CN102708353B - Flame determination method, flame determination system, and flame determination device - Google Patents
Flame determination method, flame determination system, and flame determination device Download PDFInfo
- Publication number
- CN102708353B CN102708353B CN201110432206.8A CN201110432206A CN102708353B CN 102708353 B CN102708353 B CN 102708353B CN 201110432206 A CN201110432206 A CN 201110432206A CN 102708353 B CN102708353 B CN 102708353B
- Authority
- CN
- China
- Prior art keywords
- flame
- flames
- image
- doubtful
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000004458 analytical method Methods 0.000 claims abstract description 89
- 210000001364 upper extremity Anatomy 0.000 claims description 37
- 238000006073 displacement reaction Methods 0.000 claims description 8
- 238000010205 computational analysis Methods 0.000 claims description 2
- 230000001186 cumulative effect Effects 0.000 claims 2
- 238000010586 diagram Methods 0.000 description 26
- 101100243945 Fusarium vanettenii PDAT9 gene Proteins 0.000 description 4
- 101001072191 Homo sapiens Protein disulfide-isomerase A2 Proteins 0.000 description 4
- 208000012204 PDA1 Diseases 0.000 description 4
- 102100036351 Protein disulfide-isomerase A2 Human genes 0.000 description 4
- 208000030825 patent ductus arteriosus 2 Diseases 0.000 description 4
- 101150102492 pda1 gene Proteins 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000002485 combustion reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Landscapes
- Fire-Detection Mechanisms (AREA)
- Image Analysis (AREA)
- Photometry And Measurement Of Optical Pulse Characteristics (AREA)
Abstract
The invention discloses a flame judgment method, a flame judgment system and a flame judgment device applying the method, wherein the flame judgment method comprises the following steps: continuously capturing a plurality of frames to be analyzed at a plurality of time points; judging whether a suspected flame image exists in the picture to be analyzed; taking pixels in each suspected flame image as analysis points to obtain the outline size of the suspected flame image; calculating the ratio of the outline sizes of the suspected flame images captured at every two adjacent time points; calculating the frequency of change of these ratios; and judging whether the change frequency accords with the range of the flame frequency value, and if the change frequency accords with the range of the flame frequency value, judging that each suspected flame image is an image of real flame. According to the invention, whether the flame image is the image of the real flame can be accurately judged by calculating the dynamic change of the flame image.
Description
Technical field
The flame that the present invention relates to a kind of flame determination methods and application the method judges system and flame judgment means, and particularly a kind of flame determination methods that accurately can judge flame and the flame applying the method judge system and flame judgment means.
Background technology
Traditional flame judges that system utilizes video camera to capture picture, and judges whether picture has image of flames.When picture has image of flames, flame judges that system judges that whether image of flames is the image of real flame further.
Traditionally, judge that the mode of flame is: judge that whether the color of image of flames is close haply with the color of real flame, if so, then judge that image of flames is the image of real flame.But the mode of traditional judgement flame is quite inaccurate, usually causes erroneous judgement.
Summary of the invention
(1) technical matters that will solve
The flame that technical matters to be solved by this invention is to provide a kind of flame determination methods and application the method judges system and flame judgment means, by the dynamic change of calculating image of flames, judges that whether image of flames is the image of real flame exactly.
(2) technical scheme
According to a first aspect of the invention, propose a kind of flame determination methods, this flame determination methods comprises the following steps: at multiple time point, captures multiple picture to be analyzed continuously; Judge, in these pictures to be analyzed, whether there is doubtful image of flames; Using the pixel in this doubtful image of flames as analysis site, obtain the overall size of this doubtful image of flames; Calculate the ratio of the overall size of the doubtful image of flames that every two adjacent time points capture; Calculate the change frequency of these ratios; Judge whether this change frequency meets flame frequency values scope; And, if change frequency meets flame frequency values scope, then judge that this doubtful image of flames is the image of real flame.
According to a second aspect of the invention, a kind of flame judgment means is proposed.This flame judgment means is for judging whether image capture unit has the image of real flame at the picture multiple to be analyzed that multiple time point captures continuously.Flame judgment means comprises an image of flames judging unit, a profile analysis unit, a computing unit and a flame judging unit.Whether this image of flames judging unit is for having doubtful image of flames in judgement respectively this picture to be analyzed.This profile analysis unit, using the pixel in doubtful image of flames as analysis site, obtains the overall size of this doubtful image of flames.This computing unit for calculating the ratio of the overall size of the doubtful image of flames captured at every two adjacent time points, and calculates the change frequency of these ratios.This flame judging unit, for judging whether this change frequency meets flame frequency values scope, if change frequency meets flame frequency values scope, then judges that this doubtful image of flames is the image of real flame.
According to a third aspect of the invention we, propose a kind of flame and judge system.This flame judges that system comprises an image capture unit and a flame judgment means.This image capture unit is used for capturing multiple doubtful image of flames continuously at multiple time point.This flame judgment means is electrically connected at this image capture unit.This flame judgment means is for judging whether this image capture unit has the image of real flame at the picture multiple to be analyzed that multiple time point captures continuously.This flame judgment means comprises an image of flames judging unit, a profile analysis unit, a computing unit and a flame judging unit.This image of flames judging unit is for judging whether have doubtful image of flames in these pictures to be analyzed.The pixel of this profile analysis unit in each this doubtful image of flames, as analysis site, obtains the overall size of this doubtful image of flames.This computing unit in order to calculate the ratio of the overall size of the doubtful image of flames captured in every two adjacent time points, and calculates the change frequency of these ratios.This flame judging unit, for judging whether change frequency meets flame frequency values scope, if this change frequency meets flame frequency values scope, then judges that doubtful image of flames is the image of real flame.
(3) beneficial effect
The flame of flame determination methods of the present invention and application the method judges system and flame judgment means, by the dynamic change of calculating image of flames, can judge that whether image of flames is the image of real flame exactly.
Accompanying drawing explanation
Fig. 1 is the functional block diagram judging system according to the flame of one embodiment of the invention.
Fig. 2 is the process flow diagram of the flame determination methods according to one embodiment of the invention.
Fig. 3 is the schematic diagram that in the picture to be analyzed of the present embodiment, doubtful image of flames is not overlapping with check point.
Fig. 4 is the picture Flame image to be analyzed of the present embodiment schematic diagram overlapping with analysis site.
Fig. 5 is the distribution schematic diagram of the check point of one embodiment of the invention.
Fig. 6 is the schematic diagram of analysis site and doubtful image of flames in other embodiment of the present invention.
Fig. 7 is the method flow diagram that the present embodiment obtains the length of the outline of doubtful image of flames.
Fig. 8 calculates the schematic diagram of the overall size of doubtful image of flames with a wherein analysis site in Fig. 4.
Fig. 9 is the change schematic diagram of the ratio of the overall size of the doubtful image of flames of the present embodiment.
Figure 10 is the location diagram of doubtful image of flames at two adjacent time points of Fig. 4.
Figure 11 is the schematic diagram of the height of the doubtful image of flames of one embodiment of the invention.
Figure 12 is the schematic diagram of the upper limb profile of the doubtful image of flames of another embodiment of the present invention.
Figure 13 is the schematic diagram of wherein one section of upper limb profile of Figure 12.
Figure 14 is the graph of a relation of doubtful flame degree and time point under fuzzy deduction system.
[main element symbol description]
100: flame judges system
102: image capture unit
104: flame judgment means
106: profile analysis unit
108: computing unit
110: flame judging unit
112: image of flames judging unit
114: right hand edge region
116: rectangle
A1: the first numerical value
A2: second value
A3: determine nonflame value
A4: determine flame value
D1: left direction
D2: up direction
H: overall height
L1: displacement
PD: check point
PDA1: the first pair of angle point
PDA2: the second pair of angle point
FR: picture to be analyzed
S1, S2: doubtful image of flames
S11: the first profile
S12: the second profile
S21, S22, S23: upper limb profile
PR1, PR2: analysis site
PB1, PB2, PB3, PB4: frontier point
PL1: the first point
PL2: the second point
PL3: the wide point of third round
PL4: the wide point of fourth round
PL5: upper limb point
RA: ratio
X1: the first axially
Y2: the second axially
X2: the three axially
Y1: the four-axial.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly understand, below in conjunction with specific embodiment, and with reference to accompanying drawing, the present invention is described in more detail.
Please refer to Fig. 1, it is the functional block diagram judging system according to the flame of one embodiment of the invention.Flame judges that system 100 comprises image capture unit 102 and flame judgment means 104, and image capture unit 102 is electrically connected at flame judgment means 104.Image capture unit 102 is such as portable video camera, and portable video camera is herein such as the video camera with all-directional rotation and zoom (Pan-Tilt-Zoom, PTZ) function.In other embodiments, image capture unit 102 also can be fixed video camera.Image capture unit 102, for capturing multiple picture FR to be analyzed continuously multiple time point T (not being shown in Fig. 1), is then sent to flame judgment means 104.Flame judgment means 104 is in order to judge whether these picture FR to be analyzed have the image of real flame.
Flame judgment means 104 comprises profile analysis unit 106, computing unit 108, flame judging unit 110 and image of flames judging unit 112.Image of flames judging unit 112 is in order to judge whether have doubtful image of flames S1 in the picture FR to be analyzed that image capture unit 102 captures.If picture FR to be analyzed has doubtful image of flames S1, then transmit doubtful image of flames S1 to profile analysis unit 106.Profile analysis unit 106, using the pixel (pixel) in each doubtful image of flames S1 as analysis site, obtains the overall size of corresponding doubtful image of flames S1.Computing unit 108 for calculating the ratio of the overall size acquired by every two adjacent time point T, and calculates the change frequency of these ratios.Flame judging unit 110, for judging whether change frequency meets flame frequency values scope, if this change frequency meets flame frequency values scope, then judges that this doubtful image of flames S1 is the image of real flame.Say further, flame judgment means 104, by calculating the dynamic change of doubtful image of flames S1, judges that whether doubtful image of flames S1 is the image of real flame exactly.
The flame determination methods of the present embodiment is described with Fig. 2 below.Fig. 2 is the process flow diagram of the flame determination methods according to one embodiment of the invention.
In step s 102, image capture unit 102 captures several picture FR to be analyzed.
Then, in step S104, image of flames judging unit 112 judges whether have doubtful image of flames S1 in picture FR to be analyzed.If so, then step S106 is entered; If not, then step S102 is got back to.
In step S104, image of flames judging unit 112, according to the characteristic of real flame, tentatively judges the imagery zone of doubtful real flame in picture FR to be analyzed, and is defined as doubtful image of flames S1, then performs it and further analyzes.The characteristic of above-mentioned real flame is such as at least one in the color of real flame and brightness.
Then, in step s 106, using the pixel in each doubtful image of flames S1 as analysis site, the overall size of each doubtful image of flames S1 is obtained.Wherein, the mode setting up analysis site PR1 (showing in the Fig. 4 below) has numerous embodiments, below illustrates wherein several.
Please refer to Fig. 3 and Fig. 4, Fig. 3 is the schematic diagram that in the picture to be analyzed of the present embodiment, doubtful image of flames is not overlapping with check point, and Fig. 4 is the picture Flame image to be analyzed of the present embodiment schematic diagram overlapping with analysis site.In the present embodiment, before step S104, image of flames judging unit 112 can set up multiple check point PD at the fringe region of picture FR to be analyzed, and wherein check point PD is the pixel of picture FR to be analyzed.When image capture unit 102 rotates toward the right, these check points PD is distributed in the right hand edge region 114 of picture FR to be analyzed; When image capture unit 102 rotates toward the left side, these check points PD can be distributed in the left hand edge region (sign) of picture FR to be analyzed.When doubtful image of flames S1 is not yet overlapping with check point PD, check point PD can maintain static, as shown in Figure 3.When image capture unit 102 edge direction of turning right moves, make doubtful image of flames S1 toward left direction D1 move and overlapping with at least one of these check points PD time, profile analysis unit 106 using the check point PD overlapping with image of flames S1 as analysis site PR1, and start the overall size of carrying out doubtful image of flames S1 obtain step.In addition, in other embodiments, check point PD also can be distributed in right hand edge region and the left hand edge region of picture FR to be analyzed simultaneously.
In addition, preferably but non-exclusively, the spacing of these check points PD is not more than the internal flame region of doubtful image of flames S1.Such as, when the internal flame region of doubtful image of flames S1 is N × N pixel, the spacing of two adjacent check point PD is not more than N number of pixel.Say further, the internal flame region of real flame is more stable scope of burning, and its range is less.When the spacing of check point PD is equal to or less than the internal flame region of doubtful image of flames S1 haply, no matter how real flame changes, and can guarantee that doubtful image of flames S1 is overlapping with at least one check point PD.
In another embodiment, check point PD also can be covered with whole picture FR to be analyzed.Please refer to Fig. 5, it is the distribution schematic diagram of the check point of one embodiment of the invention.Image of flames judging unit 112 sets up multiple check point PD, and is covered with whole picture FR to be analyzed with array (array) arrangement mode.In the case, as long as doubtful image of flames S1 mono-appears in picture FR to be analyzed, doubtful image of flames S1 is just overlapping with check point PD soon, to expect once find doubtful image of flames S1, promptly starts to carry out profile analysis to doubtful image of flames S1.
Or in other embodiments, image of flames judging unit 112 can directly set up analysis site PR2 in doubtful image of flames S1.Say further, image of flames judging unit 112 can set up analysis site in step s 106.Below further illustrate.Please refer to Fig. 6, it is the schematic diagram of analysis site and doubtful image of flames in other embodiment.Profile analysis unit 106 obtains first couple of angle point PDA1 and second couple angle point PDA2.Wherein, the coordinate of first couple of angle point PDA1 and second couple of angle point PDA2 is respectively (X1max, Y1min) and (X2min, Y2max).First couple of angle point PDA1 and second couple of angle point PDA2 defines a rectangle 116, and four limits of rectangle 116 all contact the outline of doubtful image of flames S1.Say further, doubtful image of flames S1 has frontier point PB1, PB2, PB3 and PB4 of four outermost, and four limits of rectangle 116 touch frontier point PB1, PB2, PB3 and PB4 respectively.Profile analysis unit 106 is according to the coordinate figure of following formula (1) computational analysis point PR2.
R2x＝(X2min+X1max)/2)
R2y＝(Y1min+3×Y2max)/4…………………………(1)
In step s 106, profile analysis unit 106 obtains the overall size of doubtful image of flames S1 relative to the coordinate of each analysis site PR1, this overall size is such as at least one in the length of whole outline in doubtful image of flames S1, the length of upper limb profile and maximum height, then using this overall size as performance analysis object, doubtful image of flames S1 is judged whether for the image of real flame with this.Although the position of each analysis site PR1 in doubtful image of flames S1 is different, the length of the whole outline of doubtful image of flames S1, the length of upper limb profile and maximum height are close or identical haply numerical value relative to the coordinate of each analysis site PR1.
Among the outline length of doubtful image of flames S1, the length of upper limb profile and maximum height, illustrate to obtain doubtful image of flames S1 China and foreign countries profile length.Please refer to Fig. 7 and Fig. 8, Fig. 7 is the method flow diagram that the present embodiment obtains the length of the outline of doubtful image of flames, and Fig. 8 calculates the schematic diagram of the overall size of doubtful image of flames with a wherein analysis site in Fig. 4.
In step S202, profile analysis unit 106 obtains the first point PL1 of the nearest doubtful image of flames of compartment analysis point PR1 along the first axial X1.Wherein, the first point PL1 is the pixel of the first profile S11 of doubtful image of flames S1.
In step S204, profile analysis unit 106 obtains the second point PL2 of the nearest doubtful image of flames S1 of compartment analysis point PR1 along the second axial Y2.Wherein, the second point PL2 is the pixel of the second profile S12 of doubtful image of flames S1.
In step S206, profile analysis unit 106 for starting point, calculates first profile length of the first profile S11 with the first point PL1.
In step S208, profile analysis unit 106 for starting point, calculates second profile length of the second profile S12 with the second point PL2.
Profile analysis unit 106 is using the maximum in the first profile length and the second profile length as the outline length of doubtful image of flames S1.
As shown in Figure 8, the present embodiment is respectively with multiple point, and namely point PL1 and PL2 is as starting point, to obtain more than one outline length, increases analytical precision.Say further, due to flame dynamic change and form hollow profile (being such as the second profile S12 of Fig. 8), the outline of the non-doubtful image of flames S1 of this hollow profile sometimes in combustion, and its profile length is usually shorter.In the case, if only with single point, be such as the second point PL2 as starting point, then only obtain the outline length of single incorrect doubtful image of flames S1 and the judgement led to errors.
Although the present embodiment illustrates for two point PL1 and PL2 as starting point, in other embodiments, can using more point as starting point, to obtain more outline length.Such as, as shown in Figure 8, profile analysis unit 106 can obtain third round exterior feature point PL3, wherein the 3rd axial X2 and the first axial X1 conllinear of the nearest doubtful image of flames S1 of compartment analysis point PR1 again along the opposite direction (i.e. the 3rd axial X2) of the first axial X1.Further, profile analysis unit 106 can obtain fourth round exterior feature point PL4, wherein the second axial Y2 and the four-axial Y1 conllinear of the nearest doubtful image of flames S1 of compartment analysis point PR1 again along the opposite direction (i.e. four-axial Y1) of the second axial Y2.Profile analysis unit 106 for starting point, calculates the wide length of third round of third round exterior feature with third round exterior feature point PL3, and with fourth round exterior feature point PL4 for starting point, calculates the wide length of fourth round of fourth round exterior feature.Afterwards, profile analysis unit 106 is using the maximum in the first profile length, the second profile length, the wide length of third round and the wide length of fourth round as doubtful image of flames S1 China and foreign countries profile length.Due to above-mentioned first profile S11, third round are wide and fourth round wide be identical outline in doubtful image of flames S1, so the wide length of the first profile length, third round and the wide length of fourth round identical haply.
After step S106, enter step S108, referring to Fig. 9, it is the change schematic diagram of the ratio of the overall size of the doubtful image of flames of the present embodiment.Computing unit 108 calculates the ratio of the overall size of the doubtful image of flames S1 captured at every two adjacent time point T.In the present embodiment, the overall size L (k-1) of an above doubtful image of flames S1 is as denominator, and the overall size L (k) of next doubtful image of flames S1, as molecule, calculates ratio R A (k), as shown in the formula (2) using it.
RA(k)＝L(k-1)/L(k)………………………………(2)
Then, in step s 110, computing unit 108 calculates the change frequency of these ratio R A.For example, as shown in Figure 9, computing unit 108 to calculate in these ratio R A on a numerical value with this numerical value under the number of times that changes back and forth, wherein, this numerical value can between a scope, and to filter corresponding noise, this noise is such as too low ratio.Above-mentioned numerical range can between the first numerical value A1 and second value A2, and the first numerical value A1 is such as 0.9, and second value A2 is such as 1.1; Or when considering low ratio, the first numerical value A1 is such as 0.95, and second value A2 is such as 1.05.Ratio R A (4) lower than the first numerical value A1 higher than second value A2, ratio R A (5), is changed to ratio R A (5) by ratio R A (4) and is calculated to be change once; Similarly, ratio R A (7) higher than second value A2 lower than the first numerical value A1, ratio R A (8), is changed to ratio R A (8) from ratio R A (7) and is calculated to be change once.From RA (2) to the change procedure of RA (8), the number of times changed back and forth on second value A2 and between the first numerical value A1 in these ratio R A is secondary.Equally, from RA (8) to RA the account form of the change frequency of (N) similar in appearance to said method.Due to acquisition time point T known in step S102 and the number capturing picture FR to be analyzed, therefore change frequency per second can be obtained, i.e. above-mentioned change frequency (Hz) after converting.
Then, in step S112, flame judging unit 110 judges whether this change frequency meets flame frequency values scope.This flame frequency values scope is such as between 2 hertz (Hz) and 10Hz.If so, then step S114 is entered; If not, then step S102 is got back to.
Then, in step S114, flame judging unit 110 judges that doubtful image of flames S1 is the image of a real flame.Say further, real flame has a variation characteristic, and such as the change frequency of its flame contours length is haply between 2Hz and 10Hz.If the change frequency of Fig. 8 meets the variation characteristic of real flame, then judge that doubtful image of flames S1 is the image of real flame.
In one embodiment, in the process of step S104 to S114, image capture unit 102 can rotate unceasingly and continue to capture picture to be analyzed.That is, flame judges that system 100 is judging whether doubtful image of flames S1 is in the process of the image of real flame, still side by side can detect the doubtful image of flames whether having other new, if having, then judges with above-mentioned steps S106 to S114.
When image capture unit 102 rotates, doubtful image of flames S1 relative movement in picture FR to be analyzed, meanwhile, analysis site PR1 moves along with doubtful image of flames S1, remains in doubtful image of flames S1 to make analysis site PR1.Say further, please refer to Figure 10, it is the location diagram of doubtful image of flames at two adjacent time points of Fig. 4.Move for the direction, the right of PTZ camera toward Fig. 9, along with PTZ camera is turned right rotation, doubtful image of flames S1 is relatively toward left direction D1 mobile displacement L1.Simultaneously, profile analysis unit 106 calculates these doubtful image of flames S1 that every two adjacent time point T capture displacement L1 along left direction D1, and move the identical displacement L1 of analysis site PR1 toward identical left direction D1, analysis site PR1 is remained in doubtful image of flames S1.
In addition, because PTZ camera has the function of magnified picture, when image capture unit 102 is PTZ camera, can small doubtful image of flames in magnified picture, then flame judgment means 104 carries out above-mentioned flame determination methods, to obtain judged result more accurately.Or, when the change frequency of doubtful image of flames S1 does not meet flame frequency values scope, image capture unit 102 amplifies doubtful image of flames S1, and then flame judgment means 104 carries out above-mentioned flame determination methods, to reaffirm doubtful image of flames S1 whether for the image of real flame.
Although the parameter of above-mentioned judgement real flame illustrates with the example that is changed to of the outline length of doubtful image of flames S1, but the present invention is not limited thereto.In one embodiment, please refer to Figure 11, it is the schematic diagram of the height of the doubtful image of flames of one embodiment of the invention.Flame judgment means 104 with the overall height H of doubtful image of flames S1 for judging parameter.Say further, in step s 106, profile analysis unit 106, using analysis site PR1 as with reference to coordinate, calculates the overall height H of doubtful image of flames S1, and using overall height H as the overall size of step S106.
In another embodiment, please refer to Figure 12, it is the schematic diagram of the upper limb profile of the doubtful image of flames of another embodiment of the present invention.Flame judgment means 104 with the upper limb profile length of doubtful image of flames S2 for judging parameter.Say further, in step s 106, profile analysis unit 106 is using analysis site PR1 as reference coordinate, calculate the upper limb profile total length of multiple upper limb profile S21, S22 and S23 of doubtful image of flames S2, and using upper limb profile total length as the overall size of step S106, carry out the object of subsequent analysis with this.Upper limb profile herein refers to the outline of top without point.Say further, please refer to Figure 13, it is the schematic diagram of wherein one section of upper limb profile of Figure 12.Illustrate to observe separately upper limb profile S21, upper limb profile S21 has multiple upper limb point PL5, and in upper limb profile S21, the up direction D2 of arbitrary upper limb point PL5 there is no other upper limb point PL5.
From the above, in step s 106, the outline length of doubtful image of flames S1, overall height and upper limb profile total length can be obtained, and using at least one in this three as overall size, carry out the analytic target of subsequent step (step S108 to S114).
Although the step S112 of above-described embodiment judges doubtful image of flames S1 whether for the image of real flame is described for change frequency, but this is not for limiting the present invention.In other embodiments, the flame determination methods parameter that can also judge using the summation of the ratio of these overall sizes as flame.Say further, go back to Fig. 8, profile length is that example illustrates as overall size in addition, and after step S108, computing unit 108 adds up the absolute value of each ratio R A, to obtain a net amplitude.Then, flame judging unit 110 judges whether this net amplitude is greater than a flame amplitude.If this net amplitude is greater than this flame amplitude, then flame judging unit 110 judges that doubtful image of flames S2 is the image of real flame.Above-mentioned flame amplitude is one of real flame characteristic in combustion.
In other embodiments, fuzzy deduction system of can arranging in pairs or groups (Fuzzy Inference System) increases the accuracy of judged result.Please refer to Figure 14, it is the graph of a relation of flame degree and time point under fuzzy deduction system.Can using the input as fuzzy deduction system of above-mentioned net amplitude and above-mentioned change frequency.Then, as shown in figure 14, the doubtful flame degree LEV of fuzzy deduction system-computed.Determine flame value A4 when doubtful flame degree LEV is greater than, flame judging unit 110 judges that doubtful image of flames S1 is the image of real flame; Determine nonflame value A3 when doubtful flame degree LEV is less than, flame judging unit 110 judges that doubtful image of flames S1 is the image of non-genuine flame; When doubtful flame degree LEV is between determining flame value A4 and determining that, between nonflame value A3, image capture unit 102 can amplify doubtful image of flames S1, and then carry out above-mentioned flame determination methods.
In addition, in one embodiment, flame judges that system 100 also comprises a warning horn (not shown).When flame judging unit 110 judges that doubtful image of flames S1 is the image of real flame, this warning horn sounds the alarm; Or, in another embodiment, flame judges that system 100 also comprises a wired or wireless communication module (not shown), and when flame judging unit 110 judges that doubtful image of flames S1 is the image of real flame, this communication module gives the alarm signal to the unit that puts out a fire.
The flame of the flame determination methods of the above embodiment of the present invention and application the method judges system and flame judgment means, by the dynamic change of calculating image of flames, judges that whether image of flames is the image of real flame exactly.
Above-described specific embodiment; object of the present invention, technical scheme and beneficial effect are further described; be understood that; the foregoing is only specific embodiments of the invention; be not limited to the present invention; within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (31)
1. a flame determination methods, comprising:
Capture multiple picture to be analyzed continuously at multiple time point, wherein each time point captures a picture to be analyzed;
Judge, in these pictures to be analyzed, whether there is doubtful image of flames;
Using the pixel in doubtful image of flames as analysis site, obtain the overall size of this doubtful image of flames;
Calculate the ratio of the overall size of the doubtful image of flames captured at every two adjacent time points;
Calculate the change frequency of these ratios;
Judge whether this change frequency meets a flame frequency values scope; And
If this change frequency meets this flame frequency values scope, then judge that this doubtful image of flames is the image of a real flame.
2. flame determination methods as claimed in claim 1, also comprises:
This picture to be analyzed sets up a check point; And
The step obtaining the overall size of this doubtful image of flames also comprises:
When this check point falls into the region of doubtful image of flames, using this check point as analysis site.
3. flame determination methods as claimed in claim 2, the step wherein setting up this check point also comprises:
Multiple check point is set up at an edge of this picture to be analyzed; And
The step obtaining the overall size of doubtful image of flames also comprises:
When at least one of described check point falls into the region of this doubtful image of flames, using at least one this check point as analysis site.
4. flame determination methods as claimed in claim 2, the step wherein setting up this analysis site also comprises:
Set up multiple check point to be covered with this picture to be analyzed; And
The step obtaining this overall size of this doubtful image of flames also comprises:
When at least one of these check points falls into the region of this doubtful image of flames, using check point described at least one as analysis site.
5. flame determination methods as claimed in claim 1, wherein after the step capturing multiple picture to be analyzed continuously, this flame determination methods also comprises:
Analysis site is set up in this picture to be analyzed.
6. flame determination methods as claimed in claim 5, wherein set up in the step of analysis site in this doubtful image of flames and also comprise:
Calculate two to angle point, these two define a rectangle to angle point, and four limits of this rectangle all contact the outline of respectively this doubtful image of flames; And
According to these two to angle point, the coordinate figure of computational analysis point.
7. flame determination methods as claimed in claim 1, the step wherein obtaining the overall size of respectively this doubtful image of flames also comprises:
Along one first axial one first point obtained from this nearest doubtful image of flames of this analysis site, wherein this first point is the pixel of one first profile of this doubtful image of flames;
Along one second axial one second point obtained from this nearest doubtful image of flames of this analysis site, wherein this second point is the pixel of one second profile of this doubtful image of flames;
With this first point for starting point, calculate one first profile length of this first profile;
With this second point for starting point, calculate one second profile length of this second profile;
Wherein, using the maximum in this first profile length and this second profile length as this overall size.
8. flame determination methods as claimed in claim 1, the step wherein obtaining the overall size of respectively this doubtful image of flames also comprises:
Calculate the upper limb profile total length of multiple upper limb profiles of this doubtful image of flames, this upper limb profile total length is as this overall size;
Wherein, this upper limb profile has multiple upper limb point, and the top of arbitrary upper limb point of this upper limb profile is without other upper limb point.
9. flame determination methods as claimed in claim 1, wherein this step obtaining this overall size of respectively this doubtful image of flames also comprises:
Calculate the height of this doubtful image of flames;
Wherein, this height is as this overall size.
10. flame determination methods as claimed in claim 1, wherein calculates this change frequency step of these ratios and comprises:
Calculate these ratios on a numerical value with this numerical value under the number of times that changes back and forth.
11. flame determination methods as claimed in claim 1, also comprise:
These ratios cumulative, to obtain a net amplitude.
12. flame determination methods as claimed in claim 11, also comprise:
Judge whether this net amplitude is greater than a flame amplitude;
Judge that this doubtful image of flames is that the step of the image of real flame also comprises:
If this net amplitude is greater than this flame amplitude, then judge that this doubtful image of flames is the image of real flame.
13. flame determination methods as claimed in claim 11, also comprise:
According to fuzzy theory, using this net amplitude and this change frequency as input, calculate a doubtful flame degree; And
When this doubtful flame degree be greater than determine flame value time, judge that this doubtful image of flames is the image of this real flame.
14. flame determination methods as claimed in claim 1, also comprise:
Calculate at the displacement of every two adjacent these doubtful image of flames of time point along a moving direction; And
This displacement of this analysis site is moved toward this moving direction.
15. flame determination methods as claimed in claim 1, also comprise:
If this change frequency does not meet this flame frequency values scope, amplify this doubtful image of flames.
16. 1 kinds of flame judgment means, for judging whether an image capture unit has the image of real flame at the picture multiple to be analyzed that multiple time point captures continuously, wherein each time point captures a picture to be analyzed, and this flame judgment means comprises:
One image of flames judging unit, in order to judge whether have a doubtful image of flames in this picture to be analyzed;
One profile analysis unit, its pixel in each this doubtful image of flames, as analysis site, obtains an overall size of respectively this doubtful image of flames;
One computing unit, in order to calculate the ratio of the overall size of the doubtful image of flames captured at every two adjacent time points, and calculates the change frequency of these ratios; And
One flame judging unit, in order to judge whether this change frequency meets a flame frequency values scope; And
If this change frequency meets this flame frequency values scope, judge that this doubtful image of flames is the image of real flame.
17. flame judgment means as claimed in claim 16, wherein this image of flames judging unit also for setting up a check point on this picture to be analyzed;
Wherein, when this check point falls into the region of this doubtful image of flames, this profile analysis unit obtains the overall size of this doubtful image of flames.
18. flame judgment means as claimed in claim 17, wherein this image of flames judging unit sets up multiple check point at an edge of picture to be analyzed; This profile analysis unit also for when at least one of these check points falls into the region of this doubtful image of flames, using at least one this check point as analysis site.
19. flame judgment means as claimed in claim 17, wherein this image of flames judging unit sets up multiple check point to be covered with this picture to be analyzed; This profile analysis unit also for when at least one of these check points falls into the region of this doubtful image of flames, using at least one this check point as analysis site.
20. flame judgment means as claimed in claim 17, wherein this image of flames judging unit sets up analysis site in this doubtful image of flames.
21. flame judgment means as claimed in claim 20, wherein this profile analysis unit is also in order to calculate two to angle point and the coordinate figure according to these two, angle point being calculated to this analysis site, wherein these two define a rectangle to angle point, and four limits of this rectangle all contact the outline of respectively this doubtful image of flames.
22. flame judgment means as claimed in claim 16, wherein this profile analysis unit is also for obtaining one first point from this nearest doubtful image of flames of this analysis site along one first axis, and wherein this first point is the pixel of one first profile of this doubtful image of flames;
This profile analysis unit is also for obtaining one second point from this nearest doubtful image of flames of this analysis site along one second axis, wherein this second point is the pixel of one second profile of this doubtful image of flames;
This profile analysis unit also in order to this first point for starting point, calculate one first profile length of this first profile; This profile analysis unit also in order to this second point for starting point, calculate one second profile length of this second profile;
Wherein, the maximum in this first profile length and this second profile length is as this overall size.
23. flame judgment means as claimed in claim 16, wherein this profile analysis unit is also for calculating a upper limb profile total length of multiple upper limb profiles of this doubtful image of flames, and this upper limb profile total length is as this overall size;
Wherein, respectively this upper limb profile has multiple upper limb point, and the top of arbitrary upper limb point of this upper limb profile is without other upper limb point.
24. flame judgment means as claimed in claim 16, wherein this profile analysis unit is also for calculating the height of this doubtful image of flames, and this height is as this overall size.
25. flame judgment means as claimed in claim 16, wherein this computing unit also for calculate in these ratios on a numerical value with this numerical value under the number of times that changes back and forth.
26. flame judgment means as claimed in claim 16, wherein this computing unit is also for these ratios cumulative, to obtain a net amplitude.
27. flame judgment means as claimed in claim 26, wherein this flame judging unit is also in order to judge whether this net amplitude is greater than a flame amplitude, if this net amplitude is greater than this flame amplitude, then judges that this doubtful image of flames is the image of this real flame.
28. flame judgment means as claimed in claim 26, wherein this flame judging unit is also for foundation fuzzy theory, using this net amplitude and this change frequency as input, calculate a doubtful flame degree, and when this doubtful flame degree be greater than determine flame value time, judge that this doubtful image of flames is the image of this real flame.
29. flame judgment means as claimed in claim 16, wherein this profile analysis unit is also for calculating every two adjacent time points, and this doubtful image of flames along a displacement of a moving direction, and moves this displacement of this analysis site toward this moving direction.
30. flame judgment means as claimed in claim 16, wherein this image capture unit is not also for meeting this flame frequency values scope when this change frequency, amplifies this doubtful image of flames.
31. 1 kinds of flames judge system, comprising:
One image capture unit, for capturing multiple picture to be analyzed continuously at multiple time point, wherein each time point captures a picture to be analyzed; And
One flame judgment means, is electrically connected at this image capture unit, and this flame judgment means comprises:
Whether one image of flames judging unit, for having a doubtful image of flames in judgement respectively this picture to be analyzed;
One profile analysis unit, using the pixel in this doubtful image of flames as analysis site, obtains the overall size of this doubtful image of flames;
One computing unit, for calculating the ratio of the overall size acquired by every two adjacent time points, and calculates the change frequency of these ratios; And
One flame judging unit, in order to judge whether this change frequency meets a flame frequency values scope, if this change frequency meets this flame frequency values scope, then judges that this doubtful image of flames is the image of a real flame.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW99146173 | 2010-12-27 | ||
| TW099146173A TWI540539B (en) | 2010-12-27 | 2010-12-27 | Determining method for fire, determining system for fire using the same and determining device for fire using the same |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102708353A CN102708353A (en) | 2012-10-03 |
| CN102708353B true CN102708353B (en) | 2015-01-07 |
Family
ID=46901099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110432206.8A Active CN102708353B (en) | 2010-12-27 | 2011-12-20 | Flame determination method, flame determination system, and flame determination device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102708353B (en) |
| TW (1) | TWI540539B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544854A (en) * | 2018-10-16 | 2019-03-29 | 平安科技（深圳）有限公司 | Fire detection method, device, electronic equipment and computer readable storage medium |
| US11651670B2 (en) | 2019-07-18 | 2023-05-16 | Carrier Corporation | Flame detection device and method |
| CN114998842B (en) * | 2022-08-03 | 2022-12-30 | 广东电网有限责任公司肇庆供电局 | Power machine room smoke detection method and system based on disturbance amplification |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6184792B1 (en) * | 2000-04-19 | 2001-02-06 | George Privalov | Early fire detection method and apparatus |
| EP0818766B1 (en) * | 1996-07-12 | 2002-03-27 | T2M Automation | Method for automatic detection of fires, particularly of forest fires |
| TW200839660A (en) * | 2006-12-12 | 2008-10-01 | Ind Tech Res Inst | Flame detecting method and device |
| CN101316371A (en) * | 2007-05-31 | 2008-12-03 | 财团法人工业技术研究院 | Flame detecting method and device |
| CN101339072A (en) * | 2008-08-12 | 2009-01-07 | 黎永前 | Flame status checking method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7680297B2 (en) * | 2004-05-18 | 2010-03-16 | Axonx Fike Corporation | Fire detection method and apparatus |
-
2010
- 2010-12-27 TW TW099146173A patent/TWI540539B/en active
-
2011
- 2011-12-20 CN CN201110432206.8A patent/CN102708353B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0818766B1 (en) * | 1996-07-12 | 2002-03-27 | T2M Automation | Method for automatic detection of fires, particularly of forest fires |
| US6184792B1 (en) * | 2000-04-19 | 2001-02-06 | George Privalov | Early fire detection method and apparatus |
| TW200839660A (en) * | 2006-12-12 | 2008-10-01 | Ind Tech Res Inst | Flame detecting method and device |
| CN101316371A (en) * | 2007-05-31 | 2008-12-03 | 财团法人工业技术研究院 | Flame detecting method and device |
| CN101339072A (en) * | 2008-08-12 | 2009-01-07 | 黎永前 | Flame status checking method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102708353A (en) | 2012-10-03 |
| TW201227618A (en) | 2012-07-01 |
| TWI540539B (en) | 2016-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7403526B2 (en) | Systems and methods for removable and attachable acoustic imaging sensors | |
| CN109752722B (en) | Multi-modal acoustic imaging tool | |
| Kuan et al. | Comparative study of intel R200, Kinect v2, and primesense RGB-D sensors performance outdoors | |
| JP6468439B2 (en) | Gas detection image processing apparatus, gas detection image processing method, and gas detection image processing program | |
| JP6635382B2 (en) | Image output device, image output method, and image output system | |
| CN102708353B (en) | Flame determination method, flame determination system, and flame determination device | |
| US20160320296A1 (en) | Gas detector | |
| CN109756720B (en) | Focus and/or parallax adjustment in acoustic imaging using distance information | |
| CN105222725B (en) | A kind of high-definition image dynamic collecting method based on spectrum analysis | |
| CN105574468B (en) | Video flame detection method, device and system | |
| CN102298816B (en) | Fire early warning method for marine engine room based on multi-source fusion | |
| KR102623567B1 (en) | Optical sensor device | |
| RU2504014C1 (en) | Method of controlling monitoring system and system for realising said method | |
| CN108335454B (en) | A kind of fire behavior detection method and device | |
| KR102630836B1 (en) | Real-time calculation of atmospheric precipitation rates from digital images of the environment in which atmospheric precipitation occurs | |
| Martínez-de Dios et al. | Laboratory fire spread analysis using visual and infrared images | |
| CN103020588A (en) | Flame detection method based on video image analysis | |
| US9691247B2 (en) | Fire detection apparatus, method for evaluating at least one fire detection feature and computer program | |
| CN205067360U (en) | Gaseous formation of image telemetry unit of sulfur dioxide | |
| Wang et al. | Optimum design of wide angle laser detecting system based on fish eye lens and sinusoidal amplitude grating | |
| KR102236939B1 (en) | Fire pre-detection device | |
| CN102679893A (en) | Method for determining thickness of focusing ring of star sensor | |
| CN203827464U (en) | An apparatus and a system based on passenger flow statistics for a video | |
| RU2315357C2 (en) | Object detection method | |
| CN201589660U (en) | Integral intelligent infrared thermal imager |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |