Detecting persons concealed in a vehicle

• US 6,873,921 B1
• Filed: 09/29/2003
• Issued: 03/29/2005
• Est. Priority Date: 09/29/2003
• Status: Active Grant

• Current Assignee: UT-Battelle LLC
• Original Assignee: UT-Battelle LLC
• Inventors: Tucker, Raymond W. Jr.
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): Barbee, Manuel L
• Application Number:
  US10/673,720
• Publication Number:
  US 20050071115A1
• Time in Patent Office: 547 Days
• Field of Search: 702/56, 702/70, 702/73, 702 74- 77, 702/108, 702189-191, 702/193, 702/195, 702/197, 702/199, 367/136, 340/425.5, 340/426.1, 340/429, 340/426.24, 340/426.26, 340/565, 340/566, 340/573.1, 340/539.12, 340/541, 116/28.R, 116/33, 180/287, 180/271, 735/70, 735/79, 736/59, 324/76.12, 324/76.14, 324/76.21, 324/76.28, 701/42
• US Class Current: 702/56
• CPC Class Codes: G01V 1/001 Acoustic presence detection
• CPC Class Codes: G08B 13/1663 using seismic sensing means
• CPC Class Codes: G08B 21/22 responsive to presence or absence of persons

Abstract

An improved method for detecting the presence of humans or animals concealed within in a vehicle uses a combination of the continuous wavelet transform and a ratio-based energy calculation to determine whether the motion detected using seismic sensors placed on the vehicle is due to the presence of a heartbeat within the vehicle or is the result of motion caused by external factors such as the wind. The method performs well in the presence of light to moderate ambient wind levels, producing far fewer false alarm indications. The new method significantly improves the range of ambient environmental conditions under which human presence detection systems can reliably operate.

Claims

1. A method for detecting the presence of humans or animals in a vehicle, comprising the steps of:

a) acquiring one or more digital channel signal representations of the motion of the vehicle under test using one or more motion sensors placed on the vehicle;

b) computing the continuous wavelet transform of each channel signal to obtain a two-dimensional wavelet-space representation image of the time-frequency distribution of energy within each input signal;

c) computing a corresponding energy image for each channel signal from the wavelet-space representation;

d) computing the total energy in the signal for each channel signal using the energy image;

e) comparing the total energy for each channel signal to a preset maximum and if the total energy in any channel signal exceeds the preset maximum, return an out of limit result for the test;

otherwise

f) comparing the total energy for each channel signal to a preset minimum and if the total energy in any channel signal is below the preset minimum, terminate the test and return a pass result;

otherwise

g) dividing the energy image for each channel signal into equal-length overlapping segments in time;

h) for each segment of each channel signal, computing the ratio of the resonant frequency energy to the total signal energy;

i) when all segments have been processed, retaining the minimum ratio value as the result of the test; and

j) comparing the result of the test to a preset threshold, and if that result is greater than the threshold value, displaying an indication to search the vehicle, otherwise displaying an indication to pass the vehicle.