CN114299067A - Underground coal wall caving early warning method and device - Google Patents
Underground coal wall caving early warning method and device Download PDFInfo
- Publication number
- CN114299067A CN114299067A CN202210205539.5A CN202210205539A CN114299067A CN 114299067 A CN114299067 A CN 114299067A CN 202210205539 A CN202210205539 A CN 202210205539A CN 114299067 A CN114299067 A CN 114299067A
- Authority
- CN
- China
- Prior art keywords
- picture
- identified
- frame
- wall caving
- coal wall
- 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
Images
Abstract
The invention provides an underground coal wall caving early warning method and device, wherein the method comprises the following steps: acquiring video data of a target coal mining area; converting video data of a target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified; obtaining candidate area information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model; obtaining each picture of the area to be identified according to the candidate area information on each frame of the picture to be identified and the screening threshold; obtaining statement description corresponding to each to-be-identified region picture according to each to-be-identified region picture and the image understanding model; and if the occurrence scene of the coal wall caving is obtained through statement description analysis corresponding to each picture of the area to be identified, performing coal wall caving early warning. The device is used for executing the method. The method and the device for early warning the caving wall caving of the underground coal wall provided by the embodiment of the invention improve the production safety of a coal mine.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an underground coal wall caving early warning method and device.
Background
At present, in coal mining, the coal wall caving is one of common hazards of a fully mechanized mining face, and due to the increase of the mining height of the working face, the coal wall caving is possibly caused, so that potential safety hazards are brought to the coal mining.
In order to evaluate the coal wall caving risk, a coal wall caving characteristic analysis method based on a background difference method appears in the prior art, and characteristic analysis of the coal wall caving is realized by taking caving time, area, height range and caving core height as research characteristics. However, due to the complexity and unpredictability of the scene, and the presence of various environmental disturbances and noises, such as sudden changes in illumination, fluctuations of some objects in the actual background image, camera shake, and the influence of moving objects entering and exiting the scene on the original scene, the modeling and simulation of the coal wall caving feature analysis method based on the background difference method on the background is difficult. In order to identify the coal wall caving, a machine vision-based fully mechanized coal mining face coal wall caving identification technology appears in the prior art, and a mixed image enhancement algorithm based on bilateral filtering and a Single Scale Retinex (SSR) algorithm is provided to realize the enhancement of the monitoring image of the fully mechanized coal mining face. However, the pre-processing image enhancement algorithm is only limited to denoising, and this method is not accurate enough for complex downhole scenarios.
Therefore, how to provide an underground coal wall caving early warning method to accurately early warn the occurrence of coal wall caving, reduce the accident rate and improve the safety production level of a coal mine becomes an important subject to be solved in the field.
Disclosure of Invention
Aiming at the problems in the prior art, the embodiment of the invention provides an underground coal wall caving early warning method and device, which can at least partially solve the problems in the prior art.
On one hand, the invention provides an underground coal wall caving early warning method, which comprises the following steps:
acquiring video data of a target coal mining area;
converting the video data of the target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified;
obtaining candidate region information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, wherein the candidate region information comprises region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training;
obtaining each picture of the area to be identified according to the candidate area information on each frame of the picture to be identified and the screening threshold;
obtaining statement description corresponding to each to-be-identified region picture according to each to-be-identified region picture and the image understanding model; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training;
and if the occurrence scene of the coal wall caving is obtained through statement description analysis corresponding to each picture of the area to be identified, performing coal wall caving early warning.
In another aspect, the present invention provides an underground coal wall caving early warning device, including:
the acquisition module is used for acquiring video data of a target coal mining area;
the preprocessing module is used for converting the video data of the target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified;
the first obtaining module is used for obtaining candidate region information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, wherein the candidate region information comprises region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training;
the second obtaining module is used for obtaining each picture of the area to be identified according to the candidate area information on each frame of picture to be identified and the screening threshold;
the semantic analysis module is used for obtaining statement description corresponding to each to-be-identified region picture according to each to-be-identified region picture and the image understanding model; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training;
and the early warning module is used for carrying out coal wall caving early warning after obtaining a coal wall caving occurrence scene according to statement description analysis corresponding to each to-be-identified area picture.
In another aspect, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the method for warning a caving wall caving in a coal wall in a well according to any of the embodiments described above.
In yet another aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method for warning the caving wall caving in a well.
The method and the device for underground coal wall caving early warning provided by the embodiment of the invention acquire video data of a target coal mining area, convert the video data of the target coal mining area into each frame of picture, perform image enhancement processing on each frame of picture to obtain a corresponding picture to be identified, acquire candidate area information on each frame of picture to be identified according to each frame of picture to be identified and a coal wall caving scene identification model, acquire each picture of the area to be identified according to the candidate area information on each frame of picture to be identified and a screening threshold value, acquire statement description corresponding to each picture of the area to be identified according to each picture of the area to be identified and an image understanding model, perform coal wall caving early warning after acquiring a coal wall caving scene according to the statement description corresponding to each picture of the area to be identified, and can more accurately judge the occurrence of coal wall caving, the coal mine production safety is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a downhole coal wall caving early warning method according to a first embodiment of the present invention.
Fig. 2 is a schematic flow chart of a downhole coal wall caving early warning method according to a second embodiment of the present invention.
Fig. 3 is a schematic flow chart of a downhole coal wall caving early warning method according to a third embodiment of the present invention.
Fig. 4 is a schematic flow chart of a downhole coal wall caving early warning method according to a fourth embodiment of the present invention.
Fig. 5 is a schematic flow chart of a downhole coal wall caving early warning method according to a fourth embodiment of the present invention.
Fig. 6 is a schematic structural diagram of a downhole coal wall caving early warning device according to a sixth embodiment of the present invention.
Fig. 7 is a schematic structural diagram of a downhole coal wall caving early warning device according to a seventh embodiment of the present invention.
Fig. 8 is a schematic structural diagram of an underground coal wall caving early warning device according to an eighth embodiment of the present invention.
Fig. 9 is a schematic structural diagram of a downhole coal wall caving early warning device according to a ninth embodiment of the present invention.
Fig. 10 is a schematic structural diagram of a downhole coal wall caving early warning device according to a tenth embodiment of the present invention.
Fig. 11 is a schematic physical structure diagram of an electronic device according to an eleventh embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention. It should be noted that the embodiments and features of the embodiments in the present application may be arbitrarily combined with each other without conflict.
The following describes an implementation process of the underground coal wall caving early warning method provided by the embodiment of the invention by taking a server as an execution subject.
Fig. 1 is a schematic flow chart of a downhole coal wall caving early warning method provided in a first embodiment of the present invention, and as shown in fig. 1, the downhole coal wall caving early warning method provided in the embodiment of the present invention includes:
s101, acquiring video data of a target coal mining area;
specifically, when coal mining is performed underground through coal mining equipment, a camera is installed to acquire video data of a target coal mining area, the camera sends the video data of the target coal mining area to a server, and the server receives the video data of the target coal mining area. Wherein the target coal mining area is a working face for coal mining operation.
For example, the camera may be mounted to a coal mine hydraulic mount.
S102, converting the video data of the target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified;
specifically, the server may perform frame cutting on the video data of the target coal mining area to obtain each frame of picture. And then, carrying out image enhancement processing on each frame of picture to improve the definition of each frame of picture and obtain the picture to be identified corresponding to each frame of picture.
For example, the server may cut frames of video data of a target coal mining area through an FFmpeg video processing tool to obtain each frame of picture.
S103, obtaining candidate region information of each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, wherein the candidate region information comprises region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training;
specifically, the server inputs each frame of picture to be identified into the coal wall caving scene identification model, divides each frame of picture to be identified into each candidate region through the coal wall caving scene identification model, and obtains candidate region information on each frame of picture to be identified, wherein the candidate region information comprises region coordinates and probability values of each candidate region. The region coordinates of the candidate region are pixel position coordinates of the candidate region on the to-be-identified picture, and the probability value of the candidate region is the probability representing that the candidate region is a coal wall caving scene. The coal wall caving scene recognition model is obtained based on historical coal mining picture sets and corresponding coal wall caving scene labels through training, the historical coal mining picture sets can be obtained through mine working face monitoring video data, and the coal wall caving scene labels corresponding to each coal mining picture included in the historical coal mining picture sets are manually marked.
S104, obtaining each to-be-identified area picture according to the candidate area information on each to-be-identified picture and the screening threshold;
specifically, the server compares the probability value of each candidate region on each frame of the to-be-identified picture with a screening threshold, if the probability value of the candidate region is greater than the screening threshold, it indicates that the probability of the occurrence of the coal wall caving in the candidate region is greater, and the picture corresponding to the candidate region can be intercepted from the corresponding to-be-identified picture according to the region coordinate of the candidate region, and is used as the to-be-identified region picture. The screening threshold is set according to actual needs, and the embodiment of the invention is not limited.
S105, obtaining statement description corresponding to each to-be-identified area picture according to each to-be-identified area picture and the image understanding model; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training;
specifically, the server inputs each picture of the area to be recognized into an image understanding model, and the sentence description corresponding to each picture of the area to be recognized can be output through the processing of the image understanding model. The image understanding model is obtained based on a historical region picture set and corresponding semantic tags in training, the historical region picture set can be obtained by screening from mine working face monitoring video data, the semantic tags are set artificially, the semantic tags are used for describing scenes of the historical region pictures by natural language and can be used as statement descriptions corresponding to the to-be-identified region pictures.
For example, the statement is described as: the side protection plate is abnormal in support, the coal wall is collapsed, and the rib is formed; the side protection plate is abnormal in support, the coal wall is not collapsed, and the side stripping can occur; the side protection plate is supported in place, the coal wall is hollow, and the rib spalling can occur. The method can relate to the supporting state of a side protection plate, the concrete condition of the coal wall, whether the edge breaking occurs or not, and the like. The statement description is set according to actual needs, and the embodiment of the invention is not limited.
And S106, if the occurrence scene of the coal wall caving is obtained according to the statement description analysis corresponding to each picture of the area to be identified, performing coal wall caving early warning.
Specifically, the server analyzes the statement description corresponding to each to-be-identified area picture to obtain a scene analysis statement corresponding to each to-be-identified area picture, and if the scene analysis statement corresponding to a certain to-be-identified area picture indicates that a scene in the to-be-identified area picture is a coal wall caving occurrence scene, coal wall caving early warning information is sent to perform coal wall caving early warning.
For example, semantic matching is performed on the scene analysis sentences corresponding to the to-be-identified area pictures and the possible scene sentences which occur to the preset coal wall caving, if the probability of the same semantic is greater than the preset value, it is determined that the scene analysis sentences corresponding to the to-be-identified area pictures are matched with the scene sentences which occur to the preset coal wall caving, and it is considered that the caving is possible to occur and early warning is given. And if the probability value is smaller than the preset value, confirming that the scene analysis sentences corresponding to the pictures of the area to be identified are not matched with the preset scene sentences of the coal wall caving, and considering that the caving cannot occur. The preset value is set according to actual needs, and the embodiment of the invention is not limited. The possible scene sentences of the coal wall caving are set according to actual needs, and the embodiment of the invention is not limited.
The underground coal wall caving early warning method provided by the embodiment of the invention obtains the video data of the target coal mining area, converts the video data of the target coal mining area into each frame of picture, performs image enhancement processing on each frame of picture to obtain the corresponding picture to be identified, obtaining candidate area information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, obtaining each picture of the area to be identified according to the candidate area information on each picture of the area to be identified and the screening threshold value, obtaining sentence description corresponding to each region picture to be identified according to each region picture to be identified and the image understanding model, after the occurrence scenes of the coal wall caving are obtained according to the statement description analysis corresponding to each picture of the area to be identified, the coal wall caving early warning is carried out, the occurrence of the coal wall caving can be accurately judged, and the coal mine production safety is improved.
Fig. 2 is a schematic flow chart of an underground coal wall caving early warning method provided in a second embodiment of the present invention, and as shown in fig. 2, further, on the basis of the foregoing embodiments, obtaining a coal wall caving scene recognition model based on a historical coal mining picture set and corresponding coal wall caving scene label training includes:
s201, obtaining a historical coal mining picture set, wherein each historical coal mining picture in the historical coal mining picture set is divided into a preset number of training picture areas, and each training picture area is provided with a corresponding coal wall caving scene label;
specifically, mine working face monitoring video data can be collected and framed to obtain a series of frame pictures, each frame picture is divided into a preset number of training picture areas, and a corresponding coal wall caving scene label is marked for each training picture area. The coal wall caving scene labels are a coal wall caving scene and a coal wall caving scene without the coal wall caving scene. The preset number is set according to actual needs, and the embodiment of the invention is not limited.
S202, training an initial model based on each training picture area of each historical coal mining picture in the historical coal mining picture set and each corresponding coal wall caving scene label to obtain the coal wall caving scene recognition model.
Specifically, the server may divide the historical coal mining picture set into a training set and a verification set, train the initial model through each training picture region of each historical coal mining picture in the training set and a corresponding coal wall caving scene tag to obtain a to-be-verified coal wall caving scene recognition model, perform model verification on the to-be-verified coal wall caving scene recognition model through the verification set, and use the to-be-verified coal wall caving scene recognition model as the coal wall caving scene recognition model. The initial model may adopt a deep learning model, such as a model established based on a deep ctr algorithm.
Fig. 3 is a schematic flow chart of an underground coal wall caving early warning method according to a third embodiment of the present invention, and as shown in fig. 3, on the basis of the foregoing embodiments, further performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified includes:
s301, acquiring dark channels of all pixel points on each frame of picture;
specifically, the server acquires a dark channel of each pixel point on each frame of picture, wherein the dark channel is the minimum value of the color channels of the pixel points.
For example, for picture I (x), there are three RGB color channels, that is, each pixel point on picture I (x) can pass through (I)R，IG，IB) Is shown by taking IR，IG，IBAnd the minimum value of the three values is used as a dark channel of the corresponding pixel point.
S302, obtaining a to-be-identified picture corresponding to each frame of picture according to the dark channel, the atmospheric light intensity value, the transmissivity and the dark channel defogging model of each pixel point on each frame of picture; wherein the dark channel defogging model is preset.
Specifically, for a frame of picture, the dark channel, the atmospheric light intensity value and the transmittance of each pixel point on the picture are substituted into the dark channel defogging model, so that the corresponding picture to be identified can be obtained. Wherein the dark channel defogging model is preset. Due to poor underground light and poor definition of video data, the definition of the picture can be obviously improved through the processing of the steps S301 and S302. The atmospheric light intensity value and the transmittance are set according to actual needs, and the embodiment of the present invention is not limited.
For example, the dark channel defogging model is
Wherein, in the step (A),
which represents the value of the light intensity of the atmosphere,
which represents the picture to be recognized and which,
the light-emitting efficiency is shown in the figure,
which represents the acquisition of one frame of picture,
the constant value can be 0.3-0.5.
On the basis of the foregoing embodiments, further before obtaining candidate region information on each frame of the picture to be identified according to each frame of the picture to be identified and the coal wall caving scene identification model, the method further includes:
and carrying out region filtering on each frame of picture to be identified.
Specifically, in order to suppress noise of the image, each frame of the image to be recognized may be subjected to region filtering, and the image to be recognized is subjected to region rate wave and then input to the coal wall caving scene recognition model for processing. And carrying out regional filtering on each frame of picture to be identified, namely recalculating the value of each pixel point of the picture to be identified, and obtaining the value of the pixel point by weighted averaging the value of the pixel point and the values of other pixel points in the neighborhood.
Fig. 4 is a schematic flow chart of a downhole coal wall caving early warning method provided in a fourth embodiment of the present invention, and as shown in fig. 4, on the basis of the foregoing embodiments, further training based on a historical region picture set and a corresponding semantic tag to obtain an image understanding model includes:
s401, acquiring a history area picture set, wherein each history area picture included in the history area picture set corresponds to a semantic tag;
specifically, mine working face monitoring video data can be collected and framed to obtain a series of frame pictures, each frame picture is divided into a preset number of picture areas, the picture areas are obtained to serve as historical area pictures, and corresponding semantic tags are manually marked for the historical area pictures. The semantic tags describe scenes of the historical region pictures by using natural language, and the historical region pictures included in the historical region picture set comprise the historical region pictures with the coal wall caving scenes. The semantic tags are set according to actual needs, and the embodiment of the invention is not limited.
For example, the preset number is 9, for each framed picture, the picture may be divided into 9 picture regions on average, and the semantic labels corresponding to the picture regions are obtained by performing semantic labeling on the 9 picture regions respectively.
S402, training the original model according to each historical region picture and the corresponding semantic label to obtain the image understanding model.
Specifically, the server may divide the historical coal mining picture set into a training set and a verification set, train the original model through each historical region picture in the training set and the corresponding semantic tag to obtain an image understanding model to be verified, perform model verification on the image understanding model to be verified through the verification set, and use the image understanding model to be verified that is verified as the image understanding model.
According to the underground coal wall caving early warning method provided by the embodiment of the invention, in the image preprocessing stage, the dedusting fog and definition enhancement algorithm is adopted to realize the quality enhancement of each frame of image of the video stream; through the video content understanding technology in the deep learning, the process of the rib caving can be accurately judged. The process of the underground coal wall caving early warning will be described with reference to fig. 5.
Firstly, video data is obtained in real time. When underground coal mining is carried out, the camera acquires video data of a working face of coal mining operation and sends the video data to the server in real time. The transmission of the video data may adopt an RTSP protocol.
And secondly, acquiring a picture to be identified. The server decodes the video data, and frames the video data through the FFmpeg video processing tool to obtain an image sequence which is a series of frame pictures.
And thirdly, performing up-sampling. The server upsamples the image training.
And fourthly, performing image enhancement processing. The server performs image enhancement processing on each frame of picture based on a computer vision dark channel defogging algorithm so as to improve the definition of each frame of picture.
And fifthly, performing regional filtering. The server performs region filtering on each frame of picture.
And sixthly, identifying the coal wall caving scene. And the server inputs each frame of image to be identified into the coal wall caving scene identification model to obtain the candidate area information on each frame of image to be identified. And then comparing the probability value of each candidate region on each frame of picture to be identified with a screening threshold, if the probability value of the candidate region is greater than the screening threshold, indicating that the probability of the coal wall caving of the candidate region is higher, and intercepting the picture corresponding to the candidate region from the corresponding picture to be identified according to the region coordinates of the candidate region to serve as the picture of the region to be identified.
And seventhly, analyzing semantic meaning. And the server inputs each picture of the area to be identified into the image understanding model, and the statement description corresponding to each picture of the area to be identified can be output through the processing of the image understanding model.
And eighthly, judging the coal wall caving. And the server analyzes the statement description corresponding to each to-be-identified area picture to obtain a scene analysis statement corresponding to each to-be-identified area picture, and if the scene analysis statement corresponding to the to-be-identified area picture indicates that the scene in the to-be-identified area picture is a coal wall caving occurrence scene, the ninth step is carried out. And if the scene analysis statement corresponding to the to-be-identified area picture indicates that the scene in the to-be-identified area picture is not the coal wall caving scene, the ninth step is not carried out. And entering the tenth step for storing the judgment result of the coal wall caving no matter whether the scene of the coal wall caving occurrence is identified.
And ninthly, carrying out coal wall caving early warning. The server can perform coal wall caving early warning by means of sending early warning information, sound and light warning and the like.
And step ten, storing the result. And storing the coal wall caving judgment result obtained in the eighth step, and sending the coal wall caving judgment result to a display screen for displaying.
According to the invention, by the underground coal wall caving early warning method, real-time accurate early warning is provided for coal wall caving detection in the fully mechanized mining face, the working safety of coal miners can be well ensured, and meanwhile, important contribution is made to reducing safety accidents.
According to the underground coal wall caving early warning method provided by the embodiment of the invention, the identification of the coal wall caving is carried out by adopting a video content understanding and identifying algorithm instead of adopting a motion detection mode, so that the coal wall caving scene problem can be more accurately judged. The defogging algorithm of the image preprocessing is fused, the image up-sampling can be better carried out, the false recognition caused by the image blurring is avoided, and the stability of the coal wall caving recognition is improved. The defogging algorithm and the video content understanding and recognizing algorithm are fused to increase the accuracy of caving recognition and reduce the false alarm rate of small batch coal caving alarm.
Fig. 6 is a schematic structural diagram of an underground coal wall caving early warning device provided in a sixth embodiment of the present invention, and as shown in fig. 6, the underground coal wall caving early warning device provided in the embodiment of the present invention includes an obtaining module 601, a preprocessing module 602, a first obtaining module 603, a second obtaining module 604, a semantic parsing module 605 and an early warning module 606, where:
the obtaining module 601 is configured to obtain video data of a target coal mining area; the preprocessing module 602 is configured to convert video data of the target coal mining area into each frame of picture, and perform image enhancement processing on each frame of picture to obtain a corresponding picture to be identified; the first obtaining module 603 is configured to obtain candidate region information on each frame of the picture to be identified according to each frame of the picture to be identified and the coal wall caving scene identification model, where the candidate region information includes region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training; the second obtaining module 604 is configured to obtain each to-be-identified region picture according to the candidate region information on each to-be-identified picture and the screening threshold; the semantic analysis module 605 is configured to obtain, according to each to-be-identified region picture and the image understanding model, a statement description corresponding to each to-be-identified region picture; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training; the early warning module 606 is configured to perform early warning on the coal wall caving after obtaining a scene of occurrence of the coal wall caving according to the statement description analysis corresponding to each image of the to-be-identified region.
Specifically, when coal mining is performed underground through coal mining equipment, a camera is installed to acquire video data of a target coal mining area, the camera sends the video data of the target coal mining area to the acquisition module 601, and the acquisition module 601 receives the video data of the target coal mining area. Wherein the target coal mining area is a working face for coal mining operation.
The preprocessing module 602 will cut frames of the video data of the target coal mining area to obtain each frame of picture. And then, carrying out image enhancement processing on each frame of picture to improve the definition of each frame of picture and obtain the picture to be identified corresponding to each frame of picture.
The first obtaining module 603 inputs each frame of picture to be identified into the coal wall caving scene identification model, divides each frame of picture to be identified into each candidate region through the coal wall caving scene identification model, and obtains candidate region information on each frame of picture to be identified, wherein the candidate region information includes region coordinates and probability values of each candidate region. The region coordinates of the candidate region are pixel position coordinates of the candidate region on the to-be-identified picture, and the probability value of the candidate region is the probability representing that the candidate region is a coal wall caving scene. The coal wall caving scene recognition model is obtained based on historical coal mining picture sets and corresponding coal wall caving scene labels through training, the historical coal mining picture sets can be obtained through mine working face monitoring video data, and the coal wall caving scene labels corresponding to each coal mining picture included in the historical coal mining picture sets are manually marked.
The second obtaining module 604 compares the probability value of each candidate region on each frame of the to-be-identified picture with a screening threshold, if the probability value of the candidate region is greater than the screening threshold, it indicates that the probability of the occurrence of the coal wall caving in the candidate region is greater, and according to the region coordinates of the candidate region, a picture corresponding to the candidate region can be intercepted from the corresponding to-be-identified picture as a to-be-identified region picture. The screening threshold is set according to actual needs, and the embodiment of the invention is not limited.
The semantic parsing module 605 inputs each to-be-identified region picture into the image understanding model, and may output a sentence description corresponding to each to-be-identified region picture through the processing of the image understanding model. The image understanding model is obtained based on a historical region picture set and corresponding semantic tags in training, the historical region picture set can be obtained by screening from mine working face monitoring video data, the semantic tags are set artificially, the semantic tags are used for describing scenes of the historical region pictures by natural language and can be used as statement descriptions corresponding to the to-be-identified region pictures.
The early warning module 606 analyzes the statement description corresponding to each picture of the area to be identified to obtain a scene analysis statement corresponding to each picture of the area to be identified, and if the scene analysis statement corresponding to a certain picture of the area to be identified indicates that a scene in the picture of the area to be identified is a scene of a coal wall caving occurrence, then coal wall caving early warning information is sent to perform coal wall caving early warning.
The underground coal wall caving early warning device provided by the embodiment of the invention obtains the video data of the target coal mining area, converts the video data of the target coal mining area into each frame of picture, performs image enhancement processing on each frame of picture to obtain the corresponding picture to be identified, obtaining candidate area information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, obtaining each picture of the area to be identified according to the candidate area information on each picture of the area to be identified and the screening threshold value, obtaining sentence description corresponding to each region picture to be identified according to each region picture to be identified and the image understanding model, after the occurrence scenes of the coal wall caving are obtained according to the statement description analysis corresponding to each picture of the area to be identified, the coal wall caving early warning is carried out, the occurrence of the coal wall caving can be accurately judged, and the coal mine production safety is improved.
Fig. 7 is a schematic structural diagram of a downhole coal wall caving early warning device provided in a seventh embodiment of the present invention, and as shown in fig. 7, on the basis of the foregoing embodiments, further, the downhole coal wall caving early warning device provided in the embodiment of the present invention further includes a first obtaining module 607 and a first training module 608, where:
the first obtaining module 607 is configured to obtain a historical coal mining picture set, where each historical coal mining picture in the historical coal mining picture set is divided into a preset number of training picture regions, and each training region has a corresponding coal wall caving scene label; the first training module 608 is configured to train an initial model based on each training picture region of each historical coal mining picture in the historical coal mining picture set and a corresponding coal wall caving scene tag, so as to obtain the coal wall caving scene recognition model.
Fig. 8 is a schematic structural diagram of a downhole coal wall caving early warning device according to an eighth embodiment of the present invention, as shown in fig. 8, on the basis of the foregoing embodiments, further, the preprocessing module 602 includes a first obtaining unit 6021 and a second obtaining unit 6022, where:
the first obtaining unit 6021 is configured to obtain a dark channel of each pixel point on each frame of picture; the second obtaining unit 6022 is configured to obtain a to-be-identified picture corresponding to each frame of picture according to the dark channel, the atmospheric light intensity value, the transmittance of each pixel point on each frame of picture and the dark channel defogging model; wherein the dark channel defogging model is preset.
Fig. 9 is a schematic structural diagram of the downhole coal wall caving early warning device provided in the ninth embodiment of the present invention, and as shown in fig. 9, on the basis of the foregoing embodiments, further, the downhole coal wall caving early warning device provided in the embodiment of the present invention further includes a region filtering module 609, where:
the region filtering module 609 is configured to perform region filtering on each frame of the picture to be identified.
Fig. 10 is a schematic structural diagram of a downhole coal wall caving early warning device provided in a tenth embodiment of the present invention, and as shown in fig. 10, on the basis of the foregoing embodiments, further, the downhole coal wall caving early warning device provided in the embodiment of the present invention further includes a second obtaining module 610 and a second training module 611, where:
the second obtaining module 610 is configured to obtain a history region picture set, where each history region picture corresponds to a semantic tag; the second training module 611 is configured to train the original model according to each historical region picture and the corresponding semantic tag, so as to obtain the image understanding model.
The embodiment of the server provided in the embodiment of the present invention may be specifically configured to execute the processing flows of the above method embodiments, and the functions of the embodiment are not described herein again, and refer to the detailed description of the above method embodiments.
Fig. 11 is a schematic physical structure diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 11, the electronic device may include: a processor (processor)1101, a communication Interface (Communications Interface)1102, a memory (memory)1103 and a communication bus 1104, wherein the processor 1101, the communication Interface 1102 and the memory 1103 are communicated with each other via the communication bus 1104. The processor 1101 may call logic instructions in the memory 1103 to perform the following method: acquiring video data of a target coal mining area; converting the video data of the target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified; obtaining candidate region information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, wherein the candidate region information comprises region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training; obtaining each picture of the area to be identified according to the candidate area information on each frame of the picture to be identified and the screening threshold; obtaining statement description corresponding to each to-be-identified region picture according to each to-be-identified region picture and the image understanding model; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training; and if the occurrence scene of the coal wall caving is obtained through statement description analysis corresponding to each picture of the area to be identified, performing coal wall caving early warning.
In addition, the logic instructions in the memory 1103 can be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The present embodiment discloses a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method provided by the above-mentioned method embodiments, for example, comprising: acquiring video data of a target coal mining area; converting the video data of the target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified; obtaining candidate region information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, wherein the candidate region information comprises region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training; obtaining each picture of the area to be identified according to the candidate area information on each frame of the picture to be identified and the screening threshold; obtaining statement description corresponding to each to-be-identified region picture according to each to-be-identified region picture and the image understanding model; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training; and if the occurrence scene of the coal wall caving is obtained through statement description analysis corresponding to each picture of the area to be identified, performing coal wall caving early warning.
The present embodiment provides a computer-readable storage medium, which stores a computer program, where the computer program causes the computer to execute the method provided by the above method embodiments, for example, the method includes: acquiring video data of a target coal mining area; converting the video data of the target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified; obtaining candidate region information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, wherein the candidate region information comprises region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training; obtaining each picture of the area to be identified according to the candidate area information on each frame of the picture to be identified and the screening threshold; obtaining statement description corresponding to each to-be-identified region picture according to each to-be-identified region picture and the image understanding model; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training; and if the occurrence scene of the coal wall caving is obtained through statement description analysis corresponding to each picture of the area to be identified, performing coal wall caving early warning.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In the description herein, reference to the description of the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," "an example," "a particular example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (12)
1. A coal wall caving early warning method in a well is characterized by comprising the following steps:
acquiring video data of a target coal mining area;
converting the video data of the target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified;
obtaining candidate region information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, wherein the candidate region information comprises region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training;
obtaining each picture of the area to be identified according to the candidate area information on each frame of the picture to be identified and the screening threshold;
obtaining statement description corresponding to each to-be-identified region picture according to each to-be-identified region picture and the image understanding model; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training;
and if the occurrence scene of the coal wall caving is obtained through statement description analysis corresponding to each picture of the area to be identified, performing coal wall caving early warning.
2. The method of claim 1, wherein obtaining a ledge scene recognition model based on historical coal mining picture sets and corresponding ledge scene label training comprises:
acquiring a historical coal mining picture set, wherein each historical coal mining picture in the historical coal mining picture set is divided into a preset number of training picture areas, and each training area is provided with a corresponding coal wall caving scene label;
training an initial model based on each training picture area of each historical coal mining picture in the historical coal mining picture set and the corresponding coal wall caving scene label to obtain the coal wall caving scene recognition model.
3. The method according to claim 1, wherein the image enhancement processing is performed on each frame of picture, and obtaining the corresponding picture to be identified comprises:
acquiring dark channels of all pixel points on each frame of picture;
obtaining a to-be-identified picture corresponding to each frame of picture according to the dark channel, the atmospheric light intensity value, the transmissivity and the dark channel defogging model of each pixel point on each frame of picture; wherein the dark channel defogging model is preset.
4. The method according to claim 1, before obtaining the candidate region information on each frame of the image to be recognized according to each frame of the image to be recognized and the coal wall caving scene recognition model, further comprising:
and carrying out region filtering on each frame of picture to be identified.
5. The method of any one of claims 1 to 4, wherein training to obtain an image understanding model based on the historical region picture set and the corresponding semantic tags comprises:
acquiring a historical region picture set, wherein each historical region picture included in the historical region picture set corresponds to a semantic tag;
and training the original model according to each historical region picture and the corresponding semantic label to obtain the image understanding model.
6. The utility model provides a coal cliff caving early warning device in pit which characterized in that includes:
the acquisition module is used for acquiring video data of a target coal mining area;
the preprocessing module is used for converting the video data of the target coal mining area into each frame of picture, and performing image enhancement processing on each frame of picture to obtain a corresponding picture to be identified;
the first obtaining module is used for obtaining candidate region information on each frame of picture to be identified according to each frame of picture to be identified and the coal wall caving scene identification model, wherein the candidate region information comprises region coordinates and probability values of each candidate region; the coal wall caving scene recognition model is obtained based on a historical coal mining picture set and corresponding coal wall caving scene label training;
the second obtaining module is used for obtaining each picture of the area to be identified according to the candidate area information on each frame of picture to be identified and the screening threshold;
the semantic analysis module is used for obtaining statement description corresponding to each to-be-identified region picture according to each to-be-identified region picture and the image understanding model; the image understanding model is obtained based on historical region picture sets and corresponding semantic label training;
and the early warning module is used for carrying out coal wall caving early warning after obtaining a coal wall caving occurrence scene according to statement description analysis corresponding to each to-be-identified area picture.
7. The apparatus of claim 6, further comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a historical coal mining picture set, each historical coal mining picture in the historical coal mining picture set is divided into a preset number of training picture areas, and each training area is provided with a corresponding coal wall caving scene label;
and the first training module is used for training an initial model based on each training picture area of each historical coal mining picture in the historical coal mining picture set and the corresponding coal wall caving scene label to obtain the coal wall caving scene recognition model.
8. The apparatus of claim 6, wherein the pre-processing module comprises:
the first acquisition unit is used for acquiring dark channels of all pixel points on each frame of picture;
the second acquisition unit is used for acquiring the to-be-identified picture corresponding to each frame of picture according to the dark channel, the atmospheric light intensity value, the transmissivity and the dark channel defogging model of each pixel point on each frame of picture; wherein the dark channel defogging model is preset.
9. The apparatus of claim 6, further comprising:
and the region filtering module is used for performing region filtering on each frame of picture to be identified.
10. The apparatus of any one of claims 6 to 9, further comprising:
the second acquisition module is used for acquiring a historical region picture set, and each historical region picture corresponds to one semantic tag;
and the second training module is used for training the original model according to each historical region picture and the corresponding semantic label to obtain the image understanding model.
11. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 5 when executing the computer program.
12. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210205539.5A CN114299067A (en) | 2022-03-04 | 2022-03-04 | Underground coal wall caving early warning method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210205539.5A CN114299067A (en) | 2022-03-04 | 2022-03-04 | Underground coal wall caving early warning method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114299067A true CN114299067A (en) | 2022-04-08 |
Family
ID=80978660
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210205539.5A Pending CN114299067A (en) | 2022-03-04 | 2022-03-04 | Underground coal wall caving early warning method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114299067A (en) |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102411774A (en) * | 2011-08-08 | 2012-04-11 | 安防科技（中国）有限公司 | Processing method, device and system based on single-image defogging |
| CN102880873A (en) * | 2012-08-31 | 2013-01-16 | 公安部第三研究所 | Personnel behavior identification implementation system and method based on image segmentation and semantic extraction |
| CN103248878A (en) * | 2013-05-23 | 2013-08-14 | 南车株洲电力机车有限公司 | Pattern recognition method, device and system of abnormal situation of fully mechanized coal mining face |
| CN104166991A (en) * | 2014-07-28 | 2014-11-26 | 西安电子科技大学 | Level set image segmentation method based on self-adaptive parameter |
| CN105809679A (en) * | 2016-03-04 | 2016-07-27 | 李云栋 | Mountain railway side slope rockfall detection method based on visual analysis |
| CN105976338A (en) * | 2016-05-12 | 2016-09-28 | 山东大学 | Dark channel prior defogging method based on sky recognition and segmentation |
| CN107454317A (en) * | 2017-07-27 | 2017-12-08 | 广东欧珀移动通信有限公司 | Image processing method, device, computer-readable recording medium and computer equipment |
| CN109934409A (en) * | 2019-03-18 | 2019-06-25 | 西安科技大学 | A kind of high working face coal wall working face wall caving prediction network and its prediction technique |
| CN110119701A (en) * | 2019-04-30 | 2019-08-13 | 东莞恒创智能科技有限公司 | The coal mine fully-mechanized mining working unsafe acts recognition methods of view-based access control model relationship detection |
| US20190332361A1 (en) * | 2018-04-26 | 2019-10-31 | Microsoft Technology Licensing, Llc | Developer and runtime environments supporting multi-input modalities |
| CN110675469A (en) * | 2019-09-06 | 2020-01-10 | 常州大学 | Image description method for detecting spatial relationship between targets in construction scene |
| CN110674772A (en) * | 2019-09-29 | 2020-01-10 | 国家电网有限公司技术学院分公司 | Intelligent safety control auxiliary system and method for electric power operation site |
| CN111325323A (en) * | 2020-02-19 | 2020-06-23 | 山东大学 | Power transmission and transformation scene description automatic generation method fusing global information and local information |
| CN111598800A (en) * | 2020-05-07 | 2020-08-28 | 辽宁师范大学 | Single image defogging method based on space domain homomorphic filtering and dark channel prior |
| CN111680535A (en) * | 2019-03-11 | 2020-09-18 | 维布络有限公司 | Method and system for real-time prediction of one or more potential threats in video surveillance |
| CN113942805A (en) * | 2021-12-22 | 2022-01-18 | 西安华创马科智能控制系统有限公司 | Coal block transportation monitoring method and device |
| CN114037831A (en) * | 2021-07-20 | 2022-02-11 | 星汉智能科技股份有限公司 | Image depth intensive description method, system and storage medium |
-
2022
- 2022-03-04 CN CN202210205539.5A patent/CN114299067A/en active Pending
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102411774A (en) * | 2011-08-08 | 2012-04-11 | 安防科技（中国）有限公司 | Processing method, device and system based on single-image defogging |
| CN102880873A (en) * | 2012-08-31 | 2013-01-16 | 公安部第三研究所 | Personnel behavior identification implementation system and method based on image segmentation and semantic extraction |
| CN103248878A (en) * | 2013-05-23 | 2013-08-14 | 南车株洲电力机车有限公司 | Pattern recognition method, device and system of abnormal situation of fully mechanized coal mining face |
| CN104166991A (en) * | 2014-07-28 | 2014-11-26 | 西安电子科技大学 | Level set image segmentation method based on self-adaptive parameter |
| CN105809679A (en) * | 2016-03-04 | 2016-07-27 | 李云栋 | Mountain railway side slope rockfall detection method based on visual analysis |
| CN105976338A (en) * | 2016-05-12 | 2016-09-28 | 山东大学 | Dark channel prior defogging method based on sky recognition and segmentation |
| CN107454317A (en) * | 2017-07-27 | 2017-12-08 | 广东欧珀移动通信有限公司 | Image processing method, device, computer-readable recording medium and computer equipment |
| US20190332361A1 (en) * | 2018-04-26 | 2019-10-31 | Microsoft Technology Licensing, Llc | Developer and runtime environments supporting multi-input modalities |
| CN111680535A (en) * | 2019-03-11 | 2020-09-18 | 维布络有限公司 | Method and system for real-time prediction of one or more potential threats in video surveillance |
| CN109934409A (en) * | 2019-03-18 | 2019-06-25 | 西安科技大学 | A kind of high working face coal wall working face wall caving prediction network and its prediction technique |
| CN110119701A (en) * | 2019-04-30 | 2019-08-13 | 东莞恒创智能科技有限公司 | The coal mine fully-mechanized mining working unsafe acts recognition methods of view-based access control model relationship detection |
| CN110675469A (en) * | 2019-09-06 | 2020-01-10 | 常州大学 | Image description method for detecting spatial relationship between targets in construction scene |
| CN110674772A (en) * | 2019-09-29 | 2020-01-10 | 国家电网有限公司技术学院分公司 | Intelligent safety control auxiliary system and method for electric power operation site |
| CN111325323A (en) * | 2020-02-19 | 2020-06-23 | 山东大学 | Power transmission and transformation scene description automatic generation method fusing global information and local information |
| CN111598800A (en) * | 2020-05-07 | 2020-08-28 | 辽宁师范大学 | Single image defogging method based on space domain homomorphic filtering and dark channel prior |
| CN114037831A (en) * | 2021-07-20 | 2022-02-11 | 星汉智能科技股份有限公司 | Image depth intensive description method, system and storage medium |
| CN113942805A (en) * | 2021-12-22 | 2022-01-18 | 西安华创马科智能控制系统有限公司 | Coal block transportation monitoring method and device |
Non-Patent Citations (4)
| Title |
|---|
| WEI CHEN ET AL: "Face Location in Dim-Lightening Coal Mine Surveillance Images", 《2009 2ND INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING》 * |
| 徐荣鑫: "综采工作面煤壁片帮识别关键技术研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
| 栗科峰著: "《人脸图像处理与识别技术》", 30 September 2018 * |
| 郑树泉等: "《工业智能技术与应用》", 31 January 2019 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107862270B (en) | Face classifier training method, face detection method and device and electronic equipment | |
| CN110084172B (en) | Character recognition method and device and electronic equipment | |
| CN111222478A (en) | Construction site safety protection detection method and system | |
| CN113159091A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN111814637A (en) | Dangerous driving behavior recognition method and device, electronic equipment and storage medium | |
| CN115359471A (en) | Image processing and joint detection model training method, device, equipment and storage medium | |
| CN113326773A (en) | Recognition model training method, recognition method, device, equipment and storage medium | |
| CN113688837A (en) | Image desensitization method, device, electronic equipment and computer readable storage medium | |
| CN113705381A (en) | Target detection method and device in foggy days, electronic equipment and storage medium | |
| CN111126372B (en) | Logo region marking method and device in video and electronic equipment | |
| CN114299067A (en) | Underground coal wall caving early warning method and device | |
| CN111783613A (en) | Anomaly detection method, model training method, device, equipment and storage medium | |
| CN115631502A (en) | Character recognition method, character recognition device, model training method, electronic device and medium | |
| CN115376137A (en) | Optical character recognition processing and text recognition model training method and device | |
| CN113344121B (en) | Method for training a sign classification model and sign classification | |
| CN115375964A (en) | Improved method and device based on YOLOv5 model | |
| CN114998275A (en) | State recognition method of target object and training method of deep learning model | |
| CN113989720A (en) | Target detection method, training method, device, electronic equipment and storage medium | |
| CN114373162A (en) | Dangerous area personnel intrusion detection method and system for transformer substation video monitoring | |
| CN113887394A (en) | Image processing method, device, equipment and storage medium | |
| CN112699754A (en) | Signal lamp identification method, device, equipment and storage medium | |
| CN111539277A (en) | Detection method and system for construction machinery in power transmission line area | |
| CN109886253A (en) | A kind of building safety detection system based on video analysis | |
| KR101859672B1 (en) | Apparatus for extracting HALO coronal mass ejection and method thereof | |
| CN114627423A (en) | Object detection and model training method, device, equipment and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220408 |