CN112614175A - Injection parameter determination method for hole sealing agent injector based on characteristic decorrelation - Google Patents
Injection parameter determination method for hole sealing agent injector based on characteristic decorrelation Download PDFInfo
- Publication number
- CN112614175A CN112614175A CN202011519163.2A CN202011519163A CN112614175A CN 112614175 A CN112614175 A CN 112614175A CN 202011519163 A CN202011519163 A CN 202011519163A CN 112614175 A CN112614175 A CN 112614175A
- Authority
- CN
- China
- Prior art keywords
- vector
- feature
- gas
- decorrelation
- inner product
- 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
- 238000002347 injection Methods 0.000 title claims abstract description 95
- 239000007924 injection Substances 0.000 title claims abstract description 95
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000007789 sealing Methods 0.000 title abstract description 31
- 239000003795 chemical substances by application Substances 0.000 title abstract description 23
- 239000013598 vector Substances 0.000 claims abstract description 325
- 239000000565 sealant Substances 0.000 claims abstract description 72
- 239000011159 matrix material Substances 0.000 claims abstract description 56
- 238000013528 artificial neural network Methods 0.000 claims abstract description 37
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 17
- 230000008569 process Effects 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 12
- 238000005065 mining Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 5
- 239000000243 solution Substances 0.000 description 5
- 239000003245 coal Substances 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000005553 drilling Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
- G06T9/002—Image coding using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a method for determining injection parameters of a hole sealing agent injector based on feature decorrelation, which comprises the following steps: passing the image of the borehole to be sealed through a convolutional neural network to obtain a characteristic map; acquiring gas parameters obtained in gas extraction and converting the gas parameters into input vectors; passing the input vector through a deep neural network to obtain a gas feature vector; calculating an inner product matrix of the feature map relative to the gas feature vector; calculating an inner product vector of the gas feature vector relative to the feature map; subtracting the inner product matrix by pixel position according to the feature map to obtain a decorrelated feature map; subtracting the inner product vector from the gas feature vector according to element positions to obtain a decorrelation feature vector; passing the decorrelated feature map through a fully connected layer to obtain a shape vector; and cascading the shape vector and the decorrelation feature vector through an encoder to obtain injection parameters for the sealant injector.
Description
Technical Field
The present application relates to the field of artificial intelligence, and more particularly, to a method for determining injection parameters for a sealant injector based on feature decorrelation, a system for determining injection parameters for a sealant injector based on feature decorrelation, and an electronic device.
Background
The existing coal mine gas mining adopts a drilling extraction mode, so that drilling and hole sealing are needed in the mining process to ensure the safe mining of the gas. At present, when holes are sealed, a hole sealing agent injector is adopted to inject hole sealing agent into the holes to seal the holes.
However, the existing hole sealing agent injectors still have some defects when used for sealing drilled holes, firstly, the hole sealing agent injectors cannot be adjusted according to the shape, size and depth of the holes, and different hole sealing agent injectors are usually selected according to experience for different holes, so that the hole sealing effect is influenced. In addition, when the hole sealing agent is injected into the hole, the drilled hole sometimes needs to be directionally sealed, so that the hole sealing agent can be effectively directionally expanded in the hole, the using effect is improved, but no method for well determining the hole sealing orientation exists at present.
Accordingly, an optimized solution for injection parameter determination of a sealant injector for coal mine gas mining boreholes is desired.
At present, deep learning and neural networks have been widely applied in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks also exhibit a level close to or even exceeding that of humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.
In recent years, deep learning and development of neural networks provide new solutions and schemes for determination of injection parameters of sealant injectors.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides an injection parameter determination method for a sealing agent injector based on feature decorrelation, an injection parameter determination system for a sealing agent injector based on feature decorrelation and an electronic device, wherein a feature map of a borehole image is extracted through a deep neural network to capture the shape and size features of the borehole, a feature vector used for representing the depth and internal structure features of the borehole is extracted through the deep neural network based on the gas parameters flowing out of the borehole, the feature map and the feature vector are subjected to decorrelation processing, and the feature vector and the feature map after the decorrelation processing are coded to determine the injection parameters of the sealing agent injector.
According to one aspect of the present application, there is provided a method of injection parameter determination for a sealant injector based on feature decorrelation, comprising:
acquiring an image of a borehole to be sealed;
passing the borehole image through a convolutional neural network to obtain a feature map;
acquiring gas parameters obtained in the process of extracting gas from a drilled hole and converting the gas parameters into input vectors;
passing the input vector through a deep neural network to obtain a gas feature vector;
calculating an inner product matrix of the feature map relative to the gas feature vector, wherein the inner product matrix is used for representing the correlation between the feature map and the gas feature vector;
calculating an inner product vector of the gas feature vector relative to the feature map, wherein the inner product vector is used for representing the correlation between the gas feature vector and the feature map;
subtracting the inner product matrix by pixel position according to the feature map to obtain a decorrelated feature map;
subtracting the inner product vector from the gas feature vector according to element positions to obtain a decorrelation feature vector;
passing the decorrelated feature map through a fully connected layer to obtain a shape vector; and
and cascading the shape vector and the decorrelation characteristic vector and then passing the concatenated vector through an encoder to obtain injection parameters for the sealant injector.
In the injection parameter determination method for a sealant injector based on feature decorrelation, the passing the input vector through a deep neural network to obtain a gas feature vector includes: and performing one-dimensional convolution processing on the input vector to obtain the gas characteristic vector.
In the injection parameter determination method for a sealant injector based on feature decorrelation, in passing the input vector through a deep neural network to obtain a gas feature vector, the deep neural network is a multilayer perceptron.
In the injection parameter determination method for a sealant injector based on feature decorrelation, an inner product matrix of the feature map relative to the gas feature vector is calculated, and the inner product matrix is used for representing the correlation between the feature map and the gas feature vector, and includes: adjusting the gas feature vector to a dimension matching the feature map; and calculating the inner product matrix of the feature map relative to the gas feature vector adjusted to the matched dimension.
In the injection parameter determination method for a sealant injector based on feature decorrelation, calculating an inner product vector of the gas feature vector relative to the feature map, the inner product vector being used for representing a correlation between the gas feature vector and the feature map, includes: adjusting the feature map to a scale matching the gas feature vector; and calculating the inner product vector of the gas feature vector relative to the feature map adjusted to the matched scale.
In the method for determining injection parameters for a sealant injector based on feature decorrelation, the shape vector and the decorrelation feature vector are cascaded and then pass through an encoder to obtain the injection parameters for the sealant injector, and the method includes: and cascading the shape vector and the decorrelation characteristic vector and then passing through one or more full connection layers to obtain injection parameters for the sealant injector, wherein the number of output bits of the last full connection layer in the one or more full connection layers is equal to the number of the injection parameters of the sealant injector. .
According to another aspect of the present application, there is provided an injection parameter determination system for a sealant injector based on feature decorrelation, comprising:
the image acquisition unit is used for acquiring an image of a drill hole to be sealed;
the characteristic map generating unit is used for enabling the borehole image obtained by the image obtaining unit to pass through a convolutional neural network so as to obtain a characteristic map;
the gas parameter acquisition unit is used for acquiring gas parameters obtained in the process of extracting gas from a drilled hole and converting the gas parameters into input vectors;
the gas characteristic vector generating unit is used for enabling the input vector obtained by the gas parameter obtaining unit to pass through a deep neural network so as to obtain a gas characteristic vector;
an inner product matrix calculation unit configured to calculate an inner product matrix of the feature map obtained by the feature map generation unit with respect to the gas feature vector obtained by the gas feature vector generation unit, the inner product matrix being used to represent a correlation between the feature map and the gas feature vector;
an inner product vector calculation unit configured to calculate an inner product vector of the gas feature vector obtained by the gas feature vector generation unit with respect to the feature map obtained by the feature map generation unit, the inner product vector representing a correlation between the gas feature vector and the feature map;
a decorrelation feature map generating unit configured to subtract the inner product matrix obtained by the inner product matrix calculating unit from the feature map obtained by the feature map generating unit by pixel position to obtain a decorrelation feature map;
a decorrelation feature vector generation unit configured to subtract the inner product vector obtained by the inner product vector calculation unit from the gas feature vector obtained by the gas feature vector generation unit by an element position to obtain a decorrelation feature vector;
a shape vector generating unit, configured to pass the decorrelation feature map obtained by the decorrelation feature map generating unit through a full connection layer to obtain a shape vector; and
and the coding unit is used for cascading the shape vector obtained by the shape vector generation unit and the decorrelation characteristic vector obtained by the decorrelation characteristic vector generation unit and then passing the concatenated vector through a coder so as to obtain the injection parameters for the sealant injector.
In the injection parameter determination system for a sealant injector based on feature decorrelation, the gas feature vector generation unit is further configured to: and performing one-dimensional convolution processing on the input vector to obtain the gas characteristic vector.
In the injection parameter determination system for a sealant injector based on feature decorrelation described above, the deep neural network is a multilayer perceptron.
In the injection parameter determination system for a sealant injector based on feature decorrelation as described above, the inner product matrix calculation unit is further configured to: adjusting the gas feature vector to a dimension matching the feature map; and calculating the inner product matrix of the feature map relative to the gas feature vector adjusted to the matched dimension.
In the injection parameter determination system for a sealant injector based on feature decorrelation as described above, the inner product vector calculation unit is further configured to: adjusting the feature map to a scale matching the gas feature vector; and calculating the inner product vector of the gas feature vector relative to the feature map adjusted to the matched scale.
In the injection parameter determination system for a sealant injector based on feature decorrelation as described above, the encoding unit is further configured to: and cascading the shape vector and the decorrelation characteristic vector and then passing through one or more full connection layers to obtain injection parameters for the sealant injector, wherein the number of output bits of the last full connection layer in the one or more full connection layers is equal to the number of the injection parameters of the sealant injector.
According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory having stored therein computer program instructions which, when executed by the processor, cause the processor to perform the method of injection parameter determination for a sealant injector based on feature decorrelation as described above.
According to yet another aspect of the present application, there is provided a computer readable medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform a method for injection parameter determination for a sealer injector based on feature decorrelation as described above.
According to the injection parameter determination method for the sealing agent injector based on the characteristic decorrelation, the injection parameter determination system for the sealing agent injector based on the characteristic decorrelation and the electronic device, the characteristic graph of the borehole image is extracted through the deep neural network to capture the shape and size characteristics of the borehole, the characteristic vector used for representing the depth and the internal structure characteristics of the borehole is extracted through the deep neural network based on the gas parameters flowing out of the borehole, the characteristic graph and the characteristic vector are subjected to the decorrelation processing, and the characteristic vector and the characteristic graph after the decorrelation processing are coded to determine the injection parameters of the sealing agent injector.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing in more detail embodiments of the present application with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings, like reference numbers generally represent like parts or steps.
Fig. 1 illustrates a scenario diagram of an injection parameter determination method for a sealant injector based on feature decorrelation according to an embodiment of the present application.
Fig. 2 illustrates a flow chart of an injection parameter determination method for a sealer injector based on feature decorrelation according to an embodiment of the present application.
Fig. 3 illustrates an architectural schematic diagram of an injection parameter determination method for a sealant injector based on feature decorrelation according to an embodiment of the present application.
Fig. 4 illustrates a flow chart of calculating an inner product matrix of the feature map with respect to the gas feature vector in a method for determining injection parameters for a sealant injector based on feature decorrelation according to an embodiment of the present application.
Fig. 5 illustrates a flow chart for calculating an inner product vector of the gas feature vector with respect to the feature map in a method for determining injection parameters for a sealer injector based on feature decorrelation according to an embodiment of the present application.
Fig. 6 illustrates a block diagram of an injection parameter determination system for a sealant injector based on feature decorrelation according to an embodiment of the present application.
FIG. 7 illustrates a block diagram of an electronic device in accordance with an embodiment of the present application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and that the present application is not limited by the example embodiments described herein.
Overview of a scene
As mentioned above, the existing coal mine gas mining adopts a drilling extraction mode, so that drilling and hole sealing are needed in the mining process to ensure the safe mining of the gas. At present, when holes are sealed, a hole sealing agent injector is adopted to inject hole sealing agent into the holes to seal the holes.
However, the existing hole sealing agent injectors still have some defects when used for sealing drilled holes, firstly, the hole sealing agent injectors cannot be adjusted according to the shape, size and depth of the holes, and different hole sealing agent injectors are usually selected according to experience for different holes, so that the hole sealing effect is influenced. In addition, when the hole sealing agent is injected into the hole, the drilled hole sometimes needs to be directionally sealed, so that the hole sealing agent can be effectively directionally expanded in the hole, the using effect is improved, but no method for well determining the hole sealing orientation exists at present.
Accordingly, an optimized solution for injection parameter determination of a sealant injector for coal mine gas mining boreholes is desired.
Based on the above problems, the applicant of the present application considers determining injection parameters of a sealant injector, such as an injection aperture of the injector, an injection flow rate, an injection orientation, and the like, through a deep learning-based feature extraction and coding technique. In particular, the solution of the present application comprises two parts, considering that the features relating to the shape and size of the hole can be obtained on the basis of feature extraction on the image of the drilled hole, whereas the depth and internal structural features of the hole can be obtained by back-deriving the parameters of the gas flowing out in the hole.
Firstly, a characteristic map is obtained after the obtained borehole image passes through a convolutional neural network. Then, gas parameters obtained in the borehole extraction process, specifically, gas outflow velocities obtained at time intervals within a certain period of time are obtained, thereby constituting input vectors for use as input parameters. This can be characterized as an attribute of the depth and internal structure of the borehole, which is then passed through a depth neural network to obtain a feature vector.
For the feature map and the feature vector, although reflecting different characteristics of the borehole itself, there may still be some correlations that, when further parameter coding is performed by an encoder comprising fully connected layers, may cause an overfitting of the coded parameters, thereby affecting the parameter accuracy. Therefore, in the scheme of the present application, the correlation between the feature map and the feature vector is further removed. Specifically, such correlation is measured as an inner product of the feature map and the feature vector, and an inner product matrix and an inner product vector for representing the correlation are subtracted from the feature map and the feature vector, respectively, to obtain a decorrelated feature map and feature vector. That is, assuming that the feature map is A and the feature vector is B, the decorrelated feature map is A-A.B and the decorrelated feature vector is B-B.A. Note that if the feature map does not match the dimensions of the feature vector, the feature map is up-sampled or down-sampled for scaling.
And then, converting the characteristic diagram into a shape vector through a full connection layer, and obtaining the injection parameters of the sealant injector through an encoder after the characteristic diagram is cascaded with the characteristic vector.
Based on this, the present application proposes a method for determining injection parameters for a sealant injector based on feature decorrelation, comprising: acquiring an image of a borehole to be sealed; passing the borehole image through a convolutional neural network to obtain a feature map; acquiring gas parameters obtained in the process of extracting gas from a drilled hole and converting the gas parameters into input vectors; passing the input vector through a deep neural network to obtain a gas feature vector; calculating an inner product matrix of the feature map relative to the gas feature vector, wherein the inner product matrix is used for representing the correlation between the feature map and the gas feature vector; calculating an inner product vector of the gas feature vector relative to the feature map, wherein the inner product vector is used for representing the correlation between the gas feature vector and the feature map; subtracting the inner product matrix by pixel position according to the feature map to obtain a decorrelated feature map; subtracting the inner product vector from the gas feature vector according to element positions to obtain a decorrelation feature vector; passing the decorrelated feature map through a fully connected layer to obtain a shape vector; and cascading the shape vector and the decorrelation feature vector through an encoder to obtain injection parameters for the sealant injector.
Fig. 1 illustrates a scenario diagram of an injection parameter determination method for a sealant injector based on feature decorrelation according to an embodiment of the present application.
As shown in fig. 1, in this application scenario, first, an image of a borehole to be closed is acquired by a camera (e.g., C as illustrated in fig. 1), and a gas parameter in the process of extracting gas from the borehole is acquired by a sensor (e.g., S1 as illustrated in fig. 1), for example, a velocity sensor; the borehole image and the gas parameters are then input into a server (e.g., S2 as illustrated in fig. 1) deployed with a feature decorrelation-based injection parameter determination algorithm for a sealant injector, wherein the server is capable of processing the borehole image and the gas parameters with the feature decorrelation-based injection parameter determination algorithm for a sealant injector to determine injection parameters of the sealant injector.
Having described the general principles of the present application, various non-limiting embodiments of the present application will now be described with reference to the accompanying drawings.
Exemplary method
Fig. 2 illustrates a flow chart of an injection parameter determination method for a sealer injector based on feature decorrelation according to an embodiment of the present application. As shown in fig. 2, the injection parameter determination method for a sealant injector based on feature decorrelation according to an embodiment of the present application includes: s110, acquiring an image of a drill hole to be sealed; s120, passing the borehole image through a convolutional neural network to obtain a characteristic map; s130, acquiring gas parameters obtained in the process of extracting gas from a borehole and converting the gas parameters into input vectors; s140, passing the input vector through a deep neural network to obtain a gas characteristic vector; s150, calculating an inner product matrix of the feature map relative to the gas feature vector, wherein the inner product matrix is used for representing the correlation between the feature map and the gas feature vector; s160, calculating an inner product vector of the gas feature vector relative to the feature map, wherein the inner product vector is used for representing the correlation between the gas feature vector and the feature map; s170, subtracting the inner product matrix from the feature map according to pixel positions to obtain a decorrelation feature map; s180, subtracting the inner product vector from the gas characteristic vector according to element positions to obtain a decorrelation characteristic vector; s190, enabling the decorrelation feature map to pass through a full connection layer to obtain a shape vector; and S200, cascading the shape vector and the decorrelation characteristic vector and then passing through an encoder to obtain injection parameters for the sealant injector.
Fig. 3 illustrates an architectural schematic diagram of an injection parameter determination method for a sealant injector based on feature decorrelation according to an embodiment of the present application. As shown IN fig. 3, IN this network architecture, an acquired image of a borehole to be sealed (e.g., IN0 as illustrated IN fig. 3) is first passed through a convolutional neural network (e.g., CNN as illustrated IN fig. 3) to obtain a signature (e.g., F1 as illustrated IN fig. 3). Meanwhile, an input vector (e.g., V0 as illustrated in fig. 3) converted based on the acquired gas parameters obtained in the process of extracting gas from the borehole is passed through a deep neural network (e.g., DN as illustrated in fig. 3) to obtain a gas feature vector (e.g., V1 as illustrated in fig. 3). Then, an inner product matrix of the feature map with respect to the gas feature vector (e.g., M1 as illustrated in fig. 3) is calculated, the inner product matrix being used to represent a correlation between the feature map and the gas feature vector, and an inner product vector of the gas feature vector with respect to the feature map (e.g., Vi as illustrated in fig. 3) is calculated, the inner product vector being used to represent a correlation between the gas feature vector and the feature map. The inner product matrix is then subtracted from the feature map by pixel location to obtain a decorrelated feature map (e.g., Fu as illustrated in fig. 3), and the inner product vector is subtracted from the gas feature vector by element location to obtain a decorrelated feature vector (e.g., Vu as illustrated in fig. 3). The decorrelated feature map is then passed through a fully connected layer (e.g., Fcl as illustrated in fig. 3) to obtain a shape vector (e.g., Vs as illustrated in fig. 3). The shape vector and the decorrelation feature vector are then concatenated through an encoder (e.g., En as illustrated in fig. 3) to obtain injection parameters for the sealant injector.
In step S110, an image of the borehole to be closed is acquired. In the technical scheme of the application, the borehole image is used as a detection image and used for extracting the characteristics of the shape and the size of the borehole.
In step S120, the borehole image is passed through a convolutional neural network to obtain a feature map. That is, the borehole image is processed with a depth convolution neural network to extract high-dimensional features of the borehole image. Those skilled in the art will appreciate that convolutional neural networks have superior performance in extracting local spatial features of an image.
Preferably, in the present embodiment, the convolutional neural network is implemented as a deep residual network, e.g., ResNet 50. Compared with the traditional convolutional neural network, the deep residual error network is an optimized network structure provided on the basis of the traditional convolutional neural network, and mainly solves the problem that the gradient disappears in the training process. The depth residual error network introduces a residual error network structure, the network layer can be made deeper through the residual error network structure, and the problem of gradient disappearance can not occur. The residual error network uses the cross-layer link thought of a high-speed network for reference, breaks through the convention that the traditional neural network only can provide N layers as input from the input layer of the N-1 layer, enables the output of a certain layer to directly cross several layers as the input of the later layer, and has the significance of providing a new direction for the difficult problem that the error rate of the whole learning model is not reduced and inversely increased by superposing multiple layers of networks.
In step S130, gas parameters obtained in the process of extracting gas from a borehole are obtained and converted into input vectors. In the technical scheme of the application, the input vector is used as source domain data and is used for extracting the characteristics of the internal structure and the depth of the borehole.
Specifically, in the embodiment of the present application, a process of acquiring gas parameters obtained in a process of extracting gas from a borehole and converting the gas parameters into input vectors includes: first, the gas-exit flow rate at time intervals over a period of time is obtained and normalized to obtain the input vector. That is, in the solution of the present application, the gas parameter is a gas outlet flow rate.
In step S140, the input vector is passed through a deep neural network to obtain a gas feature vector. That is, the input vector is processed with a deep neural network to extract high-dimensional features in the input vector.
Specifically, in one specific example of the present application, the process of passing the input vector through a deep neural network to obtain a gas feature vector includes: and performing one-dimensional convolution processing on the input vector to obtain the gas characteristic vector. That is, the deep neural network is a 1-dimensional convolutional neural network, which can extract timing correlation information between elements in the input vector and information of each element itself in the input vector to obtain the gas feature vector.
In another specific example of the present application, the process of passing the input vector through a deep neural network to obtain a gas feature vector includes: and processing the input vector by a multilayer perceptron model to obtain the gas characteristic vector. That is, in this particular example, the deep neural network is a multi-layered perceptron model that is able to leverage information of various elements in the input vector to generate the gas feature vector. Those skilled in the art will appreciate that the multi-layered perceptron is a deep neural network model that can fully utilize information of and correlation information between various positions in an audio vector to extract high-dimensional features that can express audio data. The multi-layer perceptron model is a feedforward artificial neural network model, which can have a plurality of hidden layers in the middle besides an input and output layer, and the hidden layers map input data sets to a single output data set.
In step S150, an inner product matrix of the feature map with respect to the gas feature vector is calculated, and the inner product matrix is used to represent a correlation between the feature map and the gas feature vector. It will be appreciated that for the profile and the gas eigenvectors, although they themselves reflect different characteristics of the borehole, there may still be some correlation that, when further parameter encoded by an encoder comprising fully connected layers, may cause an over-fit of the encoded parameters, affecting parameter accuracy. Therefore, in the scheme of the present application, the correlation between the feature map and the feature vector is further removed.
Accordingly, in step S150, an inner product matrix of the feature map with respect to the gas feature vector is calculated, and the inner product matrix is used to represent the correlation between the feature map and the gas feature vector. Those of ordinary skill in the art will appreciate that the inner product matrix is a function of the row to study the similarity between the profile versus the gas eigenvectors. Specifically, in the process of calculating the inner product matrix of the feature map relative to the gas feature vector, the gas feature vector needs to be adjusted to a dimension matched with the feature map, that is, the number of columns of the gas feature vector is equal to the number of rows of the feature map, and then the inner product matrix of the feature map relative to the gas feature vector after being adjusted to the matched dimension is calculated. It should be understood that setting the feature map as a and the feature vector as B, the decorrelated feature map is a-a · B.
Fig. 4 illustrates a flow chart of calculating an inner product matrix of the feature map with respect to the gas feature vector in a method for determining injection parameters for a sealant injector based on feature decorrelation according to an embodiment of the present application. As shown in fig. 4, in the embodiment of the present application, calculating an inner product matrix of the feature map with respect to the gas feature vector includes: s210, adjusting the gas characteristic vector into a dimension matched with the characteristic map; and S220, calculating the inner product matrix of the feature map relative to the gas feature vector after being adjusted to the matched dimension.
In step S160, an inner product vector of the gas feature vector with respect to the feature map is calculated, and the inner product vector is used to represent a correlation between the gas feature vector and the feature map. Those of ordinary skill in the art will appreciate that the inner product vector is a study of the similarity between the feature vectors relative to the feature map from a line perspective. Specifically, in the process of calculating the inner product vector of the gas feature vector with respect to the feature map, the feature map is first adjusted to a scale matching the gas feature vector, that is, the number of rows of the feature map is adjusted to be equal to the number of columns of the gas feature vector, and then the inner product vector of the gas feature vector with respect to the feature map after being adjusted to the matched scale is calculated. It should be understood that setting the feature map to be a and the feature vector to be B, the decorrelated feature vector is B-B · a.
Fig. 5 illustrates a flow chart for calculating an inner product vector of the gas feature vector with respect to the feature map in a method for determining injection parameters for a sealer injector based on feature decorrelation according to an embodiment of the present application. As shown in fig. 5, in the embodiment of the present application, calculating an inner product vector of the gas feature vector with respect to the feature map includes: s310, adjusting the gas characteristic vector to be a dimension matched with the characteristic map; and S320, calculating the inner product matrix of the feature map relative to the gas feature vector after being adjusted to the matched dimension.
In step S170, the inner product matrix is subtracted from the feature map by pixel position to obtain a decorrelated feature map. That is, features in the feature map that are associated with the gas feature vector are removed to obtain the decorrelated feature map.
In step S180, the inner product vector is subtracted from the gas feature vector by element position to obtain a decorrelated feature vector. That is, features of the gas feature vector that are associated with the feature map are removed to obtain the decorrelation vector.
In step S190, the decorrelated feature map is passed through a full-connected layer to obtain a shape vector. That is, the decorrelation feature map is encoded with a full-connected layer as an encoder to extract the shape vector representing the feature of the shape and size of the borehole in the decorrelation feature map.
In step S200, the shape vector and the decorrelation feature vector are concatenated and then passed through an encoder to obtain injection parameters for a sealant injector. That is, the characteristics of the shape and size of the bore hole and the characteristics of the internal structure and depth of the bore hole are integrated and encoded by an encoder to obtain the injection parameters of the sealant injector. Accordingly, in the embodiments of the present application, the injection parameters include, but are not limited to: injection aperture of the injector, injection flow rate, injection orientation, etc.
More specifically, in the embodiment of the present application, the comprehensive feature vector is encoded by using one or more fully-connected layers as an encoder to obtain injection parameters for a sealant injector, wherein the number of output bits of the last fully-connected layer in the one or more fully-connected layers is equal to the number of the injection parameters of the sealant injector. It should be appreciated that the information of each element in the integrated feature vector can be fully utilized by the one or more fully-connected layers as an encoder to improve encoding accuracy.
In summary, an injection parameter determination method for a sealant injector based on feature decorrelation according to an embodiment of the present application is illustrated, which extracts feature maps of borehole images with a deep neural network to capture shape and size features of a borehole, extracts feature vectors representing depths and internal structural features of the borehole based on gas parameters flowing out of the borehole with the deep neural network, further performs decorrelation processing on the feature maps and the feature vectors, and encodes the feature vectors and the feature maps after the decorrelation processing to determine injection parameters of the sealant injector.
Exemplary System
Fig. 6 illustrates a block diagram of an injection parameter determination system for a sealant injector based on feature decorrelation according to an embodiment of the present application.
As shown in fig. 6, an injection parameter determination system 600 for a sealant injector based on feature decorrelation according to an embodiment of the present application includes: an image acquisition unit 610 for acquiring an image of a borehole to be closed; a feature map generation unit 620, configured to pass the borehole image obtained by the image obtaining unit 610 through a convolutional neural network to obtain a feature map; the gas parameter acquiring unit 630 is configured to acquire gas parameters obtained in a process of extracting gas from a borehole and convert the gas parameters into input vectors; a gas feature vector generating unit 640, configured to pass the input vector obtained by the gas parameter obtaining unit 630 through a deep neural network to obtain a gas feature vector; an inner product matrix calculation unit 650 configured to calculate an inner product matrix of the feature map obtained by the feature map generation unit 620 with respect to the gas feature vector obtained by the gas feature vector generation unit 640, where the inner product matrix is used to represent a correlation between the feature map and the gas feature vector; an inner product vector calculation unit 660 configured to calculate an inner product vector of the gas feature vector obtained by the gas feature vector generation unit 640 with respect to the feature map obtained by the feature map generation unit 620, the inner product vector being used to represent a correlation between the gas feature vector and the feature map; a decorrelation feature map generating unit 670 configured to subtract the inner product matrix obtained by the inner product matrix calculating unit 650 from the feature map obtained by the feature map generating unit 620 by pixel position to obtain a decorrelation feature map; a decorrelation feature vector generation unit 680 configured to subtract the inner product vector obtained by the inner product vector calculation unit 660 by the gas feature vector obtained by the gas feature vector generation unit 640 by an element position to obtain a decorrelation feature vector; a shape vector generating unit 690 for passing the decorrelated feature map obtained by the decorrelated feature map generating unit 670 through a full connection layer to obtain a shape vector; and an encoding unit 700, configured to cascade the shape vector obtained by the shape vector generation unit 690 and the decorrelation feature vector obtained by the decorrelation feature vector generation unit 680 through an encoder to obtain injection parameters for the sealant injector.
In one example, in the above parameter determination system 600, the gas feature vector generation unit 640 is further configured to: and performing one-dimensional convolution processing on the input vector to obtain the gas characteristic vector.
In one example, in the above-described parameter determination system 600, the deep neural network is a multi-layered perceptron.
In one example, in the above-mentioned parameter determination system 600, the inner product matrix calculation unit 650 is further configured to: adjusting the gas feature vector to a dimension matching the feature map; and calculating the inner product matrix of the feature map relative to the gas feature vector adjusted to the matched dimension.
In one example, in the above-mentioned parameter determination system 600, the inner product vector calculation unit 660 is further configured to: adjusting the feature map to a scale matching the gas feature vector; and calculating the inner product vector of the gas feature vector relative to the feature map adjusted to the matched scale.
In an example, in the above-mentioned parameter determining system 600, the encoding unit 700 is further configured to: and cascading the shape vector and the decorrelation characteristic vector and then passing through one or more full connection layers to obtain injection parameters for the sealant injector, wherein the number of output bits of the last full connection layer in the one or more full connection layers is equal to the number of the injection parameters of the sealant injector.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described parameter determination system 600 have been described in detail in the above description of the injection parameter determination method for a sealant injector based on feature decorrelation with reference to fig. 1 to 5, and thus, a repetitive description thereof will be omitted.
As described above, the parameter determination system 600 according to the embodiment of the present application may be implemented in various terminal devices, such as a server for parameter control of a sealant injector, and the like. In one example, the parameter determination system 600 according to the embodiment of the present application may be integrated into the terminal device as one software module and/or hardware module. For example, the parameter determination system 600 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the parameter determination system 600 may also be one of many hardware modules of the terminal device.
Alternatively, in another example, the parameter determination system 600 and the terminal device may be separate devices, and the parameter determination system 600 may be connected to the terminal device through a wired and/or wireless network and transmit the interaction information according to an agreed data format.
Exemplary electronic device
Next, an electronic apparatus according to an embodiment of the present application is described with reference to fig. 7.
FIG. 7 illustrates a block diagram of an electronic device in accordance with an embodiment of the present application.
As shown in fig. 7, the electronic device 10 includes one or more processors 11 and memory 12.
The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other form of connection mechanism (not shown).
The input device 13 may include, for example, a keyboard, a mouse, and the like.
The output device 14 can output various information including injection parameters and the like to the outside. The output devices 14 may include, for example, a display, speakers, a printer, and a communication network and its connected remote output devices, among others.
Of course, for simplicity, only some of the components of the electronic device 10 relevant to the present application are shown in fig. 7, and components such as buses, input/output interfaces, and the like are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer-readable storage Medium
In addition to the above-described methods and apparatus, embodiments of the present application may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the method for feature-based decorrelation of injection parameters for a sealant injector according to various embodiments of the present application described in the "exemplary methods" section above in this specification.
The computer program product may be written with program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the steps in the method for injection parameter determination for a sealant injector based on feature decorrelation according to various embodiments of the present application described in the "exemplary methods" section above in this specification.
The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing describes the general principles of the present application in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present application are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the foregoing disclosure is not intended to be exhaustive or to limit the disclosure to the precise details disclosed.
The block diagrams of devices, apparatuses, systems referred to in this application are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
It should also be noted that in the devices, apparatuses, and methods of the present application, the components or steps may be decomposed and/or recombined. These decompositions and/or recombinations are to be considered as equivalents of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.
Claims (10)
1. A method for determining injection parameters for a sealant injector based on feature decorrelation, comprising:
acquiring an image of a borehole to be sealed;
passing the borehole image through a convolutional neural network to obtain a feature map;
acquiring gas parameters obtained in the process of extracting gas from a drilled hole and converting the gas parameters into input vectors;
passing the input vector through a deep neural network to obtain a gas feature vector;
calculating an inner product matrix of the feature map relative to the gas feature vector, wherein the inner product matrix is used for representing the correlation between the feature map and the gas feature vector;
calculating an inner product vector of the gas feature vector relative to the feature map, wherein the inner product vector is used for representing the correlation between the gas feature vector and the feature map;
subtracting the inner product matrix by pixel position according to the feature map to obtain a decorrelated feature map;
subtracting the inner product vector from the gas feature vector according to element positions to obtain a decorrelation feature vector;
passing the decorrelated feature map through a fully connected layer to obtain a shape vector; and
and cascading the shape vector and the decorrelation characteristic vector and then passing the concatenated vector through an encoder to obtain injection parameters for the sealant injector.
2. The feature decorrelation-based injection parameter determination method for a sealant injector according to claim 1, wherein passing the input vector through a deep neural network to obtain a gas feature vector comprises:
and performing one-dimensional convolution processing on the input vector to obtain the gas characteristic vector.
3. The feature decorrelation-based injection parameter determination method for a sealant injector according to claim 1, wherein in passing the input vector through a deep neural network to obtain a gas feature vector, the deep neural network is a multilayer perceptron.
4. The method of claim 1, wherein calculating an inner product matrix of the feature map relative to the gas feature vector, the inner product matrix representing a correlation between the feature map and the gas feature vector, comprises:
adjusting the gas feature vector to a dimension matching the feature map; and
calculating the inner product matrix of the feature map relative to the gas feature vector adjusted to the matched dimension.
5. The feature decorrelation-based injection parameter determination method for a sealant injector according to claim 1, wherein calculating an inner product vector of the gas feature vector with respect to the feature map, the inner product vector being used to represent a correlation between the gas feature vector and the feature map, comprises:
adjusting the feature map to a scale matching the gas feature vector; and
and calculating the inner product vector of the gas feature vector relative to the feature map after being adjusted to the matched scale.
6. The method for determining injection parameters for a sealant injector based on feature decorrelation according to claim 1, wherein cascading the shape vector and the decorrelation feature vector through an encoder to obtain injection parameters for a sealant injector comprises:
and cascading the shape vector and the decorrelation characteristic vector and then passing through one or more full connection layers to obtain injection parameters for the sealant injector, wherein the number of output bits of the last full connection layer in the one or more full connection layers is equal to the number of the injection parameters of the sealant injector.
7. An injection parameter determination system for a sealant injector based on feature decorrelation, comprising:
the image acquisition unit is used for acquiring an image of a drill hole to be sealed;
the characteristic map generating unit is used for enabling the borehole image obtained by the image obtaining unit to pass through a convolutional neural network so as to obtain a characteristic map;
the gas parameter acquisition unit is used for acquiring gas parameters obtained in the process of extracting gas from a drilled hole and converting the gas parameters into input vectors;
the gas characteristic vector generating unit is used for enabling the input vector obtained by the gas parameter obtaining unit to pass through a deep neural network so as to obtain a gas characteristic vector;
an inner product matrix calculation unit configured to calculate an inner product matrix of the feature map obtained by the feature map generation unit with respect to the gas feature vector obtained by the gas feature vector generation unit, the inner product matrix being used to represent a correlation between the feature map and the gas feature vector;
an inner product vector calculation unit configured to calculate an inner product vector of the gas feature vector obtained by the gas feature vector generation unit with respect to the feature map obtained by the feature map generation unit, the inner product vector representing a correlation between the gas feature vector and the feature map;
a decorrelation feature map generating unit configured to subtract the inner product matrix obtained by the inner product matrix calculating unit from the feature map obtained by the feature map generating unit by pixel position to obtain a decorrelation feature map;
a decorrelation feature vector generation unit configured to subtract the inner product vector obtained by the inner product vector calculation unit from the gas feature vector obtained by the gas feature vector generation unit by an element position to obtain a decorrelation feature vector;
a shape vector generating unit, configured to pass the decorrelation feature map obtained by the decorrelation feature map generating unit through a full connection layer to obtain a shape vector; and
and the coding unit is used for cascading the shape vector obtained by the shape vector generation unit and the decorrelation characteristic vector obtained by the decorrelation characteristic vector generation unit and then passing the concatenated vector through a coder so as to obtain the injection parameters for the sealant injector.
8. The injection parameter determination system for a sealant injector based on feature decorrelation according to claim 7, wherein the gas feature vector generation unit is further configured to: and performing one-dimensional convolution processing on the input vector to obtain the gas characteristic vector.
9. The feature decorrelation-based injection parameter determination system for a sealant injector according to claim 7, wherein the deep neural network is a multilayer perceptron.
10. An electronic device, comprising:
a processor; and
a memory having stored therein computer program instructions which, when executed by the processor, cause the processor to perform the method of feature decorrelation based injection parameter determination for a sealant injector according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011519163.2A CN112614175A (en) | 2020-12-21 | 2020-12-21 | Injection parameter determination method for hole sealing agent injector based on characteristic decorrelation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011519163.2A CN112614175A (en) | 2020-12-21 | 2020-12-21 | Injection parameter determination method for hole sealing agent injector based on characteristic decorrelation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112614175A true CN112614175A (en) | 2021-04-06 |
Family
ID=75244362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011519163.2A Pending CN112614175A (en) | 2020-12-21 | 2020-12-21 | Injection parameter determination method for hole sealing agent injector based on characteristic decorrelation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112614175A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304846A (en) * | 2017-09-11 | 2018-07-20 | 腾讯科技（深圳）有限公司 | Image-recognizing method, device and storage medium |
| CN108416434A (en) * | 2018-02-07 | 2018-08-17 | 复旦大学 | The circuit structure accelerated with full articulamentum for the convolutional layer of neural network |
| US20190138898A1 (en) * | 2017-11-07 | 2019-05-09 | Samsung Electronics Co., Ltd. | Method and apparatus with neural network performing deconvolution |
| CN111428761A (en) * | 2020-03-11 | 2020-07-17 | 深圳先进技术研究院 | Image feature visualization method, image feature visualization device and electronic equipment |
| CN111612024A (en) * | 2019-02-25 | 2020-09-01 | 北京嘀嘀无限科技发展有限公司 | Feature extraction method and device, electronic equipment and computer-readable storage medium |
| CN111852390A (en) * | 2020-07-21 | 2020-10-30 | 叶际平 | Coal mine gas extraction drilling hole sealing agent directional expansion device and expansion method |
-
2020
- 2020-12-21 CN CN202011519163.2A patent/CN112614175A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304846A (en) * | 2017-09-11 | 2018-07-20 | 腾讯科技（深圳）有限公司 | Image-recognizing method, device and storage medium |
| US20190138898A1 (en) * | 2017-11-07 | 2019-05-09 | Samsung Electronics Co., Ltd. | Method and apparatus with neural network performing deconvolution |
| CN108416434A (en) * | 2018-02-07 | 2018-08-17 | 复旦大学 | The circuit structure accelerated with full articulamentum for the convolutional layer of neural network |
| CN111612024A (en) * | 2019-02-25 | 2020-09-01 | 北京嘀嘀无限科技发展有限公司 | Feature extraction method and device, electronic equipment and computer-readable storage medium |
| CN111428761A (en) * | 2020-03-11 | 2020-07-17 | 深圳先进技术研究院 | Image feature visualization method, image feature visualization device and electronic equipment |
| CN111852390A (en) * | 2020-07-21 | 2020-10-30 | 叶际平 | Coal mine gas extraction drilling hole sealing agent directional expansion device and expansion method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111859912B (en) | PCNN model-based remote supervision relationship extraction method with entity perception | |
| US20180253640A1 (en) | Hybrid architecture system and method for high-dimensional sequence processing | |
| CN110334587A (en) | Training method, device and the crucial independent positioning method of face key point location model | |
| Xue et al. | A better way to attend: Attention with trees for video question answering | |
| CN111914731B (en) | Multi-mode LSTM video motion prediction method based on self-attention mechanism | |
| CN116025319A (en) | Multi-medium thermal fluid operation monitoring system and method thereof | |
| CN113961736A (en) | Method and device for generating image by text, computer equipment and storage medium | |
| CN111325660A (en) | Remote sensing image style conversion method based on text data | |
| CN113362416B (en) | Method for generating image based on text of target detection | |
| CN109344822A (en) | A kind of scene text detection method based on shot and long term memory network | |
| KR102117654B1 (en) | Method and Apparatus for Generating Video Based on Natural Language | |
| CN114581965B (en) | Finger vein recognition model training method, finger vein recognition model training system and terminal | |
| CN116563302B (en) | Intelligent medical information management system and method thereof | |
| CN116776872A (en) | Medical data structured archiving system | |
| Feng et al. | Unsupervised multimodal anomaly detection with missing sources for liquid rocket engine | |
| Li et al. | Detection of generative linguistic steganography based on explicit and latent text word relation mining using deep learning | |
| Sern et al. | PhishGAN: Data augmentation and identification of homoglyph attacks | |
| Al-Wesabi et al. | Tampering Detection Approach of Arabic-Text Based on Contents Interrelationship. | |
| CN112614175A (en) | Injection parameter determination method for hole sealing agent injector based on characteristic decorrelation | |
| CN116127019B (en) | Dynamic parameter and visual model generation WEB 2D automatic modeling engine system | |
| CN116678258A (en) | Cold-heat exchanger for pressure vessel and control method thereof | |
| CN112836752A (en) | Intelligent sampling parameter control method based on feature map fusion of depth values | |
| Zhang et al. | A rapid combined model for automatic generating web UI codes | |
| CN112819040A (en) | Intelligent control method for cement processing based on weight generation mechanism | |
| Fan et al. | Facial expression animation through action units transfer in latent space |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240103 Address after: Room 408-8, 4th Floor, Building 2, Haichuang Technology Center, Cangqian Street, Yuhang District, Hangzhou City, Zhejiang Province, 313000 Applicant after: HANGZHOU ZHUILIE TECHNOLOGY Co.,Ltd. Address before: Room 1004, building 2, new Tianxiang Commercial Plaza, 388 Suya Road, Suzhou Industrial Park, Suzhou area, China (Jiangsu) pilot Free Trade Zone, Suzhou 215000, Jiangsu Province Applicant before: Suzhou tuochi Information Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right |