System and method using complex centroids for detecting object-of-interest speeds from two dimensional pictures

Abstract

A traffic monitoring system includes a first car moving on a first path; a camera having a field of vision including at least a portion of the first path; and a computing system. The computing system receives a plurality of images from the camera. The computing system has a processor. When instructed, the processor performs circling a perimeter of the first car on each of the images with a first rectangle; composing a first set of points, each point of the first set of points representing a center of the first rectangle; finding a first centroid using the first set of points, wherein the first centroid represents the first path; and calculating a speed of the first car using the first centroid.

What is claimed is: 
     
         1 . A traffic monitoring system, comprising
 a camera having a field of vision including at least a portion of the first path; and   a computing system, the computing system receiving a plurality of images from the camera, the images includes a first car moving on a first path, the computing system having a processor, when instructed, the processor performs
 circling a perimeter of the first car on each of the images with a first rectangle; 
 composing a first set of points, each point of the first set of points representing a center of the first rectangle; 
 finding a first centroid using the first set of point, wherein the first centroid represents the first path; and 
 calculating a speed of the first car using the first centroid. 
   
     
     
         2 . The traffic monitoring system according to  claim 1 , wherein the first centroid is selected from a plurality of potential centroid candidates, the first centroid produces a least summation of distances between the first centroid and the first set of points compared to other members of the plurality of potential centroid candidates. 
     
     
         3 . The traffic monitoring system according to  claim 1 , wherein the first centroid is one selected from a point, a straight line, a curved line, and a plain. 
     
     
         4 . The traffic monitoring system according to  claim 3 , when the first centroid is a curved line, the first centroid is parabolic or polynomial. 
     
     
         5 . The traffic monitoring system according to  claim 1 , wherein the first centroid includes a slope parameter and a shift parameter. 
     
     
         6 . The traffic monitoring system according to  claim 5 , wherein the slope parameter and the shift parameter are set as to produce a least summation of distances between the first centroid and the first set of points. 
     
     
         7 . The traffic monitoring system according to  claim 1 , wherein the images are taken by the camera with a fixed time interval. 
     
     
         8 . The traffic monitoring system according to  claim 7 , wherein a first speed of the first car is calculated using the fixed time interval. 
     
     
         9 . The traffic monitoring system according to  claim 1 , comprising a second car moving on a second path, the processor finds a second centroid representing the second path. 
     
     
         10 . The traffic monitoring system according to  claim 9 , wherein a distance between the first centroid and the second centroid is set as 12 feet. 
     
     
         11 . A computing system, comprising:
 a non-transitory computer-readable medium comprising instructions which, when executed by a processor of the computing system, causes the processor to perform the steps of:
 receiving a plurality of images taken by a traffic camera, the images includes a first car moving on a first path; 
 circling a perimeter of the first car on each of the images with a first rectangle; 
 composing a first set of points, each point of the first set of points representing a center of the first rectangle; 
 finding a first centroid using the first set of point, wherein the first centroid represents the first path; 
 calculating a speed of the first car using the first centroid. 
   
     
     
         12 . The computing system according to  claim 11 , wherein the first centroid is selected from a plurality of potential centroid candidates, the first centroid produces a least summation of distances between the first centroid and the first set of points compared to other members of the plurality of potential centroid candidates. 
     
     
         13 . The computing system according to  claim 11 , wherein the first centroid is one selected from a point, a straight line, a curved line, and a plain. 
     
     
         14 . The computing system according to  claim 13 , when the first centroid is a curved line, the first centroid is parabolic or polynomial. 
     
     
         15 . The computing system according to  claim 11 , wherein the first centroid includes a slope parameter and a shift parameter. 
     
     
         16 . The computing system according to  claim 15 , wherein the slope parameter and the shift parameter are set as to produce a least summation of distances between the first centroid and the first set of points. 
     
     
         17 . The computing system according to  claim 11 , wherein the images are taken by the traffic camera with a fixed time interval. 
     
     
         18 . The computing system according to  claim 11 , wherein a first speed of the first car is calculated using the fixed time interval. 
     
     
         19 . The computing system according to  claim 11 , comprising a second car moving on a second path, the processor finds a second centroid representing the second path. 
     
     
         20 . The computing system according to  claim 19 , wherein a distance between the first centroid and the second centroid is set as 12 feet.