Distinguishing between fire and non-fire conditions using cameras

• US 7,245,315 B2
• Filed: 05/20/2002
• Issued: 07/17/2007
• Est. Priority Date: 05/20/2002
• Status: Expired due to Fees

• Current Assignee: Simmonds Precision Products, Inc. (Raytheon Technologies Corporation d/b/a RTX)
• Original Assignee: Simmonds Precision Products, Inc. (Raytheon Technologies Corporation d/b/a RTX)
• Inventors: Zakrzewski, Radoslaw Romuald, Sadok, Mokhtar
• Primary Examiner(s): Wong; Allen
• Application Number:
  US10/152,166
• Publication Number:
  US 20030214583A1
• Time in Patent Office: 1,884 Days
• Field of Search: 348/65, 348/61, 348/143, 348/154, 348/165, 340/588, 382/100, 345/473
• US Class Current: 348/165
• CPC Class Codes: G08B 17/125 by using a video camera to detect fire or smoke

Abstract

Detecting fire and non-fire conditions includes receiving a plurality of frames of video information, determining an energy indicator for each of a subset of the plurality of frames, detecting a fire condition in response to the energy indicator for each of the subset of the plurality of frames forming a pattern as a function of time corresponding to a fire condition, and detecting a non-fire condition in response to the energy indicator for each of the subset of the plurality of frames forming a pattern as a function of time corresponding to a non-fire condition. Detecting fire and non-fire conditions may also include comparing energy indicators for each of the subset of the plurality of frames to a reference frame. The reference frame may correspond to a video frame taken when no fire is present, a video frame immediately preceding each of the subset of the plurality of frames, or a video frame immediately preceding a frame that is immediately preceding each of the subset of the plurality of frames.

Claims

1. A method of detecting fire and non-fire conditions, comprising:

receiving a plurality of frames of video information;

determining an energy indicator for each of a subset of the plurality of frames to provide energy indicators as a function of time;

determining a pattern formed by the energy indicators as a function of time for a defined period of time, wherein said pattern is formed by an energy indicator for each of the frames in the subset at a different time in said defined period, wherein said determining includes calculating an energy difference for each of a plurality of pixels, i, j, in each of said frames of the subset, and wherein said energy difference associated with said each pixel, i, j, is determined using a difference between brightness of said each pixel, i, j, of said each frame and brightness of a corresponding pixel, i, j, of a reference frame;

detecting a fire condition in response to the pattern formed by the energy indicators corresponding to a fire condition; and

detecting a non-fire condition in response to the pattern formed by the energy indicators corresponding to a non-fire condition.

2. A method, according to claim 1, wherein determining a pattern includes comparing energy indicators for each of the subset of the plurality of frames to a reference frame.

3. A method, according to claim 2, wherein the reference frame corresponds to a video frame taken when no fire is present.

4. A method, according to claim 2, wherein the reference frame corresponds to a video frame immediately preceding each of the subset of the plurality of frames.

5. A method, according to claim 2, wherein the reference frame corresponds to a video frame immediately preceding a frame that is immediately preceding each of the subset of the plurality of frames.

6. A method, according to claim 1, wherein at least some of the subset of the plurality of frames are provided by a camera having a sensitivity of between 400 nm and 1000 nm.

7. A method, according to claim 6, wherein the camera generates 640×

480 pixels per frame.

8. A method, according to claim 7, wherein the camera is a CCD camera.

9. A method, according to claim 7, wherein the camera is a CMOS camera.

10. A method, according to claim 7, wherein the camera generates thirty frames per second.

11. A method, according to claim 10, wherein one in ten frames is selected for processing.

12. A method, according to claim 7, wherein the camera provides gray scale output.

13. A method, according to claim 1, wherein at least some of the subset of the plurality of frames are provided by a camera having a sensitivity of between 7 and 14 micrometers.

14. A method, according to claim 13, wherein the camera is an IR camera.

15. A method, according to claim 1, wherein determining a pattern formed by the energy indicators includes calculating energy provided by a subset of the pixels of each of the subset of frames.

16. A method, according to claim 1, further comprising:

prior to determining a pattern formed by the energy indicators, re-sizing each of the plurality of the subset of the plurality of frames.

17. A method of detecting fire and non-fire conditions, comprising:

receiving a plurality of frames of video information;

determining an edge result frame for each of a subset of the plurality of frames, said edge result frame identifying image edge locations in said each frame; and

detecting a fire condition by comparing each of the edge result frames to a reference edge frame identifying image edge locations in a reference image corresponding to a non-fire condition, wherein said detecting includes calculating an energy difference for each of a plurality of pixels, i, j, in each of said edge result frames, and wherein said energy difference associated with said each pixel, i, j, is determined using a difference between brightness of said each pixel, i, j, of said each edge result frame and brightness of a corresponding pixel, i, j, of a reference edge frame.

18. A method, according to claim 17, wherein the reference edge frame corresponds to a video frame taken in the presence of fog.

19. A method, according to claim 17, wherein at least some of the subset of the plurality of frames are provided by a camera having a sensitivity of between 400 nm and 1000 nm.

20. A method, according to claim 19, wherein the camera generates 640×

480 pixels per frame.

21. A method, according to claim 20, wherein the camera is a CCD camera.

22. A method, according to claim 20, wherein the camera is a CMOS camera.

23. A method, according to claim 17, wherein at least some of the subset of the plurality of frames are provided by a camera having a sensitivity of between 7 and 14 micrometers.

24. A method, according to claim 23, wherein the camera is an IR camera.

25. A method, according to claim 17, further comprising:

detecting a non-fire condition by comparing each of the edge result frames to the reference edge frame.

26. A computer-readable storage medium encoded with computer executable instructions stored thereon that detects fire and non-fire conditions, comprising:

executable code that receives a plurality of frames of video information;

executable code that determines an energy indicator for each of a subset of the plurality of frames to provide energy indicators as a function of time;

executable code that determines a pattern formed by the energy indicators as a function of time for a defined period of time, wherein said pattern is formed by an energy indicator for each of the frames in the subset at a different time in said defined period, said executable code that determines a pattern including executable code that calculates an energy difference for each of a plurality of pixels, i, j, in each of said frames of the subset, and wherein said energy difference associated with said each pixel, i, j, is determined using a difference between brightness of said each pixel, i, j, of said each frame and brightness of a corresponding pixel, i, j, of a reference frame;

and

executable code that detects a fire condition in response to the pattern formed by the energy indicators corresponding to a fire condition.

27. A computer readable storage medium, according to claim 26, wherein executable code that determines a pattern includes executable code that compares energy indicators for each of the subset of the plurality of frames to a reference frame.

28. A computer readable storage medium, according to claim 27, wherein the reference frame corresponds to a video frame taken when no fire is present.

29. A computer readable storage medium, according to claim 27, wherein the reference frame corresponds to a video frame immediately preceding each of the subset of the plurality of frames.

30. A computer readable storage medium, according to claim 27, wherein the reference frame corresponds to a video frame immediately preceding a frame that is immediately preceding each of the subset of the plurality of frames.

31. A computer readable storage medium, according to claim 26, further comprising:

executable code that calculates energy provided by a subset of the pixels of each of the subset of frames.

32. A computer readable storage medium, according to claim 26, further comprising:

executable code that re-sizes each of the plurality of the subset of the plurality of frames prior to determining a pattern formed by the energy indicators.

33. A computer readable medium, encoded with computer executable instructions stored thereon that detects fire and non-fire conditions, comprising:

executable code that receives a plurality of frames of video information;

executable code that determines an edge result frame for each of a subset of the plurality of frames, said edge result frame identifying image edge locations in said each frame; and

executable code that detects a fire condition by comparing each of the edge result frames to a reference edge frame identifying image edge locations in a reference image corresponding to a non-fire condition, wherein said executable code that detects a fire condition includes executable code that calculates an energy difference for each of a plurality of pixels, i, j, in each of said edge result frames, and wherein said energy difference associated with said each pixel, i, j, is determined using a difference between brightness of said each pixel, i, j, of said each edge result frame and brightness of a corresponding pixel, i, j, of a reference edge frame.

34. A computer readable storage medium, according to claim 33, wherein the reference edge frame corresponds to a video frame taken in the presence of fog.

35. A computer readable storage medium, according to claim 33, further comprising:

executable code that detects a non-fire condition by comparing each of the edge result frames to the reference edge frame.

36. An apparatus that detects fire and non-fire conditions, comprising:

a plurality of cameras that provide a plurality of frames of video information; and

a processor, coupled to the cameras, that determines an energy for each of a subset of the plurality of frames to provide energy indicators as a function of time, determines a pattern formed by the energy indicators as a function of time for a defined period of time, wherein said pattern is formed by an energy indicator for each of;

the frames in the subset at a different time in said defined period, calculates an energy difference for each of a plurality of pixels, i, j, in each of said frames of the subset, and wherein said energy difference associated with said each pixel, i, j, is determined using a difference between brightness of said each pixel, i, j, of said each frame and brightness of a corresponding pixel, i, j, of a reference frame, detects a fire condition in response to the pattern formed by the energy indicators corresponding to a fire condition, and detects a non-fire condition in response to the pattern formed by the energy indicators corresponding to a non-fire condition.

37. An apparatus, according to claim 36, wherein the processor compares an energy indicators for each of the subset of the plurality of frames to a reference frame.

38. An apparatus, according to claim 37, wherein the reference frame corresponds to a video frame taken when no fire is present.

39. An apparatus, according to claim 37, wherein the reference frame corresponds to a video frame immediately preceding each of the subset of the plurality of frames.

40. An apparatus, according to claim 37, wherein the reference frame corresponds to a video frame immediately preceding a frame that is immediately preceding each of the subset of the plurality of frames.

41. An apparatus, according to claim 36, wherein at least some of the plurality cameras have a sensitivity of between 400 nm and 1000 nm.

42. An apparatus, according to claim 41, wherein at least some of the plurality of cameras generate 640×

480 pixels per frame.

43. An apparatus, according to claim 42, wherein at least some of the plurality of cameras are CCD cameras.

44. An apparatus, according to claim 42, wherein at least some of the plurality of cameras are CMOS cameras.

45. An apparatus, according to claim 42, wherein at least some of the plurality of cameras provide gray scale output.

46. An apparatus, according to claim 36, wherein at least some of the plurality cameras have a sensitivity of between 7 and 14 micrometers.

47. An apparatus, according to claim 46, wherein at least some of the plurality of cameras are IR cameras.

48. An apparatus, according to claim 36, further comprising:

a filter that filters at least a subset of the frames using a filtering technique selected from the group consisting of;

image subtraction, image averaging, smoothing filters, low-pass filters, median filter, sharpening filters, high-pass filters, stochastic techniques, and histogram processing.

49. An apparatus, according to claim 36, wherein at least some of the plurality of cameras generate thirty frames per second and wherein one in ten frames is selected for processing.

50. An apparatus, according to claim 36, wherein the processor calculates energy provided by a subset of the pixels of each of the subset of frames.

51. An apparatus, according to claim 36, wherein the processor re-sizes each of the plurality of the subset of the plurality of frames prior to determining a pattern formed by the energy indicators.

52. An apparatus that detects fire and non-fire conditions, comprising:

a plurality of cameras that receive a plurality of frames of video information; and

a processor that determines an edge result frame for each of a subset of the plurality of frames, wherein said edge result frame identifies image edge locations in said each frame, calculates an energy difference for each of a plurality of pixels, i, j, in each of said edge result frames, and wherein said energy difference associated with said each pixel, i, j, is determined using a difference between brightness of said each pixel, i, j, of said each edge result frame and brightness of a corresponding pixel, i, j, of a reference edge frame, and detects a fire condition by comparing each of the edge result frames to said reference edge frame identifying image edge locations in a reference image corresponding to a non-fire condition.

53. An apparatus, according to claim 52, wherein the reference edge frame corresponds to a video frame taken in the presence of fog.

54. An apparatus, according to claim 52, wherein at least some of the plurality of cameras have a sensitivity of between 400 nm and 1000 nm.

55. An apparatus, according to claim 54, wherein at least some of the plurality of cameras generate 640×

480 pixels per frame.

56. An apparatus, according to claim 55, wherein at least some of the plurality of cameras are CCD cameras.

57. An apparatus, according to claim 55, wherein at least some of the plurality of cameras are CMOS cameras.

58. An apparatus, according to claim 52, wherein at least some of the plurality cameras have a sensitivity of between 7 and 14 micrometers.

59. An apparatus, according to claim 58, wherein at least some of the plurality of cameras are IR cameras.

60. An apparatus, according to claim 52, wherein the processor detects a non-fire condition by comparing each of the edge result frames to the reference edge frame.

61. A method of detecting fire and non-fire conditions comprising:

receiving a plurality of frames of video information;

determining at least one feature for each of a subset of the plurality of frames to provide the feature as a function of time;

determining a pattern formed by a selected feature as a function of time for a defined period of time, wherein said pattern is formed by said at least one feature for the frames in the subset at different times in said defined period, wherein said determining a pattern includes calculating an energy difference for each of a plurality of pixels, i, j, in each of said frames of the subset, and wherein said energy difference associated with said each pixel, i, j, is determined using a difference between brightness of said each pixel, i, j, of said each frame and brightness of a corresponding pixel, i, j, of a reference frame;

determining, by a conventional smoke detection control unit, a smoke detection signal, said conventional smoke detection control unit using a non-image based technique in connection with smoke detection;

detecting a fire condition in response to the pattern formed by the selected feature corresponding to a fire condition and said smoke detection signal; and

detecting a non-fire condition in response to the pattern formed by the selected feature corresponding to a non-fire condition and said smoke detection signal.

62. The method of claim 61, further comprising:

filtering out at least one non-fire source of heat.

63. The method of claim 62, wherein said non-fire source of heat includes one of:

cargo in a cargo bay, a mechanical cooler generating a hot spot, and an aircraft being in a warm area.

64. The method of claim 62, further comprising:

using multiple two-dimensional camera views from a plurality of cameras to synthesize a three-dimensional camera view.

65. The method of claim 64, wherein said using multiple two-dimensional camera views from a plurality of cameras to synthesize three-dimensional camera views is performed when a cargo compartment is filled such that one or more frames of video information does not provide sufficient information to determine one of a fire and non-fire state of the cargo compartment.

66. The method of claim 65, further comprising:

determining whether the cargo compartment includes at least one of;

smoke, fog, and dust.

67. The method of claim 65, further comprising:

determining whether the cargo compartment includes at least one of;

smoke and non-smoke aerosols.

68. The method of claim 61, wherein said smoke detection signal from said conventional smoke detection system performs a gating function so that a fire condition is determined only after said conventional smoke detection control unit also provides a positive fire indication signal.

69. The method of claim 61, wherein a fire condition is determined even though the conventional smoke detection control unit has not detected a fire.

70. The method of claim 61, further comprising:

processing a first portion of said plurality of frames by a first processing board; and

processing a second portion of said plurality of frames by a second processing board, wherein said first processing board is coupled to said second processing board.

71. The method of claim 70, wherein said first and second processing boards are configured such that each of said two processing board processes approximately half of the plurality of frames.

72. The method of claim 70, wherein each of said two processing boards performs at least one of:

actuating a light source, processing camera input signals in accordance with one or more different types of cameras, providing video output to be viewed by a user, and performing fire detection.

73. The method of claim 61, wherein all of said plurality of frames are processed by a first processing board and, in the event the first processing board fails, a second processing board having a hardware configuration identical to that of said first processing board;

processes all of said plurality of frames.

74. The method of claim 61, further comprising:

determining whether fire suppression has been performed; and

filtering out image distortion cause by said fire suppression if said fire suppression has been performed.

75. The method of claim 61, wherein said plurality of frames of video information are obtained using a plurality of cameras including at least one CCD camera and at least one IR camera and wherein at least one CCD camera is mounted proximate to a corresponding IR camera, each CCD camera having an LED unit mounted therewith.

76. The method of claim 61, wherein said plurality of frames of video information are obtained using a plurality of cameras mounted in upper corners of a cargo bay area being monitored.

77. The method of claim 76, wherein said plurality of cameras includes at least one CCD camera with an on-board digital signal processing hardware.

78. The method of claim 76, wherein said plurality of cameras includes at least one CCD camera with an automatic gain control to adjust an amount of light provided in a video view area.

79. The method of claim 61, further comprising:

associating a first set of features extracted with a first region of a first of said subset of frames and associating a second set of features extracted with a second region of the first frame.

80. The method of claim 79, further comprising:

extracting the first set of features; and

extracting the second set of features.

81. The method of claim 79, further comprising:

growing one of said first and said second regions by pixel aggregation and averaging.

82. The method of claim 61, further comprising:

identifying at least one feature in accordance with an image distribution map.

83. The method of claim 82, wherein said at least one feature includes at least one of:

pixel intensity, pixel grey level, a Fourier descriptor, a wavelet coefficient, a statistical moment, and a motion indicator.

84. The method of claim 82, further comprising:

using at least one of said features to identify one or more regions of interest in an image.

85. The method of claim 84, further comprising:

splitting a region into a plurality of regions.

86. The method of claim 84, further comprising:

merging a region with another region.

87. The method of claim 84, wherein a region of interest is associated with at least one of:

a fire region, a smoke region, a hotspot region.

88. The method of claim 87, wherein a region of interest is defined as a contiguous set of pixels.

89. The method of claim 87, wherein a region of interest is defined as a set of pixels having a same property.

90. The method of claim 61, further comprising:

compensating a first frame from said subset prior to said determining at least one feature for said first frame.

91. The method of claim 90, wherein said compensating includes resizing said first frame reducing a number of pixels of said frame processed to determine said at least one feature.

92. The method of claim 90, further comprising:

compensating for a camera condition.

93. The method of claim 92, wherein said camera condition is one of:

a special camera lens, a camera instability causing vibration, a bright spot in frames obtained with a camera, a dark spot in frames obtained with a camera, and a line in frames obtained with a camera.

94. The method of claim 92, wherein said compensating includes performing temperature compensation for video data obtained using an IR camera.

95. The method of claim 90, wherein compensating includes:

adjusting a frame for vibration.

96. The method of claim 95, wherein said compensating uses a Weiner filter.

97. The method of claim 90, wherein said compensating further comprises:

performing calibration in accordance with an age of a camera.

98. The method of claim 90, wherein said compensating uses at least one external input value including one of:

results from a smoke detection control unit, ambient temperature used in IR camera image compensation, an aircraft altitude signal, and a cargo bay door open signal.

99. The method of claim 90, further comprising:

processing a frame in the frequency domain using a homorphic filter to perform simultaneous brightness range compression and contrast enhancement.

100. The method of claim 90, further comprising:

applying a logarithmic transformation to a frame to split the illumination and reflection components producing a resulting image which is processed in the frequency domain where functions of brightness range compression and contrast enhancement are performed simultaneously.

101. The method of claim 90, further comprising:

using matrix multiplication on a frame to suppress a camera vibration effect wherein the elements of a matrix used in the matrix multiplication are determined and verified in accordance with at least one vibration pattern observed in an aircraft environment.

102. The method of claim 101, wherein said at least one vibration pattern includes at least one of frequency, magnitude and orientation.

103. The method of claim 90, further comprising:

enhancing a frame in a space domain using a contrast stretching technique that increases a dynamic range of said frame.

104. The method of claim 90, further comprising:

calibrating a dynamic range for at least one camera used to obtain one of said frames of said subset in accordance with a type of said at least one camera; and

compensating said one frame causing image grayscale distribution to be within a range capability of said at least one camera.

105. The method of claim 90, further comprising:

detecting a hotspot in a first frame;

enhancing said first frame using a gray level slicing technique to highlight a specific range of gray levels associated with a hotspot-related feature.

106. The method of claim 90, further comprising:

expanding a dynamic range associated with at least one of said frames in said subset in accordance with a viewing range of a human eye.

107. A computer readable medium encoded with computer executable instructions that detects fire and non-fire conditions comprising:

executable code that receives a plurality of frames of video information;

executable code that determines at least one feature for each of a subset of the plurality of frames to provide the feature as a function of time;

executable code that determines a pattern formed by a selected feature as a function of time for a defined period of time, wherein said pattern is formed by said at least one feature for the frames in the subset at different times in said defined period, wherein said executable code that determines a pattern includes code that calculates an enemy difference for each of a plurality of pixels, i, j, in each of said frames of the subset, and wherein said energy difference associated with said each pixel, i, j, is determined using a difference between brightness of said each pixel, i, j, of said each frame and brightness of a corresponding pixel, i, j, of a reference frame;

executable code that determines, by a conventional smoke detection control unit, a smoke detection signal, said conventional smoke detection control unit using a non-image based technique in connection with smoke detection;

executable code that detects a fire condition in response to the pattern formed by the selected feature corresponding to a fire condition and said smoke detection signal; and

executable code that detects a non-fire condition in response to the pattern formed by the selected feature corresponding to a non-fire condition and said smoke detection signal.

108. The computer readable medium of claim 107, further comprising:

executable code that filters out at least one non-fire source of heat.

109. The computer readable medium of claim 108, wherein said non-fire source of heat includes one of:

cargo in a cargo bay, a mechanical cooler generating a hot spot, and an aircraft being in a warm area.

110. The computer readable medium of claim 108, further comprising:

executable code that uses multiple two-dimensional camera views from a plurality of cameras to synthesize a three-dimensional camera view.

111. The computer readable medium of claim 110, wherein said executable code that uses multiple two-dimensional camera views from a plurality of cameras to synthesize three-dimensional camera views is executed when a cargo compartment is filled such that one or more frames of video information does not provide sufficient information to determine one of a fire and non-fire state of the cargo compartment.

112. The computer readable medium of claim 111, further comprising:

executable code that determines whether the cargo compartment includes at least one of;

smoke, fog, and dust.

113. The computer readable medium of claim 107, wherein said smoke detection signal from said conventional smoke detection system performs a gating function so that a fire condition is determined only after said conventional smoke detection control unit also provides a positive fire indication signal.

114. The computer readable medium of claim 107, wherein a fire condition is determined even though the conventional smoke detection control unit has not detected a fire.

115. The computer readable medium of claim 107, further comprising:

executable code that processes a first portion of said plurality of frames by a first processing board; and

executable code that processes a second portion of said plurality of frames by a second processing board, wherein said first processing board is coupled to said second processing board.

116. The computer readable medium of claim 115, wherein said first and second processing boards are configured such that each of said two processing board processes approximately half of the plurality of frames.

117. The computer readable medium of claim 115, further comprising executable code that causes each of said two processing boards to perform at least one of:

actuating a light source, processing camera input signals in accordance with one or more different types of cameras, providing video output to be viewed by a user, and performing fire detection.

118. The computer readable medium of claim 107, wherein all of said plurality of frames are processed by a first processing board and, in the event the first processing board fails, a second processing board having a hardware configuration identical to that of said first processing board, processes all of said plurality of frames.

119. The computer readable medium of claim 107, further comprising:

executable code that determines whether fire suppression has been performed; and

executable code that filters out image distortion cause by said fire suppression if said fire suppression has been performed.

120. The computer readable medium of claim 107, wherein said plurality of frames of video information are obtained using a plurality of cameras including at least one CCD camera and at least one IR camera and wherein at least one CCD camera is mounted proximate to a corresponding IR camera, each CCD camera having an LED unit. mounted therewith.

121. The computer readable medium of claim 107, wherein said plurality of frames of video information are obtained using a plurality of cameras mounted in upper corners of a cargo bay area being monitored.

122. The computer readable medium of claim 121, wherein said plurality of cameras includes at least one CCD camera with an on-board digital signal processing hardware.

123. The computer readable medium of claim 121, wherein said plurality of cameras includes at least one CCD camera with an automatic gain control to adjust an amount of light provided in a video view area.

124. The computer readable medium of claim 107, further comprising:

executable code that associates a first set of features extracted with a first region of a first of said subset of frames and associates a second set of features extracted with a second region of the first frame.

125. The computer readable medium of claim 124, further comprising:

executable code that extracts the first set of features; and

executable code that extracts the second set of features.

126. The computer readable medium of claim 124, further comprising:

executable code that grows one of said first and said second regions by pixel aggregation and averaging.

127. The computer readable medium of claim 107, further comprising:

executable code that identifies at least one feature in accordance with an image distribution map.

128. The computer readable medium of claim 127, wherein said at least one feature includes at least one of:

pixel intensity, pixel grey level, a Fourier descriptor, a wavelet coefficient, a statistical moment, and a motion indicator.

129. The computer readable medium of claim 127, further comprising:

executable code that uses at least one of said features to identify one or more regions of interest in an image.

130. The computer readable medium of claim 129, further comprising:

executable code that splits a region into a plurality of regions.

131. The computer readable medium of claim 129, further comprising:

executable code that merges a region with another region.

132. The computer readable medium of claim 129, wherein a region of interest is associated with at least one of:

a fire region, a smoke region, a hotspot region.

133. The computer readable medium of claim 132, wherein a region of interest is defined as a contiguous set of pixels.

134. The computer program product of claim 132, wherein a region of interest is defined as a set of pixels having a same property.

135. The computer readable medium of claim 107, further comprising:

executable code that compensates a first frame from said subset prior to said determining at least one feature for said first frame.

136. The computer readable medium of claim 135, wherein said executable code that compensates includes executable code that resizes said first frame reducing a number of pixels of said frame processed to determine said at least one feature.

137. The computer readable medium of claim 135, further comprising:

executable code that compensates for a camera condition.

138. The computer readable medium of claim 137, wherein said camera condition is one of:

a special camera lens, a camera instability causing vibration, a bright spot in frames obtained with a camera, a dark spot in frames obtained with a camera, and a line in frames obtained with a camera.

139. The computer readable medium of claim 137, wherein said executable code that compensates includes executable code that performs temperature compensation for video data obtained using an IR camera.

140. The computer readable medium of claim 135, wherein said executable code that compensates includes:

executable code that adjusts a frame for vibration.

141. The computer readable medium of claim 140, wherein said executable code that compensates uses a Weiner filter.

142. The computer readable medium of claim 135, wherein said executable code that compensates further comprises:

executable code that performs calibration in accordance with an age of a camera.

143. The computer readable medium of claim 135, wherein said executable code that compensates uses at least one external input value including one of:

results from a smoke detection control unit, ambient temperature used in IR camera image compensation, an aircraft altitude signal, and a cargo bay door open signal.

144. The computer readable medium of claim 135, further comprising:

executable code that processes a frame in the frequency domain using a homorphic filter to perform simultaneous brightness range compression and contrast enhancement.

145. The computer readable medium of claim 135, further comprising:

executable code that applies a logarithmic transformation to a frame to split the illumination and reflection components producing a resulting image which is processed in the frequency domain where functions of brightness range compression and contrast enhancement are performed simultaneously.

146. The computer readable medium of claim 135, further comprising:

executable code that uses matrix multiplication on a frame to suppress a camera vibration effect wherein the elements of a matrix used in the matrix multiplication are determined and verified in accordance with at least one vibration pattern observed in an aircraft environment.

147. The computer readable medium of claim 146, wherein said at least one vibration pattern includes at least one of frequency, magnitude and orientation.

148. The computer readable medium of claim 135, further comprising:

executable code that enhances a frame in a space domain using a contrast stretching technique that increases a dynamic range of said frame.

149. The computer readable medium of claim 135, further comprising:

executable code that calibrates a dynamic range for at least one camera used to obtain one of said frames of said subset in accordance with a type of said at least one camera; and

executable code that compensates said one frame causing image grayscale distribution to be within a range capability of said at least one camera.

150. The computer readable medium of claim 135, further comprising:

executable code that detects a hotspot in a first frame;

executable code that enhances said first frame using a gray level slicing technique to highlight a specific range of gray levels associated with a hotspot-related feature.

151. The computer readable medium of claim 135, further comprising:

executable code that expands a dynamic range associated with at least one of said frames in said subset in accordance with a viewing range of a human eye.

152. The computer program product of claim 107, further comprising:

executable code that determines whether the cargo compartment includes at least one of;

smoke and a non-smoke aerosols.

153. A method of detecting fire and non-fire conditions, comprising:

receiving a plurality of frames of video information;

determining an edge result frame for each of a subset of the plurality of frames, said edge result frame identifying image edge locations in said each frame; and

detecting a fire condition by comparing each of the edge result frames to a reference edge frame identifying image edge locations in a reference image corresponding to a non-fire condition, wherein said detecting includes calculating an energy difference for each of a plurality of pixels, i, j, in each of said edge result frames, wherein said energy difference associated with said each pixel, i,j, is represented as a mathematically squared difference between brightness at said each pixel i,j of said each edge result frame, and brightness at a corresponding pixel i,j of said reference edge frame.

154. A computer readable medium, encoded with computer executable instructions stored thereon that detects fire and non-fire conditions, comprising:

executable code that receives a plurality of frames of video information;

executable code that determines an edge result frame for each of a subset of the plurality of frames, said edge result frame identifying image edge locations in said each frame; and

executable code that detects a fire condition by comparing each of the edge result frames to a reference edge frame identifying image edge locations in a reference image corresponding to a non-fire condition, wherein said executable code that detects includes code that calculates an energy difference for each of a plurality of pixels, i, j, in each of said edge result frames, wherein said energy difference associated with said each pixel, i,j, is represented as a mathematically squared difference between brightness at said each pixel i,j of said each edge result frame, and brightness at a corresponding pixel i,j of said reference edge frame.

155. An apparatus that detects fire and non-fire conditions, comprising:

a plurality of cameras that receive a plurality of frames of video information; and

a processor that determines an edge result frame for each of a subset of the plurality of frames, said edge result frame identifying image edge locations in said each frame, and detects a fire condition by comparing each of the edge result frames to a reference edge frame identifying image edge locations in a reference image corresponding to a non-fire condition,

wherein the processor calculates an energy difference for each of a plurality of pixels, i, j, in each of said edge result frames, wherein said energy difference associated with said each pixel, i,j, is represented as a mathematically squared difference between brightness at said each pixel i,j of said each edge result frame, and brightness at a corresponding pixel i,j of said reference edge frame.

156. A method of detecting fire and non-fire conditions, comprising:

receiving a plurality of frames of video information;

determining an energy indicator based on pixel intensity for each of a subset of the plurality of frames to provide energy indicators as a function of time;

determining a pattern formed by the energy indicators as a function of time for a defined period of time, wherein said pattern is formed by energy indicators for the frames in the subset at different times in said defined period;

detecting a fire condition in response to the pattern formed by the energy indicators corresponding to a fire condition; and

detecting a non-fire condition in response to the pattern formed by the energy indicators corresponding to a non-fire condition, wherein at least one of said detecting steps includes calculating an energy difference for each of a plurality of pixels, i, j, in each of said frames in said subset, wherein said energy difference associated with said each pixel, i,j, is represented as a mathematically squared difference between brightness at said each pixel i,j of said each frame, and brightness at a corresponding pixel i,j of a reference frame corresponding to one of a fire condition or a non-fire condition in accordance with said at least one detecting step.

157. A computer-readable storage medium encoded with computer executable instructions stored thereon that detects fire and non-fire conditions, comprising:

executable code that receives a plurality of frames of video information;

executable code that determines an energy indicator based on pixel intensity for each of a subset of the plurality of frames to provide energy indicators as a function of time;

executable code that determines a pattern formed by the energy indicators as a function of time for a defined period of time, wherein said pattern is formed by energy indicators for the frames in the subset at different times in said defined period;

executable code that detects a fire condition in response to the pattern formed by the energy indicators corresponding to a fire condition; and

executable code that calculates an energy difference for each of a plurality of pixels, i, j, in each of said frames in said subset, wherein said energy difference associated with said each pixel, i,j, is represented as a mathematically squared difference between brightness at said each pixel i,j of said each frame, and brightness at a corresponding pixel i,j of a reference frame corresponding to a fire condition.

158. An apparatus that detects fire and non-fire conditions, comprising:

a plurality of cameras that provide a plurality of frames of video information; and

a processor, coupled to the cameras, that determines an energy indicator based on pixel intensity for each of a subset of the plurality of frames to provide energy indicators as a function of time, determines a pattern formed by the energy indicators as a function of time for a defined period of time, wherein said pattern is formed by energy indicators for the frames in the subset at different times in said defined period, detects a fire condition in response to the pattern formed by the energy indicators corresponding to a fire condition, and detects a non-fire condition in response to the pattern formed by the energy indicators corresponding to a nonfire condition,

wherein said processor calculates an energy difference for each of a plurality of pixels, i, j, in each of said frames in said subset, wherein said energy difference associated with said each pixel, i,j, is represented as a mathematically squared difference between brightness at said each pixel i,j of said each frame, and brightness at a corresponding pixel i,j of a reference frame corresponding to at least one of a fire condition or a non-fire condition in accordance with detecting performed by said processor.

159. A method of detecting fire and non-fire conditions comprising:

receiving a plurality of frames of video information;

determining at least one feature for each of a subset of the plurality of frames to provide the feature as a function of time;

determining a pattern formed by a selected feature as a function of time for a defined period of time, wherein said pattern is formed by said at least one feature for the frames in the subset at different times in said defined period;

determining, by a conventional smoke detection control unit, a smoke detection signal;

detecting a fire condition in response to the pattern formed by the selected feature corresponding to a fire condition and said smoke detection signal; and

detecting a non-fire condition in response to the pattern formed by the selected feature corresponding to a non-fire condition and said smoke detection signal,

wherein at least one of said detecting steps includes calculating an energy difference for each of a plurality of pixels, i, j, in each of said frames in said subset, wherein said energy difference associated with said each pixel, i,j, is represented as a mathematically squared difference between brightness at said each pixel i,j of said each frame, and brightness at a corresponding pixel i,j of a reference frame corresponding to one of a fire condition or a non-fire condition in accordance with said at least one detecting step.

160. A computer readable medium encoded with computer executable instructions that detects fire and non-fire conditions comprising:

executable code that receives a plurality of frames of video information;

executable code that determines at least one feature for each of a subset of the plurality of frames to provide the feature as a function of time;

executable code that determines a pattern formed by a selected feature as a function of time for a defined period of time, wherein said pattern is formed by said at least one feature for the frames in the subset at different times in said defined period;

executable code that determines, by a conventional smoke detection control unit, a smoke detection signal;

executable code that detects a fire condition in response to the pattern formed by the selected feature corresponding to a fire condition and said smoke detection signal; and

executable code that detects a non-fire condition in response to the pattern formed by the selected feature corresponding to a non-fire condition and said smoke detection signal,

wherein at least one of said executable codes that detects includes calculating an energy difference for each of a plurality of pixels, i, j, in each of said frames in said subset, wherein said energy difference associated with said each pixel, i,j, is represented as a mathematically squared difference between brightness at said each pixel i,j of said each frame, and brightness at a corresponding pixel i,j of a reference frame corresponding to one of a fire condition or a non-fire condition in accordance with said at least one executable code.