JPH0863597A - Face extracting method - Google Patents
Face extracting methodInfo
- Publication number
- JPH0863597A JPH0863597A JP6196735A JP19673594A JPH0863597A JP H0863597 A JPH0863597 A JP H0863597A JP 6196735 A JP6196735 A JP 6196735A JP 19673594 A JP19673594 A JP 19673594A JP H0863597 A JPH0863597 A JP H0863597A
- Authority
- JP
- Japan
- Prior art keywords
- face
- area
- extraction method
- determined
- candidate area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
Description
【０００１】[0001]
【産業上の利用分野】本発明は、カメラなどで人を撮影
したネガフィルムやポジフィルムなどのカラー原画像か
ら顔の部分を抽出する顔抽出方法に関する。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a face extracting method for extracting a face portion from a color original image such as a negative film or a positive film obtained by photographing a person with a camera.
【０００２】[0002]
【従来の技術】カメラで撮影したネガフィルムを陽画に
焼付ける場合のように、カラー原画像を複写材料に複写
する場合、適切な露光量で複写することが大切であり、
特に人を撮影したカラー写真の場合には、人の顔の色が
適正になるように焼付けると一般に写真を見た人に与え
る感じが良く、写真の質を高めることになる。2. Description of the Related Art When copying a color original image onto a copying material such as when printing a negative film photographed by a camera on a positive image, it is important to copy with an appropriate exposure amount,
Particularly in the case of a color photograph of a person, printing it so that the color of the person's face is appropriate will generally give the person who sees the photograph a good feeling and enhance the quality of the photograph.
【０００３】写真撮影の場合、撮影条件が常に一定であ
ればその撮影条件に応じた適切な露光量で焼付けを行え
ばよいが、実際には逆光で撮影された原画像やストロボ
を用いて撮影された原画像等が１本のフィルムの中に混
在していることがある。このため、質の良い写真を得る
ためには原画像ごとに焼付け時の露光量を変更するのが
好ましく、この焼付け時の露光量を決めるのに、人が入
った写真の場合には人の顔の色に着目するのが便利であ
る。顔の色は肌色であることが前もってわかっているた
め、焼付けられた後の写真における人の顔の色が肌色に
なるように露光量を決めることができるからである。In the case of photographing, if the photographing condition is always constant, the exposure may be performed with an appropriate exposure amount according to the photographing condition, but in reality, the original image photographed with backlight or a strobe is photographed. Original images and the like may be mixed in one film. For this reason, it is preferable to change the exposure amount during printing for each original image in order to obtain a good quality photo. It is convenient to focus on the color of the face. This is because it is known in advance that the color of the face is flesh color, and therefore the exposure amount can be determined so that the color of the face of the person in the photograph after printing becomes the flesh color.
【０００４】このように人の顔の色を基準として焼付け
時の露光量を決定する方法は従来から知られており、カ
ラーフィルムの現像、焼付けの一部作業を自動化する場
合に利用されている。As described above, a method of determining the exposure amount at the time of printing on the basis of the color of the face of a person has been conventionally known, and is used when a part of the work of developing and printing the color film is automated. .
【０００５】たとえば、カラーフィルムの原画像中の顔
領域をオペレータがライトペンで指定して人の顔の濃度
データを抽出し、この抽出した濃度データに基づいて顔
の色が適正に焼付けられるように露光量を決定する方法
がすでに知られている。For example, an operator designates a face area in an original image of a color film with a light pen to extract density data of a human face, and the face color is properly printed based on the extracted density data. A method for determining the exposure amount is already known.
【０００６】また、原画像中の人の顔を抽出する方法と
して、原画像から肌色データを抽出し、肌色範囲と判断
された測光点のクラスタを顔とする方法が知られている
（特開昭５２−１５６６２４号公報、特開昭５３−１４
５６２１号公報、特開昭５３−１４５６２２号公報）。
これは、カラー原画像を多数の測光点に分割するととも
に各測光点をＲ（赤）、Ｇ（緑）、Ｂ（青）の３色に分
解して測光し、測光データから計算した各測光点の色が
肌色範囲内か否かを判断し、肌色範囲と判断された測光
点のクラスタ（群）を顔の濃度データとする方法であ
る。As a method of extracting a human face in an original image, a method is known in which skin color data is extracted from the original image, and a cluster of photometric points determined to be in the skin color range is used as a face (Japanese Patent Laid-Open No. 2000-242242). JP-A-52-156624, JP-A-53-14
5621, and JP-A-53-145622).
This is to divide the original color image into a large number of photometric points, separate each photometric point into three colors of R (red), G (green), and B (blue), perform photometry, and calculate each photometric value from the photometric data. In this method, it is determined whether or not the color of a point is within the skin color range, and the cluster (group) of photometric points determined to be within the skin color range is used as face density data.
【０００７】さらに、特開平４−３４６３３３号公報に
は、測光したデータを色相値（Ｈ）、彩度値（Ｓ）に変
換し、Ｈ、Ｓの２次元ヒストグラムを作成し、このヒス
トグラムを単峰の山に分割し、原画像の各画素が分割さ
れた山のどれに属するかを判断して画素を分割すること
により顔の候補領域を抽出し、顔候補領域の輪郭、内部
構造から人か否かを判断する方法が開示されている。Further, in Japanese Patent Laid-Open No. 4-346333, the photometric data is converted into a hue value (H) and a saturation value (S), a two-dimensional histogram of H and S is created, and this histogram is simply A candidate area of a face is extracted by dividing each pixel of the original image into which of the divided mountains each pixel of the original image belongs, and extracting the candidate area of the face. A method of determining whether or not is disclosed.
【０００８】[0008]
【発明が解決しようとする課題】上述のオペレータが顔
領域をライトペンで指定する方法の場合には、カラー原
画像中の顔領域を間違いなく抽出することができるが、
画像毎にオペレータがライトペンで顔領域を指定しなけ
ればならないため、焼付け作業に時間がかかるという問
題がある。また、この方法の場合、オペレータが介在し
ない完全な無人化（自動化）は不可能である。In the case of the above-mentioned method in which the operator specifies the face area with the light pen, the face area in the color original image can be definitely extracted.
Since the operator must specify the face area with a light pen for each image, there is a problem that the printing operation takes time. Further, in the case of this method, complete unmanned operation (automation) without operator intervention is impossible.
【０００９】一方、原画像から肌色データを抽出し、肌
色範囲と判断された測光点のクラスタを顔とする方法の
場合には、地面、木の幹、洋服等の肌色または肌色に近
似した色をした顔以外の部位も顔の濃度データとして抽
出されてしまい、精度に欠けるという問題がある。ま
た、フィルム種、光源によって、顔を抽出できない場合
もあるという問題がある。On the other hand, in the case of the method of extracting the skin color data from the original image and using the cluster of the photometric points determined to be in the skin color range as the face, the skin color of the ground, tree trunk, clothes, or the like or a color close to the skin color. There is a problem in that accuracy is lacked because parts other than the face on which the image was taken are also extracted as face density data. There is also a problem that the face may not be extracted depending on the film type and the light source.
【００１０】そして、特開平４−３４６３３３号公報に
開示された方法の場合には、顔と手、顔と顔が接触して
いる場合などでは、肌色領域の形状が複雑になり、顔だ
けを検出することができない場合があるという問題があ
る。In the case of the method disclosed in Japanese Patent Laid-Open No. 4-346333, when the faces are in contact with each other or the faces are in contact with each other, the shape of the flesh color region becomes complicated, and only the face is removed. There is a problem that it may not be detected.
【００１１】本発明は上記の点にかんがみてなされたも
ので、ネガフィルムやポジフィルム等のカラー原画像か
ら、人手を介さず完全に自動で且つ精度よく人の顔を抽
出する顔抽出方法を提供することを目的とする。The present invention has been made in view of the above points, and provides a face extraction method for extracting a human face from a color original image such as a negative film or a positive film completely automatically and accurately without human intervention. The purpose is to provide.
【００１２】[0012]
【課題を解決するための手段】本発明は上記の目的を達
成するために、画像から人の顔を抽出するための抽出方
法において、人の顔の形状に相当する顔候補領域を決定
し、前記顔候補領域内の特徴量から顔領域を決定するよ
うにした。In order to achieve the above object, the present invention determines a face candidate area corresponding to the shape of a human face in an extraction method for extracting a human face from an image, The face area is determined from the feature amount in the face candidate area.
【００１３】また、別の方法によれば、画像から人の顔
の輪郭を抽出することにより顔候補領域を検出するよう
にした。According to another method, the face candidate area is detected by extracting the contour of the human face from the image.
【００１４】さらに、別の方法によれば、画像から人の
顔を抽出するための抽出方法において、顔の様々な形状
をした複数のテンプレートを用意しておき、このテンプ
レートと画像とのマッチング度を計算し、マッチング度
の最も高いテンプレートを選択し、最も高かったマッチ
ング度が予め定めたしきい値以上であれば、選択された
テンプレート内の領域を顔候補領域とするようにした。Further, according to another method, in an extraction method for extracting a human face from an image, a plurality of templates having various shapes of the face are prepared and the matching degree between the template and the image is prepared. Then, the template with the highest degree of matching is selected, and if the highest degree of matching is greater than or equal to a predetermined threshold value, the area within the selected template is set as the face candidate area.
【００１５】[0015]
【作用】本発明は以上の方法によって、人の顔の形状に
相当する顔候補領域を決定し、前記顔候補領域内の特徴
量から顔領域を決定することにより、カラー原画像から
人の顔を抽出する。According to the present invention, the face candidate area corresponding to the shape of the human face is determined by the above method, and the face area is determined from the feature amount in the face candidate area. To extract.
【００１６】また、別の方法によれば、画像から人の顔
の輪郭を抽出することにより顔候補領域を検出し、カラ
ー原画像から人の顔を抽出する。According to another method, the face candidate area is detected by extracting the contour of the human face from the image, and the human face is extracted from the original color image.
【００１７】さらに、別の方法によれば、複数の顔の形
状をしたテンプレートを用意しておき、このテンプレー
トと画像とのマッチング度を計算し、マッチング度の最
も高いテンプレートを選択し、最も高かったマッチング
度が予め定めたしきい値以上であれば、選択されたテン
プレート内の領域を顔候補領域とすることにより、カラ
ー原画像から人の顔を抽出する。Further, according to another method, a template having a plurality of face shapes is prepared, the degree of matching between the template and an image is calculated, and the template with the highest degree of matching is selected to obtain the highest degree of matching. If the matching degree is equal to or higher than a predetermined threshold value, the face of the person is extracted from the original color image by setting the region in the selected template as the face candidate region.
【００１８】[0018]
【実施例】以下本発明を図面に基づいて説明する。DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the drawings.
【００１９】図１は、本発明による顔抽出方法を用いた
顔領域抽出装置のブロック図である。FIG. 1 is a block diagram of a face area extracting apparatus using the face extracting method according to the present invention.
【００２０】フィルム１はカラー原画像が記録されたフ
ィルムであり、ネガフィルムであってもポジフィルムで
あってもかまわない。ポジ画像から肌色を抽出するとき
はポジの測光値から直接肌色を抽出すればよく、ネガ画
像から肌色を抽出するときはネガの測光値をポジに変換
して肌色を抽出してもよいし、ネガの測光値から直接肌
色を抽出してもよい。スキャナ２はフィルム１のカラー
原画像を光学的に読取り、色分解して各画素のＢ
（青）、Ｇ（緑）、Ｒ（赤）値を得ることができる。こ
のＢＧＲ値は、増幅器３で増幅された後にＡ／Ｄ変換器
４でデジタルデータに変換されてＣＰＵ５に入力され
る。ＣＰＵ５では、後述する顔抽出のための各処理を実
行する。The film 1 is a film on which a color original image is recorded and may be a negative film or a positive film. When extracting the skin color from the positive image, it is sufficient to extract the skin color directly from the positive photometric value, and when extracting the skin color from the negative image, the negative photometric value may be converted to positive to extract the skin color, The skin color may be directly extracted from the negative photometric value. The scanner 2 optically reads the original color image on the film 1, color-separates it, and outputs B for each pixel.
(Blue), G (green), R (red) values can be obtained. The BGR value is amplified by the amplifier 3, converted into digital data by the A / D converter 4, and input to the CPU 5. The CPU 5 executes each process for face extraction described later.
【００２１】図２は、本発明による顔抽出方法の第１の
実施例のフローチャートである。FIG. 2 is a flow chart of the first embodiment of the face extraction method according to the present invention.
【００２２】まず、スキャナ２によって得られた各画素
のＢＧＲ値から、明度、色相、彩度、色度、（Ｂ−
Ｒ）、（Ｇ−Ｒ）といった色の特徴量を求め（Ａ−
１）、これらの色の特徴量が予め定めた範囲内に入って
いれば、対象画素が肌色であると判定する（Ａ−２）。
もちろん、ネガ画像のＢＧＲ値から肌色を抽出する場合
と、ポジ画像のＢＧＲ値から肌色を抽出する場合とで
は、特徴量について予め定める範囲が異なる。また、求
めた色の特徴量をニューラルネットワークの入力値とし
て、肌色か否かを判定してもかまわない。First, from the BGR value of each pixel obtained by the scanner 2, lightness, hue, saturation, chromaticity, (B-
R), (GR), and color feature quantities such as (A-
1) If the feature amounts of these colors are within a predetermined range, it is determined that the target pixel is a skin color (A-2).
Of course, the predetermined range of the feature amount differs between the case of extracting the skin color from the BGR value of the negative image and the case of extracting the skin color from the BGR value of the positive image. Further, it may be possible to determine whether the color is a skin color by using the obtained color feature amount as an input value of the neural network.
【００２３】次に、肌色と判定された画素から成る画像
に対してエッジ抽出を行なう（Ａ−３）。エッジ抽出方
法としては、たとえば、対象画素を中心として、周囲８
画素の明度平均を取り、その平均値と対象画素の明度と
の差が所定値よりも大きければ、対象画素をエッジ画素
とし、エッジ画素であるか否かで２値化する。ここで
は、このようにして得た画像をエッジ画像という。Next, edge extraction is performed on the image composed of pixels determined to be skin color (A-3). As an edge extraction method, for example, with the target pixel as the center,
The brightness average of the pixels is calculated, and if the difference between the average value and the brightness of the target pixel is larger than a predetermined value, the target pixel is set as an edge pixel, and binarization is performed depending on whether the pixel is an edge pixel. Here, the image thus obtained is referred to as an edge image.
【００２４】ステップ（Ａ−４）では、まず、サイズ、
長軸／短軸（ｌａ／ｌｂ）の比率の異なる複数の楕円あ
るいは円形の顔テンプレート（図３参照）を作成する。
これらの顔テンプレートは予め作成しておき記憶装置６
に記憶させておいてもよい。顔テンプレートは、楕円な
いしは円の輪郭か否かで２値化されている。実際の顔の
輪郭は、正確な楕円ないしは円でないために、顔テンプ
レートの輪郭は、数画素、好ましくは２〜３画素の幅を
もたせて、実際の顔の輪郭とのマッチング度を上げても
よい。In step (A-4), first, the size,
A plurality of elliptical or circular face templates (see FIG. 3) having different major axis / minor axis (la / lb) ratios are created.
These face templates are created in advance and stored in the storage device 6.
It may be stored in. The face template is binarized depending on whether it is the outline of an ellipse or a circle. Since the actual contour of the face is not an accurate ellipse or circle, the contour of the face template may have a width of several pixels, preferably a few pixels, to improve the matching degree with the contour of the actual face. Good.
【００２５】ステップ（Ａ−４）では、続いて、エッジ
画像と顔テンプレートのマッチング度を求める。マッチ
ング度の求め方は、既存の手法を用いる。たとえば、数
１で表されるような手法でマッチング度ｍ（ｕ，ｖ）を
求める。In step (A-4), subsequently, the matching degree between the edge image and the face template is obtained. An existing method is used to obtain the matching degree. For example, the matching degree m (u, v) is obtained by the method represented by the equation 1.
【００２６】[0026]
【数１】 数１において、ｆは対象画像を表し、ｔは顔テンプレー
トを表し、Ｓはｔ（ｘ，ｙ）の値域を表す。ｆ’、ｔ’
はそれぞれｆ（ｘ＋ｕ，ｙ＋ｖ）、ｔ（ｘ，ｙ）のＳ内
での平均を表す。[Equation 1] In Expression 1, f represents a target image, t represents a face template, and S represents a range of t (x, y). f ', t'
Represents the average of f (x + u, y + v) and t (x, y) in S, respectively.
【００２７】このような手法で、数種類の顔テンプレー
トを用いてマッチングを行ない、最もよく一致するテン
プレートを対象画素に対して求め（Ａ−４）、マッチン
グ度が予め定めたしきい値以上であれば、対象画素を中
心として、最もよく一致する顔テンプレートで囲まれる
領域を顔候補領域であると判定する（Ａ−６）。With such a method, matching is performed using several types of face templates, the best matching template is obtained for the target pixel (A-4), and the matching degree is equal to or more than a predetermined threshold value. For example, the area surrounded by the best matching face template with the target pixel as the center is determined to be the face candidate area (A-6).
【００２８】また、演算回数を減らすために、最初に大
まかな走査として、一定の大きさの顔テンプレートを１
画素ずつ、あるいは、何画素かおきにずらしていって、
マッチング度を求め、マッチング度が所定値以上の対象
画素に対してのみ、大きさの違うテンプレートをあては
めて、最適な候補領域を決定してもよい。In order to reduce the number of calculations, a rough scan is first performed and a face template of a certain size is set to 1
You can shift pixels pixel by pixel or every other pixel,
The optimum matching area may be determined by finding the matching degree and applying templates of different sizes only to the target pixel having the matching degree equal to or higher than a predetermined value.
【００２９】図２に示した第１の実施例によれば、顔の
輪郭を抽出しているために、たとえば顔と手、顔と顔が
接触しているような画像であっても顔単体だけを正確に
抽出することができるという効果がある。According to the first embodiment shown in FIG. 2, since the outline of the face is extracted, even if the image is such that the face is in contact with the hand or the face is in contact with the face, the face alone is extracted. There is an effect that only the data can be accurately extracted.
【００３０】また、肌色領域の抽出にあたり肌色とみな
す色の範囲を広く設定すれば、フィルムの種類や光源の
違いをカバーすることができる。Further, when the skin color area is extracted, the range of the color regarded as the skin color is set wide so that the difference in the film type and the light source can be covered.
【００３１】さらに、肌色の範囲を、ネガ画像用、ポジ
画像用にそれぞれ設定することでネガ画像からでも、ポ
ジ画像からでも顔の抽出が可能になるという効果もあ
る。Furthermore, by setting the range of the skin color for the negative image and for the positive image respectively, it is possible to extract the face from the negative image or the positive image.
【００３２】なお、図２に示した第１の実施例では肌色
領域の抽出を行ってから顔テンプレートとのマッチング
度を求めたが、本発明はこれに限らず、対象画像全体を
( 肌色抽出することなしに) 明度画像に変換し、その後
に、第１の実施例と同じ手法で顔候補領域を決定しても
よい。In the first embodiment shown in FIG. 2, the degree of matching with the face template is obtained after extracting the skin color area, but the present invention is not limited to this, and the entire target image is
It is also possible to convert to a lightness image (without extracting the skin color) and then determine the face candidate area by the same method as in the first embodiment.
【００３３】また、図２に示した第１の実施例ではエッ
ジであるか否かで２値化した画像とエッジの顔テンプレ
ートとのマッチングを行ったが、本発明はこれに限ら
ず、肌色領域抽出後、肌色であるか否かで２値化した画
像と、肌色か否かで２値化した顔テンプレートとのマッ
チングを行ってもよい。In the first embodiment shown in FIG. 2, the binarized image is matched with the edge face template depending on whether it is an edge or not, but the present invention is not limited to this, and the skin color is not limited to this. After extracting the region, an image binarized depending on whether it is a flesh color or not may be matched with a face template binarized depending on whether it is a flesh color or not.
【００３４】ところで、第１の実施例によって顔候補領
域と判定された部分を顔とみなしても十分であるが、画
像によっては顔ではない部分を顔候補領域と判定してし
まう場合がある。そこで、以下では第１の実施例で顔候
補領域と判定された複数の部分をさらに絞り込み、実際
に顔である部分を確実に抽出する方法について説明す
る。By the way, it is sufficient to regard the portion determined as the face candidate area in the first embodiment as a face, but depending on the image, the portion other than the face may be determined as the face candidate area. Therefore, hereinafter, a method of further narrowing down a plurality of portions determined as face candidate areas in the first embodiment and surely extracting a portion that is actually a face will be described.
【００３５】図４は、本発明による顔抽出方法の第２の
実施例のフローチャートである。FIG. 4 is a flowchart of the second embodiment of the face extraction method according to the present invention.
【００３６】まず、第１の実施例によって判定された顔
候補領域内で目のある領域（目候補領域）を推定し設定
する（Ｂ−１）。このとき、画像が縦である場合（図５
（ａ）参照）と横である場合（図５（ｂ）参照）のそれ
ぞれに対して推定する。この推定方法としては、たとえ
ば、画像の向きが縦の場合には顔候補領域の上半分と下
半分を目候補領域に設定し、一方、画像の向きが横の場
合には顔候補領域の右半分と左半分を目候補領域に設定
する。すなわち、顔候補領域に対して、上半分、下半
分、右半分、左半分を目候補領域と設定する。この目候
補領域の設定方法は、もっと狭く設定してもかまわな
い。たとえば、画像の向きが縦の場合、図６に示すよう
に、顔候補領域の下端の高さを０、上端の高さを１とし
たとき、高さが０．５〜０．８の範囲を目候補領域と設
定してもよい。First, an area with eyes (eye candidate area) is estimated and set in the face candidate area determined by the first embodiment (B-1). At this time, if the image is vertical (see FIG.
(See (a)) and horizontal (see FIG. 5 (b)). As the estimation method, for example, when the orientation of the image is vertical, the upper half and the lower half of the face candidate area are set as eye candidate areas, while when the orientation of the image is horizontal, the right side of the face candidate area is set. Half and the left half are set as eye candidate areas. That is, with respect to the face candidate area, the upper half, the lower half, the right half, and the left half are set as eye candidate areas. The method of setting the eye candidate area may be set narrower. For example, when the orientation of the image is vertical, as shown in FIG. 6, when the height of the lower end of the face candidate area is 0 and the height of the upper end is 1, the height is in the range of 0.5 to 0.8. May be set as the eye candidate area.
【００３７】次に、それぞれの目候補領域内で、図７に
示すような明度に関するヒストグラムを作成する（Ｂ−
２）。ここでは、明度を８ビットで表した場合について
説明する（明度０：０、明度１００：２５５）。まず明
度領域（０〜２５５）を均等に８分割して、それぞれの
明度領域内での出現頻度を求める。黒目の部分は明度が
低いので、ヒストグラムは少なくとも黒目の明度の部分
と肌色の明度の部分にピークを有する（図７参照）。こ
こでは明度のヒストグラムを作成したが、本発明はこれ
に限らず、明度の代わりに彩度、色相、色度を用いても
かまわない。Next, in each eye candidate area, a histogram regarding lightness as shown in FIG. 7 is created (B-
2). Here, a case where the lightness is represented by 8 bits will be described (lightness 0: 0, lightness 100: 255). First, the lightness region (0 to 255) is equally divided into eight, and the appearance frequency in each lightness region is obtained. Since the brightness of the black eye is low, the histogram has peaks in at least the brightness of the black eye and the brightness of the skin color (see FIG. 7). Although a histogram of brightness is created here, the present invention is not limited to this, and saturation, hue, and chromaticity may be used instead of brightness.
【００３８】さて、こうして作成したヒストグラムの形
状認識手法の１つとして、以下のような手法がある。ま
ず、肌色の領域と思われる明度範囲（たとえば８ビット
で９６以上）で最大のピーク（第１のピーク）を探す。
次に所定値よりも低い明度範囲で最大のピーク（第２の
ピーク）を探す。この所定値は、測定機器を用いていく
つかの目のサンプルを測光し経験的に求める。（第２の
ピークの頻度）／（第１のピークの頻度）の値を計算
し、この値を特徴量とする（Ｂ−３）。Now, as one of the shape recognition methods of the histogram thus created, there is the following method. First, the maximum peak (first peak) is searched for in the lightness range (for example, 8 bits or more 96 or more) considered to be a skin color region.
Next, the maximum peak (second peak) is searched in the lightness range lower than the predetermined value. This predetermined value is determined empirically by photometrically measuring several eye samples using a measuring instrument. The value of (frequency of second peak) / (frequency of first peak) is calculated, and this value is used as a feature amount (B-3).
【００３９】この特徴量が所定範囲内であれば（Ｂ−
４）、このヒストグラムの形状は顔を表していると判断
する。ステップ（Ｂ−４）における所定範囲は、目候補
領域の大きさによって異なるため、それぞれのケースご
とに最適範囲を求めておかなければならない。If this feature amount is within the predetermined range (B-
4) It is determined that the shape of this histogram represents a face. The predetermined range in step (B-4) differs depending on the size of the eye candidate area, and therefore the optimum range must be obtained for each case.
【００４０】このようにして、それぞれの目候補領域ご
とに判断を行ない、目と判断される領域があれば、その
顔候補領域は顔であると判定し（Ｂ−４）、目と判断さ
れる領域がなければ、顔でないと判定される。In this way, the judgment is made for each eye candidate area, and if there is an area judged as an eye, it is judged that the face candidate area is a face (B-4), and it is judged as an eye. If there is no region to be identified, it is determined that it is not a face.
【００４１】また、図４に示した第２の実施例では、ス
テップ（Ｂ−１）で目候補領域を設定し、ステップ（Ｂ
−２）でその目候補領域内の明度のヒストグラムを作成
したが、本発明はこれに限らず、たとえば、目候補領域
を設定せずに、顔候補領域全体で明度のヒストグラムを
作成して顔であるか否かの判定を行うようにしてもよ
い。In the second embodiment shown in FIG. 4, the eye candidate area is set in step (B-1) and the eye candidate area is set in step (B-1).
Although the histogram of the brightness in the eye candidate area is created in -2), the present invention is not limited to this. For example, the brightness histogram of the entire face candidate area is created without setting the eye candidate area, and You may make it determine whether it is.
【００４２】図４に示した第２の実施例によれば、顔候
補領域内に複数の目設定領域を設定することにより、画
像の向きの違いに対応することができる。According to the second embodiment shown in FIG. 4, by setting a plurality of eye setting areas in the face candidate area, it is possible to cope with the difference in the orientation of the image.
【００４３】また、言い換えれば、この手法により自動
的に画像の向きを判定することができるという効果もあ
る。In other words, there is also an effect that the orientation of the image can be automatically determined by this method.
【００４４】なお、図４に示した第２の実施例では明度
の１次元のヒストグラムを作成したが、本発明はこれに
限らず、たとえば、明度と色相とを軸にした２次元ヒス
トグラムを作成し、肌色を表す領域のピークを第１ピー
ク、目の黒を表す領域のピークを第２ピークとして第２
の実施例と同じ手法で判定するようにしてもよいし、明
度と彩度の２次元ヒストグラムや、色相と彩度の２次元
ヒストグラムを用いてもよい。In the second embodiment shown in FIG. 4, a one-dimensional histogram of lightness is created, but the present invention is not limited to this, and for example, a two-dimensional histogram centered on lightness and hue is created. Then, the peak in the area representing the skin color is the first peak, and the peak in the area representing the black eye is the second peak.
The determination may be performed by the same method as that of the embodiment, or a two-dimensional histogram of lightness and saturation or a two-dimensional histogram of hue and saturation may be used.
【００４５】図８は、本発明による顔抽出方法の第３の
実施例のフローチャートである。FIG. 8 is a flowchart of the third embodiment of the face extraction method according to the present invention.
【００４６】まず、第１の実施例によって判定された顔
候補領域内の画像を明度に変換する（Ｃ−１）。本実施
例では顔候補領域内の画像を明度に変換するものとして
説明するが、本発明は明度画像でなく色度画像、色相画
像、彩度画像に変換してもかまわない。First, the image in the face candidate area determined by the first embodiment is converted into lightness (C-1). In the present embodiment, the image in the face candidate area is described as being converted into the brightness, but the present invention may be converted into a chromaticity image, a hue image, and a saturation image instead of the brightness image.
【００４７】次に、顔候補領域の大きさを基準サイズに
合わせるために拡大、縮小を行い、顔候補領域のサイズ
を規格化する（Ｃ−２）。そして、規格化された顔候補
領域に対して２次元フーリエ変換を行う（Ｃ−３）。Next, the size of the face candidate area is enlarged and reduced to match the reference size, and the size of the face candidate area is standardized (C-2). Then, the two-dimensional Fourier transform is performed on the standardized face candidate area (C-3).
【００４８】ここで、このフーリエ変換の結果におい
て、そのパワースペクトラムを最も大きなピーク値で規
格化しておく。そして、実際に顔を表すいくつかのパタ
ーン（以下「顔の基準データ」という）に対して同様に
２次元フーリエ変換を行ったものを予め記憶装置６に記
憶させておくことにより用意しておき、顔候補領域のフ
ーリエ変換結果と顔の基準データフーリエ変換結果との
間でマッチングをとり、このマッチング度の最も高い値
を特徴量とする（Ｃ−４）。Here, in the result of the Fourier transform, the power spectrum is standardized with the largest peak value. Then, some patterns that actually represent faces (hereinafter referred to as “face reference data”) are similarly subjected to two-dimensional Fourier transform and stored in the storage device 6 in advance to be prepared. Then, matching is performed between the Fourier transform result of the face candidate region and the face reference data Fourier transform result, and the value with the highest matching degree is set as the feature amount (C-4).
【００４９】そして、この特徴量がしきい値以上であれ
ば（Ｃ−５）、その顔候補領域は顔であると判定する
（Ｃ−６）。If this feature amount is equal to or greater than the threshold value (C-5), it is determined that the face candidate area is a face (C-6).
【００５０】ところで、図８に示した第３の実施例では
ステップ（Ｃ−３）で２次元フーリエ変換を行ったが、
本発明はこれに限らず、１次元フーリエ変換を行っても
よい。ただし、この場合には、顔の基準データとして、
正面顔、横向き顔について、それぞれ画像の向きが上、
下、右、左の条件を組み合わせたパターンを準備する必
要がある。By the way, in the third embodiment shown in FIG. 8, the two-dimensional Fourier transform is performed in step (C-3).
The present invention is not limited to this, and one-dimensional Fourier transform may be performed. However, in this case, as the face reference data,
For the front face and the sideways face, the image orientation is up,
It is necessary to prepare a pattern that combines the conditions of bottom, right, and left.
【００５１】図８に示した第３の実施例によれば、顔候
補領域の抽出の不正確さにより顔の部位（目や口）の位
置がずれてしまったような場合にも、その影響を受けに
くく、顔の高い抽出率が得られるという効果がある。According to the third embodiment shown in FIG. 8, even when the position of the face part (eye or mouth) is displaced due to the inaccuracy of the extraction of the face candidate area, the influence is exerted. There is an effect that it is hard to receive and a high extraction rate of the face can be obtained.
【００５２】なお、図８に示した第３の実施例では２次
元フーリエ変換を行ったが、本発明は２次元フーリエ変
換を行わなくても、周波数特性に準ずるものであればか
まわない。たとえば、次に示すような方法であってもか
まわない。Although the two-dimensional Fourier transform is performed in the third embodiment shown in FIG. 8, the present invention does not need to perform the two-dimensional Fourier transform as long as it conforms to the frequency characteristic. For example, the following method may be used.
【００５３】まず、画像の横軸、縦軸をそれぞれｘ軸、
ｙ軸とし、ｘ軸上ｘ0 の位置の明度の値をｙ軸に沿って
加算しｘ0 での値とする明度のｘ軸への投影と、同様に
明度のｙ軸への投影とを行なう。そして、ｘ軸への投影
とｙ軸への投影のそれぞれについてそれぞれのピーク値
で規格化し、このデータと実際に顔画像から同様にして
得られた基準データとの間でマッチングを取り、しきい
値より高ければ顔と判断する。もちろん、明度画像でな
くてもよく、色相、色度、彩度等の色を表す特徴量のｘ
軸、ｙ軸への投影を用いればよい。First, the horizontal axis and the vertical axis of the image are the x-axis,
Let the y-axis be the projection of the lightness on the x-axis, which is the sum of the lightness values at the position x 0 on the x-axis along the y-axis, and the value at x 0 , and the projection of the lightness on the y-axis. To do. Then, the projection on the x-axis and the projection on the y-axis are standardized by their respective peak values, and this data and reference data obtained in the same way from the face image are matched to obtain a threshold value. If it is higher than the value, it is judged as a face. Of course, it is not necessary to use a lightness image, and x of a feature amount representing a color such as hue, chromaticity, and saturation may be used.
The projection on the axis and the y-axis may be used.
【００５４】この場合、２次元フーリエ変換に比べて計
算時間が短くてすむので、高速に処理することができる
という効果がある。In this case, since the calculation time is shorter than that of the two-dimensional Fourier transform, there is an effect that the processing can be performed at high speed.
【００５５】図９は、本発明による顔抽出方法の第４の
実施例のフローチャートである。FIG. 9 is a flow chart of the fourth embodiment of the face extraction method according to the present invention.
【００５６】まず、第１の実施例によって判定された顔
候補領域内の各画素の色相、明度を求め（Ｄ−１）、色
相、明度の２次元空間の共分散楕円の面積を求め、これ
を特徴量とする（Ｄ−２）。First, the hue and lightness of each pixel in the face candidate area determined by the first embodiment is obtained (D-1), and the area of the covariance ellipse of the two-dimensional space of hue and lightness is obtained. Is a feature amount (D-2).
【００５７】ステップ（Ｄ−２）で求めた共分散楕円の
面積が所定範囲内であれば（Ｄ−３）、その顔候補領域
は顔であると判定する（Ｄ−４）。If the area of the covariance ellipse obtained in step (D-2) is within the predetermined range (D-3), it is determined that the face candidate area is a face (D-4).
【００５８】なお、共分散楕円の代わりに、色空間（Ｌ
* ａ* ｂ* 、Ｌ* ｕ* ｖ* 等）での分散楕円体の体積、
色相の分散、明度の分散、彩度の分散を特徴量としても
かまわない。あるいはこれらの少なくとも2 つ以上を組
み合わせて判断すればより認識率が上がる。Instead of the covariance ellipse, the color space (L
* a * b * , L * u * v *, etc.) volume of the dispersion ellipsoid,
The dispersion of hue, the dispersion of lightness, and the dispersion of saturation may be used as the feature amount. Alternatively, if at least two of these are combined and judged, the recognition rate will increase.
【００５９】図９に示した第４の実施例によれば、画像
の向きに関係なく特徴抽出することができる。According to the fourth embodiment shown in FIG. 9, features can be extracted regardless of the orientation of the image.
【００６０】また、図４に示した第２の実施例で求めら
れた特徴量を組み合わせて判断すれば、より認識率を上
げることができる。Further, the recognition rate can be further increased by combining and determining the feature amounts obtained in the second embodiment shown in FIG.
【００６１】図１０は本発明による顔抽出方法の第５の
実施例のフローチャートである。FIG. 10 is a flowchart of the fifth embodiment of the face extraction method according to the present invention.
【００６２】まず、第１の実施例によって判定された顔
候補領域を小領域に分割する（Ｅ−１）。分割方法とし
て、以下の（１）〜（３）の３つの方法が考えられる。 （１）顔候補領域に外接する長方形を３×３の長方形に
分割し、顔候補領域と重なっている領域を小領域とす
る。上段３つの小領域を上の小領域、下段の真ん中の小
領域を下の小領域、中段と下段の両端の小領域を横の小
領域とする（図１１（ａ）参照）。 （２）顔候補領域に外接する長方形を対角線状に４つの
三角形に分割し、顔候補領域と重なっている領域を小領
域とする。左右の領域を横の小領域とする（図１１
（ｂ）参照）。 （３）顔候補領域に外接する長方形の４つの辺近傍と、
候補領域とが重なっている領域を小領域とする。左右の
領域を横の小領域とする（図１１（ｃ）参照）。First, the face candidate area determined by the first embodiment is divided into small areas (E-1). The following three methods (1) to (3) can be considered as the dividing method. (1) The rectangle circumscribing the face candidate area is divided into 3 × 3 rectangles, and the area overlapping the face candidate area is defined as a small area. The three small areas in the upper row are upper small areas, the small area in the middle of the lower row is a lower small area, and the small areas at both ends of the middle and lower rows are horizontal small areas (see FIG. 11A). (2) A rectangle circumscribing the face candidate area is diagonally divided into four triangles, and an area overlapping the face candidate area is defined as a small area. The left and right areas are lateral small areas (see FIG. 11).
(B)). (3) Near four sides of a rectangle circumscribing the face candidate area,
An area where the candidate area overlaps is defined as a small area. The left and right regions are horizontal small regions (see FIG. 11C).
【００６３】次に、特徴量として各小領域での色の分布
量を求める（Ｅ−２）。色の分布量としては以下の種類
が考えられる。 ・色相、明度の２次元空間の共分散楕円の面積。 ・色空間（Ｌ* ａ* ｂ* 、Ｌ* ｕ* ｖ* 等）での分散楕
円体の体積。 ・色相の分散。 ・明度の分散。 ・彩度の分散。Next, the color distribution amount in each small area is obtained as the feature amount (E-2). The following types can be considered as the color distribution amount. -The area of the covariance ellipse in the two-dimensional space of hue and lightness. The volume of the dispersion ellipsoid in the color space (L * a * b * , L * u * v *, etc.). -Hue dispersion. -Dispersion of brightness. -Saturation dispersion.
【００６４】ここでは、目や口や髪の生え際に相当する
小領域では色の分布量が大きく、ほほに相当する小領域
では色の分布量が小さいことを利用して顔の認識を行
う。すなわち、上下の領域（上、下の小領域）のそれぞ
れの分布量もしくは分布量の平均が、左右の領域（横の
小領域）のそれぞれの分布量もしくは分布量の平均より
大きいとき（Ｅ−３）にその顔候補領域は顔であると判
定する（Ｅ−４）。そして、この作業を画像を９０°ず
つ回転させて行なう。Here, face recognition is performed by utilizing the fact that the amount of color distribution is large in a small region corresponding to the hairline of the eyes, mouth and hair, and the amount of color distribution is small in a small region corresponding to the hairline. That is, when the distribution amount of each of the upper and lower regions (upper and lower small regions) or the average of the distribution amounts is larger than the respective distribution amount of each of the left and right regions (horizontal small regions) or the average of the distribution amounts (E− In 3), it is determined that the face candidate area is a face (E-4). Then, this work is performed by rotating the image by 90 °.
【００６５】ある一方向についての条件を式であらわす
と数２のようになる。数２において、v(X1) は小領域X1
での色の分布量を表している。When the condition for a certain one direction is expressed by an equation, the following equation 2 is obtained. In equation 2, v (X1) is the small area X1
Represents the amount of color distribution in.
【００６６】[0066]
【数２】 v(X1) > v(X2) all X1, X2 or average(v(X1)) > average(v(X2)) or v(X1) > average(v(X2)) all X1 or average(v(X1)) > v(X2) all X2 ここで、X1: (x,y)=上の小領域、下の小領域であり、X
2: (x,y)=横の小領域である。[Formula 2] v (X1)> v (X2) all X1, X2 or average (v (X1))> average (v (X2)) or v (X1)> average (v (X2)) all X1 or average (v (X1))> v (X2) all X2 where X1: (x, y) = small area above, small area below, and X
2: (x, y) = small horizontal area.
【００６７】なお、上、下の小領域のそれぞれの色の分
布量もしくは分布量の平均があるしきい値以上、横の小
領域のそれぞれの色の分布量もしくは分布量の平均があ
るしきい値以下、この二つの条件をともに満たすとき、
もしくは、どちらかを満たすときにその顔候補領域は顔
であると判定する方法もある。この場合も作業を９０°
ずつ回転させて行なう。この場合のある一方向について
条件を式であらわすと数３のようになる。数３におい
て、v(X1) は小領域X1での色の分布量を表している。It should be noted that the distribution amount of each color in the upper and lower small regions or the average of the distribution amount is above a certain threshold, and the distribution amount of each color in the horizontal small region or the average of the distribution amounts is a threshold. Below the value, when these two conditions are both satisfied,
Alternatively, there is also a method of determining that the face candidate area is a face when either of them is satisfied. Also in this case work 90 °
Rotate each time. In this case, the condition for one direction is expressed by a formula as shown in Formula 3. In Expression 3, v (X1) represents the color distribution amount in the small area X1.
【００６８】[0068]
【数３】{ v(X1) > C1 all X1 and/or v(X2) < C2 all X2 } or { average(v(X1)) > C3 and/or average(v(X2)) < C4 } ここで、X1: (x,y)=上の小領域、下の小領域であり、X
2: (x,y)=横の小領域である。[Formula 3] {v (X1)> C1 all X1 and / or v (X2) <C2 all X2} or {average (v (X1))> C3 and / or average (v (X2)) <C4} here Where X1: (x, y) = small area above, small area below, and X
2: (x, y) = small horizontal area.
【００６９】図１０に示した第５の実施例によれば、顔
の部位ごとの特徴を生かして認識するため認識率が高い
という効果がある。According to the fifth embodiment shown in FIG. 10, there is an effect that the recognition rate is high because the features of each face part are used for recognition.
【００７０】なお、図１０に示した第５の実施例では特
徴量として色の分布量を用いたが、特徴量として、小領
域の平均色の顔候補領域全体の平均色に対する色の差の
量と、方向とを用いる方法もある（図１２参照）。ここ
で、この方法について説明する。In the fifth embodiment shown in FIG. 10, the color distribution amount is used as the feature amount. However, as the feature amount, the difference in color between the average color of the small regions and the average color of the entire face candidate region is used. There is also a method of using the quantity and the direction (see FIG. 12). Here, this method will be described.
【００７１】まず、第５の実施例と同様に顔候補領域を
小領域に分割し、特徴量として、小領域の平均色の顔候
補領域全体の平均色に対する色の差の量と方向を求め
る。目や髪を判断する際には明度、口を判断する際には
色相を用いるとよい。First, as in the fifth embodiment, the face candidate area is divided into small areas, and the amount and direction of the color difference of the average color of the small areas from the average color of the entire face candidate area is obtained as the feature amount. . It is good to use lightness when judging eyes and hair, and to use hue when judging mouth.
【００７２】差としては以下の種類が考えられる。 ・色相、明度の２次元空間での二つの平均色の差ベクト
ル。 ・色空間（Ｌ* ａ* ｂ* 、Ｌ* ｕ* ｖ* 等）での二つの
平均色の差ベクトル。 ・色相の平均の差。 ・明度の平均の差。 ・彩度の平均の差。The following types of differences can be considered. A difference vector of two average colors in a two-dimensional space of hue and lightness. A difference vector of two average colors in a color space (L * a * b * , L * u * v *, etc.). -Difference in average hue. -The difference in average brightness. -The difference in average saturation.
【００７３】平均の計算方法としては以下の種類が考え
られる。 ・小領域内で均一な平均。 ・小領域の中央に重みをつけた平均。The following types can be considered as the average calculation method. -A uniform average within a small area. -Average weighted on the center of the small area.
【００７４】ここでは、目や口や髪の生え際に相当する
小領域で小領域の平均色と候補領域の平均色との色の差
の方向がそれぞれ黒、赤、黒の方向であるとき、顔と認
識する。具体的には、以下の３つの条件を同時にもしく
はいずれかを満たしているとき、その顔候補領域は顔で
あると判定する。 （１）上の小領域の顔候補領域の平均色に対するそれぞ
れの色の差方向もしくは色の差の平均の方向が明度が小
さくなる方向であること。 （２）下の小領域の顔候補領域の平均色に対するそれぞ
れの色の差方向もしくは色の差量の平均方向が肌色から
赤の方向であること。 （３）横の小領域の顔候補領域の平均色に対するそれぞ
れの色の差量もしくは色の差量の平均があるしきい値よ
り小さいこと。Here, when the direction of the color difference between the average color of the small area and the average color of the candidate area in the small area corresponding to the hairline of the eyes, mouth and hair is black, red and black respectively, Recognize as a face. Specifically, when the following three conditions are satisfied at the same time or any one of the following three conditions is satisfied, the face candidate area is determined to be a face. (1) The direction of difference in color or the direction of average of color differences with respect to the average color of the face candidate area of the small area above is the direction in which the lightness decreases. (2) The difference direction of each color or the average direction of the amount of color difference with respect to the average color of the face candidate area of the lower small area is from the skin color to the red direction. (3) The amount of difference of each color or the average of the amount of color difference with respect to the average color of the face candidate region in the lateral small region is smaller than a certain threshold.
【００７５】そして、この作業を画像を９０°ずつ回転
させて行なう。Then, this work is performed by rotating the image by 90 °.
【００７６】また、図１０に示した第５の実施例では特
徴量として色の分布量を用いたが、特徴量として空間周
波数的性質を用いる方法もある（図１３参照）。ここ
で、この方法について説明する。Although the color distribution amount is used as the feature amount in the fifth embodiment shown in FIG. 10, there is also a method of using the spatial frequency characteristic as the feature amount (see FIG. 13). Here, this method will be described.
【００７７】まず、第５の実施例と同様に顔候補領域を
小領域に分割し、特徴量として空間周波数的性質の値を
求める。First, as in the fifth embodiment, the face candidate area is divided into small areas, and the value of the spatial frequency characteristic is obtained as the feature quantity.
【００７８】空間周波数の対象としては以下の種類が考
えられる。 ・色相。 ・明度。 ・彩度。The following types can be considered as targets of the spatial frequency. -Hue. ·brightness. ·saturation.
【００７９】そして、特徴量およびその求め方としては
以下の種類が考えられる。 ・２次元フーリエ変換を行い、高周波領域内での応答の
積分値と空間周波数領域全体での応答の積分値との割合
を特徴量とする。 ・２次元フーリエ変換を行いパワースペクトラムを求
め、高周波領域でのピークの高さを特徴量とする。 ・四方の隣接する画素との差の平均値のヒストグラムを
求め、差の平均値が所定値以上である画素の割合を特徴
量とする。The following types can be considered as the feature amount and the method of obtaining the feature amount. A two-dimensional Fourier transform is performed, and the ratio of the integrated value of the response in the high frequency region to the integrated value of the response in the entire spatial frequency region is used as the feature amount. -The power spectrum is obtained by performing the two-dimensional Fourier transform, and the height of the peak in the high frequency region is used as the feature amount. -The histogram of the average value of the differences from the four adjacent pixels is obtained, and the ratio of the pixels having the average value of the differences equal to or greater than the predetermined value is set as the feature amount.
【００８０】ここでは、目や口や髪の生え際に相当する
小領域で特徴量が大きいとき、その顔候補領域は顔であ
ると判定する。すなわち、上、下の小領域の特徴量より
も横の小領域の特徴量が大きいときその顔候補領域は顔
であると判定する。やはり、この作業は画像を９０°ず
つ回転させて行なう。Here, when the feature amount is large in a small area corresponding to the hairline of the eyes, mouth or hair, the face candidate area is determined to be a face. That is, when the feature amount of the horizontal small regions is larger than the feature amount of the upper and lower small regions, it is determined that the face candidate region is a face. Again, this work is performed by rotating the image by 90 °.
【００８１】図１４は本発明による顔抽出方法の第６の
実施例のフローチャートである。FIG. 14 is a flowchart of the sixth embodiment of the face extraction method according to the present invention.
【００８２】まず、第１の実施例によって判定された顔
候補領域のサイズに基づいて目の大きさを推定し、図１
５（ａ）に示すような目のテンプレートおよび図１５
（ｂ）に示すような眼鏡をかけた目のテンプレートを作
成する（Ｆ−１）。First, the size of the eyes is estimated based on the size of the face candidate area determined by the first embodiment.
5 (a) eye template and FIG.
A template of eyeglasses is created as shown in (b) (F-1).
【００８３】そして、顔候補領域内の対象画像を目を表
す画像であるか否かで２値化する（Ｆ−２）。通常どお
りに明度、彩度、色相等の色情報で目を表す黒画素を検
出してもかまわないが、次に示す手法の方が好ましい。 （１）まず、顔候補領域内で肌色画素の明度の平均を求
める。 （２）この平均値との差（もちろん、目を表す画素の方
が小さくなる）がしきい値以上である場合に目を表す黒
画素であると判定する。Then, the target image in the face candidate area is binarized depending on whether it is an image showing eyes (F-2). The black pixel representing the eye may be detected by color information such as lightness, saturation, and hue as usual, but the following method is preferable. (1) First, the average of the luminosity of the flesh color pixels in the face candidate area is calculated. (2) If the difference from this average value (of course, the pixel representing the eye is smaller) is greater than or equal to the threshold value, it is determined to be a black pixel representing the eye.
【００８４】こうすることで、顔の明るさの違いによる
目の黒画素の明るさの違いをカバーすることができる。By doing so, it is possible to cover the difference in the brightness of the black pixels of the eye due to the difference in the brightness of the face.
【００８５】ステップ（Ｆ−２）で２値化された顔候補
領域に対して、垂直方向、水平方向にそれぞれの目テン
プレートを走査させ（図１５（ａ）参照）、最も高いマ
ッチング度を求め、これを特徴量とする（Ｆ−３）。For the face candidate area binarized in step (F-2), the eye templates are scanned in the vertical and horizontal directions (see FIG. 15A) to obtain the highest matching degree. , And let this be a feature amount (F-3).
【００８６】そして、この特徴量がしきい値以上であれ
ば（Ｆ−４）、その顔候補領域は顔であると判定する
（Ｆ−５）。If this feature amount is equal to or larger than the threshold value (F-4), it is determined that the face candidate area is a face (F-5).
【００８７】図１４に示した第６の実施例によれば、対
象者が眼鏡をかけているかどうかの判定が可能であり、
また、画像の向きの判定も可能となる効果がある。According to the sixth embodiment shown in FIG. 14, it is possible to judge whether or not the subject wears glasses.
Further, there is an effect that it is possible to determine the orientation of the image.
【００８８】図１６は本発明による顔抽出方法の第７の
実施例のフローチャートである。FIG. 16 is a flow chart of the seventh embodiment of the face extraction method according to the present invention.
【００８９】本実施例では、図１４に示した第６の実施
例の処理を行った後に、目のほかに口に関しても判断を
行う。従って、ステップ（Ｇ−１）〜（Ｇ−４）は図１
４に示したフローチャートと同様なので説明は省略す
る。In this embodiment, after the processing of the sixth embodiment shown in FIG. 14 is performed, judgment is made not only for the eyes but also for the mouth. Therefore, steps (G-1) to (G-4) are shown in FIG.
Since it is the same as the flowchart shown in FIG.
【００９０】ステップ（Ｇ−５）では、目の位置情報か
ら口の位置を推定する。たとえば、両目の位置がともに
顔の上半分にあった場合、口の位置は、顔の下半分にあ
ると推定する。また、両目の位置が共に顔の右半分にあ
った場合、口の位置は、顔の左半分にあると推定する。
また、通常行なわれているように、両目を結ぶ線分の中
点からの距離で口の位置を推定すればより高い検出精度
が得られる。In step (G-5), the mouth position is estimated from the eye position information. For example, when the positions of both eyes are both in the upper half of the face, it is estimated that the position of the mouth is in the lower half of the face. If both eyes are in the right half of the face, it is estimated that the mouth is in the left half of the face.
Further, if the position of the mouth is estimated by the distance from the midpoint of the line segment connecting both eyes, as is usually done, higher detection accuracy can be obtained.
【００９１】次に、口の候補領域に対して、色情報から
口が存在するかどうかを判定する。具体的には、たとえ
ば、ＢＧＲ信号を８ビット（０〜２５５）で表し、顔候
補領域全体でＧ−Ｒを計算し、その平均値を求める。次
に、口の候補領域に対して同様にＧ−Ｒを計算し、この
計算した値と先に求めた平均値との差を求め、この差が
しきい値以上であれば口画素としてカウントする。Next, with respect to the mouth candidate area, it is determined from the color information whether or not a mouth is present. Specifically, for example, the BGR signal is represented by 8 bits (0 to 255), GR is calculated for the entire face candidate area, and the average value thereof is obtained. Next, GR is similarly calculated for the mouth candidate area, the difference between the calculated value and the previously calculated average value is calculated, and if this difference is equal to or greater than the threshold value, it is counted as a mouth pixel. To do.
【００９２】そして、口画素の割合（口の画素とカウン
トされた画素数／顔候補領域内の全画素数）を求め、こ
れを第２の特徴量とする（Ｇ−６）。Then, the ratio of the mouth pixels (the number of pixels counted as mouth pixels / the total number of pixels in the face candidate area) is obtained, and this is set as the second feature amount (G-6).
【００９３】この第２の特徴量が所定範囲内であれば
（Ｇ−７）、口であると判定し、この顔候補領域は顔で
あると判定する（Ｇ−８）。各しきい値はそれぞれ測定
機器を用いて経験的に求めればよい。また、Ｇ−Ｒの代
わりに他の色情報、たとえば明度、色相、彩度、色度を
用いてもかまわない。If the second feature amount is within the predetermined range (G-7), it is determined to be a mouth, and this face candidate area is determined to be a face (G-8). Each threshold may be empirically obtained using a measuring device. Further, other color information such as lightness, hue, saturation, and chromaticity may be used instead of GR.
【００９４】図１６に示した第７の実施例によれば、口
の位置が顔の領域のどの部分にあるかがわかり、画像の
向きを判定することができる。また、より精度の高い顔
抽出ができるという効果もある。According to the seventh embodiment shown in FIG. 16, it is possible to determine which part of the face area the mouth is located in and to determine the orientation of the image. There is also an effect that face extraction can be performed with higher accuracy.
【００９５】なお、図１６に示した第７の実施例では目
のほかに口を検出することにより、顔であるか否かを判
定したが、本発明はこれに限らず、たとえば、目のほか
に髪を検出し顔であるかどうかを判定してもよい。以下
に、具体的に説明する。In the seventh embodiment shown in FIG. 16, whether or not it is a face is determined by detecting the mouth in addition to the eyes. However, the present invention is not limited to this, and for example, the eyes can be used. Alternatively, hair may be detected to determine whether or not it is a face. The details will be described below.
【００９６】目の位置が決定したら、その目の位置情報
から髪の位置を推定する。具体的には、図１７のよう
に、たとえば両目が顔候補領域の上半分にある場合、両
目を結ぶ線より上の境界領域を髪候補領域と推定し、こ
の領域内で髪を示す黒画素をカウントする。また、両目
が顔候補領域の右半分にある場合、両目を結ぶ線より右
の境界領域を髪候補領域と推定する。When the eye position is determined, the hair position is estimated from the eye position information. Specifically, as shown in FIG. 17, for example, when both eyes are in the upper half of the face candidate area, the boundary area above the line connecting both eyes is estimated as the hair candidate area, and the black pixel indicating the hair in this area is estimated. To count. If both eyes are in the right half of the face candidate area, the boundary area to the right of the line connecting both eyes is estimated to be the hair candidate area.
【００９７】そして、髪画素の割合（黒画素数／髪候補
領域の面積）がしきい値以上であれば髪があると判定
し、この顔候補領域は顔であると判定する。ここで、顔
候補領域のサイズを数画素拡張してから髪を検出する
と、より検出精度が上がる。If the ratio of hair pixels (the number of black pixels / area of hair candidate area) is equal to or greater than the threshold value, it is determined that there is hair, and this face candidate area is determined to be a face. Here, if hair is detected after the size of the face candidate area is expanded by several pixels, the detection accuracy is further improved.
【００９８】この場合には、髪の領域が顔の上半分にあ
るか、下半分にあるか、左半分にあるか、あるいは右半
分にあるかで、画像の向きを判定することができるとい
う効果がある。In this case, the orientation of the image can be determined depending on whether the hair region is in the upper half, the lower half, the left half, or the right half of the face. effective.
【００９９】また、図１６に示した第７の実施例では目
のほかに口を検出することにより、顔であるか否かを判
定したが、本発明はこれに限らず、目のほかに首を検出
して顔であるかどうかを判定してもよい。以下に、具体
的に説明する。In addition, in the seventh embodiment shown in FIG. 16, it is determined whether or not it is a face by detecting the mouth in addition to the eyes. However, the present invention is not limited to this, and other than the eyes. You may detect a neck and determine whether it is a face. The details will be described below.
【０１００】目の位置が決定したら、その目の位置情報
から首の位置を推定する。図１７のように、たとえば両
目が顔候補領域の上半分にある場合、下１／３を首候補
領域とする。下半分というように更に広い領域に限定し
てもかまわない。When the eye position is determined, the neck position is estimated from the eye position information. As shown in FIG. 17, for example, when both eyes are in the upper half of the face candidate area, the lower 1/3 is set as the neck candidate area. It may be limited to a wider area such as the lower half.
【０１０１】次に、顔候補領域のサイズを３〜４画素拡
張し、首候補領域と接触する領域で、首を表す肌色の画
素をカウントする。首画素の割合（カウント数／顔の候
補領域の面積）がしきい値以上であれば首であると判定
し、この顔候補領域は顔であると判定する。Next, the size of the face candidate area is expanded by 3 to 4 pixels, and the skin-colored pixels representing the neck are counted in the area in contact with the neck candidate area. If the ratio of the number of neck pixels (count number / area of face candidate area) is greater than or equal to the threshold value, it is determined to be a neck, and this face candidate area is determined to be a face.
【０１０２】図１８は本発明による顔抽出方法の第７の
実施例のフローチャートである。FIG. 18 is a flow chart of the seventh embodiment of the face extraction method according to the present invention.
【０１０３】本実施例は顔候補領域内でマッチドフィル
タを用いて、その顔候補領域が顔であるか否かを判定す
るものである。In this embodiment, a matched filter is used in the face candidate area to determine whether or not the face candidate area is a face.
【０１０４】まず、第１の実施例によって判定された顔
候補領域内の画像を明度画像に変換する（Ｈ−１）。次
に、顔テンプレート画像を作成する。この画像は明度画
像で、目、鼻、口が備わり、顔の向きも正面、横向きで
上、下、左、右の画像の向きにそれぞれ対応したものが
好ましい。また、いくつかの大きさの異なるテンプレー
トを作成しておくことが好ましい。これらの顔テンプレ
ート画像は予め作成しておき、記憶装置６に記憶させて
おいてもよい。また、本実施例では、この顔テンプレー
ト画像に対して２次元フーリエ変換を行うので、予め２
次元フーリエ変換を行っておき、その結果を記憶装置６
に記憶させておいてもよい。First, the image in the face candidate area determined by the first embodiment is converted into a brightness image (H-1). Next, a face template image is created. This image is a brightness image, and it is preferable that the image has eyes, a nose, and a mouth, and the orientation of the face is front, sideways, and corresponds to the orientations of the upper, lower, left, and right images, respectively. In addition, it is preferable to create several templates having different sizes. These face template images may be created in advance and stored in the storage device 6. In addition, in the present embodiment, since the two-dimensional Fourier transform is performed on this face template image, it is possible to perform 2
The three-dimensional Fourier transform is performed and the result is stored in the storage device 6.
It may be stored in.
【０１０５】そして、ステップ（Ｈ−１）の明度画像に
対して２次元フーリエ変換を行ない（Ｈ−２）、このス
テップ（Ｈ−２）の結果の実部と、顔テンプレート画像
の２次元フーリエ変換結果の実部とを足し合わせ（Ｈ−
３）、さらに、ステップ（Ｈ−２）の結果の虚部と、顔
テンプレート画像の２次元フーリエ変換結果の虚部とを
足し合わせる（Ｈ−４）。Then, two-dimensional Fourier transform is performed on the brightness image of step (H-1) (H-2), and the real part of the result of this step (H-2) and the two-dimensional Fourier transform of the face template image. Add the real part of the conversion result (H-
3) Furthermore, the imaginary part of the result of step (H-2) and the imaginary part of the two-dimensional Fourier transform result of the face template image are added (H-4).
【０１０６】次に、足し合わせた結果得られた実部と虚
部とを逆フーリエ変換し（ここまでの過程をマッチドフ
ィルタをかけるという）、この値を特徴量とする（Ｈ−
５）。Next, the real part and the imaginary part obtained as a result of the addition are subjected to inverse Fourier transform (the process up to this point is called a matched filter), and this value is used as a feature amount (H-
5).
【０１０７】逆フーリエ変換した値は、マッチング度が
高いほど大きくなるので、この特徴量がしきい値以上で
あれば（Ｈ−６）、その顔候補領域は顔であると判定す
る。The value obtained by the inverse Fourier transform becomes larger as the matching degree becomes higher. Therefore, if this feature amount is equal to or larger than the threshold value (H-6), it is determined that the face candidate area is a face.
【０１０８】なお、図１８に示した第７の実施例では、
顔候補領域内の画像を明度画像に変換したが、これを、
色相画像や彩度画像にしてもかまわない。Incidentally, in the seventh embodiment shown in FIG.
I converted the image in the face candidate area into a brightness image.
It may be a hue image or a saturation image.
【０１０９】図１８に示した第７の実施例によれば、周
波数特性を利用しているため、顔の各部位の個人差によ
る位置の違いに対する許容度が高いという効果がある。According to the seventh embodiment shown in FIG. 18, since the frequency characteristic is used, there is an effect that the tolerance for the position difference due to the individual difference of each part of the face is high.
【０１１０】図１９は本発明による顔抽出方法の第８の
実施例のフローチャートである。FIG. 19 is a flow chart of the eighth embodiment of the face extraction method according to the present invention.
【０１１１】本実施例は、顔候補領域内で顔の各部位の
特徴量をニューラルネットワークに入力し、その顔候補
領域が顔であるか否かを判定するものである。In this embodiment, the feature amount of each part of the face in the face candidate area is input to the neural network, and it is determined whether or not the face candidate area is a face.
【０１１２】まず、他の実施例において説明した顔の各
部位の特徴量を求める（Ｊ−１）。この特徴量としては
目のマッチング度、口画素の割合、髪画素の割合、首画
素の割合などが挙げられる。そして、これらの値をニュ
ーラルネットワークに入力する（Ｊ−２）。First, the feature amount of each part of the face described in the other embodiments is obtained (J-1). Examples of this feature amount include the degree of eye matching, the ratio of mouth pixels, the ratio of hair pixels, and the ratio of neck pixels. Then, these values are input to the neural network (J-2).
【０１１３】用いるニューラルネットワークは、入力
層、中間層、出力層から成る３層構造で、それぞれの要
素数は、たとえば４、３、１とする。ニューラルネット
ワークにおける教師データは、正面顔、横向き顔と、画
像の向きが上、下、右、左の組み合わせをカバーするよ
うに、それぞれのパターンについて作成しておく。教師
データの出力は顔であれば１、そうでなければ０とす
る。ニューラルネットワークには、上記２×４＝８パタ
ーンについて、それぞれの教師データを用いてバックプ
ロパゲーション法によって学習させ、各係数が決定され
ている。また、用いるニューラルネットワークの構成
は、中間層の数が２層以上であってもかまわないし、ま
た、中間層の要素数も３でなくてもかまわず、最も認識
率が高くなるように決定すればよい。The neural network used has a three-layer structure including an input layer, an intermediate layer, and an output layer, and the number of elements of each is, for example, 4, 3, 1. The teacher data in the neural network is created for each pattern so that the front face, the lateral face, and the combination of the image orientations of up, down, right, and left are covered. The output of the teacher data is set to 1 if the face and 0 otherwise. The neural network is trained on the 2 × 4 = 8 patterns by the back-propagation method using each teacher data, and each coefficient is determined. The configuration of the neural network to be used does not matter even if the number of intermediate layers is two or more, and the number of elements in the intermediate layer does not have to be three, and is determined so that the recognition rate is the highest. Good.
【０１１４】このようにして求められた出力層からの出
力がしきい値以上であれば（Ｊ−３）、その顔候補領域
は顔であると判定する（Ｊ−４）。If the output from the output layer thus obtained is equal to or greater than the threshold value (J-3), it is determined that the face candidate area is a face (J-4).
【０１１５】なお、ニューラルネットワークへの入力と
して、色を表すパラメータ、たとえば、明度、色相、彩
度等を用いてもかまわない。It is also possible to use parameters representing color, such as lightness, hue, and saturation, as inputs to the neural network.
【０１１６】また、この抽出方法を組み込んだ機器を出
荷した際、環境の違いにより、微調整が必要な場合に
は、新たにいくつかの教師データを加えることにより微
調整をするようにすればよい。When a device incorporating this extraction method is shipped and fine adjustment is necessary due to a difference in environment, some fine adjustment data may be added to make fine adjustment. Good.
【０１１７】図１９に示した第８の実施例によれば、ニ
ューラルネットワークをいくつかの教師データにより学
習させることにより、それぞれの入力値に対して最適な
重み係数を選択することができ、高い認識率を確保する
ことができるという効果がある。According to the eighth embodiment shown in FIG. 19, by learning the neural network with some teacher data, the optimum weighting coefficient can be selected for each input value, which is high. There is an effect that the recognition rate can be secured.
【０１１８】また、測定システム等の環境の違いによる
認識率低下が発生した場合にも、いくつかの教師データ
を加えることにより簡単に微調整が可能であるという効
果もある。Further, even if the recognition rate is lowered due to the difference in the environment of the measurement system or the like, there is an effect that the fine adjustment can be easily performed by adding some teaching data.
【０１１９】本発明による顔抽出方法は、カラー原画像
を複写材料に複写する際の露光量を求めるためだけでは
なく、その他各種の画像処理に用いることができること
は言うまでもない。また、本発明は抽出対象を顔以外の
ものにすることも可能である。It goes without saying that the face extraction method according to the present invention can be used not only for obtaining the exposure amount when a color original image is copied onto a copying material, but also for various other image processing. Further, in the present invention, the extraction target can be something other than a face.
【０１２０】[0120]
【発明の効果】以上説明したように、本発明によれば、
ネガフィルムやポジフィルム等のカラー原画像から、人
手を介さず完全に自動で且つ精度よく人間の顔を抽出す
ることができる。As described above, according to the present invention,
It is possible to extract a human face from a color original image such as a negative film or a positive film completely automatically and accurately without human intervention.
【図１】本発明による顔抽出方法を用いた顔領域抽出装
置のブロック図である。FIG. 1 is a block diagram of a face area extracting apparatus using a face extracting method according to the present invention.
【図２】本発明による顔抽出方法の第１の実施例のフロ
ーチャートである。FIG. 2 is a flowchart of a first embodiment of a face extraction method according to the present invention.
【図３】顔テンプレートの一例を示す図である。FIG. 3 is a diagram showing an example of a face template.
【図４】本発明による顔抽出方法の第２の実施例のフロ
ーチャートである。FIG. 4 is a flowchart of a second embodiment of the face extraction method according to the present invention.
【図５】顔候補領域内で目候補領域を設定する際の説明
図であり、（ａ）は縦画像の例、（ｂ）は横画像の例で
ある。5A and 5B are explanatory diagrams for setting eye candidate regions in a face candidate region, where FIG. 5A is an example of a vertical image and FIG. 5B is an example of a horizontal image.
【図６】顔候補領域内で目候補領域を設定する際の説明
図である。FIG. 6 is an explanatory diagram when setting an eye candidate area within a face candidate area.
【図７】明度に関するヒストグラムである。FIG. 7 is a histogram of lightness.
【図８】本発明による顔抽出方法の第３の実施例のフロ
ーチャートである。FIG. 8 is a flowchart of a third embodiment of the face extraction method according to the present invention.
【図９】本発明による顔抽出方法の第４の実施例のフロ
ーチャートである。FIG. 9 is a flowchart of a fourth embodiment of the face extraction method according to the present invention.
【図１０】本発明による顔抽出方法の第５の実施例のフ
ローチャートである。FIG. 10 is a flowchart of a fifth embodiment of the face extraction method according to the present invention.
【図１１】顔候補領域を小領域に分割する際の説明図で
あり、（ａ）、（ｂ）、（ｃ）は３つの分割方法のそれ
ぞれに対応した説明図である。FIG. 11 is an explanatory diagram when the face candidate region is divided into small regions, and (a), (b), and (c) are explanatory diagrams corresponding to each of the three division methods.
【図１２】特徴量として、小領域の平均色の顔候補領域
全体の平均色に対する色の差の量と、方向とを用いる方
法の説明図である。FIG. 12 is an explanatory diagram of a method of using the amount of color difference with respect to the average color of the entire face candidate area of the average color of the small areas and the direction as the feature quantity.
【図１３】特徴量として、空間周波数的性質を用いる方
法の説明図である。FIG. 13 is an explanatory diagram of a method of using a spatial frequency property as a feature amount.
【図１４】本発明による顔抽出方法の第６の実施例のフ
ローチャートである。FIG. 14 is a flowchart of a sixth embodiment of the face extraction method according to the present invention.
【図１５】目テンプレートの例を示す図であり、（ａ）
は目テンプレートとその走査方向について説明する図で
あり、（ｂ）は眼鏡をかけた目テンプレートの例を示す
図である。FIG. 15 is a diagram showing an example of an eye template, (a)
FIG. 4A is a diagram for explaining an eye template and its scanning direction, and FIG. 7B is a diagram showing an example of an eye template with glasses.
【図１６】本発明による顔抽出方法の第７の実施例のフ
ローチャートである。FIG. 16 is a flowchart of a seventh embodiment of the face extraction method according to the present invention.
【図１７】髪候補領域と首候補領域を推定する場合の説
明図である。FIG. 17 is an explanatory diagram for estimating a hair candidate area and a neck candidate area.
【図１８】本発明による顔抽出方法の第７の実施例のフ
ローチャートである。FIG. 18 is a flowchart of a seventh embodiment of the face extraction method according to the present invention.
【図１９】本発明による顔抽出方法の第８の実施例のフ
ローチャートである。FIG. 19 is a flowchart of an eighth embodiment of the face extraction method according to the present invention.
１ フィルム ２ スキャナ ３ 増幅器 ４ Ａ／Ｄ変換器 ５ ＣＰＵ ６ 記憶装置 1 Film 2 Scanner 3 Amplifier 4 A / D Converter 5 CPU 6 Storage Device
───────────────────────────────────────────────────── フロントページの続き (72)発明者 河合 まどか 東京都日野市さくら町１番地 コニカ株式 会社内 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Madoka Kawai No. 1 Sakura-cho, Hino City, Tokyo Konica Stock Company
Claims (28)
法において、人の顔の形状に相当する顔候補領域を決定
し、前記顔候補領域内の特徴量から顔領域を決定するこ
とを特徴とする顔抽出方法。1. An extraction method for extracting a human face from an image, wherein a face candidate area corresponding to the shape of a human face is determined, and a face area is determined from a feature amount in the face candidate area. A characteristic face extraction method.
より顔候補領域を検出することを特徴とする顔抽出方
法。2. A face extraction method, comprising detecting a face candidate area by extracting a contour of a human face from an image.
法において、複数の顔の形状をしたテンプレートと画像
とのマッチング度を計算し、マッチング度の最も高いテ
ンプレートを選択し、最も高かったマッチング度が予め
定めたしきい値以上であれば、選択されたテンプレート
内の領域を顔候補領域とすることを特徴とする顔抽出方
法。3. In an extraction method for extracting a human face from an image, the degree of matching between a template having a plurality of face shapes and an image is calculated, and the template with the highest degree of matching is selected, A face extraction method, wherein a region in the selected template is set as a face candidate region if the matching degree is equal to or more than a predetermined threshold value.
タの１次元のヒストグラムを求め、該ヒストグラムの形
状から顔領域を決定することを特徴とする請求項１ない
し３のいずれか１項に記載の顔抽出方法。4. The one-dimensional histogram of parameters representing color is obtained in the face candidate area, and the face area is determined from the shape of the histogram. The face extraction method described.
タの２次元のヒストグラムを求め、該ヒストグラムの形
状から顔領域を決定することを特徴とする請求項１ない
し３のいずれか１項に記載の顔抽出方法。5. A face area is determined from a shape of the histogram by obtaining a two-dimensional histogram of parameters representing colors in the face candidate area. The face extraction method described.
トグラムとして、明度のヒストグラムを用いることを特
徴とした請求項４に記載の顔抽出方法。6. The face extraction method according to claim 4, wherein a brightness histogram is used as a one-dimensional histogram of the parameters representing the colors.
タの分布量から顔領域を決定することを特徴とする請求
項１ないし３のいずれか１項に記載の顔抽出方法。7. The face extraction method according to claim 1, wherein the face area is determined from the distribution amount of a parameter representing a color in the face candidate area.
元空間の共分散楕円の面積を用いることを特徴とする請
求項７に記載の顔抽出方法。8. The face extraction method according to claim 7, wherein an area of a covariance ellipse of a two-dimensional space of hue and lightness is used as the color distribution amount.
領域を決定することを特徴とする請求項１ないし３のい
ずれか１項に記載の顔抽出方法。9. The face extraction method according to claim 1, wherein the face area is determined from the spatial frequency response of the face candidate area.
れぞれの小領域での特徴量を求め、顔候補領域内での特
徴量分布を用いて顔領域を決定することを特徴とする請
求項１ないし３のいずれか１項に記載の顔抽出方法。10. The face candidate area is divided into small areas, the feature amount in each small area is obtained, and the face area is determined using the feature amount distribution in the face candidate area. The face extraction method according to any one of items 1 to 3.
目、口、ほほが分離されることを特徴とする請求項１０
に記載の顔抽出方法。11. By dividing into the small regions,
11. The eye, mouth and cheek are separated.
The face extraction method described in.
を用いることを特徴とする請求項１０または１１に記載
の顔抽出方法。12. The face extraction method according to claim 10, wherein a color distribution amount of a small area is used as the feature amount.
次元空間の共分散楕円の面積を用いることを特徴とする
請求項１２に記載の顔抽出方法。13. The hue and the lightness of 2 as the distribution amount of the color.
13. The face extraction method according to claim 12, wherein the area of the covariance ellipse of the dimensional space is used.
色の顔候補領域における平均色に対する色の差の量と方
向とを用いることを特徴とする請求項１０または１１に
記載の顔抽出方法。14. The face extraction method according to claim 10, wherein an amount and a direction of a color difference with respect to the average color in the face candidate area of the average color in the small area is used as the feature amount.
域における平均色に対する色の差の量と方向として、色
相、明度の２次元空間での２つの平均色の差ベクトルを
用いることを特徴とする請求項１４に記載の顔抽出方
法。15. A difference vector of two average colors in a two-dimensional space of hue and lightness is used as the amount and direction of the color difference with respect to the average color in the face candidate area of the average color in the small area. The face extraction method according to claim 14.
的性質を用いることを特徴とする請求項１０または１１
に記載の顔抽出方法。16. The spatial frequency characteristic of a small area is used as the feature quantity.
The face extraction method described in.
色相の空間周波数応答を用いることを特徴とする請求項
１６に記載の顔抽出方法。17. The face extraction method according to claim 16, wherein a spatial frequency response of hue is used as a spatial frequency property of the small area.
領域を決定することを特徴とする請求項１ないし３のい
ずれか１項に記載の顔抽出方法。18. The face extraction method according to claim 1, wherein eyes are extracted from the face candidate area to determine the face area.
し、目の位置情報から口を抽出し、顔領域を決定するこ
とを特徴とする請求項１ないし３のいずれか１項に記載
の顔抽出方法。19. The face area is determined by extracting an eye area from the face candidate area, extracting a mouth from eye position information, and determining a face area. The face extraction method described.
し、目の位置情報から髪を抽出し、顔領域を決定するこ
とを特徴とする請求項１ないし３のいずれか１項に記載
の顔抽出方法。20. The face area is determined by extracting an eye area from the face candidate area, extracting hair from eye position information, and determining the face area. The face extraction method described.
し、目の位置情報から首を抽出し、顔領域を決定するこ
とを特徴とする請求項１ないし３のいずれか１項に記載
の顔抽出方法。21. The face area is determined by extracting an eye area from the face candidate area, extracting a neck from eye position information, and determining a face area. The face extraction method described.
拡大し、髪の領域を抽出することにより顔領域を決定す
ることを特徴とする請求項１ないし３のいずれか１項に
記載の顔抽出方法。22. The face area is determined by enlarging the face candidate area and extracting the hair area after the face candidate area is determined. Face extraction method.
拡大し、首の領域を抽出することにより顔領域を決定す
ることを特徴とする請求項１ないし３のいずれか１項に
記載の顔抽出方法。23. The face area is determined by enlarging the face candidate area and extracting a neck area after the face candidate area is determined. Face extraction method.
ターを使用して顔領域を決定することを特徴とする請求
項１ないし３のいずれか１項に記載の顔抽出方法。24. The face extracting method according to claim 1, wherein a face area is determined in the face candidate area using a matched filter.
特徴量をニューラルネットワークに入力し、顔領域を決
定することを特徴とする請求項１ないし３のいずれか１
項に記載の顔抽出方法。25. The facial area is determined by obtaining a characteristic amount in the candidate area and inputting the characteristic amount into a neural network.
The face extraction method described in the item.
と、口を表す特徴量と、髪を表す特徴量と、首を表す特
徴量とを用いることを特徴とした請求項２５に記載の顔
抽出方法。26. The feature quantity representing an eye, the feature quantity representing a mouth, the feature quantity representing a hair, and the feature quantity representing a neck are used as the feature quantities. Face extraction method.
方法を組み込んだ機器を出荷後、前記ニューラルネット
ワークにおける新たな教師データを入力することにより
微調整可能なことを特徴とする顔認識装置。27. A face recognition apparatus, characterized in that it can be finely adjusted by inputting new teacher data in the neural network after shipping a device incorporating the face extraction method according to claim 25 or 26.
記載の顔抽出方法によって抽出された人の顔のデータに
基づいて複写材料への露光量を決定するカラープリン
タ。28. A color printer for determining the exposure amount to a copy material based on the data of a human face extracted by the face extraction method according to claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP19673594A JP3557659B2 (en) | 1994-08-22 | 1994-08-22 | Face extraction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP19673594A JP3557659B2 (en) | 1994-08-22 | 1994-08-22 | Face extraction method |
Related Child Applications (4)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2004110883A Division JP2004206738A (en) | 2004-04-05 | 2004-04-05 | Face extraction method |
| JP2004110881A Division JP2004234688A (en) | 2004-04-05 | 2004-04-05 | Face extracting method |
| JP2004110882A Division JP2004234689A (en) | 2004-04-05 | 2004-04-05 | Face extracting method |
| JP2004110880A Division JP2004265431A (en) | 2004-04-05 | 2004-04-05 | Face extraction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPH0863597A true JPH0863597A (en) | 1996-03-08 |
| JP3557659B2 JP3557659B2 (en) | 2004-08-25 |
Family
ID=16362730
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP19673594A Expired - Lifetime JP3557659B2 (en) | 1994-08-22 | 1994-08-22 | Face extraction method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP3557659B2 (en) |
Cited By (82)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000123185A (en) * | 1998-10-14 | 2000-04-28 | Canon Inc | Object recognizing device |
| JP2000215319A (en) * | 1999-01-22 | 2000-08-04 | Canon Inc | Method and device for extracting image, and storage medium |
| JP2000311248A (en) * | 1999-04-28 | 2000-11-07 | Sharp Corp | Image processor |
| KR100311952B1 (en) * | 1999-01-11 | 2001-11-02 | 구자홍 | Method of face territory extraction using the templates matching with scope condition |
| JP2002216129A (en) * | 2001-01-22 | 2002-08-02 | Honda Motor Co Ltd | Face area detector, its method and computer readable recording medium |
| JP2002358522A (en) * | 2001-05-31 | 2002-12-13 | Olympus Optical Co Ltd | Device, method and program for supporting image selection |
| KR20030012193A (en) * | 2001-07-31 | 2003-02-12 | 주식회사 드림미르 | Method of eye position detection for the image with various background |
| JP2003507796A (en) * | 1999-08-17 | 2003-02-25 | ナショナル・インストゥルメンツ・コーポレーション | System and method for locating regions of matching color and pattern in a target image |
| KR20030040680A (en) * | 2001-11-15 | 2003-05-23 | 삼성에스디에스 주식회사 | Method for detecting face from video data and apparatus thereof |
| KR100422709B1 (en) * | 2001-10-05 | 2004-03-16 | 엘지전자 주식회사 | Face detecting method depend on image |
| KR20040049893A (en) * | 2002-12-05 | 2004-06-14 | 김의선 | Automatically Face Avatar Generation Method of Vector Image Type from an Image |
| KR100439377B1 (en) * | 2002-01-17 | 2004-07-09 | 엘지전자 주식회사 | Human area detection for mobile video telecommunication system |
| KR100450793B1 (en) * | 2001-01-20 | 2004-10-01 | 삼성전자주식회사 | Apparatus for object extraction based on the feature matching of region in the segmented images and method therefor |
| KR100474771B1 (en) * | 2002-08-29 | 2005-03-10 | 엘지전자 주식회사 | Face detection method based on templet |
| JP2005108195A (en) * | 2003-09-09 | 2005-04-21 | Fuji Photo Film Co Ltd | Object identification unit, method, and program |
| GB2408887A (en) * | 2003-12-02 | 2005-06-08 | Hewlett Packard Development Co | Digital camera providing selective removal and addition of an imaged object |
| US7123754B2 (en) | 2001-05-22 | 2006-10-17 | Matsushita Electric Industrial Co., Ltd. | Face detection device, face pose detection device, partial image extraction device, and methods for said devices |
| JP2006285959A (en) * | 2005-03-07 | 2006-10-19 | Fuji Photo Film Co Ltd | Learning method of face recognition device, and method, device and program for face recognition |
| JP2007110528A (en) * | 2005-10-14 | 2007-04-26 | Pentax Corp | Imaging apparatus |
| JP2007122561A (en) * | 2005-10-31 | 2007-05-17 | Kddi Corp | Portable terminal device and portrait data providing system |
| US7224823B2 (en) | 2001-09-25 | 2007-05-29 | Matsushita Electric Industrial Co., Ltd. | Parameter estimation apparatus and data matching apparatus |
| KR100779171B1 (en) * | 2006-08-18 | 2007-11-26 | 학교법인 포항공과대학교 | Real time robust face detection apparatus and method |
| WO2008013050A1 (en) * | 2006-07-25 | 2008-01-31 | Nikon Corporation | Image processing method, image processing program, and image processing device |
| JP2008504060A (en) * | 2004-06-24 | 2008-02-14 | ヴィジョンゲイト，インコーポレーテッド | Method for correcting the relative movement (MOTION) of an object detector (OBJECT-DETECTOR) generated between successive views |
| US7356190B2 (en) | 2002-07-02 | 2008-04-08 | Canon Kabushiki Kaisha | Image area extraction method, image reconstruction method using the extraction result and apparatus thereof |
| US7397955B2 (en) | 2003-05-20 | 2008-07-08 | Fujifilm Corporation | Digital camera and method of controlling same |
| JP2008191785A (en) * | 2007-02-01 | 2008-08-21 | Toyota Motor Corp | Eye closure detection apparatus, doze detection apparatus, eye closure detection method, and program for eye closure detection |
| US7564486B2 (en) | 2003-09-29 | 2009-07-21 | Canon Kabushiki Kaisha | Image sensing apparatus with feature extraction mechanism and its control method |
| JP2009289295A (en) * | 2009-09-10 | 2009-12-10 | Canon Inc | Image extraction method and apparatus |
| US7643742B2 (en) | 2005-11-02 | 2010-01-05 | Olympus Corporation | Electronic camera, image processing apparatus, image processing method and image processing computer program |
| US7650068B2 (en) | 2006-02-14 | 2010-01-19 | Nikon Corporation | Camera and method of imaging |
| JP2010020602A (en) * | 2008-07-11 | 2010-01-28 | Nikon Corp | Image matching device and camera |
| US7706675B2 (en) | 2006-04-14 | 2010-04-27 | Nikon Corporation | Camera |
| US7791656B2 (en) | 2005-08-16 | 2010-09-07 | Konica Minolta Holdings, Inc. | Image sensing apparatus and image processing method |
| US7791668B2 (en) | 2005-01-18 | 2010-09-07 | Nikon Corporation | Digital camera |
| US7852395B2 (en) | 2006-10-10 | 2010-12-14 | Nikon Corporation | Camera with two image pickup sections |
| US7876373B2 (en) | 2006-10-10 | 2011-01-25 | Nikon Corporation | Camera with viewfinder |
| US7881600B2 (en) | 2006-02-15 | 2011-02-01 | Hoya Corporation | Photographing device |
| US7920725B2 (en) | 2003-09-09 | 2011-04-05 | Fujifilm Corporation | Apparatus, method, and program for discriminating subjects |
| WO2011089893A1 (en) | 2010-01-20 | 2011-07-28 | キヤノン株式会社 | Displacement detection device, and error correction method for scale for displacement detection device |
| US8009976B2 (en) | 2006-04-04 | 2011-08-30 | Nikon Corporation | Camera having face detection and with automatic focus control using a defocus amount |
| EP2391115A2 (en) | 2010-05-24 | 2011-11-30 | Canon Kabushiki Kaisha | Image processing apparatus, control method, and program |
| US8107768B2 (en) | 2004-06-09 | 2012-01-31 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, program for implementing the method, and storage medium storing the program |
| US8184192B2 (en) | 2008-03-28 | 2012-05-22 | Canon Kabushiki Kaisha | Imaging apparatus that performs an object region detection processing and method for controlling the imaging apparatus |
| EP2458846A1 (en) | 2010-11-30 | 2012-05-30 | Canon Kabushiki Kaisha | Image pickup apparatus that automatically determines shooting mode most suitable for shooting scene, control method therefor, and storage medium |
| JP2012120799A (en) * | 2010-12-10 | 2012-06-28 | Olympus Corp | Image processing apparatus, image processing method, and image processing program |
| US8253847B2 (en) | 2007-04-04 | 2012-08-28 | Nikon Corporation | Digital camera having an automatic focus |
| JP2012221148A (en) * | 2011-04-07 | 2012-11-12 | Hitachi Ltd | Image processing method and image processing system |
| US8374439B2 (en) | 2008-06-25 | 2013-02-12 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, program, and computer-readable print medium |
| JP2013041493A (en) * | 2011-08-18 | 2013-02-28 | Nippon Hoso Kyokai <Nhk> | Face image feature amount generation apparatus and program for generating face image feature amount |
| US8391595B2 (en) | 2006-05-26 | 2013-03-05 | Canon Kabushiki Kaisha | Image processing method and image processing apparatus |
| US8411159B2 (en) | 2006-10-25 | 2013-04-02 | Fujifilm Corporation | Method of detecting specific object region and digital camera |
| US8442315B2 (en) | 2010-07-16 | 2013-05-14 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and computer-readable medium |
| US8457397B2 (en) | 2010-07-16 | 2013-06-04 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and computer-readable medium |
| JP2013533563A (en) * | 2010-07-28 | 2013-08-22 | インターナショナル・ビジネス・マシーンズ・コーポレーション | Semantic analysis of objects in video |
| US8532431B2 (en) | 2007-05-08 | 2013-09-10 | Canon Kabushiki Kaisha | Image search apparatus, image search method, and storage medium for matching images with search conditions using image feature amounts |
| US8610817B2 (en) | 2010-12-28 | 2013-12-17 | Canon Kabushiki Kaisha | Image-capturing apparatus and control method therefor |
| US8780231B2 (en) | 2008-12-09 | 2014-07-15 | Canon Kabushiki Kaisha | Imaging apparatus and zoom control method |
| JP2015032991A (en) * | 2013-08-02 | 2015-02-16 | 株式会社ニコン | Imaging apparatus |
| US9002107B2 (en) | 2010-07-16 | 2015-04-07 | Canon Kabushiki Kaisha | Color balance correction based on skin color and highlight color |
| US9014487B2 (en) | 2012-07-09 | 2015-04-21 | Canon Kabushiki Kaisha | Image processing apparatus and image processing method |
| US9134399B2 (en) | 2010-07-28 | 2015-09-15 | International Business Machines Corporation | Attribute-based person tracking across multiple cameras |
| US9189681B2 (en) | 2012-07-09 | 2015-11-17 | Canon Kabushiki Kaisha | Image processing apparatus, method thereof, and computer-readable storage medium |
| US9208595B2 (en) | 2012-07-09 | 2015-12-08 | Canon Kabushiki Kaisha | Apparatus, image processing method and storage medium storing program |
| US9214027B2 (en) | 2012-07-09 | 2015-12-15 | Canon Kabushiki Kaisha | Apparatus, method, and non-transitory computer-readable medium |
| US9275270B2 (en) | 2012-07-09 | 2016-03-01 | Canon Kabushiki Kaisha | Information processing apparatus and control method thereof |
| US9280720B2 (en) | 2012-07-09 | 2016-03-08 | Canon Kabushiki Kaisha | Apparatus, method, and computer-readable storage medium |
| US9292760B2 (en) | 2012-07-09 | 2016-03-22 | Canon Kabushiki Kaisha | Apparatus, method, and non-transitory computer-readable medium |
| US9299177B2 (en) | 2012-07-09 | 2016-03-29 | Canon Kabushiki Kaisha | Apparatus, method and non-transitory computer-readable medium using layout similarity |
| US9330312B2 (en) | 2010-07-28 | 2016-05-03 | International Business Machines Corporation | Multispectral detection of personal attributes for video surveillance |
| US9436706B2 (en) | 2013-09-05 | 2016-09-06 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and storage medium for laying out images |
| US9501688B2 (en) | 2012-07-09 | 2016-11-22 | Canon Kabushiki Kaisha | Apparatus, processing method and storage medium storing program |
| US9509870B2 (en) | 2013-09-05 | 2016-11-29 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and storage medium enabling layout varations |
| US9519842B2 (en) | 2012-07-09 | 2016-12-13 | Canon Kabushiki Kaisha | Apparatus and method for managing an object extracted from image data |
| US9542594B2 (en) | 2013-06-28 | 2017-01-10 | Canon Kabushiki Kaisha | Information processing apparatus, method for processing information, and program |
| US9558212B2 (en) | 2012-07-09 | 2017-01-31 | Canon Kabushiki Kaisha | Apparatus, image processing method and computer-readable storage medium for object identification based on dictionary information |
| US9563823B2 (en) | 2012-07-09 | 2017-02-07 | Canon Kabushiki Kaisha | Apparatus and method for managing an object extracted from image data |
| US9846681B2 (en) | 2012-07-09 | 2017-12-19 | Canon Kabushiki Kaisha | Apparatus and method for outputting layout image |
| US9904879B2 (en) | 2013-09-05 | 2018-02-27 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and storage medium |
| WO2018110680A1 (en) * | 2016-12-15 | 2018-06-21 | ナーブ株式会社 | Image processing device, image processing method, and image processing program |
| US10013395B2 (en) | 2012-07-09 | 2018-07-03 | Canon Kabushiki Kaisha | Apparatus, control method thereof, and storage medium that determine a layout image from a generated plurality of layout images by evaluating selected target images |
| US10424342B2 (en) | 2010-07-28 | 2019-09-24 | International Business Machines Corporation | Facilitating people search in video surveillance |
-
1994
- 1994-08-22 JP JP19673594A patent/JP3557659B2/en not_active Expired - Lifetime
Cited By (103)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000123185A (en) * | 1998-10-14 | 2000-04-28 | Canon Inc | Object recognizing device |
| KR100311952B1 (en) * | 1999-01-11 | 2001-11-02 | 구자홍 | Method of face territory extraction using the templates matching with scope condition |
| JP2000215319A (en) * | 1999-01-22 | 2000-08-04 | Canon Inc | Method and device for extracting image, and storage medium |
| JP2000311248A (en) * | 1999-04-28 | 2000-11-07 | Sharp Corp | Image processor |
| JP2003507796A (en) * | 1999-08-17 | 2003-02-25 | ナショナル・インストゥルメンツ・コーポレーション | System and method for locating regions of matching color and pattern in a target image |
| KR100450793B1 (en) * | 2001-01-20 | 2004-10-01 | 삼성전자주식회사 | Apparatus for object extraction based on the feature matching of region in the segmented images and method therefor |
| JP2002216129A (en) * | 2001-01-22 | 2002-08-02 | Honda Motor Co Ltd | Face area detector, its method and computer readable recording medium |
| JP4597391B2 (en) * | 2001-01-22 | 2010-12-15 | 本田技研工業株式会社 | Facial region detection apparatus and method, and computer-readable recording medium |
| US7123754B2 (en) | 2001-05-22 | 2006-10-17 | Matsushita Electric Industrial Co., Ltd. | Face detection device, face pose detection device, partial image extraction device, and methods for said devices |
| US8055030B2 (en) | 2001-05-31 | 2011-11-08 | Olympus Corporation | Image selection support system for supporting selection of well-photographed image from plural images |
| JP2002358522A (en) * | 2001-05-31 | 2002-12-13 | Olympus Optical Co Ltd | Device, method and program for supporting image selection |
| KR20030012193A (en) * | 2001-07-31 | 2003-02-12 | 주식회사 드림미르 | Method of eye position detection for the image with various background |
| US7224823B2 (en) | 2001-09-25 | 2007-05-29 | Matsushita Electric Industrial Co., Ltd. | Parameter estimation apparatus and data matching apparatus |
| KR100422709B1 (en) * | 2001-10-05 | 2004-03-16 | 엘지전자 주식회사 | Face detecting method depend on image |
| KR20030040680A (en) * | 2001-11-15 | 2003-05-23 | 삼성에스디에스 주식회사 | Method for detecting face from video data and apparatus thereof |
| KR100439377B1 (en) * | 2002-01-17 | 2004-07-09 | 엘지전자 주식회사 | Human area detection for mobile video telecommunication system |
| US7542615B2 (en) | 2002-07-02 | 2009-06-02 | Canon Kabushiki Kaisha | Image area extraction method, image reconstruction method using the extraction result and apparatus thereof |
| US7356190B2 (en) | 2002-07-02 | 2008-04-08 | Canon Kabushiki Kaisha | Image area extraction method, image reconstruction method using the extraction result and apparatus thereof |
| KR100474771B1 (en) * | 2002-08-29 | 2005-03-10 | 엘지전자 주식회사 | Face detection method based on templet |
| KR20040049893A (en) * | 2002-12-05 | 2004-06-14 | 김의선 | Automatically Face Avatar Generation Method of Vector Image Type from an Image |
| US7397955B2 (en) | 2003-05-20 | 2008-07-08 | Fujifilm Corporation | Digital camera and method of controlling same |
| US8184870B2 (en) | 2003-09-09 | 2012-05-22 | Fujifilm Corporation | Apparatus, method, and program for discriminating subjects |
| JP2005108195A (en) * | 2003-09-09 | 2005-04-21 | Fuji Photo Film Co Ltd | Object identification unit, method, and program |
| US7920725B2 (en) | 2003-09-09 | 2011-04-05 | Fujifilm Corporation | Apparatus, method, and program for discriminating subjects |
| US7564486B2 (en) | 2003-09-29 | 2009-07-21 | Canon Kabushiki Kaisha | Image sensing apparatus with feature extraction mechanism and its control method |
| GB2408887A (en) * | 2003-12-02 | 2005-06-08 | Hewlett Packard Development Co | Digital camera providing selective removal and addition of an imaged object |
| US8107768B2 (en) | 2004-06-09 | 2012-01-31 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, program for implementing the method, and storage medium storing the program |
| US9275436B2 (en) | 2004-06-09 | 2016-03-01 | Canon Kabushiki Kaisha | Image processing apparatus for editing data in accordance with an user operation, image processing method, program for implementing the method, and storage medium storing the program |
| JP2008504060A (en) * | 2004-06-24 | 2008-02-14 | ヴィジョンゲイト，インコーポレーテッド | Method for correcting the relative movement (MOTION) of an object detector (OBJECT-DETECTOR) generated between successive views |
| US7791668B2 (en) | 2005-01-18 | 2010-09-07 | Nikon Corporation | Digital camera |
| JP2006285959A (en) * | 2005-03-07 | 2006-10-19 | Fuji Photo Film Co Ltd | Learning method of face recognition device, and method, device and program for face recognition |
| US7990443B2 (en) | 2005-08-16 | 2011-08-02 | Konica Minolta Holdings, Inc. | Image sensing apparatus and image processing method |
| US7791656B2 (en) | 2005-08-16 | 2010-09-07 | Konica Minolta Holdings, Inc. | Image sensing apparatus and image processing method |
| JP2007110528A (en) * | 2005-10-14 | 2007-04-26 | Pentax Corp | Imaging apparatus |
| JP2007122561A (en) * | 2005-10-31 | 2007-05-17 | Kddi Corp | Portable terminal device and portrait data providing system |
| US7643742B2 (en) | 2005-11-02 | 2010-01-05 | Olympus Corporation | Electronic camera, image processing apparatus, image processing method and image processing computer program |
| US7865075B2 (en) | 2005-11-02 | 2011-01-04 | Olympus Corporation | Electronic camera, image processing apparatus, image processing method and image processing computer program |
| US7650068B2 (en) | 2006-02-14 | 2010-01-19 | Nikon Corporation | Camera and method of imaging |
| US7881600B2 (en) | 2006-02-15 | 2011-02-01 | Hoya Corporation | Photographing device |
| US8009976B2 (en) | 2006-04-04 | 2011-08-30 | Nikon Corporation | Camera having face detection and with automatic focus control using a defocus amount |
| US7706675B2 (en) | 2006-04-14 | 2010-04-27 | Nikon Corporation | Camera |
| US8391595B2 (en) | 2006-05-26 | 2013-03-05 | Canon Kabushiki Kaisha | Image processing method and image processing apparatus |
| US8040389B2 (en) | 2006-07-25 | 2011-10-18 | Nikon Corporation | Image processing method, image processing program and image processing apparatus for detecting object of an image |
| JP4692632B2 (en) * | 2006-07-25 | 2011-06-01 | 株式会社ニコン | Image processing method, image processing program, and image processing apparatus |
| JPWO2008013050A1 (en) * | 2006-07-25 | 2009-12-17 | 株式会社ニコン | Image processing method, image processing program, and image processing apparatus |
| WO2008013050A1 (en) * | 2006-07-25 | 2008-01-31 | Nikon Corporation | Image processing method, image processing program, and image processing device |
| KR100779171B1 (en) * | 2006-08-18 | 2007-11-26 | 학교법인 포항공과대학교 | Real time robust face detection apparatus and method |
| US7876373B2 (en) | 2006-10-10 | 2011-01-25 | Nikon Corporation | Camera with viewfinder |
| US7852395B2 (en) | 2006-10-10 | 2010-12-14 | Nikon Corporation | Camera with two image pickup sections |
| US8411159B2 (en) | 2006-10-25 | 2013-04-02 | Fujifilm Corporation | Method of detecting specific object region and digital camera |
| JP2008191785A (en) * | 2007-02-01 | 2008-08-21 | Toyota Motor Corp | Eye closure detection apparatus, doze detection apparatus, eye closure detection method, and program for eye closure detection |
| US8253847B2 (en) | 2007-04-04 | 2012-08-28 | Nikon Corporation | Digital camera having an automatic focus |
| US8532431B2 (en) | 2007-05-08 | 2013-09-10 | Canon Kabushiki Kaisha | Image search apparatus, image search method, and storage medium for matching images with search conditions using image feature amounts |
| US8184192B2 (en) | 2008-03-28 | 2012-05-22 | Canon Kabushiki Kaisha | Imaging apparatus that performs an object region detection processing and method for controlling the imaging apparatus |
| US9894280B2 (en) | 2008-03-28 | 2018-02-13 | Canon Kabushiki Kaisha | Imaging apparatus for displaying selected face in in-focus state and method for controlling imaging apparatus |
| US8374439B2 (en) | 2008-06-25 | 2013-02-12 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, program, and computer-readable print medium |
| JP2010020602A (en) * | 2008-07-11 | 2010-01-28 | Nikon Corp | Image matching device and camera |
| US8780231B2 (en) | 2008-12-09 | 2014-07-15 | Canon Kabushiki Kaisha | Imaging apparatus and zoom control method |
| JP2009289295A (en) * | 2009-09-10 | 2009-12-10 | Canon Inc | Image extraction method and apparatus |
| WO2011089893A1 (en) | 2010-01-20 | 2011-07-28 | キヤノン株式会社 | Displacement detection device, and error correction method for scale for displacement detection device |
| US8639030B2 (en) | 2010-05-24 | 2014-01-28 | Canon Kabushiki Kaisha | Image processing using an adaptation rate |
| EP2391115A2 (en) | 2010-05-24 | 2011-11-30 | Canon Kabushiki Kaisha | Image processing apparatus, control method, and program |
| US9398282B2 (en) | 2010-05-24 | 2016-07-19 | Canon Kabushiki Kaisha | Image processing apparatus, control method, and computer-readable medium |
| US9002107B2 (en) | 2010-07-16 | 2015-04-07 | Canon Kabushiki Kaisha | Color balance correction based on skin color and highlight color |
| US9406003B2 (en) | 2010-07-16 | 2016-08-02 | Canon Kabushiki Kaisha | Image processing with color balance correction |
| US8457397B2 (en) | 2010-07-16 | 2013-06-04 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and computer-readable medium |
| US8442315B2 (en) | 2010-07-16 | 2013-05-14 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and computer-readable medium |
| US8934712B2 (en) | 2010-07-16 | 2015-01-13 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and computer-readable medium |
| US8842914B2 (en) | 2010-07-16 | 2014-09-23 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and computer-readable medium |
| US10424342B2 (en) | 2010-07-28 | 2019-09-24 | International Business Machines Corporation | Facilitating people search in video surveillance |
| US9002117B2 (en) | 2010-07-28 | 2015-04-07 | International Business Machines Corporation | Semantic parsing of objects in video |
| US9330312B2 (en) | 2010-07-28 | 2016-05-03 | International Business Machines Corporation | Multispectral detection of personal attributes for video surveillance |
| US9134399B2 (en) | 2010-07-28 | 2015-09-15 | International Business Machines Corporation | Attribute-based person tracking across multiple cameras |
| JP2013533563A (en) * | 2010-07-28 | 2013-08-22 | インターナショナル・ビジネス・マシーンズ・コーポレーション | Semantic analysis of objects in video |
| US8994783B2 (en) | 2010-11-30 | 2015-03-31 | Canon Kabushiki Kaisha | Image pickup apparatus that automatically determines shooting mode most suitable for shooting scene, control method therefor, and storage medium |
| EP2458846A1 (en) | 2010-11-30 | 2012-05-30 | Canon Kabushiki Kaisha | Image pickup apparatus that automatically determines shooting mode most suitable for shooting scene, control method therefor, and storage medium |
| JP2012120799A (en) * | 2010-12-10 | 2012-06-28 | Olympus Corp | Image processing apparatus, image processing method, and image processing program |
| US8610817B2 (en) | 2010-12-28 | 2013-12-17 | Canon Kabushiki Kaisha | Image-capturing apparatus and control method therefor |
| JP2012221148A (en) * | 2011-04-07 | 2012-11-12 | Hitachi Ltd | Image processing method and image processing system |
| JP2013041493A (en) * | 2011-08-18 | 2013-02-28 | Nippon Hoso Kyokai <Nhk> | Face image feature amount generation apparatus and program for generating face image feature amount |
| US9214027B2 (en) | 2012-07-09 | 2015-12-15 | Canon Kabushiki Kaisha | Apparatus, method, and non-transitory computer-readable medium |
| US9014487B2 (en) | 2012-07-09 | 2015-04-21 | Canon Kabushiki Kaisha | Image processing apparatus and image processing method |
| US9292760B2 (en) | 2012-07-09 | 2016-03-22 | Canon Kabushiki Kaisha | Apparatus, method, and non-transitory computer-readable medium |
| US9299177B2 (en) | 2012-07-09 | 2016-03-29 | Canon Kabushiki Kaisha | Apparatus, method and non-transitory computer-readable medium using layout similarity |
| US9275270B2 (en) | 2012-07-09 | 2016-03-01 | Canon Kabushiki Kaisha | Information processing apparatus and control method thereof |
| US9208595B2 (en) | 2012-07-09 | 2015-12-08 | Canon Kabushiki Kaisha | Apparatus, image processing method and storage medium storing program |
| US9189681B2 (en) | 2012-07-09 | 2015-11-17 | Canon Kabushiki Kaisha | Image processing apparatus, method thereof, and computer-readable storage medium |
| US10127436B2 (en) | 2012-07-09 | 2018-11-13 | Canon Kabushiki Kaisha | Apparatus, image processing method and storage medium storing program |
| US9501688B2 (en) | 2012-07-09 | 2016-11-22 | Canon Kabushiki Kaisha | Apparatus, processing method and storage medium storing program |
| US10055640B2 (en) | 2012-07-09 | 2018-08-21 | Canon Kabushiki Kaisha | Classification of feature information into groups based upon similarity, and apparatus, image processing method, and computer-readable storage medium thereof |
| US9519842B2 (en) | 2012-07-09 | 2016-12-13 | Canon Kabushiki Kaisha | Apparatus and method for managing an object extracted from image data |
| US10013395B2 (en) | 2012-07-09 | 2018-07-03 | Canon Kabushiki Kaisha | Apparatus, control method thereof, and storage medium that determine a layout image from a generated plurality of layout images by evaluating selected target images |
| US9558212B2 (en) | 2012-07-09 | 2017-01-31 | Canon Kabushiki Kaisha | Apparatus, image processing method and computer-readable storage medium for object identification based on dictionary information |
| US9563823B2 (en) | 2012-07-09 | 2017-02-07 | Canon Kabushiki Kaisha | Apparatus and method for managing an object extracted from image data |
| US9846681B2 (en) | 2012-07-09 | 2017-12-19 | Canon Kabushiki Kaisha | Apparatus and method for outputting layout image |
| US9852325B2 (en) | 2012-07-09 | 2017-12-26 | Canon Kabushiki Kaisha | Apparatus, image processing method and storage medium storing program |
| US9280720B2 (en) | 2012-07-09 | 2016-03-08 | Canon Kabushiki Kaisha | Apparatus, method, and computer-readable storage medium |
| US9542594B2 (en) | 2013-06-28 | 2017-01-10 | Canon Kabushiki Kaisha | Information processing apparatus, method for processing information, and program |
| JP2015032991A (en) * | 2013-08-02 | 2015-02-16 | 株式会社ニコン | Imaging apparatus |
| US9904879B2 (en) | 2013-09-05 | 2018-02-27 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and storage medium |
| US9509870B2 (en) | 2013-09-05 | 2016-11-29 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and storage medium enabling layout varations |
| US9436706B2 (en) | 2013-09-05 | 2016-09-06 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and storage medium for laying out images |
| WO2018110680A1 (en) * | 2016-12-15 | 2018-06-21 | ナーブ株式会社 | Image processing device, image processing method, and image processing program |
Also Published As
| Publication number | Publication date |
|---|---|
| JP3557659B2 (en) | 2004-08-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP3557659B2 (en) | Face extraction method | |
| US7609908B2 (en) | Method for adjusting the brightness of a digital image utilizing belief values | |
| US6529630B1 (en) | Method and device for extracting principal image subjects | |
| US7333653B2 (en) | Detecting and correcting redeye in an image | |
| JP4153108B2 (en) | Image processing method, image processing apparatus, and recording medium | |
| JP4335476B2 (en) | Method for changing the number, size, and magnification of photographic prints based on image saliency and appeal | |
| US8391595B2 (en) | Image processing method and image processing apparatus | |
| US7454040B2 (en) | Systems and methods of detecting and correcting redeye in an image suitable for embedded applications | |
| US6898316B2 (en) | Multiple image area detection in a digital image | |
| JP4050842B2 (en) | Image processing method | |
| JP3506958B2 (en) | Image processing method, image processing apparatus, and recording medium | |
| US6445819B1 (en) | Image processing method, image processing device, and recording medium | |
| US20040114797A1 (en) | Method for automatic determination of color-density correction values for the reproduction of digital image data | |
| JP2005310123A (en) | Apparatus for selecting image of specific scene, program therefor and recording medium with the program recorded thereon | |
| JP3499305B2 (en) | Face area extraction method and exposure amount determination method | |
| JP2004265431A (en) | Face extraction method | |
| JP3576654B2 (en) | Exposure determination method, figure extraction method, and face area determination method | |
| JP2006259974A (en) | Image-processing method and device | |
| JP3576025B2 (en) | Main subject extraction method and apparatus | |
| JP3516786B2 (en) | Face area extraction method and copy condition determination method | |
| JPH08138024A (en) | Picture direction discriminating method | |
| JP2848749B2 (en) | Feature image data extraction method | |
| JPH04346332A (en) | Exposure deciding method | |
| JP2004234689A (en) | Face extracting method | |
| JP3575976B2 (en) | Main subject extraction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A02 | Decision of refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A02 Effective date: 20040210 |
|
| A521 | Written amendment |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20040318 |
|
| A911 | Transfer of reconsideration by examiner before appeal (zenchi) |
Free format text: JAPANESE INTERMEDIATE CODE: A911 Effective date: 20040323 |
|
| TRDD | Decision of grant or rejection written | ||
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 Effective date: 20040420 |
|
| A61 | First payment of annual fees (during grant procedure) |
Free format text: JAPANESE INTERMEDIATE CODE: A61 Effective date: 20040510 |
|
| R150 | Certificate of patent or registration of utility model |
Free format text: JAPANESE INTERMEDIATE CODE: R150 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20080528 Year of fee payment: 4 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20090528 Year of fee payment: 5 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20090528 Year of fee payment: 5 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20100528 Year of fee payment: 6 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20110528 Year of fee payment: 7 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20110528 Year of fee payment: 7 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20120528 Year of fee payment: 8 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20120528 Year of fee payment: 8 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20130528 Year of fee payment: 9 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20130528 Year of fee payment: 9 |
|
| S531 | Written request for registration of change of domicile |
Free format text: JAPANESE INTERMEDIATE CODE: R313531 |
|
| S533 | Written request for registration of change of name |
Free format text: JAPANESE INTERMEDIATE CODE: R313533 |
|
| R350 | Written notification of registration of transfer |
Free format text: JAPANESE INTERMEDIATE CODE: R350 |
|
| EXPY | Cancellation because of completion of term |