公开(公告)号：US11210348B2

公开(公告)日：2021-12-28

申请号：US16396682

申请日：2019-04-27

Applicant: HUIZHOU UNIVERSITY

Inventor： Jinqiu Huang ,  Deming Xu ,  Changlin Wan

IPC: G06F16/906 ,  G06F17/18 ,  G06K9/62

Abstract: The present disclosure provides a data clustering method based on K-nearest neighbor, which sorts data points to be clustered in ascending order according to the maximum radiuses of K-nearest neighbors of the data points, that is, according to the density, and perform the first pass across the data points after sorting the data points in ascending order to incorporate the data points that conform to the statistical similarity into the same cluster; then perform the second pass across the data points with smaller cluster density according to the scale required during the clustering to find out all noise points and incorporate non-noise points into the nearest large-density cluster, so as to realize data clustering, which has the benefits of no need to preset the number of clusters and know the probability distribution of the data and convenience to set parameters.

公开(公告)号：US20190251121A1

公开(公告)日：2019-08-15

申请号：US16396682

申请日：2019-04-27

Applicant: HUIZHOU UNIVERSITY

Inventor： Jinqiu Huang ,  Deming Xu ,  Changlin Wan

IPC: G06F16/906 ,  G06K9/62 ,  G06F17/18

CPC classification number: G06F16/906 ,  G06F17/18 ,  G06K9/6215 ,  G06K9/6221 ,  G06K9/6223

Abstract: The present disclosure provides a data clustering method based on K-nearest neighbor, which sorts data points to be clustered in ascending order according to the maximum radiuses of K-nearest neighbors of the data points, that is, according to the density, and perform the first pass across the data points after sorting the data points in ascending order to incorporate the data points that conform to the statistical similarity into the same cluster; then perform the second pass across the data points with smaller cluster density according to the scale required during the clustering to find out all noise points and incorporate non-noise points into the nearest large-density cluster, so as to realize data clustering, which has the benefits of no need to preset the number of clusters and know the probability distribution of the data and convenience to set parameters.

公开(公告)号：US20180174328A1

公开(公告)日：2018-06-21

申请号：US15553697

申请日：2016-05-19

Applicant: HUIZHOU UNIVERSITY

Inventor： Changlin Wan ,  Deming Xu ,  Jianzhong Cao ,  Xiaohui Wei

IPC: G06T7/90 ,  G06T7/13 ,  G06T5/20 ,  G06T5/00

CPC classification number: G06T7/90 ,  G06K9/4604 ,  G06K9/4671 ,  G06T5/002 ,  G06T5/20 ,  G06T7/13

Abstract: The invention relates to a turning radius-based corner detection algorithm, comprising: S1: removing noise by Gaussian filtering and computing a gradient value of each pixel of an original image; S2: locating a neighboring pixel with closest grayscale to the pixel within given neighborhoods therearound; S3: computing a turning radius between the pixel and the closest neighboring pixel thereof; S4: computing a turning radius threshold; S5: marking a pixel with the turning radius which is greater than the threshold and maximum in the given neighborhoods as a corner. By the solution, the present invention can locate corners in images accurately and restrain fake corners resulting from noises and textures, and also can simplify the computation of the threshold and raise the computation efficiency, whereby automatic detection is realized and effect of corner detection is improved. The invention is applicable to 3D reproduction, visual locating, measurement, etc.