CN111062281A - Abnormal event monitoring method and device, storage medium and electronic equipment - Google Patents
Abnormal event monitoring method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN111062281A CN111062281A CN201911235714.XA CN201911235714A CN111062281A CN 111062281 A CN111062281 A CN 111062281A CN 201911235714 A CN201911235714 A CN 201911235714A CN 111062281 A CN111062281 A CN 111062281A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- recognition model
- abnormal event
- abnormal
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 148
- 238000012544 monitoring process Methods 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 46
- 239000013598 vector Substances 0.000 claims abstract description 55
- 239000000779 smoke Substances 0.000 claims description 31
- 238000004590 computer program Methods 0.000 claims description 9
- 238000013145 classification model Methods 0.000 claims description 7
- 238000012806 monitoring device Methods 0.000 abstract 1
- 238000001514 detection method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 241001465754 Metazoa Species 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 238000012706 support-vector machine Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001850 reproductive effect Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000005068 transpiration Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06V10/56—Extraction of image or video features relating to colour
-
- 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
- G06V10/60—Extraction of image or video features relating to illumination properties, e.g. using a reflectance or lighting model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- 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
- G06T2207/10024—Color image
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Fire-Detection Mechanisms (AREA)
- Image Analysis (AREA)
Abstract
The disclosure relates to an abnormal event monitoring method, an abnormal event monitoring device, a storage medium and an electronic device, wherein the method comprises the following steps: if the image of the target area acquired by the camera contains a moving target, generating a feature vector corresponding to the image according to the image feature information of the image; determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image; and determining whether the target area has abnormal events or not according to the feature vector and the target recognition model. The method and the device can select the model required for identifying the abnormal event in the image according to the characteristics of the image, so that the abnormal event can be identified through different models at different time, and the accuracy and the flexibility of monitoring the abnormal event are improved.
Description
Technical Field
The present disclosure relates to the field of image monitoring, and in particular, to a method and an apparatus for monitoring an abnormal event, a storage medium, and an electronic device.
Background
At present, forest fires are one of globally important forest disasters. The forest fire disaster recovery system is wide in distribution and high in occurrence frequency, forest fires are the most dangerous enemies of forests, and are the most terrible disasters of the forests, and the forest fire disaster recovery system can bring the most harmful and destructive consequences to the forests. Forest fires not only burn pieces of forest to damage animals in the forest, but also reduce the reproductive capacity of the forest, cause soil impoverishment and destroy the effect of conserving water sources of the forest, even cause the ecological environment to lose balance and cause global environmental pollution. Therefore, how to identify the fire in the early stage of forest fire and take countermeasures in time is very important. In the prior art, an image of a forest is usually acquired through a monitoring camera, and then a certain characteristic (for example, an abnormal object or a color) of the image is monitored through a single image recognition model to determine whether a fire disaster occurs, so that the accuracy of fire disaster recognition is low, and the flexibility is poor.
Disclosure of Invention
To overcome the problems in the related art, an object of the present disclosure is to provide an abnormal event monitoring method, apparatus, storage medium, and electronic device.
In order to achieve the above object, according to a first aspect of embodiments of the present disclosure, there is provided an abnormal event monitoring method, including:
if the image of the target area acquired by the camera contains a moving target, generating a feature vector corresponding to the image according to the image feature information of the image;
determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image;
determining whether an abnormal event occurs in the target area according to the feature vector and the target identification model, wherein the abnormal event is a fire event, and the plurality of abnormal event monitoring models comprise: a smoke recognition model and a flame recognition model.
Optionally, the determining a target recognition model from a plurality of fire recognition models trained in advance according to the brightness information of the image includes:
taking the smoke recognition model as the target recognition model when the brightness of the image is greater than or equal to a preset first threshold value;
taking the flame recognition model as the target recognition model when the brightness of the image is less than or equal to a preset second threshold value; alternatively, the first and second electrodes may be,
taking the smoke recognition model and the flame recognition model as the target recognition model if the brightness of the image is less than the first threshold and greater than the second threshold.
Optionally, if the image of the target area acquired by the camera includes a moving target, generating a feature vector corresponding to the image according to the image feature information of the image, including:
acquiring a color space image, a binary image and a gray level image corresponding to the image;
extracting first image characteristic information of the color space image;
extracting second image characteristic information of the binary image;
extracting third image characteristic information of the gray-scale image;
and generating the feature vector according to the first image feature information, the second image feature information and the third image feature information.
Optionally, the determining, by the target recognition model, whether an abnormal event occurs in the target region according to the feature vector and the target recognition model includes:
taking the feature vector as the input of the target recognition model to obtain a classification result output by the target recognition model;
determining whether the abnormal object appears in the target area according to the classification result;
and if the abnormal object appears in the target area, determining that the abnormal event appears in the target area.
Optionally, the method further includes:
after the target area is determined to have the abnormal event, determining the target position of the abnormal object in the target area;
acquiring the temperature of the target position as the target temperature of the abnormal object;
and outputting alarm information containing the target temperature and the target position.
According to a second aspect of the embodiments of the present disclosure, there is provided an abnormal event monitoring apparatus, the apparatus including:
the vector generation module is used for generating a characteristic vector corresponding to the image according to the image characteristic information of the image if the image of the target area acquired by the camera contains a moving target;
the model determining module is used for determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image;
an abnormal event monitoring module, configured to determine whether an abnormal event occurs in the target area according to the feature vector and the target identification model, where the abnormal event is a fire event, and the multiple abnormal event monitoring models include: a smoke recognition model and a flame recognition model.
Optionally, the model determining module is configured to:
taking the smoke recognition model as the target recognition model when the brightness of the image is greater than or equal to a preset first threshold value;
taking the flame recognition model as the target recognition model when the brightness of the image is less than or equal to a preset second threshold value; alternatively, the first and second electrodes may be,
taking the smoke recognition model and the flame recognition model as the target recognition model if the brightness of the image is less than the first threshold and greater than the second threshold.
Optionally, the vector generating module is configured to:
acquiring a color space image, a binary image and a gray level image corresponding to the image;
extracting first image characteristic information of the color space image;
extracting second image characteristic information of the binary image;
extracting third image characteristic information of the gray-scale image;
and generating the feature vector according to the first image feature information, the second image feature information and the third image feature information.
Optionally, the target identification model is a classification model, and the abnormal event monitoring module is configured to:
taking the feature vector as the input of the target recognition model to obtain a classification result output by the target recognition model;
determining whether the abnormal object appears in the target area according to the classification result;
and if the abnormal object appears in the target area, determining that the abnormal event appears in the target area.
Optionally, the apparatus further comprises:
the position determining module is used for determining the target position of the abnormal object in the target area after determining that the abnormal event occurs in the target area;
the temperature determining module is used for acquiring the temperature of the target position as the target temperature of the abnormal object;
and the alarm output module is used for outputting alarm information containing the target temperature and the target position.
According to a third aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the steps of the abnormal event monitoring method provided by the first aspect of the embodiments of the present disclosure.
According to a fourth aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including:
a memory having a computer program stored thereon;
a processor, configured to execute the computer program in the memory, so as to implement the steps of the abnormal event monitoring method provided in the first aspect of the embodiment of the present disclosure.
In summary, according to the technical solution provided by the embodiment of the present disclosure, a moving target can be included in an image of a target area acquired by a camera, and a feature vector corresponding to the image is generated according to image feature information of the image; determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image; and determining whether the target area has abnormal events or not according to the feature vector and the target recognition model. The method and the device can select the model required for identifying the abnormal event in the image according to the characteristics of the image, so that the abnormal event can be identified through different models at different time, and the accuracy and the flexibility of monitoring the abnormal event are improved.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow diagram illustrating a method of abnormal event monitoring in accordance with an exemplary embodiment;
FIG. 2 is a flow chart of a model determination method according to the one shown in FIG. 1;
FIG. 3 is a flow chart of a method of vector generation according to that shown in FIG. 2;
FIG. 4 is a flow chart of a method of abnormal event monitoring according to the method shown in FIG. 1;
FIG. 5 is a flow chart of another abnormal event monitoring method according to FIG. 1;
FIG. 6 is a block diagram illustrating an abnormal event monitoring apparatus in accordance with an exemplary embodiment;
FIG. 7 is a block diagram of another abnormal event monitoring apparatus according to FIG. 6;
FIG. 8 is a block diagram illustrating an electronic device in accordance with an example embodiment.
Detailed Description
The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
FIG. 1 is a flow chart illustrating a method of abnormal event monitoring, as shown in FIG. 1, according to an exemplary embodiment, the method comprising:
in step 101, if the image of the target area acquired by the camera includes a moving target, a feature vector corresponding to the image is generated according to the image feature information of the image.
Illustratively, the embodiment of the present disclosure identifies images collected by a camera to determine whether an abnormal event occurs, so as to monitor the abnormal event. Before the image is identified, whether the image contains a moving object needs to be determined. For example, when identifying whether a fire event occurs in a certain forest (in a target area), it can be considered that when an abnormal event does not occur (i.e., under normal conditions), all objects in an image of the forest acquired by a camera are still, if a moving object occurs in the image, an abnormal event is likely to occur, and the moving object is a moving target. And if the image contains the moving target, extracting the characteristic information of the image, and identifying the image according to the characteristic vector generated by the characteristic information.
For example, whether the image contains the moving target may be detected by a background difference method, that is, an image of a target area acquired by the camera under a normal condition is first used as a reference model, and then the image of the target area acquired by the camera is compared with the reference model, and if an unequal part is detected in the image, the unequal part is the moving target.
In step 102, a target recognition model is determined from a plurality of abnormal event monitoring models trained in advance according to the brightness of the image.
For example, after the feature vector of the image is obtained, a target recognition model suitable for the image is selected from a plurality of preset abnormal event recognition models according to the characteristics of the image, so that the recognition accuracy is increased. In the present disclosure, the target recognition model is selected according to the brightness of the image, that is, when the brightness of the image is different, different target recognition models are selected to detect whether an abnormal event occurs in the target area corresponding to the image.
In step 103, it is determined whether an abnormal event occurs in the target area according to the feature vector and the target recognition model.
Wherein the abnormal event is a fire event, and the plurality of abnormal event monitoring models include: a smoke recognition model and a flame recognition model.
For example, after the target model is determined, the feature vector of the image is identified according to the target model to determine whether the image contains the identified object specified in the target identification model, so as to determine whether the target area corresponding to the image has an abnormal event. The abnormal event monitoring method in the embodiment of the disclosure is mainly applied to identifying whether a fire event occurs in a forest, and after a target area is determined, whether smoke or flame occurs in an image of the forest (the target area) is determined through a determined smoke identification model and/or a flame identification model so as to determine whether the fire event occurs in the forest.
In summary, according to the technical solution provided by the embodiment of the present disclosure, a moving target can be included in an image of a target area acquired by a camera, and a feature vector corresponding to the image is generated according to image feature information of the image; determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image; and determining whether the target area has abnormal events or not according to the feature vector and the target recognition model. The method and the device can select the model required for identifying the abnormal event in the image according to the characteristics of the image, so that the abnormal event can be identified through different models at different time, and the accuracy and the flexibility of monitoring the abnormal event are improved.
Fig. 2 is a flow chart of a model determination method according to fig. 1, as shown in fig. 2, the step 102 comprising: step 1021, step 1022, or step 1023.
In step 1021, in the case that the brightness of the image is greater than or equal to a preset first threshold, the smoke recognition model is taken as the target recognition model.
For example, because the geographical environment in the forest is complex, in the case of good light in the daytime, flames are not obvious in the image, and the difficulty of identifying the flames is high, and the difficulty of identifying the smoke is relatively reduced due to the fact that the colors of the smoke are different from the colors of most objects in the forest. Therefore, if the brightness of the image is determined to be greater than or equal to the first brightness threshold (the light in the area corresponding to the image is determined to be good), the smoke model is selected as the target identification model, and whether a fire event occurs in the forest is determined by identifying whether smoke occurs in the image.
In addition, since the embodiment of the present disclosure determines whether a fire event occurs in the target area by recognizing smoke or flame, the abnormal event monitoring model may include: a smoke recognition model and a flame recognition model, and the abnormal event monitoring model may include, among other things, an animal recognition model for recognizing whether a person or animal passes within the target area.
In step 1022, in the case that the brightness of the image is less than or equal to a preset second threshold, the flame recognition model is taken as the target recognition model.
In the example, under the poor condition of night light, if flame appears in the image that the camera was gathered, the colour of flame is obviously different from dark image background colour, and the degree of difficulty of discerning flame is less this moment, and can appear a large amount of water smoke because of the transpiration of plant in the forest at night, disturbs the discernment that whether smog appears in the forest, increases the degree of difficulty of discerning smog. Therefore, when the brightness of the image is less than or equal to the second threshold, the flame recognition model is selected as the target recognition model, and whether a fire occurs in the forest is determined by recognizing whether a flame occurs in the image.
In step 1023, the smoke recognition model and the flame recognition model are used as the target recognition model if the brightness of the image is smaller than the first threshold and larger than the second threshold.
For example, the first threshold is greater than the second threshold, and therefore, if the brightness of the image is between the first threshold and the second threshold, it indicates that the light in the forest is between light and dark, and is in a dim state, and there is a certain difficulty in identifying both the flame and the smoke, so that it is necessary to use both the flame identification model and the smoke identification model as the target identification model to determine whether a fire event occurs by identifying whether smoke or flame occurs in the image.
Fig. 3 is a flow chart of a vector generation method according to fig. 2, as shown in fig. 3, the step 101 includes:
in step 1011, a color space image, a binarized image and a grayscale image corresponding to the image are acquired.
For example, when acquiring image features to generate feature vectors of an image, feature vectors of a color space image, a binary image and a gray scale image obtained by performing a series of processing on an image collected by a camera should be extracted respectively to improve accuracy in image recognition. The method comprises the steps of acquiring an unprocessed original image in a target area by a camera, namely the color space image, denoising the color space image by a preset image preprocessing method (such as a median filtering method), carrying out gray level conversion on the preprocessed image to acquire a gray level image, determining that the image contains a moving target, taking the moving target as a target object, and carrying out binarization conversion on a part of the image which does not contain the moving target as a background to acquire a binarization image.
It should be noted that, in order to reduce noise interference in an image and reduce the data processing amount in calculation, in general, when determining whether a moving object is included in a target region acquired by imaging in step 101, it is determined whether a moving object is included in an image acquired by the camera by detecting a detection result when determining whether a moving object is included in the grayscale image.
In step 1012, first image feature information of the color space image is extracted.
In step 1013, second image feature information of the binarized image is extracted.
In step 1014, third image feature information of the grayscale image is extracted.
In step 1015, a feature vector is generated according to the first image feature information, the second image feature information and the third image feature information.
Illustratively, the first image feature of the color space image includes a third-order color moment feature based on a HIS (Hue & saturation & Intensity) color space, the second image feature information of the binarized image includes a circularity feature and an area growth rate feature, the third image feature information of the grayscale image includes energy, entropy, an inverse difference matrix, and the like of a grayscale co-occurrence matrix, and feature vectors corresponding to the first image feature, the second image feature, and the third image feature are generated through a preset feature vector generation rule.
Fig. 4 is a flowchart of an abnormal event monitoring method according to fig. 1, where the target recognition model is a classification model, as shown in fig. 4, and the step 103 includes:
in step 1031, the feature vectors are used as the input of the target recognition model to obtain the classification result output by the target recognition model.
In step 1032, it is determined whether an abnormal object is present in the target area according to the classification result.
For example, the target recognition model may be a classification model such as a support vector machine, and in general, the multiple frames of continuous images collected by the camera are classified at the same time, that is, after the feature vector of each frame of image is extracted through the above steps 101 and 102, the feature vector of each frame of image is used as the input of the support vector machine, and the support vector machine classifies the image with the abnormal object and the image without the abnormal object, and outputs the classification result. And determining whether abnormal objects such as smoke, flame and the like appear in the target area according to the classification result output by the support vector machine.
In step 1033, if it is determined that the abnormal object is present in the target area, it is determined that an abnormal event is present in the target area.
For example, as described in step 1032 above, when multiple frames of continuous images are classified at the same time, if the image frame sequence of the image including the abnormal object in the classification result output by the classification model is continuous and the number of image frames is greater than the preset frame number threshold, it may be determined that an abnormal event occurs in the target region. For example, if flames appear in a plurality of continuous images, it can be determined that a fire event appears in a target area corresponding to the images.
Fig. 5 is a flow chart of another abnormal event monitoring method according to fig. 1, as shown in fig. 5, after step 103, the method further includes:
in step 104, after determining that the target area has the abnormal event, determining a target position of the abnormal object in the target area.
In step 105, the temperature of the target position is acquired as the target temperature of the abnormal object.
In step 106, alarm information including the target temperature and the target position is output.
Illustratively, a processor for executing the event monitoring method is connected with the camera and a temperature detection unit (typically an infrared scanning unit) respectively, which acquires the temperature at all positions within the target area. After the target position where the abnormal object is located in the target area is determined, the temperature at the target position is acquired by the temperature detection unit and is regarded as the target temperature of the abnormal object. Therefore, when the abnormal event occurs in the target area and the alarm information of the abnormal event is output, the alarm information also comprises the abnormal position where the abnormal event occurs and the target temperature at the abnormal position, so that when a user receives the alarm information, the severity of the abnormal event can be judged through the target temperature, and a treatment measure for dealing with the abnormal event can be taken through the target position in time.
In addition, the temperature detection unit can also be used alone to detect whether an abnormal event occurs in the target area. For example, a larger value (e.g., 300 ℃) is set as the temperature threshold, and whenever the temperature detection unit detects that the temperature at the first position in the target area exceeds the temperature threshold, an alarm message is immediately issued, regardless of whether an abnormal substance is present in the image of the target area, in which the first position information and the temperature at the first position are displayed. Or, when the temperature detection unit detects that the temperature at the second position in the target area is higher than the outdoor temperature (for example, the outdoor temperature is 20 ℃, and the temperature detection unit detects that the temperature at the second position is 37 ℃), combining the result of the image recognition of the second position by the animal recognition model in the abnormal event monitoring model, if it is determined that a person or an animal passes through the second position, no alarm information is sent, and if no person or animal passes through the second position, it is proved that the temperature at the second position belongs to an abnormal phenomenon, and alarm information including second position information and the temperature at the second position is output.
In summary, according to the technical solution provided by the embodiment of the present disclosure, a moving target can be included in an image of a target area acquired by a camera, and a feature vector corresponding to the image is generated according to image feature information of the image; determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image; and determining whether the target area has abnormal events or not according to the feature vector and the target recognition model. The method and the device can select the model required for identifying the abnormal event in the image according to the characteristics of the image, so that the abnormal event can be identified through different models at different time, and the accuracy and the flexibility of monitoring the abnormal event are improved.
Fig. 6 is a block diagram illustrating an abnormal event monitoring apparatus according to an exemplary embodiment, and as shown in fig. 6, the apparatus 600 may include:
the vector generation module 610 is configured to generate a feature vector corresponding to an image according to image feature information of the image if the image of the target area acquired by the camera includes a moving target;
a model determining module 620, configured to determine a target recognition model from a plurality of abnormal event monitoring models trained in advance according to brightness of the image;
an abnormal event monitoring module 630, configured to determine whether an abnormal event occurs in the target area according to the feature vector and the target identification model, where the abnormal event is a fire event, and the plurality of abnormal event monitoring models include: a smoke recognition model and a flame recognition model.
Optionally, the model determining module 620 is configured to:
taking the smoke recognition model as the target recognition model when the brightness of the image is greater than or equal to a preset first threshold value;
taking the flame recognition model as the target recognition model under the condition that the brightness of the image is less than or equal to a preset second threshold value; alternatively, the first and second electrodes may be,
and taking the smoke recognition model and the flame recognition model as the target recognition model under the condition that the brightness of the image is smaller than the first threshold and larger than the second threshold.
Optionally, the vector generating module 610 is configured to:
acquiring a color space image, a binary image and a gray level image corresponding to the image;
extracting first image characteristic information of the color space image;
extracting second image characteristic information of the binary image;
extracting third image characteristic information of the gray image;
and generating the feature vector according to the first image feature information, the second image feature information and the third image feature information.
Optionally, the target recognition model is a classification model, and the abnormal event monitoring module 630 is configured to:
taking the feature vector as the input of the target recognition model to obtain the classification result output by the target recognition model;
determining whether the abnormal object appears in the target area according to the classification result;
and if the abnormal object is determined to appear in the target area, determining that the abnormal event appears in the target area.
Fig. 7 is a block diagram of another abnormal event monitoring apparatus according to fig. 6, and as shown in fig. 7, the apparatus 600 further includes:
the position determining module 640 is configured to determine a target position of the abnormal object in the target area after determining that the target area has the abnormal event;
a temperature determining module 650, configured to obtain a temperature of the target location as a target temperature of the abnormal object;
and an alarm output module 660, configured to output alarm information including the target temperature and the target position.
In summary, according to the technical solution provided by the embodiment of the present disclosure, a moving target can be included in an image of a target area acquired by a camera, and a feature vector corresponding to the image is generated according to image feature information of the image; determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image; and determining whether the target area has abnormal events or not according to the feature vector and the target recognition model. The method and the device can select the model required for identifying the abnormal event in the image according to the characteristics of the image, so that the abnormal event can be identified through different models at different time, and the accuracy and the flexibility of monitoring the abnormal event are improved.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 8 is a block diagram illustrating an electronic device 800 in accordance with an example embodiment. For example, the electronic device 800 may be provided as a server. Referring to fig. 8, an electronic device 800 includes a processor 822, which may be one or more in number, and a memory 832 for storing computer programs executable by the processor 822. The computer programs stored in memory 832 may include one or more modules that each correspond to a set of instructions. Further, the processor 822 may be configured to execute the computer program to perform the above-described abnormal event monitoring method.
Additionally, the electronic device 800 may also include a power component 826 and a communication component 850, the power component 826 may be configured to perform power management of the electronic device 800, and the communication component 850 may be configured to enable communication, e.g., wired or wireless communication, of the electronic device 800. The electronic device 800 may also include input/output (I/O) interfaces 858. The electronic device 800 may operate based on an operating system stored in the memory 832, such as Windows Server, Mac OSXTM, UnixTM, LinuxTM, and the like.
In another exemplary embodiment, a computer readable storage medium is also provided, which comprises program instructions, which when executed by a processor, implement the steps of the above-described abnormal event monitoring method. For example, the computer readable storage medium may be the memory 832 including program instructions executable by the processor 822 of the electronic device 800 to perform the above-described abnormal event monitoring method.
Preferred embodiments of the present disclosure are described in detail above with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and other embodiments of the present disclosure may be easily conceived by those skilled in the art within the technical spirit of the present disclosure after considering the description and practicing the present disclosure, and all fall within the protection scope of the present disclosure.
It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. Meanwhile, any combination can be made between various different embodiments of the disclosure, and the disclosure should be regarded as the disclosure of the disclosure as long as the combination does not depart from the idea of the disclosure. The present disclosure is not limited to the precise structures that have been described above, and the scope of the present disclosure is limited only by the appended claims.
Claims (12)
1. An abnormal event monitoring method, characterized in that the method comprises:
if the image of the target area acquired by the camera contains a moving target, generating a feature vector corresponding to the image according to the image feature information of the image;
determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image;
determining whether an abnormal event occurs in the target area according to the feature vector and the target identification model, wherein the abnormal event is a fire event, and the plurality of abnormal event monitoring models comprise: a smoke recognition model and a flame recognition model.
2. The method of claim 1, wherein determining a target recognition model from a plurality of pre-trained fire recognition models based on brightness information of the image comprises:
taking the smoke recognition model as the target recognition model when the brightness of the image is greater than or equal to a preset first threshold value;
taking the flame recognition model as the target recognition model when the brightness of the image is less than or equal to a preset second threshold value; alternatively, the first and second electrodes may be,
taking the smoke recognition model and the flame recognition model as the target recognition model if the brightness of the image is less than the first threshold and greater than the second threshold.
3. The method according to claim 2, wherein if the image of the target area acquired by the camera contains a moving target, generating a feature vector corresponding to the image according to the image feature information of the image comprises:
acquiring a color space image, a binary image and a gray level image corresponding to the image;
extracting first image characteristic information of the color space image;
extracting second image characteristic information of the binary image;
extracting third image characteristic information of the gray-scale image;
and generating the feature vector according to the first image feature information, the second image feature information and the third image feature information.
4. The method of claim 2, wherein the object recognition model is a classification model, and the determining whether the target region has an abnormal event according to the feature vector and the object recognition model comprises:
taking the feature vector as the input of the target recognition model to obtain a classification result output by the target recognition model;
determining whether the abnormal object appears in the target area according to the classification result;
and if the abnormal object appears in the target area, determining that the abnormal event appears in the target area.
5. The method of claim 2, further comprising:
after the target area is determined to have the abnormal event, determining the target position of the abnormal object in the target area;
acquiring the temperature of the target position as the target temperature of the abnormal object;
and outputting alarm information containing the target temperature and the target position.
6. An abnormal event monitoring apparatus, comprising:
the vector generation module is used for generating a characteristic vector corresponding to the image according to the image characteristic information of the image if the image of the target area acquired by the camera contains a moving target;
the model determining module is used for determining a target recognition model from a plurality of pre-trained abnormal event monitoring models according to the brightness of the image;
an abnormal event monitoring module, configured to determine whether an abnormal event occurs in the target area according to the feature vector and the target identification model, where the abnormal event is a fire event, and the multiple abnormal event monitoring models include: a smoke recognition model and a flame recognition model.
7. The apparatus of claim 6, wherein the model determination module is configured to:
taking the smoke recognition model as the target recognition model when the brightness of the image is greater than or equal to a preset first threshold value;
taking the flame recognition model as the target recognition model when the brightness of the image is less than or equal to a preset second threshold value; alternatively, the first and second electrodes may be,
taking the smoke recognition model and the flame recognition model as the target recognition model if the brightness of the image is less than the first threshold and greater than the second threshold.
8. The apparatus of claim 7, wherein the vector generation module is configured to:
acquiring a color space image, a binary image and a gray level image corresponding to the image;
extracting first image characteristic information of the color space image;
extracting second image characteristic information of the binary image;
extracting third image characteristic information of the gray-scale image;
and generating the feature vector according to the first image feature information, the second image feature information and the third image feature information.
9. The apparatus of claim 7, wherein the object recognition model is a classification model, and wherein the abnormal event monitoring module is configured to:
taking the feature vector as the input of the target recognition model to obtain a classification result output by the target recognition model;
determining whether the abnormal object appears in the target area according to the classification result;
and if the abnormal object appears in the target area, determining that the abnormal event appears in the target area.
10. The apparatus of claim 7, further comprising:
the position determining module is used for determining the target position of the abnormal object in the target area after determining that the abnormal event occurs in the target area;
the temperature determining module is used for acquiring the temperature of the target position as the target temperature of the abnormal object;
and the alarm output module is used for outputting alarm information containing the target temperature and the target position.
11. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the exceptional monitoring method according to any one of claims 1-5.
12. An electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the abnormal event monitoring method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911235714.XA CN111062281A (en) | 2019-12-05 | 2019-12-05 | Abnormal event monitoring method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911235714.XA CN111062281A (en) | 2019-12-05 | 2019-12-05 | Abnormal event monitoring method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111062281A true CN111062281A (en) | 2020-04-24 |
Family
ID=70299889
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911235714.XA Pending CN111062281A (en) | 2019-12-05 | 2019-12-05 | Abnormal event monitoring method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111062281A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652184A (en) * | 2020-06-19 | 2020-09-11 | 成都通甲优博科技有限责任公司 | Smoke identification method and device, storage medium and data processing equipment |
| CN112084963A (en) * | 2020-09-11 | 2020-12-15 | 中德（珠海）人工智能研究院有限公司 | Monitoring early warning method, system and storage medium |
| CN112200170A (en) * | 2020-12-07 | 2021-01-08 | 北京沃东天骏信息技术有限公司 | Image recognition method and device, electronic equipment and computer readable medium |
| CN112272288A (en) * | 2020-10-23 | 2021-01-26 | 岭东核电有限公司 | Nuclear power operation monitoring method and device, computer equipment and storage medium |
| CN112364715A (en) * | 2020-10-23 | 2021-02-12 | 岭东核电有限公司 | Nuclear power operation abnormity monitoring method and device, computer equipment and storage medium |
| CN114002751A (en) * | 2021-10-26 | 2022-02-01 | 国网北京市电力公司 | Abnormal position identification method, system and device |
| CN114093117A (en) * | 2021-10-11 | 2022-02-25 | 北京精英系统科技有限公司 | Fire control management and control method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236947A (en) * | 2010-04-29 | 2011-11-09 | 中国建筑科学研究院 | Flame monitoring method and system based on video camera |
| US20130279803A1 (en) * | 2010-01-15 | 2013-10-24 | Ahmet Enis Cetin | Method and system for smoke detection using nonlinear analysis of video |
| CN106997461A (en) * | 2017-03-28 | 2017-08-01 | 浙江大华技术股份有限公司 | A kind of firework detecting method and device |
-
2019
- 2019-12-05 CN CN201911235714.XA patent/CN111062281A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130279803A1 (en) * | 2010-01-15 | 2013-10-24 | Ahmet Enis Cetin | Method and system for smoke detection using nonlinear analysis of video |
| CN102236947A (en) * | 2010-04-29 | 2011-11-09 | 中国建筑科学研究院 | Flame monitoring method and system based on video camera |
| CN106997461A (en) * | 2017-03-28 | 2017-08-01 | 浙江大华技术股份有限公司 | A kind of firework detecting method and device |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652184A (en) * | 2020-06-19 | 2020-09-11 | 成都通甲优博科技有限责任公司 | Smoke identification method and device, storage medium and data processing equipment |
| CN112084963A (en) * | 2020-09-11 | 2020-12-15 | 中德（珠海）人工智能研究院有限公司 | Monitoring early warning method, system and storage medium |
| CN112084963B (en) * | 2020-09-11 | 2022-11-01 | 中德（珠海）人工智能研究院有限公司 | Monitoring early warning method, system and storage medium |
| CN112272288A (en) * | 2020-10-23 | 2021-01-26 | 岭东核电有限公司 | Nuclear power operation monitoring method and device, computer equipment and storage medium |
| CN112364715A (en) * | 2020-10-23 | 2021-02-12 | 岭东核电有限公司 | Nuclear power operation abnormity monitoring method and device, computer equipment and storage medium |
| CN112364715B (en) * | 2020-10-23 | 2024-05-24 | 岭东核电有限公司 | Nuclear power operation abnormity monitoring method and device, computer equipment and storage medium |
| CN112200170A (en) * | 2020-12-07 | 2021-01-08 | 北京沃东天骏信息技术有限公司 | Image recognition method and device, electronic equipment and computer readable medium |
| CN114093117A (en) * | 2021-10-11 | 2022-02-25 | 北京精英系统科技有限公司 | Fire control management and control method and device |
| CN114093117B (en) * | 2021-10-11 | 2023-07-25 | 北京精英系统科技有限公司 | Fire control management and control method and device thereof |
| CN114002751A (en) * | 2021-10-26 | 2022-02-01 | 国网北京市电力公司 | Abnormal position identification method, system and device |
| CN114002751B (en) * | 2021-10-26 | 2023-11-03 | 国网北京市电力公司 | Abnormal position identification method, system and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111062281A (en) | Abnormal event monitoring method and device, storage medium and electronic equipment | |
| US10854062B2 (en) | Fire monitoring system | |
| US20220292944A9 (en) | Fire monitoring system | |
| CN111339997B (en) | Fire point area determination method and device, storage medium and electronic device | |
| Krstinić et al. | Histogram-based smoke segmentation in forest fire detection system | |
| US7804522B2 (en) | Image analysis for smoke detection | |
| CN111898581B (en) | Animal detection method, apparatus, electronic device, and readable storage medium | |
| Lestari et al. | Fire hotspots detection system on CCTV videos using you only look once (YOLO) method and tiny YOLO model for high buildings evacuation | |
| JP2019079445A (en) | Fire monitoring system | |
| JP2019079446A (en) | Fire monitoring system | |
| CN110348390B (en) | Training method, computer readable medium and system for flame detection model | |
| CN111489342A (en) | Video-based flame detection method and system and readable storage medium | |
| KR20210097500A (en) | Apparatus for Monitoring Fire And System having the same | |
| CN110519560B (en) | Intelligent early warning method, device and system | |
| CN114743341A (en) | Fire-fighting detection alarm method and device based on edge calculation | |
| KR102233679B1 (en) | Apparatus and method for detecting invader and fire for energy storage system | |
| CN110428579B (en) | Indoor monitoring system, method and device based on image recognition | |
| CN116543241A (en) | Detection method and device for leakage gas cloud, storage medium and electronic equipment | |
| CN111753587A (en) | Method and device for detecting falling to ground | |
| CN113361469A (en) | Method, device and equipment for identifying coverage state and storage medium | |
| CN115170894B (en) | Method and device for detecting smoke and fire | |
| CN111582183A (en) | Mask identification method and system in public place | |
| CN112580549A (en) | Fire point detection method and device and storage medium | |
| CN112347874A (en) | Fire detection method, device, equipment and storage medium | |
| CN116110045A (en) | Image labeling method and system |
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 |