JP2015138448A5

JP2015138448A5 -

Info

Publication number: JP2015138448A5
Application number: JP2014010441A
Authority: JP
Filing date: 2014-01-23
Publication date: 2017-02-23
Anticipated expiration: 2034-01-23

Description

本発明に係る画像処理装置は、入力画像を複数の分割領域に分割する分割手段と、前記複数の分割領域のそれぞれについて輝度を含む少なくとも１つの特徴量を算出する特徴量算出手段と、前記特徴量算出手段が算出した特徴量に基づいて前記入力画像における光源の位置を推定する光源推定手段と、前記複数の分割領域のそれぞれが特定の被写体を示す領域であるか否かを判別する領域判別手段とを備え、前記領域判別手段はさらに、前記光源推定手段により検出された光源の位置に基づいて、前記特定の被写体を示す領域であると判別した分割領域に隣接する分割領域のそれぞれが前記特定の被写体を示す領域であるか否かを判別することを特徴とする。 An image processing apparatus according to the present invention includes a dividing unit that divides an input image into a plurality of divided regions, a feature amount calculating unit that calculates at least one feature amount including luminance for each of the plurality of divided regions, and the feature region discriminating the light source estimation unit amount calculation means for estimating the position of the light source in the input image based on the calculated features, whether the respective front Symbol plurality of divided regions is a region indicating a particular subject and a discrimination means, said area discriminating means further on the basis of the position of the light source detected by said light source estimation means, each of the divided areas adjacent to the divided region is determined that it is a region indicating the particular object It is determined whether or not the area indicates the specific subject.

ステップＳ２００では、画像取得部１００が、撮像素子から出力されるアナログ電気信号を取得し、取得したアナログ電気信号にＡ／Ｄ変換を施し、生成されたデジタル信号に種々の信号処理を施すことで撮影画像の画像データを生成する。図３（Ａ）は、ステップＳ２００で得られた撮影画像３００の例を示す図である。撮影画像３００は、被写体として、逆光下の人物と、空（青空）と、川等を含むものとする。 In step S200, the image acquisition unit 100 acquires an analog electrical signal output from the image sensor, performs A / D conversion on the acquired analog electrical signal, and performs various signal processing on the generated digital signal. Image data of the captured image is generated. FIG. 3A is a diagram illustrating an example of the captured image 300 obtained in step S200. Photographed image 300, as an object scene body, the person of backlit, empty (blue sky), is intended to include river or the like.

１００画像取得部
１１０画像分割部
１２０特徴量算出部
１３０領域判別部
１４０光源推定部 100 image acquisition unit 110 image segmentation unit 120 feature quantity calculation unit 130 region discrimination unit 140 light source estimation unit

Claims

A dividing means for dividing the input image into a plurality of divided regions;
Feature quantity calculating means for calculating at least one feature quantity including luminance for each of the plurality of divided regions;
Light source estimation means for estimating the position of the light source in the input image based on the feature quantity calculated by the feature quantity calculation means ;
And a region discriminating means for each of the previous SL plurality of divided regions is determined whether or not the area indicating the specific object,
The area discriminating unit further indicates that each of the divided areas adjacent to the divided area determined to be an area indicating the specific subject based on the position of the light source detected by the light source estimating unit indicates the specific subject. An image processing apparatus for determining whether or not an area.

The image processing apparatus according to claim 1, wherein the dividing unit divides the input image into a plurality of regions each having a block size of equal size.

The image processing apparatus according to claim 1, wherein the dividing unit detects an edge from the input image and divides the input image into a plurality of regions based on the detected edge.

The feature amount includes at least one of a luminance gradient and its gradient direction, hue, hue gradient and its gradient direction, edge amount, frequency, and coordinates in addition to the luminance. The image processing apparatus according to any one of the above.

Said area determination means, any of claims 1 to 4 wherein each of the plurality of divided areas, wherein the determine specific to Rukoto whether the region showing the face of the person as one of the specific subject The image processing apparatus according to claim 1.

The region determination unit further determines whether each of the plurality of divided regions is a region showing the sky as one of the specific subjects based on the feature amount calculated by the feature amount calculation unit, 6. The light source estimation unit estimates the position of the light source based on a luminance gradient of an area obtained by integrating the divided areas determined to indicate the sky. The image processing apparatus according to item 1.

Before Symbol light source estimation means, characterized in that estimating the position of the light source based on the luminance gradient determined from the brightness value, wherein the feature quantity calculating means is calculated for each of the divided areas is determined that indicates the specific subject The image processing apparatus according to any one of claims 1 to 5.

The dividing means further includes a plurality of subdivided areas each of a divided area determined by the area determining means as the specific subject and a divided area adjacent to the divided area determined as the area indicating the specific subject. Divided into
The region determination unit determines whether the plurality of subdivision regions are regions indicating the specific subject based on the feature amount calculated by the feature amount calculation unit, and then determines the plurality of subdivision regions. Based on the position of the light source detected by the light source estimation means, it is determined again whether the subdivision area that is determined not to be the area indicating the specific subject is the area indicating the specific subject. The image processing apparatus according to claim 1, wherein:

An image processing method executed in an image processing apparatus,
A division step of dividing the input image into a plurality of divided regions;
A calculation step of calculating at least one feature amount including luminance for the plurality of divided regions;
A determining step of determining whether each of the plurality of divided regions is a region indicating a specific subject based on the feature amount calculated in the calculating step;
An estimation step for estimating a position of a light source in the input image based on the feature amount calculated in the calculation step ;
Based on the position of the light source detected by the previous SL estimation step, or each of the divided areas adjacent to the divided region extracted as a region indicating the specific subject in the determining step is a region indicating the specific subject An image processing method comprising: a redetermination step for redetermining whether or not.

A program that causes a computer included in an image processing apparatus to execute the image processing method according to claim 9 .