IDENTIFYING ACTIVITY IN AN AREA UTILIZING SOUND DETECTION AND COMPARISON

• US 20080025477A1
• Filed: 07/25/2007
• Published: 01/31/2008
• Est. Priority Date: 07/25/2006
• Status: Active Grant

• Current Assignee: Intelehealth, Inc.
• Original Assignee: Intelehealth, Inc.
• Inventors: Farhan, Fariborz M
• Granted Patent: US 7,825,813 B2
• Time in Patent Office: Days
• US Class Current: 379/38
• CPC Class Codes: G08B 21/0423 detecting deviation from an expected pattern of behaviour or schedule
• CPC Class Codes: G08B 21/0469 Presence detectors to detect unsafe condition, e.g. infrared sensor, microphone G08B21/0476 takes precedence
• CPC Class Codes: G16H 40/60 for the operation of medical equipment or devices

Abstract

Microprocessor technology is used to detect routine sounds in a substantially enclosed environment to determine normal or abnormal activity or noise within the environment (i.e., habitual behavior of an individual) A device is situated in a substantially enclosed environment with audio input device similar to a microphone and an optional visual display for interaction with the local users. The device has the ability to be controlled locally via key-pad or USB port connected to a local laptop computer, or remotely via a phone line or Ethernet connection to internet. The device further performs audio pattern recognition using waveform matching scheme to detect the occurrence of pre-programmed sounds representing routine activities. The microprocessor counts the number occurrence of each recognizable sound for a particular interval and over a length of a day or other duration and reports to a remote server. The remote server performs tracking and trending analysis to a web-based caregiver client. Significant changes are detected and reported to caregiver or family web-based client.

Claims

1. A device for monitoring and detecting sounds to identify activity of a person in a substantially enclosed environment, the device comprising the components of:

an audio receiver operable to detect sounds within the substantially enclosed environment and convert the sounds into electronic signals;

a speech processor for receiving the electronic signals and forming digital audio signals;

a digital audio waveform analyzer operable to store digital audio waveforms and compare two or more digital audio waveforms to perform pattern recognition; and

a processing unit operable to;

place the digital audio waveform analyzer into a learning mode for receiving and storing digital audio waveforms for various sounds,

place the digital audio waveform analyzer into a detecting mode for receiving digital audio waveforms and comparing them to previously recorded digital audio waveforms; and

performing behavioral trend analysis of the noises in the substantially closed environment over a course of time

2. The device of claim 1, wherein the processing unit is a software controllable processing unit.

3. The device of claim 1, wherein said waveform analyzer is capable of digitizing audio signals.

4. The device of claim 1, wherein said waveform analyzer is operable to filter out speech periods and silence periods in the digital audio waveforms.

5. The device of claim 1, wherein said waveform analyzer is capable of comparing an audio pattern with a previously stored pattern after power normalizing the received waveform using pattern recognition techniques.

6. The device of claim 5, wherein the pattern recognition technique comprises root mean square calculations.

7. The device of claim 1, wherein said processor is capable of operating in a learning mode to store digitized audio signal for subsequent retrieval and comparison.

8. The device of claim 1, wherein said processor is operable to count the number of occurrences of matches to each uniquely stored audio pattern during a first period of time and retain the count for a second period of time.

9. The device of claim 8, wherein the first period is less than an hour and the second period is 24 hours.

10. The device of claim 8, wherein said processor is operable to perform trend analysis by comparing the match counts to monthly averages to determine, normal activity, below normal activity or zero activity and report back to a centralized server.

11. A method for detecting abnormal activity in an environment, the method comprising the steps of:

recording sounds expected to occur in the environment;

storing information representing each unique expected sound;

recording sounds occurring in the environment over a first period of time;

identify each sound by comparing the sound to the stored information of unique expected sounds;

maintain a count of the number of occurrences for each identified sound over the first period of time; and

comparing the counts to expected values for each count within that first period of time.

12. The method of claim 11, further comprising the step of, if the maintained counts deviate from the expected counts by a threshold amount, reporting such information to a centralized server.

13. The method of claim 11, further comprising the steps of:

for each count reporting if the count exceeds the expected count by a first threshold amount, is less than the expected count by a second threshold amount or is within a threshold distance from the expected count.