the scene determination unit 35 includes distance information to the subject (subject magnification), luminance distribution information, light source detection information, contour detection information, and person detection information (collectively, in FIG. It is possible to determine the scene of the image from the various information 38).
[0046]
In the method of determining a scene based on the subject magnification, the scene determination unit 35 acquires information on the shooting magnification of the main subject to be focused from the shooting unit of the digital camera 1, and performs scene determination based on this information. . For example, when the main subject magnification is large, that is, when the distance to the subject is short, it is determined as the “macro” mode or the “portrait” mode according to the distance, and when the subject magnification is small, “ The mode is determined to be “landscape” mode.
[0047]
As another method, an area in which the main subject exists is detected from image edge (edge) detection information and edge centroid detection information. Then, the occupation ratio between the main subject area and the background area is obtained, and when the occupation ratio of the main subject is high, the “macro” mode is determined, when the main subject area is low, the “landscape” mode is determined. It may be. As described above, the scene determination unit 35 has a function of determining a scene (mode) by performing image analysis using various information 38.
[0048]
Then, the processing order determination unit 40 controls the processing order of the gradation conversion unit 42 and the edge enhancement unit 41 based on the scene determination result of the scene determination unit 35. That is, as shown in FIG. 6, each scene (photographing mode) is subjected to edge enhancement with a different degree, and the gradation conversion unit 42 and the edge enhancement unit 41 perform processing according to the degree of edge enhancement. The order is switched.
[0049]
In the present embodiment, edge enhancement of “weak” is applied to the three shooting modes of “portrait”, “night view”, and “slow sync”, and “landscape”, “macro”, “character mode” are applied. "Strong" edge enhancement is applied to the three shooting modes. The reason for associating each scene with the degree of edge enhancement will be described below.
[0050]
The “portrait” mode is a shooting mode in which a person is a subject, and it is generally preferred that the person's skin is finished smoothly and softly.
[0051]
The “landscape” mode is a shooting mode for photographing a distant landscape. Since the subject is fine, it appears that the sharpness has been lowered. Therefore, it is preferable to enhance the contrast.
[0052]
The “night view” mode is a mode for shooting a night view or a person against the background of the night view. Since there are many dark portions, noise removal is necessary.
[0053]
The “slow sync” mode is a flash photography mode with a slow shutter, and noise increases as the exposure time is increased, so noise removal is necessary.
[0054]
The “macro” mode is a mode for taking a close-up of a subject, and it is preferable to make the subject clear.
[0055]
A “character” is an image in which characters such as magazines, books, and white boards are used as subjects. It is preferred to show the characters clearly.
[0056]
Then, the processing order control unit 40 switches the processing order of gradation conversion processing and edge enhancement processing according to the edge enhancement degree.
[0057]
Here, in which processing is performed first, the degree of edge enhancement becomes strong depends on the conversion characteristics in the gradation conversion unit 42. That is, when the γ value (cγ) indicating the conversion characteristic of the gradation conversion unit 42 is cγ <1, the edge enhancement degree can be increased by performing the edge enhancement process first. On the other hand, when cγ> 1, the reverse phenomenon occurs. That is, the edge enhancement degree can be increased by performing the gradation conversion process first.
[0058]
In the present embodiment, the conversion characteristic of the gradation converting unit 42 is cγ <1, and in order to set the degree of edge enhancement to “weak”, the edge enhancement process is performed after the gradation conversion process is performed. In order to set the degree of edge enhancement to “strong”, the tone conversion processing is performed after the edge enhancement processing. This correspondence is also shown in FIG. As a result, the processing order is switched according to the required level of edge enhancement for the processing target image.
[0059]
{3. Processing flow}
In the digital camera 1 configured as described above, a processing flow in the edge enhancement adjustment unit 36 will be described with reference to FIG.
[0060]
First, the RGB image data subjected to the interpolation processing in the pixel interpolation unit 34 is input to the edge enhancement adjustment unit 36 (step S101).
[0061]
Next, the processing order control unit 40 determines the processing order (step S102). As described above, this determination is made based on the selected shooting mode when the shooting mode is explicitly selected by an operator operation. Alternatively, it is automatically determined based on luminance distribution information.
[0062]
When the gradation conversion process is performed first (Yes in step S103), the gradation enhancement process is performed on the input RGB image data (step S104), and then the edge enhancement process is performed (step S105). ). As a result, an image subjected to weak edge enhancement is output (step S108).
[0063]
On the other hand, when the gradation conversion process is not performed first (No in step S103), after the edge enhancement process is performed on the input RGB image data (step S106), the gradation conversion process is performed. It performs (step S107). As a result, an image subjected to strong edge enhancement processing is output (step S108).
[0064]
{4. Processing comparison}
Next, how the degree of edge enhancement changes by switching the processing order of gradation conversion processing and edge enhancement processing will be described by actual calculation using sample data.
[0065]
Here, for simplicity, edge enhancement processing is performed on a one-dimensional edge shape. The image data is 8 bits (0 to 255), and the γ value (cγ) that is the conversion characteristic of the gradation conversion is cγ = 0.45. Edge enhancement is performed by applying a spatial filter as shown in FIG. 8 (here, coring is not performed).
[0066]
<4-1 When edge enhancement is performed after gradation conversion>
First, a processing result when edge enhancement is performed after gradation conversion is calculated. FIG. 9A shows an edge shape before processing, in which the horizontal axis represents the pixel position and the vertical axis represents the digital value. The digital values at the pixel positions X1, X2, X3, X4, X5, and X6 are respectively expressed as f (X1), f (X2), f (X3), f (X4), f (X5), and f (X6). To do. Also, f (X1), f (X2), f (X3), f (X4), f (X5), and f (X6) in the initial state (before gradation conversion and edge enhancement) are set to d1, d2, d3, d4, d5, d6. Here, as an example, consider the case of d1 = d2 = d3 = 60, d4 = d5 = d6 = 100.
[0067]
Then, gradation conversion processing is performed for each pixel. Here, since the conversion characteristic is cγ = 0.45, f (X1) = 133.2 and f (X2) = 133.2 are obtained when gradation conversion calculation is performed for each pixel using the conversion formula shown in Formula 1. , F (X3) = 133.2, f (X4) = 167.6, f (X5) = 167.6, f (X6) = 167.6. However, in Equation 1, d1, d2, d3, etc. correspond to dn, n = 1, 2, 3 ..., and Xn, n = 1, 2, 3 ... .., X1, X2, X3... Correspond (the same applies to equations 2 to 4 described later).
[0068]
[Expression 1]

[0069]
However, A in Equation 1 is a constant calculated from 255 / (255cγ), and in this example, A = 21.1. FIG. 9B is a plot of f (X1), f (X2)... Calculated in this way again.
[0070]
Next, edge enhancement processing is performed for each pixel. Here, since the edge enhancement process uses the spatial filter shown in FIG. 8, the calculation formula shown in Formula 2 is given.
[0071]
[Expression 2]

[0072]
When the edge enhancement calculation is performed using Equation 2, f (X2) = 133.2, f (X3) = 98.8, f (X4) = 202.0, and f (X5) = 167.6. FIG. 9C is a plot of f (X2), f (X3)... Calculated in this way again.
[0073]
<4-2 When gradation conversion is performed after edge enhancement processing>
Next, a processing result in the case of performing gradation conversion after the edge enhancement processing is calculated. FIG. 10A shows an edge shape before processing, in which the horizontal axis represents the pixel position and the vertical axis represents the digital value. Similarly, digital values at pixel positions X1, X2, X3, X4, X5, and X6 are respectively expressed as f (X1), f (X2), f (X3), f (X4), f (X5), and f (X X6), and f (X1), f (X2), f (X3), f (X4), f (X5), f (X6) in the initial state are d1, d2, d3, d4, d5, d6, respectively. (d1 = d2 = d3 = 60, d4 = d5 = d6 = 100).
[0074]
Then, edge enhancement processing is performed for each pixel. Here, since the edge enhancement process uses the spatial filter shown in FIG. 8, the calculation formula shown in Formula 3 is given.
[0075]
[Equation 3]

[0076]
When edge enhancement calculation is performed using Equation 2, f (X2) = 60, f (X3) = 20, f (X4) = 140, and f (X5) = 100. FIG. 10B is a plot of f (X2), f (X3)... Calculated in this way again.
[0077]
Next, gradation conversion processing is performed. Here, since the conversion characteristic is cγ = 0.45, when the gradation conversion calculation is performed for each pixel using the conversion formula shown in Formula 4, f (X2) = 133.2, f (X3) = 81.2, f (X4) = 195.0 and f (X5) = 167.6. However, A = 255 / (255cγ) = 21.1 in equation (1).
[0078]
[Expression 4]

[0079]
FIG. 10C is a plot of f (X2), f (X3)... Calculated in this way again.
[0080]
<4-3 Comparison of processing results>
As described above, when the processing is performed by switching the processing order of the gradation conversion processing and the edge enhancement processing, as shown in FIGS. 9C and 10C, edge enhancement is performed in either case. Although it is done, the degree of edge enhancement is different.
[0081]
In other words, when the contrast of the edge rising portion is obtained using the calculation formula shown in Formula 5, when the gradation conversion process is performed first (4-1 described above), Co = 0.34 and the edge enhancement process is performed first. When it is performed (described above 4-2), Co = 0.41, and the larger the value of Co, the higher the contrast. Therefore, the edge contrast is higher when the edge enhancement processing is performed first. You can see that In other words, the edge emphasis differs depending on the processing order because the edge rise shape differs between the one that has undergone edge enhancement after tone conversion and the one that has undergone tone transformation after edge enhancement. It can be done.
[0082]
[Equation 5]

[0083]
For this reason, in the above-described embodiment, when it is desired to weaken edge enhancement such as in the “portrait” mode, gradation conversion processing is performed first, and edge enhancement such as “macro” mode is desired to be strengthened. In this case, the edge enhancement processing is performed first.
[0084]
In this embodiment, since the γ value (cγ), which is the conversion characteristic of the gradation conversion, is set to cγ <1, as shown in the above-described calculation, the edge enhancement processing is performed first, so that the degree of edge enhancement is achieved. In contrast, when the relationship of cγ> 1 is satisfied, in order to increase the edge enhancement degree, it is only necessary to perform control such that gradation conversion processing is performed first. . The calculation in the case of cγ> 1 is obtained by a similar calculation, but is omitted here.
[0085]
This means that the degree of edge enhancement can be changed by switching the order of gradation conversion and edge enhancement regardless of whether the characteristics of gradation conversion are cγ <1 or cγ> 1. doing.
[0086]
As described above, the comparison between <4-1> and <4-2> is a calculation process using a one-dimensional signal (edge shape) as an example. If this is applied to the G data (two-dimensional signal) of the digital camera 1 in the above-described embodiment, the same effect can be obtained. That is, by changing the gradation conversion characteristic γ value (cγ) value for G data to a value other than 1 (for example, 0.45) and switching the order of gradation conversion and edge enhancement, the degree of enhancement of G data is changed. Can do. As shown in FIG. 4, since the high frequency components of the G data having different emphasis levels are added to the R and B data, the emphasis degree of the output image can be changed. .
[0087]
The present embodiment is characterized in that the degree of edge enhancement is changed by switching between gradation conversion and edge enhancement processing, but it is also used in combination with the conventional edge enhancement degree switching method. You may do it.
[0088]
A combination of a method of holding a plurality of gain values of the gain adjustment unit 413 in the edge enhancement unit 41 and changing the degree of edge enhancement while switching the gain value and a method of switching the processing order of gradation conversion and edge enhancement are used together. To do. As a result, it is possible to switch between two types of edge enhancement degrees with respect to the type of gain value.
[0089]
{5. Modification}
In the embodiment described above, each processing unit (WB correction unit 33, pixel interpolation unit 34, edge enhancement adjustment unit 36, etc.) provided in the digital camera 1 may be realized by processing by hardware, You may implement | achieve by the process by software.
[0090]
Then, by realizing each processing by software processing, it is possible to cause a general computer to execute processing as executed in the digital camera 1 according to the present embodiment. A program for executing such processing can be distributed as a recording medium.
[0091]
FIG. 11 is an overview diagram of the computer 5. The computer 5 is a personal computer, for example, and can read a recording medium 51 such as a floppy disk or a CD-ROM.
[0092]
Then, by recording the program 6 capable of executing the same image processing as that of the digital camera 1 according to the present embodiment on the recording medium 51, the computer 5 executes the program 6 by installing it on the hard disk. Is possible. Alternatively, the program 6 recorded on the recording medium 51 may be directly executed.
[0093]
FIG. 12 is a processing flow of the program 6. The program 6 first reads data (step S201). For example, the program 6 reads image data stored in a predetermined directory. Here, it is assumed that the image data is RGB data output from a Bayer array CCD.
[0094]
Next, the program 6 performs WB correction processing (step S202) and pixel interpolation processing (step S203).
[0095]
Then, in the embodiment shown in FIG. 3, the function corresponding to the edge enhancement adjustment unit 36 is executed, and the processing order is determined (step S204). Here, for example, the shooting mode information is read from the shooting mode information included in the image header information, another definition file, or the like.
[0096]
When the gradation conversion process is performed first (Yes in step S205), the gradation conversion process (step S206) is performed, and then the edge enhancement process (step S207) is performed. However, the conversion characteristic of the gradation conversion process is cγ <1, and the order of the gradation conversion process and the edge enhancement process is switched in the same relationship as that shown in FIG. 6 in the above-described embodiment.
[0097]
When gradation conversion processing is not performed first (No in step S205), edge enhancement processing (step S208) is performed, and then gradation conversion processing (step S209) is performed. As described above, when the image data subjected to the gradation conversion process and the edge enhancement process is output in any one of the processing orders, the image compression process is performed (step S210), and the data is stored in the hard disk or the recording medium. Writing is performed (step S211).
[0098]
【The invention's effect】
  As explained above,The invention according to any one of claims 1 to 3 is a method of switching the processing order of gradation conversion processing and edge enhancement processing, and changes the degree of edge enhancement without having a plurality of conversion characteristics of edge enhancement processing. It is possible.
[0109]
  Claim4In the described invention, the claims1Or claims3By distributing a program that can execute the method described in (1), the image processing method of the present invention can be executed in a computer that can read the recording medium.
[Brief description of the drawings]
FIG. 1 is a schematic perspective view of a digital camera according to an embodiment.
FIG. 2 is a rear view of the digital camera.
FIG. 3 is an internal block diagram of the digital camera.
FIG. 4 is a processing block diagram of an edge enhancement unit.
FIG. 5 is a diagram illustrating an output level controlled by a coring unit and a gain adjusting unit.
FIG. 6 is a diagram illustrating a relationship between a photographing mode of the digital camera, a degree of edge enhancement, and a processing order.
FIG. 7 is a flowchart showing edge enhancement adjustment processing in the digital camera.
FIG. 8 is a diagram illustrating a spatial filter in an edge enhancement unit.
FIG. 9 is a diagram illustrating a calculation result of each pixel when gradation conversion processing is performed first.
FIG. 10 is a diagram illustrating a calculation result of each pixel when edge enhancement processing is performed first.
FIG. 11 is an overview of a computer.
FIG. 12 is a flowchart showing edge enhancement adjustment processing executed on a computer.
[Explanation of symbols]
1 Digital camera
16 Setting key
34 Pixel interpolation unit
35 Scene judgment part
36 Edge enhancement adjustment unit
40 Processing order control unit
41 Edge enhancement part
42 Gradation converter

Claims (5)

A processing method for performing gradation conversion processing and edge enhancement processing on an image,
A processing order determination step for determining a processing order of gradation conversion processing and edge enhancement processing for the processing target image according to a required level of the edge enhancement degree for the processing target image;
A step of sequentially performing gradation conversion processing and edge enhancement processing on the processing target image according to the order determined in the processing order determination step;
An image processing method comprising: The image processing method according to claim 1,
An image processing method characterized in that the conversion characteristic of the gradation conversion process is expressed by a γ value, and the γ value is a value other than 1. The image processing method according to claim 1 or 2, further comprising:
A scene determination step of determining a scene of the processing target image;
Including
The image processing method characterized in that the processing order determination step determines the processing order for the processing target image based on a determination result of the scene determination step. A computer-readable recording medium having a program recorded thereon for causing the computer to execute the image processing method according to any one of claims 1 to 3. Gradation conversion means for performing gradation conversion of an image;
Edge enhancement means for performing edge enhancement of an image;
A scene determination unit that determines a scene of a processing target image by the gradation conversion unit and the edge enhancement unit;
Means for switching the processing order of images to be processed by the gradation conversion means and the edge enhancement means based on the determination result of the scene determination means;
An image processing apparatus comprising:

Images