公开(公告)号：US09924104B2

公开(公告)日：2018-03-20

申请号：US14890150

申请日：2014-04-28

Applicant: THE UNIVERSITY OF TOKYO

Inventor： Hiromasa Oku ,  Keiko Yokoyama ,  Masatoshi Ishikawa

IPC: H04N5/232 ,  G06K9/46 ,  H04N5/225 ,  G06T7/20 ,  G01S3/786 ,  G06T7/11 ,  G06T7/174 ,  G06T7/194

CPC classification number: H04N5/23296 ,  G01S3/7864 ,  G06K9/4609 ,  G06K9/4652 ,  G06T7/11 ,  G06T7/174 ,  G06T7/194 ,  G06T7/20 ,  G06T2207/20224 ,  G06T2207/30244 ,  H04N5/2259 ,  H04N5/23216

Abstract: This invention provides a technology that allows the acquisition of background differentials using images acquired by a camera, the viewpoint of which moves. First, said camera acquires a background image, i.e. an image that does not contain a target object. Using information on the line-of-sight direction and viewpoint position of the camera, position information for pixels in the background image is transformed to polar-coordinate information Next, the camera acquires an object image. Using information on the line-of-sight direction and viewpoint position of the camera, position information for pixels in the object image is transformed to polar-coordinate information. The polar-coordinate information for the pixels in the object image is used to identify a target background, said target background being the part of the background image in an angle region corresponding to the object image. By comparing the object image and the target background, the differential therebetween is extracted.

公开(公告)号：US20160088231A1

公开(公告)日：2016-03-24

申请号：US14890150

申请日：2014-04-28

Applicant: THE UNIVERSITY OF TOKYO

Inventor： Hiromasa Oku ,  Keiko Yokoyama ,  Masatoshi Ishikawa

IPC: H04N5/232 ,  G06K9/46 ,  G06T7/20 ,  G06T7/00

CPC classification number: H04N5/23296 ,  G01S3/7864 ,  G06K9/4609 ,  G06K9/4652 ,  G06T7/11 ,  G06T7/174 ,  G06T7/194 ,  G06T7/20 ,  G06T2207/20224 ,  G06T2207/30244 ,  H04N5/2259 ,  H04N5/23216

Abstract: This invention provides a technology that allows the acquisition of background differentials using images acquired by a camera, the viewpoint of which moves. First, said camera acquires a background image, i.e. an image that does not contain a target object. Using information on the line-of-sight direction and viewpoint position of the camera, position information for pixels in the background image is transformed to polar-coordinate information Next, the camera acquires an object image. Using information on the line-of-sight direction and viewpoint position of the camera, position information for pixels in the object image is transformed to polar-coordinate information. The polar-coordinate information for the pixels in the object image is used to identify a target background, said target background being the part of the background image in an angle region corresponding to the object image. By comparing the object image and the target background, the differential therebetween is extracted.

Abstract translation: 本发明提供一种技术，其允许使用由其视点移动的相机获取的图像来获取背景差异。 首先，所述相机获取背景图像，即不包含目标对象的图像。 使用摄像机的视线方向和视点位置的信息，将背景图像中的像素的位置信息变换为极坐标信息。接下来，摄像机获取对象图像。 使用摄像机的视线方向和视点位置的信息，将对象图像中的像素的位置信息变换为极坐标信息。 用于对象图像中的像素的极坐标信息用于识别目标背景，所述目标背景是与对象图像对应的角度区域中的背景图像的一部分。 通过比较对象图像和目标背景，提取其间的差分。