CN112115766A - Flame identification method, device, equipment and storage medium based on video picture - Google Patents
Flame identification method, device, equipment and storage medium based on video picture Download PDFInfo
- Publication number
- CN112115766A CN112115766A CN202010740834.1A CN202010740834A CN112115766A CN 112115766 A CN112115766 A CN 112115766A CN 202010740834 A CN202010740834 A CN 202010740834A CN 112115766 A CN112115766 A CN 112115766A
- Authority
- CN
- China
- Prior art keywords
- flame
- picture
- pictures
- frame
- determining
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract 16
- 238000000605 extraction Methods 0.000 claims abstract 7
- 238000007781 pre-processing Methods 0.000 claims abstract 6
- 238000001914 filtration Methods 0.000 claims 1
- 238000003709 image segmentation Methods 0.000 claims 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- 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
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the application provides a flame identification method, a device, equipment and a storage medium based on a video picture, wherein the flame identification method based on the video picture comprises the following steps: the method comprises the steps of obtaining a target video picture, extracting a plurality of frame picture pictures from the target video picture, conducting image preprocessing on the frame picture pictures, conducting feature extraction on the frame picture pictures to obtain a flame feature value, determining the flame feature value as the input of a flame recognition model, and determining that flame exists in the target video picture when the flame recognition model recognizes that the flame feature value meets a preset condition. The method, the device and the equipment for recognizing the flame based on the video picture and the storage medium have the advantages of being low in false alarm rate.
Description
Technical Field
The application relates to the field of artificial intelligence, in particular to a flame identification method, device, equipment and storage medium based on a video image.
Background
At present, every year, fire causes huge losses of people and property. Especially, with the continuous expansion of the range of human activities, the fire prevention problem in large space and field is more and more prominent. If artificial fire protection is adopted for such spaces, a large amount of manpower and material resources are required.
Although the existing fire detection system can achieve unmanned monitoring on flame, the existing fire detection system mainly detects whether a fire exists through a smoke detection sensor and an infrared sensor, and the detection system has the defects of untimely detection and high false alarm rate.
Disclosure of Invention
The object of the present application includes, for example, providing a method, an apparatus, a device and a storage medium for flame identification based on video pictures, which can detect flames in time and perform a fire alarm in time, and at the same time, the method, the apparatus, the device and the storage medium for flame identification based on video pictures of the present application also have the advantage of low false alarm rate.
A first aspect of the present application discloses a flame recognition method based on a video picture, the flame recognition method comprising:
acquiring a target video picture;
extracting a plurality of frame picture pictures from the target video picture;
carrying out image preprocessing on the plurality of frame picture pictures;
extracting the characteristics of the plurality of frame picture pictures to obtain a flame characteristic value;
determining the flame characteristic value as an input of a flame identification model;
and when the flame recognition model recognizes that the flame characteristic value meets a preset condition, determining that flame exists in the target video picture.
In the first aspect of the application, a plurality of frames of picture pictures can be extracted from a target video picture by obtaining the target video picture, image preprocessing can be performed on the plurality of frames of picture pictures, feature extraction can be performed on the plurality of frames of picture pictures to obtain a flame feature value, and the flame feature value can be determined as an input of a flame recognition model, so that when the flame feature value is recognized by the flame recognition model to meet a preset condition, it can be determined that flames exist in the target video picture.
In the first aspect of the present application, it is determined that an optional implementation manner is that the performing image preprocessing on the plurality of frame picture pictures includes:
performing median filtering on the plurality of frames of picture pictures to remove noise of each frame of picture;
and sharpening the plurality of frames of picture pictures by adopting an exponential high-pass filter.
In this optional embodiment, median filtering is performed on the plurality of frame picture pictures to remove noise of each frame of the picture pictures, and at the same time, an exponential high-pass filter is adopted to sharpen the plurality of frame picture pictures.
In the first aspect of the present application, which is determined as an optional implementation manner, after the sharpening the plurality of frame picture pictures by using the exponential high-pass filter, the image preprocessing the plurality of frame picture pictures further includes:
and carrying out image segmentation on the plurality of frames of picture pictures by adopting a difference method to obtain a binary image.
In this optional embodiment, the image segmentation can be performed on the plurality of frame picture pictures by using a difference method, so that a binary image can be obtained.
In the first aspect of the present application, which is defined as an alternative embodiment, the flame characteristic value includes a flame area variation characteristic;
and the characteristic extraction is carried out on the plurality of frame picture pictures to obtain a flame characteristic value, and the method comprises the following steps:
determining two frame pictures from the plurality of frame pictures, and respectively determining the two frame pictures as a first picture and a second picture, wherein the shooting time of the first picture is prior to that of the second picture;
calculating the number of flame pixels in the second picture and the number of flame pixels in the second picture;
calculating the difference value of the number of flame pixels in the first picture and the number of flame pixels in the second picture;
and calculating the flame area change characteristic according to the difference value of the number of flame pixels in the first picture and the number of flame pixels in the second picture.
In this alternative embodiment, the flame area variation characteristic can be calculated by the number of flame pixels of at least two of the several frame picture pictures.
In the first aspect of the present application, which is defined as an optional implementation, the flame characteristic value further includes a flame shape characteristic;
and the said characteristic extraction to several frames of picture pictures, in order to obtain the flame characteristic value, also include:
removing picture parts with regular-shaped objects in the plurality of frame picture pictures, and obtaining picture parts with irregular-shaped objects;
calculating the total length of the edges of the irregular-shaped object and the area of the area where the edges are located to obtain a circularity value of the irregular-shaped object;
determining a circularity value of the irregularly shaped object as the flame shape characteristic.
In this optional embodiment, the frame part of the plurality of frame pictures with the regular-shaped object is removed, so that the frame part with the irregular-shaped object can be obtained, the total length of the edge of the irregular-shaped object and the area of the area where the edge of the irregular-shaped object is located can be calculated, the circularity value of the irregular-shaped object can be obtained, and finally the circularity value of the irregular-shaped object can be determined as the flame shape feature.
In the first aspect of the present application, which is defined as an optional implementation, the flame characteristic value further includes a flame strobe characteristic;
and the said characteristic extraction to several frames of picture pictures, in order to obtain the flame characteristic value, also include:
calculating the flame height in the first picture and the flame height in the second picture;
calculating the difference between the flame height in the first picture and the flame height in the second picture;
calculating the flicker frequency of the flame according to the difference between the flame height in the first picture and the flame height in the second picture, and determining the flicker frequency of the flame as the flame stroboscopic feature.
In this optional embodiment, by calculating the flame height in the first image picture and the flame height in the second image picture, the difference between the flame height in the first image picture and the flame height in the second image picture can be calculated, and then the flicker frequency of the flame can be calculated according to the difference between the flame height in the first image picture and the flame height in the second image picture, and the flicker frequency of the flame is determined as the flame stroboscopic feature.
In the first aspect of the present application, which is defined as an optional implementation, the flame characteristic value further includes a color characteristic;
and the said characteristic extraction to several frames of picture pictures, in order to obtain the flame characteristic value, also include:
calculating RGB values of the plurality of frame picture images;
and determining the color characteristics according to the RGB values of the plurality of frame pictures.
In this alternative embodiment, the color characteristics can be determined by calculating RGB values of the several frame pictures.
A second aspect of the present application discloses a flame identification device based on video pictures, the device comprising:
the acquisition module is used for acquiring a target video picture;
the extraction module is used for extracting a plurality of frame picture pictures from the target video picture;
the image preprocessing module is used for preprocessing the plurality of frame picture pictures;
the characteristic extraction module is used for extracting the characteristics of the plurality of frame picture pictures to obtain a flame characteristic value;
the identification module is used for determining the flame characteristic value as the input of a flame identification model;
and the determining module is used for determining that flame exists in the target video picture when the flame recognition model recognizes that the flame characteristic value meets the preset condition.
In the second aspect of the present application, the flame recognition device based on the video frame can acquire the target video frame by executing the flame recognition method based on the video frame, and then can extract a plurality of frame picture pictures from the target video frame, and then can perform image preprocessing on the plurality of frame picture pictures, and then can perform feature extraction on the plurality of frame picture pictures to obtain the flame feature value, and then can determine the flame feature value as the input of the flame recognition model, so that when the flame feature value is recognized by the flame recognition model to meet the preset condition, it can be determined that flame exists in the target video frame.
A third aspect of the present application discloses a flame identification device based on video pictures, the device comprising:
a memory storing executable program code;
a processor coupled with the memory;
the processor calls the executable program code stored in the memory to execute the video picture based flame identification method as disclosed in the first aspect of the application.
In the third aspect of the present application, the flame recognition device based on the video frame can acquire the target video frame by executing the flame recognition method based on the video frame, and then can extract a plurality of frame picture pictures from the target video frame, and then can perform image preprocessing on the plurality of frame picture pictures, and then can perform feature extraction on the plurality of frame picture pictures to obtain the flame feature value, and then can determine the flame feature value as the input of the flame recognition model, so that when the flame feature value is recognized by the flame recognition model to meet the preset condition, it can be determined that flame exists in the target video frame.
A fourth aspect of the present application discloses a storage medium storing computer instructions for executing the video-picture-based flame recognition method of the first aspect of the present application when the computer instructions are invoked.
In the fourth aspect of the present application, the storage medium may acquire a target video picture by executing a flame recognition method based on a video picture, and may further extract a plurality of frame picture pictures from the target video picture, and may further perform image preprocessing on the plurality of frame picture pictures, and may further perform feature extraction on the plurality of frame picture pictures to obtain a flame feature value, and may further determine the flame feature value as an input of a flame recognition model, so that when the flame recognition model recognizes that the flame feature value satisfies a preset condition, it may be determined that flame exists in the target video picture.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
FIG. 1 is a schematic flow chart of a method for identifying flames based on video frames according to an embodiment of the present application;
FIG. 2 is a schematic structural diagram of a flame identification device based on video pictures according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a flame identification device based on a video frame, which is disclosed in an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. 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 application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
In the description of the present application, it should be noted that if the terms "upper", "lower", "inside", "outside", etc. are used for indicating the orientation or positional relationship based on the orientation or positional relationship shown in the drawings or the orientation or positional relationship which the present invention product is usually put into use, it is only for convenience of describing the present application and simplifying the description, but it is not intended to indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and be operated, and thus, should not be construed as limiting the present application.
Furthermore, the appearances of the terms "first," "second," and the like, if any, are used solely to distinguish one from another and are not to be construed as indicating or implying relative importance.
It should be noted that the features of the embodiments of the present application may be combined with each other without conflict.
Example one
Referring to fig. 1, fig. 1 is a schematic flowchart illustrating a flame identification method based on a video frame according to an embodiment of the present disclosure. As shown in fig. 1, a flame identification method according to an embodiment of the present application includes:
101. acquiring a target video picture;
102. extracting a plurality of frame picture pictures from a target video picture;
103. carrying out image preprocessing on a plurality of frame pictures;
104. extracting the characteristics of a plurality of frame picture pictures to obtain a flame characteristic value;
105. determining the flame characteristic value as the input of a flame identification model;
106. and when the flame identification model identifies that the flame characteristic value meets the preset condition, determining that flame exists in the target video picture.
In the embodiment of the application, the target video picture is obtained, then the plurality of frame picture pictures can be extracted from the target video picture, then the plurality of frame picture pictures can be subjected to image preprocessing, then the plurality of frame picture pictures can be subjected to feature extraction, so that the flame feature value can be obtained, and then the flame feature value can be determined as the input of the flame identification model, so that when the flame feature value identified by the flame identification model meets the preset condition, the flame in the target video picture can be determined.
In the first embodiment of the present application, which is determined as an optional implementation, step 103: the image preprocessing is carried out on a plurality of frame picture pictures, and the image preprocessing comprises the following substeps:
performing median filtering on a plurality of frame picture pictures to remove noise of each frame picture;
and sharpening a plurality of frames of picture pictures by adopting an exponential high-pass filter.
In this optional embodiment, median filtering is performed on a plurality of frame picture pictures to remove noise of each frame picture, and meanwhile, an exponential high-pass filter is adopted to sharpen the plurality of frame picture pictures.
In the embodiment of the present application, it is determined that as an optional implementation manner, after sharpening several frames of picture pictures by using an exponential high-pass filter, step 103: the image preprocessing of the plurality of frame picture pictures further comprises the substeps of:
and carrying out image segmentation on the plurality of frames of picture pictures by adopting a difference method to obtain a binary image.
In this optional embodiment, the difference method can be used to perform image segmentation on a plurality of frame picture pictures, so as to obtain a binary image.
In the examples of the present application, it is determined that in an alternative embodiment, the flame characteristic value includes a flame area variation characteristic;
and, the step: the method comprises the following steps of extracting the characteristics of a plurality of frame picture pictures to obtain a flame characteristic value, and comprises the following substeps:
determining two frame pictures from the plurality of frame pictures, and respectively determining the two frame pictures as a first picture and a second picture, wherein the shooting time of the first picture is prior to that of the second picture;
calculating the number of flame pixels in the second picture and the number of flame pixels in the second picture;
calculating the difference value of the number of flame pixels in the first picture and the number of flame pixels in the second picture;
and calculating the flame area change characteristic according to the difference value of the number of flame pixels in the first picture and the number of flame pixels in the second picture.
In this alternative embodiment, the flame area variation characteristic can be calculated by the number of flame pixels of at least two of the several frame picture pictures.
In the example of the present application, it is determined that in an alternative embodiment, the flame characteristic value further includes a flame shape characteristic;
and, the step: the method comprises the following steps of extracting the characteristics of a plurality of frame picture pictures to obtain a flame characteristic value, and further comprising the following substeps:
removing picture parts with regular-shaped objects in a plurality of frames of picture pictures, and obtaining picture parts with irregular-shaped objects;
calculating the total length of the edges of the irregular-shaped object and the area of the area where the irregular-shaped object is located to obtain a circularity value of the irregular-shaped object;
the circularity values of the irregularly shaped object are determined as the flame shape characteristic.
In this optional embodiment, the picture portion of the irregular-shaped object can be obtained by removing the picture portion of the regular-shaped object in the plurality of frames of picture pictures, and then the circularity value of the irregular-shaped object can be obtained by calculating the total length of the edge of the irregular-shaped object and the area of the area where the irregular-shaped object is located, and finally the circularity value of the irregular-shaped object can be determined as the flame shape feature.
In the example of the present application, it is determined that in an alternative implementation, the flame characteristic value further includes a flame strobe characteristic;
and, the step: the method comprises the following steps of extracting the characteristics of a plurality of frame picture pictures to obtain a flame characteristic value, and further comprising the following substeps:
calculating the flame height in the first picture and the flame height in the second picture;
calculating the difference between the flame height in the first picture and the flame height in the second picture;
and calculating the flicker frequency of the flame according to the difference between the flame height in the first picture and the flame height in the second picture, and determining the flicker frequency of the flame as the flame stroboscopic characteristic.
In this optional embodiment, by calculating the flame height in the first image picture and the flame height in the second image picture, the difference between the flame height in the first image picture and the flame height in the second image picture can be calculated, and then the flicker frequency of the flame can be calculated according to the difference between the flame height in the first image picture and the flame height in the second image picture, and the flicker frequency of the flame is determined as the flame stroboscopic feature.
In the example of the present application, it is determined that in an alternative embodiment, the flame characteristic value further includes a color characteristic;
and, the step: the method comprises the following steps of extracting the characteristics of a plurality of frame picture pictures to obtain a flame characteristic value, and further comprising the following substeps:
calculating RGB values of a plurality of frame picture images;
and determining color characteristics according to the RGB values of the plurality of frame pictures.
In this alternative embodiment the color characteristics can be determined by calculating the RGB values of several frame pictures.
Example two
Referring to fig. 2, fig. 2 is a schematic structural diagram of a flame recognition device based on a video frame according to an embodiment of the present disclosure. As shown in fig. 2, a flame recognition device according to an embodiment of the present application includes:
an obtaining module 201, configured to obtain a target video frame;
an extraction module 202, configured to extract a plurality of frame pictures from a target video picture;
the image preprocessing module 203 is used for preprocessing the images of the plurality of frames of picture pictures;
the feature extraction module 204 is configured to perform feature extraction on the plurality of frame image pictures to obtain a flame feature value;
a recognition module 205 for determining the flame feature values as input to a flame recognition model;
and the determining module 206 is configured to determine that flame exists in the target video frame when the flame recognition model recognizes that the flame characteristic value satisfies the preset condition.
In the embodiment of the application, the flame recognition device based on the video picture can acquire the target video picture by executing the flame recognition method based on the video picture, and then can extract a plurality of frame picture pictures from the target video picture, and then can perform image preprocessing on the plurality of frame picture pictures, and then can perform feature extraction on the plurality of frame picture pictures to obtain the flame feature value, and further can determine the flame feature value as the input of the flame recognition model, so that when the flame feature value recognized by the flame recognition model meets the preset condition, the flame in the target video picture can be determined.
In the first embodiment of the present application, it is determined that an optional implementation manner is provided, where the image preprocessing module 203 includes a filtering sub-module and a sharpening sub-module, where:
the filtering submodule is used for carrying out median filtering on a plurality of frame picture pictures so as to remove noise of each frame picture;
and the sharpening submodule is used for sharpening the plurality of frames of picture pictures by adopting an index high-pass filter.
In this optional embodiment, median filtering is performed on a plurality of frame picture pictures to remove noise of each frame picture, and meanwhile, an exponential high-pass filter is adopted to sharpen the plurality of frame picture pictures.
In the embodiment of the present application, it is determined that in an optional implementation manner, the image preprocessing module 203 includes an image segmentation module, where:
and the image segmentation module is used for carrying out image segmentation on the plurality of frames of picture pictures by adopting a difference method to obtain a binary image.
In this optional embodiment, the difference method can be used to perform image segmentation on a plurality of frame picture pictures, so as to obtain a binary image.
In the embodiment of the present application, the flame feature value is determined as an optional implementation manner, the flame feature value includes a flame area variation feature, and the feature extraction module 204 includes a first determination sub-module, a first calculation sub-module, a second calculation sub-module, and a third calculation sub-module, where:
the first determining submodule is used for determining two frame pictures from the plurality of frame pictures, the two frame pictures are respectively determined as a first picture and a second picture, and the shooting time of the first picture is prior to that of the second picture;
the first calculation submodule is used for calculating the number of flame pixels in the second picture and the number of flame pixels in the second picture;
the second calculation submodule calculates the difference value of the number of flame pixels in the first picture and the number of flame pixels in the second picture;
and the third calculation submodule is used for calculating the flame area change characteristics according to the difference value of the number of flame pixels in the first picture and the number of flame pixels in the second picture.
In this alternative embodiment, the flame area variation characteristic can be calculated by the number of flame pixels of at least two of the several frame picture pictures.
In the embodiment of the present application, it is determined that in an optional implementation, the flame feature value further includes a flame shape feature, and the feature extraction module 204 further includes a removal sub-module and a fourth calculation sub-module, where:
the removing submodule is used for removing the picture parts of the objects with the regular shapes in the plurality of frames of picture pictures and obtaining the picture parts of the objects with the irregular shapes;
the fourth calculation submodule is used for calculating the total length of the edge of the object with the irregular shape and the area of the area where the edge of the object with the irregular shape is located, and obtaining the circularity value of the object with the irregular shape;
a second determination submodule for determining circularity values of objects having irregular shapes as flame shape characteristics.
In this optional embodiment, the picture portion of the irregular-shaped object can be obtained by removing the picture portion of the regular-shaped object in the plurality of frames of picture pictures, and then the circularity value of the irregular-shaped object can be obtained by calculating the total length of the edge of the irregular-shaped object and the area of the area where the irregular-shaped object is located, and finally the circularity value of the irregular-shaped object can be determined as the flame shape feature.
In the embodiment of the present application, it is determined that in an optional implementation, the flame feature value further includes a flame strobe feature, and the feature extraction module 204 further includes a fifth computation submodule, a sixth computation submodule, and a seventh computation submodule, where:
the fifth calculation submodule is used for calculating the flame height in the first picture and the flame height in the second picture;
the sixth calculating submodule is used for calculating the difference between the flame height in the first picture and the flame height in the second picture;
and the seventh calculation submodule is used for calculating the flicker frequency of the flame according to the difference between the flame height in the first picture and the flame height in the second picture, and determining the flicker frequency of the flame as the flame stroboscopic characteristic.
In this optional embodiment, by calculating the flame height in the first image picture and the flame height in the second image picture, the difference between the flame height in the first image picture and the flame height in the second image picture can be calculated, and then the flicker frequency of the flame can be calculated according to the difference between the flame height in the first image picture and the flame height in the second image picture, and the flicker frequency of the flame is determined as the flame stroboscopic feature.
In the embodiment of the present application, the determination is an optional implementation manner, the flame feature value further includes a color feature, and the feature extraction module 204 further includes an eighth calculation sub-module and a third determination sub-module, where:
the eighth calculation submodule is used for calculating RGB values of a plurality of frame picture images;
and the third determining sub-module is used for determining color characteristics according to the RGB values of the plurality of frame pictures.
In this alternative embodiment the color characteristics can be determined by calculating the RGB values of several frame pictures.
EXAMPLE III
Referring to fig. 3, fig. 3 is a schematic structural diagram of a flame identification device based on a video frame according to an embodiment of the present disclosure. As shown in FIG. 3, the flame identification device of the embodiment of the present application includes
A memory 301 storing executable program code;
a processor 302 coupled to the memory;
the processor 302 calls the executable program code stored in the memory 301 to execute the flame recognition method based on video frames as disclosed in the first embodiment of the present application.
In the embodiment of the application, the flame recognition device based on the video picture can acquire the target video picture by executing the flame recognition method based on the video picture, and then can extract a plurality of frame picture pictures from the target video picture, and further can perform image preprocessing on the plurality of frame picture pictures, and further can perform feature extraction on the plurality of frame picture pictures to obtain the flame feature value, and further can determine the flame feature value as the input of the flame recognition model, so that when the flame feature value recognized by the flame recognition model meets the preset condition, the flame in the target video picture can be determined.
Example four
The embodiment of the application discloses a storage medium, wherein a computer instruction is stored in the storage medium, and when the computer instruction is called, the storage medium is used for executing the flame identification method based on the video image.
In the embodiment of the application, the storage medium can acquire a target video picture by executing a flame identification method based on the video picture, and then can extract a plurality of frame picture pictures from the target video picture, and then can perform image preprocessing on the plurality of frame picture pictures, and then can perform feature extraction on the plurality of frame picture pictures to obtain a flame feature value, and then can determine the flame feature value as the input of a flame identification model, so that when the flame feature value identified by the flame identification model meets a preset condition, it can be determined that flames exist in the target video picture.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A flame identification method based on a video picture is characterized by comprising the following steps:
acquiring a target video picture;
extracting a plurality of frame picture pictures from the target video picture;
carrying out image preprocessing on the plurality of frame picture pictures;
extracting the characteristics of the plurality of frame picture pictures to obtain a flame characteristic value;
determining the flame characteristic value as an input of a flame identification model;
and when the flame recognition model recognizes that the flame characteristic value meets a preset condition, determining that flame exists in the target video picture.
2. The method of claim 1, wherein the image pre-processing the frames of picture pictures comprises:
performing median filtering on the plurality of frames of picture pictures to remove noise of each frame of picture;
and sharpening the plurality of frames of picture pictures by adopting an exponential high-pass filter.
3. The method of claim 2, wherein after the sharpening the number of frame picture pictures with the exponential high-pass filter, the image pre-processing the number of frame picture pictures further comprises:
and carrying out image segmentation on the plurality of frames of picture pictures by adopting a difference method to obtain a binary image.
4. The method of claim 3, wherein the flame signature value comprises a flame area variation signature;
and the characteristic extraction is carried out on the plurality of frame picture pictures to obtain a flame characteristic value, and the method comprises the following steps:
determining two frame pictures from the plurality of frame pictures, and respectively determining the two frame pictures as a first picture and a second picture, wherein the shooting time of the first picture is prior to that of the second picture;
calculating the number of flame pixels in the second picture and the number of flame pixels in the second picture;
calculating the difference value of the number of flame pixels in the first picture and the number of flame pixels in the second picture;
and calculating the flame area change characteristic according to the difference value of the number of flame pixels in the first picture and the number of flame pixels in the second picture.
5. The method of claim 4, wherein the flame signature value further comprises a flame shape characteristic;
and the said characteristic extraction to several frames of picture pictures, in order to obtain the flame characteristic value, also include:
removing picture parts with regular-shaped objects in the plurality of frame picture pictures, and obtaining picture parts with irregular-shaped objects;
calculating the total length of the edges of the irregular-shaped object and the area of the area where the edges are located to obtain a circularity value of the irregular-shaped object;
determining a circularity value of the irregularly shaped object as the flame shape characteristic.
6. The method of claim 5, wherein the flame signature value further comprises a flame strobe signature;
and the said characteristic extraction to several frames of picture pictures, in order to obtain the flame characteristic value, also include:
calculating the flame height in the first picture and the flame height in the second picture;
calculating the difference between the flame height in the first picture and the flame height in the second picture;
calculating the flicker frequency of the flame according to the difference between the flame height in the first picture and the flame height in the second picture, and determining the flicker frequency of the flame as the flame stroboscopic feature.
7. The method of claim 1, wherein the flame signature value further comprises a color signature;
and the said characteristic extraction to several frames of picture pictures, in order to obtain the flame characteristic value, also include:
calculating RGB values of the plurality of frame picture images;
and determining the color characteristics according to the RGB values of the plurality of frame pictures.
8. A flame identification device based on video pictures, the device comprising:
the acquisition module is used for acquiring a target video picture;
the extraction module is used for extracting a plurality of frame picture pictures from the target video picture;
the image preprocessing module is used for preprocessing the plurality of frame picture pictures;
the characteristic extraction module is used for extracting the characteristics of the plurality of frame picture pictures to obtain a flame characteristic value;
the identification module is used for determining the flame characteristic value as the input of a flame identification model;
and the determining module is used for determining that flame exists in the target video picture when the flame recognition model recognizes that the flame characteristic value meets the preset condition.
9. A flame identification device based on video pictures, characterized in that the device comprises:
a memory storing executable program code;
a processor coupled with the memory;
the processor calls the executable program code stored in the memory to execute the video picture-based flame identification method according to any one of claims 1 to 7.
10. A storage medium storing computer instructions which, when invoked, perform a method for video-picture-based flame recognition according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010740834.1A CN112115766A (en) | 2020-07-28 | 2020-07-28 | Flame identification method, device, equipment and storage medium based on video picture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010740834.1A CN112115766A (en) | 2020-07-28 | 2020-07-28 | Flame identification method, device, equipment and storage medium based on video picture |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112115766A true CN112115766A (en) | 2020-12-22 |
Family
ID=73799182
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010740834.1A Pending CN112115766A (en) | 2020-07-28 | 2020-07-28 | Flame identification method, device, equipment and storage medium based on video picture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112115766A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115841731A (en) * | 2023-02-16 | 2023-03-24 | 广东汇通信息科技股份有限公司 | Infrared-monitoring park fire early warning method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103886344A (en) * | 2014-04-14 | 2014-06-25 | 西安科技大学 | Image type fire flame identification method |
| US20150169984A1 (en) * | 2013-12-13 | 2015-06-18 | Michael Newton | Flame detection system and method |
| CN108038510A (en) * | 2017-12-22 | 2018-05-15 | 湖南源信光电科技股份有限公司 | A kind of detection method based on doubtful flame region feature |
| CN108074234A (en) * | 2017-12-22 | 2018-05-25 | 湖南源信光电科技股份有限公司 | A kind of large space flame detecting method based on target following and multiple features fusion |
| CN109145689A (en) * | 2017-06-28 | 2019-01-04 | 南京理工大学 | A kind of robot fire detection method |
| CN109886227A (en) * | 2019-02-27 | 2019-06-14 | 哈尔滨工业大学 | Indoor fire video recognition method based on multi-channel convolutional neural network |
| CN110276284A (en) * | 2019-06-11 | 2019-09-24 | 暨南大学 | Flame identification method, device, equipment and storage medium based on video quality assessment |
-
2020
- 2020-07-28 CN CN202010740834.1A patent/CN112115766A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150169984A1 (en) * | 2013-12-13 | 2015-06-18 | Michael Newton | Flame detection system and method |
| CN103886344A (en) * | 2014-04-14 | 2014-06-25 | 西安科技大学 | Image type fire flame identification method |
| CN109145689A (en) * | 2017-06-28 | 2019-01-04 | 南京理工大学 | A kind of robot fire detection method |
| CN108038510A (en) * | 2017-12-22 | 2018-05-15 | 湖南源信光电科技股份有限公司 | A kind of detection method based on doubtful flame region feature |
| CN108074234A (en) * | 2017-12-22 | 2018-05-25 | 湖南源信光电科技股份有限公司 | A kind of large space flame detecting method based on target following and multiple features fusion |
| CN109886227A (en) * | 2019-02-27 | 2019-06-14 | 哈尔滨工业大学 | Indoor fire video recognition method based on multi-channel convolutional neural network |
| CN110276284A (en) * | 2019-06-11 | 2019-09-24 | 暨南大学 | Flame identification method, device, equipment and storage medium based on video quality assessment |
Non-Patent Citations (3)
| Title |
|---|
| TI HAN等: "Flame detection method based on feature recognition", 《COMMUNICATIONS, SIGNAL PROCESSING, AND SYSTEMS》, pages 1 - 8 * |
| 唐岩岩: "基于视频图像的火灾检测方法研究", 《中国优秀硕士学位论文全文数据库 (信息科技辑)》, no. 2013, pages 138 - 613 * |
| 王媛彬: "基于计算机视觉的锅炉燃烧状态检测方法研究", 《中国博士学位论文全文数据库 (信息科技辑)》, no. 2017, pages 138 - 203 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115841731A (en) * | 2023-02-16 | 2023-03-24 | 广东汇通信息科技股份有限公司 | Infrared-monitoring park fire early warning method |
| CN115841731B (en) * | 2023-02-16 | 2023-05-05 | 广东汇通信息科技股份有限公司 | Infrared monitoring park fire early warning method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106934803B (en) | method and device for detecting surface defects of electronic device | |
| CN112052797A (en) | MaskRCNN-based video fire identification method and system | |
| CN106056079B (en) | A kind of occlusion detection method of image capture device and human face five-sense-organ | |
| CN111783744A (en) | Operation site safety protection detection method and device | |
| WO2023039781A1 (en) | Method for detecting abandoned object, apparatus, electronic device, and storage medium | |
| JP4999794B2 (en) | Still region detection method and apparatus, program and recording medium | |
| CN109784322B (en) | Method, equipment and medium for identifying vin code based on image processing | |
| KR20180074556A (en) | Face detection method and apparatus thereof | |
| CN118840785B (en) | Intelligent identification method and system for ascending operation risk based on industrial scene | |
| CN111652043A (en) | Object state identification method and device, image acquisition equipment and storage medium | |
| CN111797726A (en) | Flame detection method and device, electronic equipment and storage medium | |
| CN111263955A (en) | Method and device for determining movement track of target object | |
| CN110175553B (en) | Method and device for establishing feature library based on gait recognition and face recognition | |
| CN115019152A (en) | Image shooting integrity judgment method and device | |
| CN107203638B (en) | Monitoring video processing method, device and system | |
| CN112115766A (en) | Flame identification method, device, equipment and storage medium based on video picture | |
| US20200394802A1 (en) | Real-time object detection method for multiple camera images using frame segmentation and intelligent detection pool | |
| CN115311623A (en) | Equipment oil leakage detection method and system based on infrared thermal imaging | |
| CN113065454B (en) | High-altitude parabolic target identification and comparison method and device | |
| KR101395666B1 (en) | Surveillance apparatus and method using change of video image | |
| CN111860326B (en) | Building site article movement detection method, device, equipment and storage medium | |
| CN117974701A (en) | Gas leakage detection method and device, electronic equipment and readable storage medium | |
| CN112329572B (en) | Rapid static living body detection method and device based on frame and flash point | |
| CN114332082B (en) | Definition evaluation method and device, electronic equipment and computer storage medium | |
| CN111353954A (en) | Video image processing method and device and electronic equipment |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20201222 |
|
| WD01 | Invention patent application deemed withdrawn after publication |