Intelligent image smoke/flame sensor and detection system

• US 7,991,187 B2
• Filed: 08/29/2007
• Issued: 08/02/2011
• Est. Priority Date: 08/29/2007
• Status: Active Grant

• Current Assignee: InnoSys Industries Limited
• Original Assignee: Billy Hou
• Inventors: Hou, Billy
• Primary Examiner(s): Tabatabai; Abolfazl
• Application Number:
  US11/892,992
• Publication Number:
  US 20090060260A1
• Time in Patent Office: 1,434 Days
• Field of Search: 382/100, 340/577, 340/578, 340/584, 340/693.12, 340/425.5, 340/426.18, 340/438, 340/463, 340/506, 340/628, 340/693.6, 340/901, 379/88.07, 704/273
• US Class Current: 382/100
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G08B 17/12 Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions
• CPC Class Codes: G08B 29/043 of fire detection circuits
• CPC Class Codes: G08B 29/24 Self-calibration, e.g. compensating for environmental drift or ageing of components

Abstract

An intelligent image smoke/flame/complex intelligent image smoke/flame sensor, and an intelligent image smoke/flame detection system formed of such an intelligent image smoke/flame/complex intelligent image smoke/flame sensor. The intelligent image smoke/flame sensor is formed of an image sensor, for example, CCD (Charge-coupled Device), a DSP Processor, an input/output circuit, an infrared lamp and filter assembly, a casing, a window glass, a front cover, and a rear cover. The intelligent image smoke/flame detection system accurately detects fire characteristics including smoke and flame, producing an early alarm to lower catastrophe.

Claims

1. An intelligent image smoke/flame sensor, comprised of an image sensor, a DSP processor, an input/output circuit, an infrared lamp and filter assembly, a casing, a window glass, a front cover, and a rear cover, said DSP processor providing an operation algorithm of statistics of t time serial image characteristics matching and probability fusion calculation, which includes the steps of:

(i) the system picks up a length L of serial images for at t time;

(ii) the system makes sure of fixed background characteristics by means of self-learning, in which the cycle for long cycle background to be T_L and the cycle for short cycle background to be T_S;

(iii) the system analyzes the serial images of length m obtained at t time, obtaining the differences and variation between each two images, the differences and variation between each image and the short cycle background image, and the differences and variation between each image and the long cycle background image;

(iv) the system calculates every characteristic parameter and matches the calculated characteristic parameters with flame/smoke characteristics, so as to set FCi and SCj to be 1 if the radio of successful matching images reaches a predetermined value or to be −

1 if the radio of successful matching images is below the predetermined value;

(v) the system starts a first stage data fusion subject to different characteristic weights, thereby obtaining flame probability incremental amount and smoke probability incremental amount to be Δ

P_F(t) and Δ

P_S(t) respectively; and

(vi) the system starts a second stage data fusion to obtain a fire probability P(t) at t time.

2. The intelligent image smoke/flame sensor as claimed in claim 1, wherein said image sensor is a charge-coupled device.

3. The intelligent image smoke/flame sensor as claimed in claim 1, wherein the surrounding light of the field of view is below a predetermined illumination, said DSP processor turns on the infrared lamp of said infrared lamp and filter assembly and switch said image sensor to a black-and-white operation mode.

4. The intelligent image smoke/flame sensor as claimed in claim 1, which comprises pre-alarm/fire alarm/failure relay outputs, a RS232/RS485 serial communication interface, a compound video output interface, and a TCP/IP Ethernet output interface.

6. The intelligent image smoke/flame sensor as claimed in claim 1 or 5, which divides different probability zones including warning probability P_a1<

P(t)≦

P_ah, pre-alarm probability P_p1<

P(t)≦

P_ph, and alarm probability P_F1<

P(t)≦

P_Fh to give different grades of alarm information subject to t time fire probability P(t).

7. An intelligent image smoke/flame detection system comprised of an intelligent image smoke/flame sensor, an information management and digital recording system, a learning and setting adjustment system, and a display and operation panel obtained subject to claim 1 or 5.

8. The intelligent image smoke/flame detection system as claimed in claim 7, wherein said learning and setting adjustment system is connected with said intelligent image smoke/flame sensor and said information management and digital recording system to form a loop self-learning system for producing updated system decision parameters when said intelligent smoke/flame sensor made a false alarm/alarm miss.

5. An intelligent image smoke/flame sensor, comprised of an image sensor (charge-coupled device), a DSP processor, an input/output circuit, an infrared lamp and filter assembly, a casing, a window glass, a front cover, a rear cover, and a signal processors of ultraviolet/infrared fire sensors, said DSP processor providing an operation algorithm with a fire probability fusion data obtained from different fire sensors for obtaining at time fire probability by means of adding to data fusion flow the fire characteristic probability incremental amount obtained through a probability incremental amount and image analysis.