公开(公告)号：US07702156B2

公开(公告)日：2010-04-20

申请号：US11399684

申请日：2006-04-05

Applicant: Christopher Galea

Inventor： Christopher Galea

IPC: G06K9/46 ,  G06K9/34

CPC classification number: G06K9/4609 ,  G06T7/12

Abstract: A method for the extraction of shadows in noisy images includes capture of an image, subdivision of the image into a plurality of pixels, definition of a square odd matrix for every pixel having said pixel in the center, application of an algorithm to the matrixes of every pixel, and displaying of an error map of the image. The algorithm is a contour algorithm that includes the operations of =computation of the averages for each corner of the matrix, computation of the averages for the top, bottom, left and right sides, computation of the maximum variance between these eight averages, and comparison with a present value (max variance).

Abstract translation: 用于在噪声图像中提取阴影的方法包括图像的捕获，将图像细分为多个像素，对于具有中心具有所述像素的每个像素的平方奇数矩阵的定义，将算法应用于 每个像素，以及显示图像的错误图。 该算法是一种轮廓算法，包括运算矩阵的每个角的平均值的运算，上，下，左和右两边的平均值的计算，这八个平均值之间的最大方差的计算，以及比较 具有现值（最大方差）。

公开(公告)号：US20060280369A1

公开(公告)日：2006-12-14

申请号：US11399684

申请日：2006-04-05

Applicant: Christopher Galea

Inventor： Christopher Galea

IPC: G06K9/48

CPC classification number: G06K9/4609 ,  G06T7/12

Abstract: A method for the extraction of shadows in noisy images includes capture of an image, subdivision of the image into a plurality of pixels, definition of a square odd matrix for every pixel having said pixel in the center, application of an algorithm to the matrixes of every pixel, and displaying of an error map of the image. The algorithm is a contour algorithm that includes the operaitons of = computation of the averages for each corner of the matrix, computation of the averages for the top, bottom, left and right sides, computation of the maximum variance between these eight averages, and comparison with a present value (max variance).

Abstract translation: 用于在噪声图像中提取阴影的方法包括图像的捕获，将图像细分为多个像素，对于具有中心具有所述像素的每个像素的平方奇数矩阵的定义，将算法应用于 每个像素，以及显示图像的错误图。 该算法是一种轮廓算法，其包括矩阵的每个角的平均值的计算的运算，顶部，底部，左侧和右侧的平均值的计算，这八个平均值之间的最大方差的计算以及比较 具有现值（最大方差）。