CN114998842B - Power machine room smoke detection method and system based on disturbance amplification - Google Patents
Power machine room smoke detection method and system based on disturbance amplification Download PDFInfo
- Publication number
- CN114998842B CN114998842B CN202210924276.3A CN202210924276A CN114998842B CN 114998842 B CN114998842 B CN 114998842B CN 202210924276 A CN202210924276 A CN 202210924276A CN 114998842 B CN114998842 B CN 114998842B
- Authority
- CN
- China
- Prior art keywords
- video frame
- frame
- formula
- image
- frequency
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of video image processing, and discloses a smoke detection method and a smoke detection system for an electric power machine room based on disturbance amplification.
Description
Technical Field
The invention relates to the technical field of video image processing, in particular to a method and a system for detecting smoke in an electric power machine room based on disturbance amplification.
Background
The smoke detector is an ion smoke detector, and a radioactive source americium 241 is arranged in an inner ionization chamber and an outer ionization chamber, and positive ions and negative ions generated by ionization move towards positive and negative electrodes under the action of an electric field. Under normal conditions, the current and voltage of the inner ionization chamber and the outer ionization chamber are stable. Once smoke escapes from the outer ionization chamber. The normal movement of the charged particles is disturbed, the current and voltage are changed, the balance between the inner and outer ionization chambers is broken, and a signal is sent out.
The existing smoke detection system has an industrial problem. Smoke and heat may not actually reach highly installed detectors due to a physical phenomenon known as thermal stratification. The smoke cools as it rises, reducing its buoyancy relative to the surrounding air. Thus, in some environments, smoke may cool and stop rising before reaching the detector. In this case, the smoke detector cannot grasp the change of smoke or heat to perform judgment and early warning, thereby reducing the accuracy of smoke detection.
Disclosure of Invention
The invention provides a method and a system for detecting smoke in an electric power machine room based on disturbance amplification, which solve the technical problem of low accuracy of smoke detection.
In view of this, a first aspect of the present invention provides a smoke detection method for an electric power room based on disturbance amplification, including the following steps:
s1, framing a smoke video to obtain a plurality of video frames, and constructing a video frame sequence with a time sequence;
s2, performing Laplacian pyramid spatial decomposition on each video frame in the video frame sequence to obtain frame images under different spatial frequencies;
s3, acquiring a smog-free video frame as a background frame, dividing the background frame and each video frame into a plurality of image blocks respectively, carrying out graying processing on each image block, and calculating total block variation of the image blocks corresponding to the background frame and each video frame respectively;
s4, fourier transform is carried out on image blocks corresponding to the background frame and each video frame respectively, and the frequency energy ratio of the background frame and the frequency energy ratio of the video frame are calculated;
and S5, determining the smoke area of each video frame according to the comparison result of the total block variation of the image blocks corresponding to the background frame and the total block variation of the image blocks corresponding to each video frame and the comparison result of the frequency energy ratio of the background frame and the frequency energy ratio of the video frame.
Preferably, step S2 is followed by:
s201, performing time domain filtering on the video frame.
Preferably, step S3 specifically includes:
s301, acquiring a smog-free video frame as a background frame, dividing the background frame and each video frame into 16x16 image blocks respectively, and assuming that the image of the background frame and each video frame before division is
Suppose that the background frame and each video frame are divided into images
，
，
Respectively represent the row and column numbers of the image block
Go to the first
The total block variation of the image blocks of the rows is:
In the formula 1, the reaction mixture is,
the total block variation is shown, and x and y respectively represent the abscissa and the ordinate of the pixel.
Preferably, step S4 specifically includes:
s401, performing two-dimensional Fourier transform on the image block corresponding to the video frame to obtain:
In the formula 2, the first step is,
representing a spectral function of the image block, u, v representing coordinate values of pixels of the image block;
s402, carrying out spectrum function on image block
Carrying out normalization processing to obtain:
S403, acquiring the pixel point with the lowest frequency of the image block based on the normalized frequency spectrum function, and calculating the absolute distance from the pixel point with the lowest frequency of the image block to other pixel points according to the following formula 4:
In the formula 4, the first step is,
which represents the absolute distance of the vehicle,
representing the coordinates of the pixel points of the image block with the lowest frequency;
s404, screening out low-frequency pixels by adopting a gain-free unit step function in the following formula 5:
In the case of the formula 5, the compound,
，
the unit step function of the offset on the R-axis,
if the frequency separation threshold is preset, the separated low-frequency pixel part is:
s405, performing decentralized and inverse Fourier transform processing on the separated low-frequency pixel part to obtain a low-frequency image:
S406, calculating images according to the following formulas 7 and 8
And image
The frequency energy of (a) is:
In the formula 7~8,
representing images
The energy of the frequency of (a) is,
representing images
The frequency energy of (a);
s407, calculating the frequency energy ratio of the video frame by the following equation 9:
In the formula (9), the first and second groups,
representing the frequency energy ratio of the video frame of the t-th frame;
s408, calculating the frequency energy ratio of the background frame according to the steps S401 to S407
。
Preferably, step S5 specifically includes:
s501, taking a logarithm of the total block variation by the following formula, to obtain a logarithm value of the total block variation:
In the formula (10), the compound represented by the formula (10),
a logarithmic value representing the total block variation;
s502, judging whether the total block variation of the image blocks corresponding to the video frame is smaller than that of the image blocks corresponding to the background frame, if so, judging whether the frequency energy occupation ratio of the video frame is larger than that of the background frame, and if so, determining that the corresponding image blocks are smoke areas of the video frame.
In a second aspect, the present invention provides a smoke detection system for an electric power room based on disturbance amplification, including:
the framing module is used for framing the smoke video to obtain a plurality of video frames and constructing a video frame sequence with a time sequence;
the spatial decomposition module is used for carrying out Laplacian pyramid spatial decomposition on each video frame in the video frame sequence to obtain frame images under different spatial frequencies;
the first calculation module is used for acquiring a smog-free video frame as a background frame, dividing the background frame and each video frame into a plurality of image blocks respectively, carrying out graying processing on each image block, and calculating total block variation of the image blocks corresponding to the background frame and each video frame respectively;
the second calculation module is used for performing Fourier transform on image blocks corresponding to the background frame and each video frame respectively, and calculating the frequency energy ratio of the background frame and the frequency energy ratio of the video frame;
and the smoke determining module is used for determining the smoke area of each video frame according to the comparison result of the total block variation of the image blocks corresponding to the background frame and the total block variation of the image blocks corresponding to each video frame and the comparison result of the frequency energy ratio of the background frame and the frequency energy ratio of the video frame.
Preferably, the system further comprises:
and the filtering module is used for carrying out time domain filtering on the video frame.
According to the technical scheme, the invention has the following advantages:
the method comprises the steps of performing framing processing on a smoke video to obtain a plurality of video frames, performing Laplacian pyramid spatial decomposition on each video frame to obtain frame images under different spatial frequencies to make the images more obvious, calculating total block variation of image blocks corresponding to a background frame and each video frame by using the video frames without smoke as the background frame, performing Fourier transform on the image blocks corresponding to the background frame and each video frame, calculating the frequency energy occupation ratio of the background frame and the frequency energy occupation ratio of the video frame, and determining a smoke area of each video frame according to a comparison result of the total block variation of the image blocks corresponding to the background frame and the total block variation of the image blocks corresponding to each video frame and a comparison result of the frequency energy occupation ratio of the background frame and the frequency energy occupation ratio of the video frame, so that the accuracy of smoke detection is improved.
Drawings
Fig. 1 is a flowchart of a smoke detection method for an electric power room based on disturbance amplification according to an embodiment of the present invention;
fig. 2 is a specific flow block diagram of a smoke detection method for an electric power room based on disturbance amplification according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a smoke detection system of an electric power room based on disturbance amplification according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
For convenience of understanding, please refer to fig. 1~2, the method for detecting smoke in a power machine room based on disturbance amplification provided by the present invention includes the following steps:
s1, performing framing processing on the smoke video to obtain a plurality of video frames, and constructing a video frame sequence with a time sequence.
It can be understood that, in the present embodiment, the input video is processed into a group of video frames, so that the change situation of the tiny elements of the image can be displayed more obviously, which is beneficial for the subsequent calculation of the total block variation and the low-frequency energy ratio, so as to better reveal the hidden information.
And S2, performing Laplacian pyramid spatial decomposition on each video frame in the video frame sequence to obtain frame images under different spatial frequencies.
It should be noted that, the input video frames are spatially decomposed, that is, a laplacian pyramid is established on each frame, and different levels of pyramids are established according to different spatial frequencies and signal-to-noise ratios, so as to improve the spatial high frequency applicability. The laplacian pyramid is obtained from a gaussian pyramid, and the original image is convolved with a convolution kernel and subjected to interlaced and alternate downsampling (scale reduction), and the complete gaussian pyramid can be obtained by repeating the process. And performing alternate interlacing on the topmost layer of the Gaussian pyramid and performing convolution with the same convolution kernel to obtain an up-sampling image, performing subtraction on the Gaussian pyramid in the same layer and the up-sampling image to obtain a Laplacian pyramid in the layer, and repeating the process to obtain all Laplacian pyramids. If the upsampled image is added with the corresponding layer of the Laplacian pyramid, the Gaussian pyramid at the same layer can be obtained, and the original image can be obtained by repeating the process.
And S3, acquiring a smog-free video frame as a background frame, dividing the background frame and each video frame into a plurality of image blocks respectively, carrying out graying processing on each image block, and calculating total block variation of the image blocks corresponding to the background frame and each video frame respectively.
And S4, performing Fourier transform on image blocks corresponding to the background frame and each video frame respectively, and calculating the frequency energy ratio of the background frame and the frequency energy ratio of the video frame.
And S5, determining the smoke area of each video frame according to the comparison result of the total block variation of the image blocks corresponding to the background frame and the total block variation of the image blocks corresponding to each video frame and the comparison result of the frequency energy ratio of the background frame and the frequency energy ratio of the video frame.
The embodiment provides a smoke detection method for an electric power machine room based on disturbance amplification, which comprises the steps of performing framing processing on a smoke video to obtain a plurality of video frames, performing Laplacian pyramid spatial decomposition on each video frame to obtain frame images under different spatial frequencies so as to make the images more obvious, using a non-smoke video frame as a background frame, calculating total block variation of image blocks corresponding to the background frame and each video frame respectively, performing Fourier transform on the image blocks corresponding to the background frame and each video frame respectively, calculating a frequency energy ratio of the background frame and a frequency energy ratio of the video frame, and determining a smoke area of each video frame according to a comparison result of the total block variation of the image blocks corresponding to the background frame and the total block variation of the image blocks corresponding to each video frame and a comparison result of the frequency energy ratio of the background frame and the frequency energy ratio of the video frame, so as to improve the accuracy of smoke detection.
In a specific embodiment, after step S2, the method further includes:
s201, performing time domain filtering on the video frame.
The time domain filtering amplifies the frequency band of a certain fixed pixel point and keeps other frequency bands unchanged. And (4) for the part of each level which is amplified and superposed to the part before filtering, and finally reconstructing the amplified pyramid. The filtering is classified into low-pass filtering (filtering out a high-frequency component and retaining a low-frequency component), high-pass filtering (highlighting a high-frequency component and relatively suppressing a low-frequency component), and band-pass filtering (retaining a wave of a specific frequency band while shielding other frequency bands).
In a specific embodiment, step S3 specifically includes:
s301, acquiring a smog-free video frame as a background frame, and dividing the background frame and each video frame into 16x16 image blocks respectively, wherein the background frame and each video frame are assumedThe image of the video frame before segmentation is
Suppose that the background frame and each video frame are divided into images
，
，
Respectively represent the row and column numbers of the image block
Go to the first
The total block variation of the image blocks of the rows is:
In the formula 1, the reaction mixture is,
the total block variation is shown, and x and y respectively represent the abscissa and the ordinate of the pixel.
The frame image is divided into 16 × 16 image blocks, which can be set as dynamically variable parameters, and when a finer smoke region needs to be determined, the image can be divided into more blocks, so as to improve the recognition accuracy.
In a specific embodiment, step S4 specifically includes:
s401, performing two-dimensional Fourier transform on the image block corresponding to the video frame to obtain:
In the formula 2, the first step is,
representing a spectral function of the image block, u, v representing coordinate values of pixels of the image block;
according to the theory of two-dimensional fourier transform, a two-dimensional spectrogram of an image is a superposition of all one-dimensional fourier transforms in both the horizontal and vertical directions of an input image.
S402, spectrum function of image block
Carrying out normalization processing to obtain:
S403, acquiring the pixel point with the lowest frequency of the image block based on the normalized frequency spectrum function, and calculating the absolute distance from the pixel point with the lowest frequency of the image block to other pixel points according to the following formula 4:
In the formula 4, the first step is,
which represents the absolute distance of the vehicle,
representing the coordinates of the pixel points of the image block with the lowest frequency;
among them, the brighter pixels of the digital spectrogram are relatively low-frequency pixels, because the energy of the image is generally concentrated in the low-frequency part. After the frequency spectrum is centered, the low frequency is in the middle, and the high frequency is around, that is, the brightest point in the middle of the image block is the lowest frequency, and belongs to the direct current component.
S404, screening out low-frequency pixels by adopting a gain-free unit step function in the following formula 5:
In the case of the formula 5, the compound,
，
the unit step function of the offset on the R-axis,
if the frequency separation threshold is preset, the low-frequency pixel part separated is:
wherein the frequency is dependent on
The frequency spectrum after the centering is higher towards the edge frequency, the four corners of the spectrogram and the extreme ends of the X and Y axes are high frequencies, and the low frequency part in the frequency can be screened out.
S405, performing decentralized and inverse Fourier transform processing on the separated low-frequency pixel part to obtain a low-frequency image:
S406, calculating images according to the following formulas 7 and 8
And image
The frequency energy of (a) is:
In the formula 7~8,
representing images
The energy of the frequency of (a) is,
representing images
The frequency energy of (a);
the image can be regarded as a set of one-dimensional signals on all scanning lines of the x axis, and the energy of the one-dimensional signals is also the superposition of the energy of all the one-dimensional signals.
S407, calculating the frequency energy ratio of the video frame by the following formula 9:
In the formula (9), the first and second groups,
representing the frequency energy ratio of the video frame of the t-th frame;
s408, calculating the frequency energy ratio of the background frame according to the steps S401 to S407
。
In addition, according to the energy level distribution of the spectrum, the energy contained in the dc component is the largest, and the energy contained in the dc component is smaller as the frequency is higher. For the smoke image, the higher the smoke density is, the more blurred the image is, the difference of the gray levels of the adjacent areas of the corresponding image is reduced, the high frequency part is reduced, and the low frequency part is increased. Therefore, the fuzzy degree of the image, namely whether smoke exists or not can be judged by calculating and comparing the low-frequency energy ratio of the image.
In a specific embodiment, step S5 specifically includes:
s501, taking logarithm of the total block variation by the following formula to obtain a logarithm value of the total block variation:
In the formula (10), the compound represented by the formula (10),
a logarithmic value representing the total block variation;
s502, judging whether the total block variation of the image blocks corresponding to the video frame is smaller than that of the image blocks corresponding to the background frame, if so, judging whether the frequency energy occupation ratio of the video frame is larger than that of the background frame, and if so, determining that the corresponding image blocks are smoke areas of the video frame.
It should be noted that, in the present embodiment, the total block variation is adopted to determine the smoke region in the video frame, and the principle is as follows:
clear images due to defocusing, motion, transmission, encoding, transcoding and the like
Becoming blurred and a new blurred image is obtained:
In the formula (11), the first and second groups,
is degeneration ofSpatial description of functions, as a function of blurring filters on images
Carrying out fuzzy processing; ∗ represents a spatial convolution;
for the noise term, the noise term is ignored for simplicity, and then equation 11 becomes:
This ensures image-to-image ratio
Blurring
. Blurring an image according to equation 1
The local block variance of (a) is:
The offset derivative of x in the formula 12 is obtained by using a variable limit integral derivative rule:
By
The constraint (see formulas 13 and 14) indicates that:
Combining formula 16 with formula 17, we obtain:
Similarly, the partial derivative of y in equation 12 can be obtained:
Therefore, combining equations 18 and 19, equation 15 is transformed into:
When the image is clearest
，
，
From the Cauchy inequality for equation 15,
Therefore, it is not only easy to use
The total block variation of the blurred image is smaller than that of the sharp image, and the small images after blocking are obviously true, so that the total block variation is adopted as a judgment condition for the blurred image such as smoke.
Meanwhile, according to the energy level distribution of the frequency spectrum, the energy contained in the direct current component is the largest, and the energy contained in the direct current component is smaller as the frequency is higher. For the smoke image, the higher the smoke density is, the more blurred the image is, the difference of the gray levels of the adjacent areas of the corresponding image is reduced, the high frequency part is reduced, and the low frequency part is increased. Therefore, the fuzzy degree of the image, namely whether smoke exists or not can be judged by calculating and comparing the low-frequency energy ratio of the image. And when the two conditions are simultaneously satisfied, the image block can be judged to be a smoke area.
The above is a detailed description of an embodiment of the smoke detection method for the electric power room based on disturbance amplification provided by the invention, and the following is a detailed description of an embodiment of the smoke detection system for the electric power room based on disturbance amplification provided by the invention.
For convenience of understanding, referring to fig. 3, the smoke detection system for an electric power room based on disturbance amplification provided by the present invention includes:
the framing module 100 is configured to perform framing processing on the smoke video to obtain a plurality of video frames, and construct a video frame sequence with a time sequence;
a spatial decomposition module 200, configured to perform laplacian pyramid spatial decomposition on each video frame in the video frame sequence to obtain frame images at different spatial frequencies;
the first calculation module 300 is configured to acquire a smog-free video frame as a background frame, divide the background frame and each video frame into a plurality of image blocks, perform graying processing on each image block, and calculate total block variation of the image blocks corresponding to the background frame and each video frame;
the second calculating module 400 is configured to perform fourier transform on image blocks corresponding to the background frame and each video frame, and calculate a frequency energy ratio of the background frame and a frequency energy ratio of the video frame;
the smoke determining module 500 is configured to determine a smoke region of each video frame according to a comparison result of the total block variation of the image block corresponding to the background frame and the total block variation of the image block corresponding to each video frame and a comparison result of the frequency energy ratio of the background frame and the frequency energy ratio of the video frame.
In one embodiment, the system further comprises:
and the filtering module is used for carrying out time domain filtering on the video frame.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is only a logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (5)
1. A smoke detection method for an electric power machine room based on disturbance amplification is characterized by comprising the following steps:
s1, performing framing processing on a smoke video to obtain a plurality of video frames, and constructing a video frame sequence with a time sequence;
s2, performing Laplacian pyramid spatial decomposition on each video frame in the video frame sequence to obtain frame images under different spatial frequencies;
s3, obtaining a smog-free video frame as a background frame, dividing the background frame and each video frame obtained after the decomposition of the Laplacian pyramid space in the step S2 into a plurality of image blocks, carrying out graying processing on each image block, and calculating total block variation of the image blocks corresponding to the background frame and each video frame respectively;
step S3 specifically includes:
s301, acquiring a smog-free video frame as a background frame, dividing the background frame and each video frame into 16x16 image blocks respectively, and assuming that the image of the background frame and each video frame before division is
Suppose that the background frame and each video frame are divided into images
，
，
Respectively represent the row and column numbers of the image block
Go to the first
The total block variation of the image blocks of the rows is:
In the formula 1, the reaction mixture is,
the total block variation is represented, and x and y respectively represent the abscissa and the ordinate of the pixel;
s4, fourier transform is carried out on image blocks corresponding to the background frame and each video frame respectively, and the frequency energy ratio of the background frame and the frequency energy ratio of the video frame are calculated;
s5, determining a smoke area of each video frame according to a comparison result of the total block variation of the image blocks corresponding to the background frame and the total block variation of the image blocks corresponding to each video frame and a comparison result of the frequency energy ratio of the background frame and the frequency energy ratio of the video frame;
step S5 specifically includes:
s501, taking logarithm of the total block variation by the following formula to obtain a logarithm value of the total block variation:
In the formula (10), the compound represented by the formula (10),
a logarithmic value representing the total block variation;
s502, judging whether the total block variation of the image blocks corresponding to the video frame is smaller than that of the image blocks corresponding to the background frame, if so, judging whether the frequency energy occupation ratio of the video frame is larger than that of the background frame, and if so, determining that the corresponding image blocks are smoke areas of the video frame.
2. The electric machine room smoke detection method based on disturbance amplification according to claim 1, wherein step S2 is followed by further comprising:
s201, performing time domain filtering on the video frame.
3. The electric machine room smoke detection method based on disturbance amplification as claimed in claim 1, wherein the step S4 specifically includes:
s401, performing two-dimensional Fourier transform on the image block corresponding to the video frame to obtain:
In the formula 2, the first step is,
representing a spectral function of the image block, u, v representing coordinate values of pixels of the image block;
s402, spectrum function of image block
Carrying out normalization processing to obtain:
S403, acquiring the pixel point with the lowest frequency of the image block based on the normalized frequency spectrum function, and calculating the absolute distance from the pixel point with the lowest frequency of the image block to other pixel points according to the following formula 4:
In the formula 4, the first step is,
which represents the absolute distance of the vehicle,
representing the coordinates of the pixel points of the image block with the lowest frequency;
s404, screening out low-frequency pixels by adopting a gain-free unit step function in the following formula 5:
In the formula 5, the first step is,
，
the unit step function of the offset on the R-axis,
if the frequency separation threshold is preset, the low-frequency pixel part separated is:
s405, performing decentralized and inverse Fourier transform processing on the separated low-frequency pixel part to obtain a low-frequency image:
S406, calculating images according to the following formulas 7 and 8
And image
The frequency energy of (a) is:
In the formula 7~8,
representing images
The energy of the frequency of (a) is,
representing images
The frequency energy of (a);
s407, calculating the frequency energy ratio of the video frame by the following equation 9:
In the formula (9), the first and second groups,
representing the frequency energy ratio of the video frame of the t-th frame;
s408, calculating the frequency energy ratio of the background frame according to the steps S401 to S407
。
4. The utility model provides an electric power computer lab smog detecting system based on disturbance is enlarged which characterized in that includes:
the framing module is used for framing the smoke video to obtain a plurality of video frames and constructing a video frame sequence with a time sequence;
the spatial decomposition module is used for carrying out Laplacian pyramid spatial decomposition on each video frame in the video frame sequence to obtain frame images under different spatial frequencies;
the first calculation module is used for acquiring a smog-free video frame as a background frame, dividing the background frame and each video frame obtained after spatial decomposition of a Laplacian pyramid in the spatial decomposition module into a plurality of image blocks respectively, carrying out graying processing on each image block, and calculating total block variation of the image blocks corresponding to the background frame and each video frame respectively; specifically, the method is used for acquiring a smog-free video frame as a background frame, dividing the background frame and each video frame into 16 × 16 image blocks respectively, and assuming that an image of the background frame and an image of each video frame before division are
Suppose that the background frame and each video frame are divided into images
，
，
Respectively represent the row and column numbers of the image block
Go to the first
The total block variation of the image blocks of the rows is:
In the formula 1, the reaction mixture is,
the total block variation is represented, and x and y respectively represent a pixel abscissa and an ordinate;
the second calculation module is used for performing Fourier transform on image blocks corresponding to the background frame and each video frame respectively, and calculating the frequency energy ratio of the background frame and the frequency energy ratio of the video frame;
the smoke determining module is used for determining a smoke area of each video frame according to a comparison result of the total block variation of the image blocks corresponding to the background frame and the total block variation of the image blocks corresponding to each video frame and a comparison result of the frequency energy ratio of the background frame and the frequency energy ratio of the video frame; the smoke determining module is specifically configured to log the total block variation by the following formula to obtain a log value of the total block variation:
In the formula (10), the compound represented by the formula (10),
a logarithmic value representing the total block variation;
the smoke determining module is further configured to determine whether a total block variation of an image block corresponding to the video frame is smaller than a total block variation of an image block corresponding to a background frame, if the determination is yes, determine whether a frequency energy duty ratio of the video frame is larger than a frequency energy duty ratio of the background frame, and if the determination is yes, determine that the corresponding image block is a smoke region of the video frame.
5. The electrical machine room smoke detection system based on disturbance amplification of claim 4, further comprising:
and the filtering module is used for carrying out time domain filtering on the video frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210924276.3A CN114998842B (en) | 2022-08-03 | 2022-08-03 | Power machine room smoke detection method and system based on disturbance amplification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210924276.3A CN114998842B (en) | 2022-08-03 | 2022-08-03 | Power machine room smoke detection method and system based on disturbance amplification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114998842A CN114998842A (en) | 2022-09-02 |
| CN114998842B true CN114998842B (en) | 2022-12-30 |
Family
ID=83021325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210924276.3A Active CN114998842B (en) | 2022-08-03 | 2022-08-03 | Power machine room smoke detection method and system based on disturbance amplification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114998842B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116563743B (en) * | 2022-12-09 | 2023-12-01 | 南京图格医疗科技有限公司 | Detection method based on deep learning and smoke removal system |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7280696B2 (en) * | 2002-05-20 | 2007-10-09 | Simmonds Precision Products, Inc. | Video detection/verification system |
| JP5286342B2 (en) * | 2010-09-29 | 2013-09-11 | 能美防災株式会社 | Smoke detection frequency component identification method and smoke detection device |
| TWI540539B (en) * | 2010-12-27 | 2016-07-01 | 財團法人工業技術研究院 | Determining method for fire, determining system for fire using the same and determining device for fire using the same |
| TWI420423B (en) * | 2011-01-27 | 2013-12-21 | Chang Jung Christian University | Machine vision flame identification system and method |
| CN105319183A (en) * | 2015-11-13 | 2016-02-10 | 哈尔滨工程大学 | Detector and detection method for real-time on-line detection of emission smoke intensity of diesel engine |
| CN107316022B (en) * | 2017-06-27 | 2020-12-01 | 歌尔光学科技有限公司 | Dynamic gesture recognition method and device |
| CN110135374A (en) * | 2019-05-21 | 2019-08-16 | 吉林大学 | It is identified using image block characteristics and returns the fire hazard smoke detecting method classified |
| US11720999B2 (en) * | 2019-10-11 | 2023-08-08 | Kayrros | Method, device and non-transitory computer-readable storage medium for increasing the resolution and dynamic range of a sequence of respective top view images of a same terrestrial location |
| CN112884720B (en) * | 2021-02-01 | 2023-08-11 | 广东电网有限责任公司广州供电局 | Distribution line pollution flashover insulator detection method and system |
-
2022
- 2022-08-03 CN CN202210924276.3A patent/CN114998842B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114998842A (en) | 2022-09-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1230805B1 (en) | Video signal noise level estimator | |
| US8064718B2 (en) | Filter for adaptive noise reduction and sharpness enhancement for electronically displayed pictures | |
| CN114998842B (en) | Power machine room smoke detection method and system based on disturbance amplification | |
| KR100677133B1 (en) | Method and apparatus for detecting and processing noisy edges in image detail enhancement | |
| WO2000048117A1 (en) | Method and apparatus for adaptive filter tap selection according to a class | |
| KR20140070216A (en) | Method and Apparatus for processing image | |
| CN114528887A (en) | Bridge monitoring method, system and device based on micro-vibration amplification technology | |
| US20100238354A1 (en) | Method and system for adaptive noise reduction filtering | |
| US20050110907A1 (en) | Apparatus and method of measuring noise in a video signal | |
| CN112435182B (en) | Image noise reduction method and device | |
| CN112714308B (en) | Method and device for detecting video rolling stripes | |
| CN106846262B (en) | Method and system for removing mosquito noise | |
| JP4661775B2 (en) | Image processing apparatus, image processing method, program for image processing method, and recording medium recording program for image processing method | |
| US7031551B2 (en) | Noise reduction apparatus, noise reduction method, program and medium | |
| Martinez et al. | Implicit motion compensated noise reduction of motion video scenes | |
| Novoselac et al. | Image noise reduction by vector median filter | |
| CN112150385B (en) | Infrared image filtering method and device | |
| JPH07236074A (en) | Picture noise elimination device | |
| KR102027886B1 (en) | Image Resizing apparatus for Large Displays and method thereof | |
| CN111131661B (en) | Image processing circuit and related image processing method | |
| Ghazal et al. | A fast directional sigma filter for noise reduction in digital TV signals | |
| KR20060080551A (en) | Device and method for analysing images | |
| KR102658842B1 (en) | Image processing apparatus | |
| Nakamura et al. | Noise-level estimation from single color image using correlations between textures in RGB channels | |
| Jain et al. | An improved image denoising algorithm using robust estimation for high density salt and pepper noise |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |