Method of operating three-dimensional video surveillance system
First Claim
Patent Images
1. A method of video surveillance, the method comprising:
- generating a disparity map according to two sets of image data captured using a left camera and a right camera;
identifying at least one object in the disparity map;
mapping the at least one object onto a plane view, wherein the plane view has a three dimensional section of interest, a bottom area of the section of interest is enclosed by an artificial boundary and a physical boundary, the section of interest comprises a plurality of sub-sections each having cross-sectional areas identical to the bottom area, and a sum of heights of the sub-sections is equal to a height of the section of interest;
tracking the at least one object in the plane view to identify a height of each of the at least one object;
when detecting an object of the at least one object has crossed the artificial boundary, comparing a height of the object with at least one height range of at least one sub-section of the section of interest, and determining if the object has crossed a surface of each of the at least one sub-section according to a comparison result;
after determining the object has crossed the surface according to the comparison result, detecting number of components of the object crossing the surface, a direction the object crossing the surface, a ratio of a portion of the object crossing the surface to the object, and/or a period of time the object has crossed the surface;
comparing the ratio of the portion of the object crossing the surface to the object and a ratio threshold of the sub-section; and
determining the object has intruded the sub-section when the ratio of the portion of the object crossing the surface to the object is higher than the ratio threshold of the sub-section.
1 Assignment
0 Petitions
Accused Products
Abstract
A method includes generating a disparity map according to two sets of image data, identifying at least one object in the disparity map, mapping the at least one object onto a plane view, tracking at least one object on the plane view, and providing a robust algorithm about a cross-line time interval and a cross-line degree. The robust algorithm includes detecting whether the at least one object enters a predetermined region when the at least one object crosses a predetermined boundary.
1 Citation
6 Claims
-
1. A method of video surveillance, the method comprising:
-
generating a disparity map according to two sets of image data captured using a left camera and a right camera; identifying at least one object in the disparity map; mapping the at least one object onto a plane view, wherein the plane view has a three dimensional section of interest, a bottom area of the section of interest is enclosed by an artificial boundary and a physical boundary, the section of interest comprises a plurality of sub-sections each having cross-sectional areas identical to the bottom area, and a sum of heights of the sub-sections is equal to a height of the section of interest; tracking the at least one object in the plane view to identify a height of each of the at least one object; when detecting an object of the at least one object has crossed the artificial boundary, comparing a height of the object with at least one height range of at least one sub-section of the section of interest, and determining if the object has crossed a surface of each of the at least one sub-section according to a comparison result; after determining the object has crossed the surface according to the comparison result, detecting number of components of the object crossing the surface, a direction the object crossing the surface, a ratio of a portion of the object crossing the surface to the object, and/or a period of time the object has crossed the surface; comparing the ratio of the portion of the object crossing the surface to the object and a ratio threshold of the sub-section; and determining the object has intruded the sub-section when the ratio of the portion of the object crossing the surface to the object is higher than the ratio threshold of the sub-section. - View Dependent Claims (2)
-
-
3. A method of a video surveillance, the method comprising:
-
generating a disparity map according to two sets of image data captured using a left camera and a right camera; identifying at least one object in the disparity map; mapping the at least one object onto a plane view, wherein the plane view has a three dimensional section of interest, a bottom area of the section of interest is enclosed by an artificial boundary and a physical boundary, the section of interest comprises a plurality of sub-sections each having cross-sectional areas identical to the bottom area, and a sum of heights of the sub-sections is equal to a height of the section of interest; tracking the at least one object in the plane view to identify a height of each of the at least one object; when detecting an object of the at least one object has crossed the artificial boundary, comparing a height of the object with at least one height range of at least one sub-section of the section of interest, and determining if the object has crossed a surface of each of the at least one sub-section according to a comparison result; after determining the object has crossed the surface according to the comparison result, detecting number of components of the object crossing the surface, a direction the object crossing the surface, a ratio of a portion of the object crossing the surface to the object, and/or a period of time the object has crossed the surface; comparing a period of time the object has crossed the surface and a time threshold of the sub-section; and determining the object has intruded the sub-section when the period of time the object has crossed the surface is longer than the time threshold of the sub-section. - View Dependent Claims (4)
-
-
5. A method of video surveillance, the method comprising:
-
generating a disparity map according to two sets of image data captured using a left camera and a right camera; identifying at least one object in the disparity map; mapping the at least one object onto a plane view, wherein the plane view has a three dimensional section of interest, a bottom area of the section of interest is enclosed by an artificial boundary and a physical boundary, the section of interest comprises a plurality of sub-sections each having cross-sectional areas identical to the bottom area, and a sum of heights of the sub-sections is equal to a height of the section of interest; tracking the at least one object in the plane view to identify a height of each of the at least one object; when detecting an object of the at least one object has crossed the artificial boundary, comparing a height of the object with at least one height range of at least one sub-section of the section of interest, and determining if the object has crossed a surface of each of the at least one sub-section according to a comparison result; after determining the object has crossed the surface according to the comparison result, detecting number of components of the object crossing the surface, a direction the object crossing the surface, a ratio of a portion of the object crossing the surface to the object, and/or a period of time the object has crossed the surface; comparing number of components of the object crossing the surface and a number threshold of the sub-section; and determining the object has intruded the sub-section when the number of components of the object crossing the surface is larger than the number threshold of the sub-section. - View Dependent Claims (6)
-
Specification