CN112767342A - Intelligent gas detection method based on double-branch inference mechanism - Google Patents
Intelligent gas detection method based on double-branch inference mechanism Download PDFInfo
- Publication number
- CN112767342A CN112767342A CN202110050001.7A CN202110050001A CN112767342A CN 112767342 A CN112767342 A CN 112767342A CN 202110050001 A CN202110050001 A CN 202110050001A CN 112767342 A CN112767342 A CN 112767342A
- Authority
- CN
- China
- Prior art keywords
- feature
- image
- gas
- feature map
- detected
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/85—Investigating moving fluids or granular solids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06N3/048—Activation functions
-
- 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
- G06N3/08—Learning methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/85—Investigating moving fluids or granular solids
- G01N2021/8578—Gaseous flow
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Quality & Reliability (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
Abstract
The application relates to intelligent detection in the field of intelligent mining, and particularly discloses a gas intelligent detection method based on a double-branch inference mechanism. Particularly, in the detection process, a first branch for deducing the water drop content and a second branch for deducing the water vapor content are adopted, wherein in the second branch for deducing the water vapor content, a gas image without water vapor is adopted as a reference image, and a foreground mask is extracted from the detection image to be used as a characteristic basis for deducing the water vapor content, so that the influence of the water vapor and the water drops in the gas is comprehensively considered in the classification process, and the classification precision is improved.
Description
Technical Field
The present invention relates to intelligent detection in the field of intelligent mining, and more particularly, to an intelligent gas detection method based on a dual-branch inference mechanism, an intelligent gas detection system based on a dual-branch inference mechanism, and an electronic device.
Background
A lot of gas can appear in the mining process of colliery, metal mine, and gas easily causes the explosion in the pit on the one hand, and is extremely dangerous, and on the other hand can regard as renewable energy, improves the energy utilization efficiency in the mining process. Therefore, in the mining process, gas is increasingly being extracted from the mining site and transported to a device where the gas can be recycled. However, since the gas may contain water vapor and water droplets, the water vapor content needs to be determined before the gas is delivered to a recycling device as a recyclable energy source to improve the safety of the use of the gas.
Therefore, it is desirable to provide a technical solution for intelligent detection of gas.
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.
The deep learning and the development of the neural network provide a new solution idea and scheme for the intelligent detection of the gas.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides a gas intelligent detection method based on a double-branch inference mechanism, a gas intelligent detection system based on the double-branch inference mechanism and electronic equipment, wherein a computer vision technology based on deep learning is adopted to perform feature extraction and classification on an image of gas in a conveying process so as to determine whether the water vapor content in the current gas is suitable for being directly conveyed to a recycling device for recycling. Particularly, in the detection process, a first branch for deducing the water drop content and a second branch for deducing the water vapor content are adopted, wherein in the second branch for deducing the water vapor content, a gas image without water vapor is adopted as a reference image, and a foreground mask is extracted from the detection image to be used as a characteristic basis for deducing the water vapor content, so that the influence of the water vapor and the water drops in the gas is comprehensively considered in the classification process, and the classification precision is improved.
According to an aspect of the present application, there is provided a gas intelligent detection method based on a dual-branch inference mechanism, which includes:
acquiring an image to be detected, wherein the image to be detected is an image of the gas in the conveying process;
passing the image to be detected through a first convolutional neural network to obtain a first characteristic diagram;
passing the first feature map through a plurality of fully-connected layers to obtain a first feature vector, wherein the first feature vector is a feature representation of the water droplet content in the gas in a high-dimensional feature space;
respectively enabling the image to be detected and the reference image to pass through a second convolutional neural network so as to extract a second characteristic diagram from the image to be detected and a third characteristic diagram from the reference image, wherein the reference image is a gas image without water vapor, and the last layer of the second convolutional neural network is activated by a Sigmoid function;
for each position in the second feature map and the third feature map, calculating an absolute value of a difference value between feature values of corresponding positions of the second feature map and the third feature map and dividing the absolute value by the feature value of the corresponding position of the third feature map to obtain a differential proportion feature map, wherein the differential proportion feature map is used for representing a foreground mask of the image to be detected relative to the reference image;
passing the differential proportion characteristic diagram through a plurality of fully-connected layers to obtain a second characteristic vector, wherein the second characteristic vector is a characteristic representation of the water vapor content in the gas in a high-dimensional characteristic space; and
and enabling the first characteristic vector and the second characteristic vector to pass through a classifier so as to obtain a classification result, wherein the classification result is used for representing whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
In the above intelligent gas detection method based on the dual-branch inference mechanism, the first convolutional neural network and the second convolutional neural network have the same network structure.
In the above-mentioned gas intelligent detection method based on the double-branch inference mechanism, for each position in the second feature map and the third feature map, calculating an absolute value of a difference between feature values of corresponding positions of the second feature map and the third feature map and dividing the absolute value by the feature value of the corresponding position of the third feature map to obtain a differential ratio feature map, including: and performing up-sampling or down-sampling processing on the third feature map to adjust the scale of the third feature map to be consistent with the scale of the second feature map.
In the above intelligent gas detection method based on a dual-branch inference mechanism, passing the first eigenvector and the second eigenvector through a classifier to obtain a classification result includes: concatenating the first feature vector with the second feature vector; and passing the first feature vector and the second feature vector after the cascade connection through a classification function to obtain the classification result.
In the above intelligent gas detection method based on a dual-branch inference mechanism, passing the first eigenvector and the second eigenvector through a classifier to obtain a classification result includes: calculating a position-weighted sum of the first feature vector and the second feature vector with a preset weight to obtain a classified feature vector; and passing the classified feature vector through a classification function to obtain the result.
In the above intelligent gas detection method based on the dual-branch inference mechanism, the first convolutional neural network and the second convolutional neural network are deep residual error networks.
According to another aspect of the present application, there is provided a gas intelligent detection system based on a dual-branch inference mechanism, comprising:
the device comprises an image acquisition unit to be detected, a data acquisition unit and a data processing unit, wherein the image acquisition unit to be detected is used for acquiring an image to be detected, and the image to be detected is an image of the gas in the conveying process;
the first characteristic diagram generating unit is used for enabling the image to be detected obtained by the image to be detected obtaining unit to pass through a first convolutional neural network so as to obtain a first characteristic diagram;
a first feature vector generation unit, configured to pass the first feature map obtained by the first feature map generation unit through a plurality of fully-connected layers to obtain a first feature vector, where the first feature vector is a feature representation of water droplet content in the gas in a high-dimensional feature space;
the characteristic diagram extraction unit is used for respectively enabling the image to be detected and the reference image obtained by the image to be detected acquisition unit to pass through a second convolutional neural network so as to extract a second characteristic diagram from the image to be detected and a third characteristic diagram from the reference image, wherein the reference image is a gas image without water vapor, and the last layer of the second convolutional neural network is activated by a Sigmoid function;
a difference proportion feature map generation unit, configured to calculate, for each position in the second feature map and the third feature map obtained by the feature map extraction unit, an absolute value of a difference between feature values of corresponding positions of the second feature map and the third feature map and divide the absolute value by the feature value of the corresponding position of the third feature map to obtain a difference proportion feature map, where the difference proportion feature map is used to represent a foreground mask of the image to be detected with respect to the reference image;
the second eigenvector generating unit is used for enabling the differential proportion characteristic diagram obtained by the differential proportion characteristic diagram generating unit to pass through a plurality of full-connection layers so as to obtain a second eigenvector, and the second eigenvector is the characteristic representation of the water vapor content in the gas in a high-dimensional characteristic space;
and the classification result generating unit is used for enabling the first characteristic vector obtained by the first characteristic vector generating unit and the second characteristic vector obtained by the second characteristic vector generating unit to pass through a classifier so as to obtain a classification result, and the classification result is used for indicating whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
In the above intelligent gas detection system based on the dual-branch inference mechanism, the first convolutional neural network and the second convolutional neural network have the same network structure.
In the above gas intelligent detection system based on the two-branch inference mechanism, the differential proportion characteristic map generating unit is further configured to: and performing up-sampling or down-sampling processing on the third feature map to adjust the scale of the third feature map to be consistent with the scale of the second feature map.
In the above-mentioned gas intelligent detection system based on the two-branch inference mechanism, the classification result generating unit includes: a concatenation subunit, configured to concatenate the first feature vector and the second feature vector; and the first classification subunit is used for enabling the first feature vector and the second feature vector after the cascade connection to pass through a classification function so as to obtain the classification result.
In the above-mentioned gas intelligent detection system based on the two-branch inference mechanism, the classification result generating unit includes: a classification feature vector generation subunit, configured to calculate a position-weighted sum of the first feature vector and the second feature vector with a preset weight to obtain a classification feature vector; and the second classification subunit is used for enabling the classification feature vector obtained by the classification feature vector generation subunit to pass through a classification function so as to obtain the result.
In the above gas intelligent detection system based on the double-branch inference mechanism, the first convolutional neural network and the second convolutional neural network are deep residual error networks.
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 dual branch inference mechanism based gas intelligent detection method 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 execute the intelligent gas detection method based on a dual-branch inference mechanism as described above.
Compared with the prior art, the intelligent gas detection method based on the double-branch inference mechanism, the intelligent gas detection system based on the double-branch inference mechanism and the electronic equipment provided by the application adopt the computer vision technology based on deep learning to perform feature extraction and classification on the image of the gas in the conveying process so as to determine whether the water vapor content in the current gas is suitable for being directly conveyed to a recycling device for recycling. Particularly, in the detection process, a first branch for deducing the water drop content and a second branch for deducing the water vapor content are adopted, wherein in the second branch for deducing the water vapor content, a gas image without water vapor is adopted as a reference image, and a foreground mask is extracted from the detection image to be used as a characteristic basis for deducing the water vapor content, so that the influence of the water vapor and the water drops in the gas is comprehensively considered in the classification process, and the classification precision is improved.
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 an application scenario of a gas intelligent detection method based on a double-branch inference mechanism according to an embodiment of the present application;
fig. 2 illustrates a flowchart of a gas intelligent detection method based on a double-branch inference mechanism according to an embodiment of the present application;
fig. 3 is a schematic diagram illustrating a system architecture of a gas intelligent detection method based on a dual-branch inference mechanism according to an embodiment of the present application;
fig. 4 is a flowchart illustrating that the first feature vector and the second feature vector are passed through a classifier to obtain a classification result in the gas intelligent detection method based on the double-branch inference mechanism according to the embodiment of the present application;
fig. 5 illustrates another flowchart of passing the first feature vector and the second feature vector through a classifier to obtain a classification result in the gas intelligent detection method based on the double-branch inference mechanism according to the embodiment of the present application;
fig. 6 illustrates a block diagram of a gas intelligent detection system based on a dual-branch inference mechanism according to an embodiment of the present application;
fig. 7 illustrates a block diagram of a classification result generation unit in a gas intelligent detection system based on a dual-branch inference mechanism according to an embodiment of the present application;
fig. 8 illustrates another block diagram of a classification result generation unit in a gas intelligent detection system based on a dual-branch inference mechanism according to an embodiment of the present application;
FIG. 9 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 previously mentioned, in mining processes, gas is increasingly being extracted from a mining site and transported to a means of recycling the gas. However, since the gas may contain water vapor and water droplets, the water vapor content needs to be determined before the gas is delivered to a recycling device as a recyclable energy source to improve the safety of the use of the gas.
Because the vision technology can judge the moisture content from the gas more intuitively, the inventor of the application expects to adopt the computer vision technology based on deep learning to carry out feature extraction on the image of the gas in the conveying process, and classify based on the extracted features, to determine whether the moisture content in the current gas is suitable for being directly conveyed to a recycling device for recycling.
In a specific practical process, water drops can be directly detected from an image of gas as suspended particles in the conveyed gas, and water vapor is mixed with the gas as the gas and cannot be used as a detectable object. Therefore, in the technical scheme of the application, a first branch for deducing the water drop content and a second branch for deducing the water vapor content are respectively adopted, wherein in the second branch for deducing the water vapor content, a gas image without water vapor is adopted as a reference image, and a foreground mask is extracted from a current image to be used as a characteristic basis for deducing the water vapor content.
Specifically, in the technical solution of the present application, an image of the gas in the conveying process is obtained first, a first feature map is obtained through a first convolutional neural network, and the first feature map is converted into a first feature vector for expressing the content of water droplets in the gas through a plurality of fully connected layers, and here, because it is not necessary to directly detect water droplets in the gas in the present application, it is not necessary to perform object detection of each object based on the first feature map, but only a feature vector capable of expressing the distribution characteristics of the object is extracted.
Then, the image of the gas in the conveying process and the reference image are respectively passed through a second convolution neural network to obtain a second characteristic diagram and a third characteristic diagram, wherein the last layer of the second convolution neural network is activated by a sigmoid function, so that the characteristic value of each position in the second characteristic diagram and the third characteristic diagram is in the interval of 0 to 1, and for each position, the difference value of the characteristic value of the second characteristic diagram and the third characteristic diagram is calculated and divided by the difference value of the third characteristic diagram, so as to obtain a difference ratio characteristic diagram for expressing the foreground mask of the current image relative to the reference image. Similarly, because the exact value of the moisture content does not need to be determined, only a certain ratio of the moisture content relative to a reference needs to be obtained, and thus the differential ratio signature can be converted through a plurality of fully connected layers into a second signature vector for expressing the moisture content in the gas.
And then, cascading the first characteristic vector and the second characteristic vector, and obtaining a classification result through a classifier, wherein the classification result indicates whether the current gas is suitable for being directly conveyed to a recycling device for recycling.
Based on this, this application has proposed a gas intelligent detection method based on two branch inference mechanisms, it includes: acquiring an image to be detected, wherein the image to be detected is an image of the gas in the conveying process; passing the image to be detected through a first convolutional neural network to obtain a first characteristic diagram; passing the first feature map through a plurality of fully-connected layers to obtain a first feature vector, wherein the first feature vector is a feature representation of the water droplet content in the gas in a high-dimensional feature space; respectively enabling the image to be detected and the reference image to pass through a second convolutional neural network so as to extract a second characteristic diagram from the image to be detected and a third characteristic diagram from the reference image, wherein the reference image is a gas image without water vapor, and the last layer of the second convolutional neural network is activated by a Sigmoid function; for each position in the second feature map and the third feature map, calculating an absolute value of a difference value between feature values of corresponding positions of the second feature map and the third feature map and dividing the absolute value by the feature value of the corresponding position of the third feature map to obtain a differential proportion feature map, wherein the differential proportion feature map is used for representing a foreground mask of the image to be detected relative to the reference image; passing the differential proportion characteristic diagram through a plurality of fully-connected layers to obtain a second characteristic vector, wherein the second characteristic vector is a characteristic representation of the water vapor content in the gas in a high-dimensional characteristic space; and enabling the first characteristic vector and the second characteristic vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
Fig. 1 illustrates an application scenario of a gas intelligent detection method based on a double-branch inference mechanism according to an embodiment of the present application.
As shown in fig. 1, in the application scenario, an image of the gas to be detected in the conveying process is acquired through a camera (e.g., as indicated by C in fig. 1); then, the image is input into a server (for example, S as illustrated in fig. 1) deployed with a gas intelligent detection algorithm based on a double-branch inference mechanism, wherein the server can process the image based on the gas intelligent detection algorithm based on the double-branch inference mechanism to generate a detection result indicating whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
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 a gas intelligent detection method based on a double-branch inference mechanism. As shown in fig. 2, the intelligent gas detection method based on the dual-branch inference mechanism according to the embodiment of the present application includes: s110, acquiring an image to be detected, wherein the image to be detected is an image of the gas in the conveying process; s120, passing the image to be detected through a first convolutional neural network to obtain a first characteristic diagram; s130, passing the first characteristic diagram through a plurality of fully-connected layers to obtain a first characteristic vector, wherein the first characteristic vector is a characteristic representation of the water drop content in the gas in a high-dimensional characteristic space; s140, the image to be detected and the reference image respectively pass through a second convolutional neural network to extract a second characteristic diagram from the image to be detected and a third characteristic diagram from the reference image, wherein the reference image is a gas image without water vapor, and the last layer of the second convolutional neural network is activated by a Sigmoid function; s150, calculating the absolute value of the difference between the feature values of the corresponding positions of the second feature map and the third feature map and dividing the absolute value by the feature value of the corresponding position of the third feature map to obtain a differential proportion feature map for each position in the second feature map and the third feature map, wherein the differential proportion feature map is used for representing the foreground mask of the image to be detected relative to the reference image; s160, passing the differential proportion characteristic diagram through a plurality of full-connection layers to obtain a second characteristic vector, wherein the second characteristic vector is a characteristic representation of the water vapor content in the gas in a high-dimensional characteristic space; and S170, enabling the first feature vector and the second feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
Fig. 3 is a schematic diagram illustrating an architecture of a gas intelligent detection method based on a dual-branch inference mechanism according to an embodiment of the present application. As shown IN fig. 3, IN the network architecture of the gas intelligent detection method based on the dual-branch inference mechanism, firstly, the acquired image of the gas to be detected IN the conveying process (for example, IN1 as illustrated IN fig. 3) is input into a first convolution neural network (for example, CNN1 as illustrated IN fig. 3) to obtain a first characteristic map (for example, F1 as illustrated IN fig. 3); then, passing the first feature map through a plurality of fully connected layers (e.g., Fcl as illustrated in fig. 3) to obtain a first feature vector (e.g., V1 as illustrated in fig. 3); next, the image to be detected and the reference image (e.g., R1 as illustrated in fig. 3) are respectively passed through a second convolutional neural network (e.g., CNN2 as illustrated in fig. 3) to extract a second feature map (e.g., F2 as illustrated in fig. 3) from the image to be detected and a third feature map (e.g., F3 as illustrated in fig. 3) from the reference image; then, calculating an absolute value of a difference between the feature values of the corresponding positions of the second feature map and the third feature map and dividing by the feature value of the corresponding position of the third feature map to obtain a difference scale feature map (e.g., Fd as illustrated in fig. 3); then, passing the differential scale feature map through a plurality of fully connected layers (e.g., Fcl as illustrated in fig. 3) to obtain a second feature vector (e.g., V2 as illustrated in fig. 3); then, the first feature vector and the second feature vector are passed through a classifier (for example, as indicated by a circle S in fig. 3) to obtain a classification result, and the classification result is used for indicating whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
In step S110, an image to be detected is obtained, where the image to be detected is an image of the gas in the conveying process. As described above, since the visual technology can more intuitively determine the moisture content from the gas, the inventors of the present application expect to adopt the computer vision technology based on deep learning to perform feature extraction on the image of the gas in the conveying process, and classify the image based on the extracted features to determine whether the current moisture content in the gas is suitable for being directly conveyed to a recycling device for recycling. Specifically, in the embodiment of the present application, an image of the gas during transportation is first acquired by a camera as an image to be detected.
In step S120, the image to be detected is passed through a first convolutional neural network to obtain a first feature map. Namely, extracting each high-dimensional feature in the image to be detected by using a first convolution neural network.
In particular, in the embodiment of the present application, the first convolutional neural network may employ a deep residual neural network, for example, ResNet 50. It should be known to those skilled in the art that, compared to the conventional convolutional neural network, the deep residual network is an optimized network structure proposed on the basis of the conventional convolutional neural network, which mainly solves the problem of gradient disappearance during 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, the first feature map is passed through a plurality of fully connected layers to obtain a first feature vector, where the first feature vector is a feature representation of the water droplet content in the gas in a high-dimensional feature space. That is, the learned "distributed feature representation" is mapped to the sample label space through the full connectivity layer. It should be understood that, since it is not necessary to directly detect water droplets in the gas in the present application, it is not necessary to perform object detection for each object based on the first feature map, but it is sufficient to extract only feature vectors that can express the distribution characteristics of the objects.
In step S140, the image to be detected and the reference image are respectively passed through a second convolutional neural network to extract a second feature map from the image to be detected and a third feature map from the reference image, where the reference image is a gas image without water vapor, and the last layer of the second convolutional neural network is activated by a Sigmoid function. It will be appreciated that the last layer of the second convolutional neural network is activated with a sigmoid function such that the feature value for each position in the second and third feature maps is in the interval 0 to 1.
In particular, in embodiments of the present application, the second convolutional neural network may employ a deep residual neural network, e.g., ResNet 50. It is worth mentioning that the second convolutional neural network and the first convolutional neural network have the same network structure, that is, part of the weights and the hyper-parameters of the first convolutional neural network and the second convolutional neural network can be shared, so that the calculated amount in the training process is reduced, and the disappearance of the gradient is avoided. Meanwhile, the first characteristic diagram and the second characteristic diagram output by the first convolutional neural network and the second convolutional neural network with the same network structure have the same scale, so that subsequent calculation is facilitated.
In step S150, for each position in the second feature map and the third feature map, calculating an absolute value of a difference between feature values of corresponding positions of the second feature map and the third feature map and dividing the absolute value by the feature value of the corresponding position of the third feature map to obtain a differential scale feature map, where the differential scale feature map is used to represent a foreground mask of the image to be detected relative to the reference image. As will be appreciated by those skilled in the art, a foreground mask, also referred to as a foreground transparency or transparency mask, is a result of separating a foreground from a background, and is a gray scale map, where a gray scale value of each pixel represents a degree to which each pixel of an original image belongs to a foreground object, white represents a certain pixel and determines to belong to a foreground, and black represents a certain pixel and determines to belong to a background.
As described above, in the transported gas, water droplets can be directly detected as suspended particles from the image of the gas, and water vapor is mixed with the gas as a gas and cannot be detected. Therefore, in the technical scheme of the application, the gas image without water vapor is used as the reference image, and the foreground mask is extracted from the image to be detected and used as the characteristic basis for water vapor content inference.
Specifically, in this embodiment of the present application, for each position in the second feature map and the third feature map, a process of calculating an absolute value of a difference between feature values of corresponding positions of the second feature map and the third feature map and dividing the absolute value by the feature value of the corresponding position of the third feature map to obtain a differential ratio feature map includes: and performing up-sampling or down-sampling processing on the third feature map to adjust the scale of the third feature map to be consistent with the scale of the second feature map. That is, the third feature map is scaled by up-sampling or down-sampling between calculating the differential scale feature map, so that features between the second feature map and the third feature map can be aligned as much as possible, so as to improve the characterization capability of the differential scale feature map.
In step S160, the differential proportional characteristic map is passed through a plurality of fully connected layers to obtain a differential proportional characteristic map, and the second eigenvector is a characteristic representation of the moisture content in the gas in a high-dimensional characteristic space. That is, the differential scale feature map is encoded by a plurality of fully-connected layers as an encoder to generate a second feature vector corresponding to the differential scale feature map. It will be appreciated that the differential ratio profile may be converted through a plurality of fully connected layers into a second signature vector for expressing the moisture content in the gas, as the exact value of the moisture content need not be determined, only a certain ratio of the moisture content relative to a reference need be obtained. The learned distributed feature representation is mapped to a sample mark space through the full connection layer, and the obtained second feature vector can express the distribution characteristics of the water vapor object.
In step S170, the first feature vector and the second feature vector are passed through a classifier to obtain a classification result, where the classification result is used to indicate whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
Specifically, in this embodiment of the present application, the process of passing the first feature vector and the second feature vector through a classifier to obtain a classification result includes: first, the first eigenvector and the second eigenvector are concatenated, and it should be understood that the concatenation is essentially that the weights of the first eigenvector and the second eigenvector are 1, that is, the influence weights of water vapor and water drops in the gas are considered equally. And then, passing the first feature vector and the second feature vector after cascading through a classification function to obtain the classification result. It should be understood that the classification function is a Softmax classification function, and the first characteristic vector and the second characteristic vector are cascaded and then pass through the Softmax classification function, so that the influence of water vapor and water drops in the gas is comprehensively considered in the classification, and the classification result is more accurate.
Fig. 4 is a flowchart illustrating that the first feature vector and the second feature vector are passed through a classifier to obtain a classification result in the gas intelligent detection method based on the double-branch inference mechanism according to the embodiment of the present application, and as shown in fig. 4, the first feature vector and the second feature vector are passed through a classifier to obtain a classification result, including: s210, cascading the first feature vector and the second feature vector; and S220, the first feature vector and the second feature vector after the cascade connection are processed through a classification function to obtain the classification result.
It should be noted that in other examples of the present application, the first feature vector and the second feature vector may also be passed through a classifier in other ways to obtain the classification result, for example, in another example of the present application, the process of passing the first feature vector and the second feature vector through a classifier to obtain the classification result includes: first, a position-weighted sum of the first eigenvector and the second eigenvector is calculated with a preset weight to obtain a classification eigenvector, that is, influence weights of water vapor and water droplets in the gas are unequally considered. In particular, the preset weights may be involved in the training process as hyper-parameters. The classified feature vector is then passed through a Softmax classification function to obtain the result.
Fig. 5 is a flowchart illustrating that the first feature vector and the second feature vector are passed through a classifier to obtain a classification result in a gas intelligent detection method based on a double-branch inference mechanism according to an embodiment of the present application, and as shown in fig. 5, the first feature vector and the second feature vector are passed through a classifier to obtain a classification result, including: s310, calculating a position-weighted sum of the first feature vector and the second feature vector according to preset weight to obtain a classified feature vector; and S320, passing the classification feature vector through a classification function to obtain the result.
In summary, the intelligent gas detection method based on the dual-branch inference mechanism according to the embodiment of the present application is elucidated, and the computer vision technology based on deep learning is adopted to perform feature extraction and classification on the image of the gas in the conveying process, so as to determine whether the water vapor content in the current gas is suitable for being directly conveyed to a recycling device for recycling. Particularly, in the detection process, a first branch for deducing the water drop content and a second branch for deducing the water vapor content are adopted, wherein in the second branch for deducing the water vapor content, a gas image without water vapor is adopted as a reference image, and a foreground mask is extracted from the detection image to be used as a characteristic basis for deducing the water vapor content, so that the influence of the water vapor and the water drops in the gas is comprehensively considered in the classification process, and the classification precision is improved.
Exemplary System
Fig. 6 illustrates a block diagram of a gas intelligent detection system based on a double-branch inference mechanism according to an embodiment of the present application.
As shown in fig. 6, the gas intelligent detection system 600 based on the dual-branch inference mechanism according to the embodiment of the present application includes: the image acquiring unit 610 to be detected is used for acquiring an image to be detected, wherein the image to be detected is an image of gas in a conveying process; a first feature map generating unit 620, configured to pass the image to be detected obtained by the image to be detected obtaining unit 610 through a first convolutional neural network to obtain a first feature map; a first feature vector generation unit 630, configured to pass the first feature map obtained by the first feature map generation unit 620 through a plurality of fully-connected layers to obtain a first feature vector, where the first feature vector is a feature representation of the water droplet content in the gas water in a high-dimensional feature space; the feature map extraction unit 640 is configured to respectively pass the to-be-detected image and the reference image obtained by the to-be-detected image obtaining unit 610 through a second convolutional neural network to extract a second feature map from the to-be-detected image and a third feature map from the reference image, where the reference image is a gas image without water vapor, and a last layer of the second convolutional neural network is activated by a Sigmoid function; a difference scale feature map generation unit 650, configured to calculate, for each position in the second feature map and the third feature map obtained by the feature map extraction unit 640, an absolute value of a difference between feature values of corresponding positions of the second feature map and the third feature map and divide the absolute value by the feature value of the corresponding position of the third feature map to obtain a difference scale feature map, where the difference scale feature map is used to represent a foreground mask of the image to be detected relative to the reference image; a second eigenvector generating unit 660, configured to pass the differential proportion characteristic map obtained by the differential proportion characteristic map generating unit 650 through a plurality of fully-connected layers to obtain a second eigenvector, where the second eigenvector is a characteristic representation of the water vapor content in the gas in a high-dimensional characteristic space; and a classification result generating unit 670, configured to pass the first feature vector obtained by the first feature vector generating unit 630 and the second feature vector obtained by the second feature vector generating unit 660 through a classifier to obtain a classification result, where the classification result is used to indicate whether the gas in the image to be detected is suitable for being directly delivered to a recycling device for recycling.
In one example, in the above-described intelligent detection system 600, the first convolutional neural network and the second convolutional neural network have the same network structure.
In an example, in the above-mentioned smart detection system 600, the differential scale characteristic map generating unit 650 is further configured to: and performing up-sampling or down-sampling processing on the third feature map to adjust the scale of the third feature map to be consistent with the scale of the second feature map.
In one example, in the above-mentioned intelligent detection system 600, as shown in fig. 7, the classification result generating unit 670 includes: a concatenation subunit 671 for concatenating the first eigenvector with the second eigenvector; and a first classification subunit 672, configured to pass the concatenated first feature vector and the concatenated second feature vector through a classification function to obtain the classification result.
In another example, in the above-mentioned smart detection system 600, as shown in fig. 8, the classification result generating unit 670 includes: a classification feature vector generation subunit 673, configured to calculate a position-weighted sum of the first feature vector and the second feature vector with a preset weight to obtain a classification feature vector; and a second classification subunit 674 configured to pass the classification feature vector obtained by the classification feature vector generation subunit 673 through a classification function to obtain the result.
In one example, in the above-described intelligent detection system 600, the first convolutional neural network and the second convolutional neural network are deep residual error networks.
Here, it can be understood by those skilled in the art that the detailed functions and operations of the respective units and modules in the above-described smart detection system 600 have been described in detail in the above description of the gas smart detection method based on the dual branch inference mechanism with reference to fig. 1 to 5, and thus, a repetitive description thereof will be omitted.
As described above, the smart detection system 600 according to the embodiment of the present application may be implemented in various terminal devices, such as a server for smart detection of gas, and the like. In one example, the smart detection system 600 according to the embodiment of the present application may be integrated into a terminal device as a software module and/or a hardware module. For example, the smart detection 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 smart detection system 600 may also be one of many hardware modules of the terminal device.
Alternatively, in another example, the smart detection system 600 and the terminal device may be separate devices, and the smart detection 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. 9.
FIG. 9 illustrates a block diagram of an electronic device in accordance with an embodiment of the present application.
As shown in fig. 9, the electronic device 10 includes one or more processors 11 and a 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 system 13 and an output system 14, which are interconnected by a bus system and/or other form of connection mechanism (not shown).
The input system 13 may comprise, for example, a keyboard, a mouse, etc.
The output system 14 may output various information including classification results and the like to the outside. The output system 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. 9, 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 functions of the dual-branch inference mechanism-based gas intelligent detection method 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 intelligent gas detection method based on a two-branch inference mechanism 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, or device, or any combination 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 gas intelligent detection method based on a double-branch inference mechanism is characterized by comprising the following steps:
acquiring an image to be detected, wherein the image to be detected is an image of the gas in the conveying process;
passing the image to be detected through a first convolutional neural network to obtain a first characteristic diagram;
passing the first feature map through a plurality of fully-connected layers to obtain a first feature vector, wherein the first feature vector is a feature representation of the water droplet content in the gas in a high-dimensional feature space;
respectively enabling the image to be detected and the reference image to pass through a second convolutional neural network so as to extract a second characteristic diagram from the image to be detected and a third characteristic diagram from the reference image, wherein the reference image is a gas image without water vapor, and the last layer of the second convolutional neural network is activated by a Sigmoid function;
for each position in the second feature map and the third feature map, calculating an absolute value of a difference value between feature values of corresponding positions of the second feature map and the third feature map and dividing the absolute value by the feature value of the corresponding position of the third feature map to obtain a differential proportion feature map, wherein the differential proportion feature map is used for representing a foreground mask of the image to be detected relative to the reference image;
passing the differential proportion characteristic diagram through a plurality of fully-connected layers to obtain a second characteristic vector, wherein the second characteristic vector is a characteristic representation of the water vapor content in the gas in a high-dimensional characteristic space; and
and enabling the first characteristic vector and the second characteristic vector to pass through a classifier so as to obtain a classification result, wherein the classification result is used for representing whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
2. The intelligent gas detection method based on the double-branch inference mechanism according to claim 1, wherein the first convolutional neural network and the second convolutional neural network have the same network structure.
3. The intelligent gas detection method based on the double-branch inference mechanism according to claim 2, wherein for each position in the second feature map and the third feature map, calculating an absolute value of a difference between feature values of corresponding positions of the second feature map and the third feature map and dividing by the feature value of corresponding position of the third feature map to obtain a differential scale feature map, comprising:
and performing up-sampling or down-sampling processing on the third feature map to adjust the scale of the third feature map to be consistent with the scale of the second feature map. .
4. The intelligent gas detection method based on the double-branch inference mechanism according to claim 1, wherein the step of passing the first feature vector and the second feature vector through a classifier to obtain a classification result comprises:
concatenating the first feature vector with the second feature vector; and
and passing the first feature vector and the second feature vector after cascading through a classification function to obtain the classification result.
5. The intelligent gas detection method based on the dual-branch inference mechanism as claimed in claim, wherein passing the first feature vector and the second feature vector through a classifier to obtain a classification result comprises:
calculating a position-weighted sum of the first feature vector and the second feature vector with a preset weight to obtain a classified feature vector; and
and passing the classified feature vector through a classification function to obtain the result.
6. The intelligent gas detection method based on the double-branch inference mechanism according to claim 2, wherein the first convolutional neural network and the second convolutional neural network are deep residual error networks.
7. The utility model provides a gaseous intellectual detection system of gas based on two branch inference mechanisms which characterized in that includes:
the device comprises an image acquisition unit to be detected, a data acquisition unit and a data processing unit, wherein the image acquisition unit to be detected is used for acquiring an image to be detected, and the image to be detected is an image of the gas in the conveying process;
the first characteristic diagram generating unit is used for enabling the image to be detected obtained by the image to be detected obtaining unit to pass through a first convolutional neural network so as to obtain a first characteristic diagram;
a first feature vector generation unit, configured to pass the first feature map obtained by the first feature map generation unit through a plurality of fully-connected layers to obtain a first feature vector, where the first feature vector is a feature representation of water droplet content in the gas in a high-dimensional feature space;
the characteristic diagram extraction unit is used for respectively enabling the image to be detected and the reference image obtained by the image to be detected acquisition unit to pass through a second convolutional neural network so as to extract a second characteristic diagram from the image to be detected and a third characteristic diagram from the reference image, wherein the reference image is a gas image without water vapor, and the last layer of the second convolutional neural network is activated by a Sigmoid function;
a difference proportion feature map generation unit, configured to calculate, for each position in the second feature map and the third feature map obtained by the feature map extraction unit, an absolute value of a difference between feature values of corresponding positions of the second feature map and the third feature map and divide the absolute value by the feature value of the corresponding position of the third feature map to obtain a difference proportion feature map, where the difference proportion feature map is used to represent a foreground mask of the image to be detected with respect to the reference image;
the second eigenvector generating unit is used for enabling the differential proportion characteristic diagram obtained by the differential proportion characteristic diagram generating unit to pass through a plurality of full-connection layers so as to obtain a second eigenvector, and the second eigenvector is the characteristic representation of the water vapor content in the gas in a high-dimensional characteristic space;
and the classification result generating unit is used for enabling the first characteristic vector obtained by the first characteristic vector generating unit and the second characteristic vector obtained by the second characteristic vector generating unit to pass through a classifier so as to obtain a classification result, and the classification result is used for indicating whether the gas in the image to be detected is suitable for being directly conveyed to a recycling device for recycling.
8. The intelligent gas detection system based on the dual-branch inference mechanism as claimed in claim 7, wherein the classification result generation unit comprises:
a concatenation subunit, configured to concatenate the first feature vector and the second feature vector; and
and the first classification subunit is used for enabling the first feature vector and the second feature vector after the cascade connection to pass through a classification function so as to obtain the classification result.
9. The intelligent gas detection system based on the dual-branch inference mechanism as claimed in claim 7, wherein the classification result generation unit comprises:
a classification feature vector generation subunit, configured to calculate a position-weighted sum of the first feature vector and the second feature vector with a preset weight to obtain a classification feature vector; and
and the second classification subunit is used for enabling the classification feature vector obtained by the classification feature vector generation subunit to pass through a classification function so as to obtain the result.
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 dual branch inference mechanism based gas intelligent detection method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110050001.7A CN112767342A (en) | 2021-01-14 | 2021-01-14 | Intelligent gas detection method based on double-branch inference mechanism |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110050001.7A CN112767342A (en) | 2021-01-14 | 2021-01-14 | Intelligent gas detection method based on double-branch inference mechanism |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112767342A true CN112767342A (en) | 2021-05-07 |
Family
ID=75700583
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110050001.7A Withdrawn CN112767342A (en) | 2021-01-14 | 2021-01-14 | Intelligent gas detection method based on double-branch inference mechanism |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112767342A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115526865A (en) * | 2022-09-30 | 2022-12-27 | 深圳市创瑞鑫科技有限公司 | Insulation testing method and system for heat dissipation module of notebook computer |
| CN115901794A (en) * | 2023-02-17 | 2023-04-04 | 广州达普绅智能设备有限公司 | System and method for detecting bottle opening flaws through strip-shaped light source |
-
2021
- 2021-01-14 CN CN202110050001.7A patent/CN112767342A/en not_active Withdrawn
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115526865A (en) * | 2022-09-30 | 2022-12-27 | 深圳市创瑞鑫科技有限公司 | Insulation testing method and system for heat dissipation module of notebook computer |
| CN115901794A (en) * | 2023-02-17 | 2023-04-04 | 广州达普绅智能设备有限公司 | System and method for detecting bottle opening flaws through strip-shaped light source |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ahmad et al. | Object detection through modified YOLO neural network | |
| CN115375691B (en) | Image-based semiconductor diffusion paper source defect detection system and method thereof | |
| CN115203380A (en) | Text processing system and method based on multi-mode data fusion | |
| CN115761642A (en) | Image processing-based crushing operation monitoring method and system | |
| CN112508041A (en) | Training method of neural network for spray control based on classification result label | |
| CN112767342A (en) | Intelligent gas detection method based on double-branch inference mechanism | |
| Alsanad et al. | YOLO-V3 based real-time drone detection algorithm | |
| CN115456789B (en) | Abnormal transaction detection method and system based on transaction pattern recognition | |
| CN115620303B (en) | Personnel file intelligent management system | |
| CN116247824B (en) | Control method and system for power equipment | |
| CN115082745A (en) | Image-based cable strand quality detection method and system | |
| CN116152611B (en) | Multistage multi-scale point cloud completion method, system, equipment and storage medium | |
| CN117077075A (en) | Water quality monitoring system and method for environmental protection | |
| CN111611386A (en) | Text classification method and device | |
| CN115471216A (en) | Data management method of intelligent laboratory management platform | |
| CN113971750A (en) | Key information extraction method, device, equipment and storage medium for bank receipt | |
| CN112960213A (en) | Intelligent package quality detection method using characteristic probability distribution representation | |
| CN113255752A (en) | Solid material consistency sorting method based on feature clustering | |
| CN112733751A (en) | Intelligent water inlet control method for waste heat recovery | |
| CN112733853A (en) | Method for detecting separation and associated mineral separation effect based on foreground and background | |
| CN116342390A (en) | Fluorescent pigment composition and preparation method thereof | |
| CN111340139A (en) | Method and device for judging complexity of image content | |
| CN113139520B (en) | Equipment diaphragm performance monitoring method for industrial Internet | |
| CN112800995A (en) | Intelligent particle size detection method using multi-scale feature weighting | |
| CN112819040A (en) | Intelligent control method for cement processing based on weight generation mechanism |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210507 |