CN118122723A - Pipeline cleaning method, device and equipment based on convolutional neural network - Google Patents
Pipeline cleaning method, device and equipment based on convolutional neural network Download PDFInfo
- Publication number
- CN118122723A CN118122723A CN202410551474.9A CN202410551474A CN118122723A CN 118122723 A CN118122723 A CN 118122723A CN 202410551474 A CN202410551474 A CN 202410551474A CN 118122723 A CN118122723 A CN 118122723A
- Authority
- CN
- China
- Prior art keywords
- pipeline
- neural network
- convolutional neural
- preset
- pipeline image
- 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
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 161
- 238000004140 cleaning Methods 0.000 title claims abstract description 95
- 238000000034 method Methods 0.000 title claims abstract description 61
- 230000008569 process Effects 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims description 113
- 238000011176 pooling Methods 0.000 claims description 64
- 238000012549 training Methods 0.000 claims description 57
- 230000006870 function Effects 0.000 claims description 29
- 238000007781 pre-processing Methods 0.000 claims description 25
- 238000000605 extraction Methods 0.000 claims description 14
- 238000013519 translation Methods 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 6
- 238000007689 inspection Methods 0.000 abstract description 4
- 238000004590 computer program Methods 0.000 description 18
- 238000010586 diagram Methods 0.000 description 8
- 230000009466 transformation Effects 0.000 description 4
- 239000000306 component Substances 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000010408 sweeping Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004205 output neuron Anatomy 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The application discloses a pipeline cleaning method, a device and equipment based on a convolutional neural network, which can automatically process the current pipeline image through a pre-trained preset convolutional neural network model, thereby realizing the automatic discovery of a cleaning scene and the automatic full-automatic operation of cleaning by a pipeline cleaning robot; simultaneously, effectively combine the degree of depth learning technique and pipeline cleaning scene, can realize effectively discernment pipeline in need clean the scene, improve the automation level and the operating efficiency that pipeline maintained, reduce the risk of manual inspection simultaneously.
Description
Technical Field
The present application relates to the field of robotics, and in particular, to a method, an apparatus, and a device for cleaning a pipeline based on a convolutional neural network.
Background
With the continuous development of science and technology in China, the demand for electric power is rapidly increased, and the urban electric power transmission form is changed from an overhead transmission line to underground cable laying, so that a large amount of overground space can be saved, and urban construction can be beautified. The main mode of underground cable laying is calandria cable laying, because factors such as construction mode and cable itself long-time high load operation, underground cable easily produces a large amount of heat and can not give off in the calandria, shortens cable life, and even easily takes place the conflagration, and cable duct environment is complicated even, and the pipeline surface is grey easily, so need to wash and maintain cable and pipeline.
Currently, during the construction and maintenance of pipelines, the periodic cleaning process relies on manual inspection and judgment, which is time-consuming and labor-consuming, and can present a safety risk. Therefore, how to safely and efficiently clean the pipeline is a problem to be solved.
Disclosure of Invention
The application aims at least solving the technical problems existing in the prior art, and therefore, the first aspect of the application provides a pipeline cleaning method based on a convolutional neural network, which comprises the following steps:
Acquiring a current pipeline image acquired by a pipeline cleaning robot;
Inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result; the method comprises the steps that a preset convolutional neural network model is generated based on preset training samples, wherein the preset training samples comprise a preset number of pipeline image samples and marking information, and each pipeline image sample is generated by carrying out data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample;
and controlling the pipeline cleaning robot to clean the pipeline according to the processing result.
In one possible implementation manner, a construction process of the convolutional neural network model is preset, including:
Acquiring a preset number of pipeline image samples and marking information corresponding to each pipeline image sample;
inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a preset convolutional neural network model.
In one possible embodiment, obtaining a preset number of pipeline image samples includes:
acquiring a preset number of original pipeline images and preset marking information;
For each original pipeline image, carrying out data preprocessing on the original pipeline image to generate a preprocessed pipeline image; the data preprocessing at least comprises format conversion processing and normalization processing;
Carrying out data enhancement processing on the preprocessed pipeline image to generate a pipeline image sample; the data enhancement processing at least comprises rotation processing, translation processing and scaling processing.
In one possible implementation manner, inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a preset convolutional neural network model, including:
Inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a prediction result;
substituting the prediction result and the marking information corresponding to each pipeline image sample into a preset loss function, and calculating the value of the loss function;
and adjusting parameters of the initial convolutional neural network model based on the value of the loss function until a preset convergence condition is reached, and generating a preset convolutional neural network model.
In one possible implementation manner, the initial convolutional neural network model includes an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer and a full-connection layer, and the method includes inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into the initial convolutional neural network model for training, so as to generate a prediction result, including:
Receiving a preset number of pipeline image samples and marking information corresponding to each pipeline image sample through an input layer, inputting the received data into a first convolution layer, performing feature extraction on the pipeline image samples through the first convolution layer, generating first feature data, and inputting the first feature data into a first pooling layer;
The first characteristic data is subjected to downsampling treatment through a first pooling layer, second characteristic data subjected to maximum pooling is generated, and the second characteristic data is input into a second convolution layer;
Performing feature extraction on the second feature data through the second convolution layer to generate third feature data, and inputting the third feature data into the second pooling layer;
The third characteristic data is subjected to downsampling treatment through the second pooling layer, fourth characteristic data subjected to maximum pooling is generated, and the fourth characteristic data is input into the full-connection layer;
And processing the fourth characteristic data through the full connection layer to generate a prediction result.
In one possible implementation manner, a preset number of pipeline image samples and marking information corresponding to each pipeline image sample are input into an initial convolutional neural network model for training, and a preset convolutional neural network model is generated, and the method further includes:
Inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating an intermediate convolutional neural network model;
and evaluating and optimizing the middle convolutional neural network model according to the preset verification set to generate a preset convolutional neural network model.
In one possible embodiment, the processing result includes an output value, and controlling the pipe cleaning robot to perform pipe cleaning according to the processing result includes:
and if the output value meets the preset condition, controlling the pipeline cleaning robot to clean the pipeline according to the output value.
The second aspect of the present application proposes a pipe cleaning device based on a convolutional neural network, the device comprising:
the acquisition module is used for acquiring the current pipeline image acquired by the pipeline cleaning robot;
The processing module is used for inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result; the method comprises the steps that a preset convolutional neural network model is generated based on preset training samples, wherein the preset training samples comprise a preset number of pipeline image samples and marking information, and each pipeline image sample is generated by carrying out data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample;
and the control module is used for controlling the pipeline cleaning robot to clean the pipeline according to the processing result.
In one possible embodiment, the pipe cleaning device based on a convolutional neural network is further used for:
Acquiring a preset number of pipeline image samples and marking information corresponding to each pipeline image sample;
inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a preset convolutional neural network model.
In one possible embodiment, the pipe cleaning device based on a convolutional neural network is further used for:
acquiring a preset number of original pipeline images and preset marking information;
For each original pipeline image, carrying out data preprocessing on the original pipeline image to generate a preprocessed pipeline image; the data preprocessing at least comprises format conversion processing and normalization processing;
Carrying out data enhancement processing on the preprocessed pipeline image to generate a pipeline image sample; the data enhancement processing at least comprises rotation processing, translation processing and scaling processing.
In one possible embodiment, the pipe cleaning device based on a convolutional neural network is further used for:
Inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a prediction result;
substituting the prediction result and the marking information corresponding to each pipeline image sample into a preset loss function, and calculating the value of the loss function;
and adjusting parameters of the initial convolutional neural network model based on the value of the loss function until a preset convergence condition is reached, and generating a preset convolutional neural network model.
In one possible implementation, the initial convolutional neural network model includes an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, and a full-connection layer, and the pipe cleaning device based on the convolutional neural network is further configured to:
Receiving a preset number of pipeline image samples and marking information corresponding to each pipeline image sample through an input layer, inputting the received data into a first convolution layer, performing feature extraction on the pipeline image samples through the first convolution layer, generating first feature data, and inputting the first feature data into a first pooling layer;
The first characteristic data is subjected to downsampling treatment through a first pooling layer, second characteristic data subjected to maximum pooling is generated, and the second characteristic data is input into a second convolution layer;
Performing feature extraction on the second feature data through the second convolution layer to generate third feature data, and inputting the third feature data into the second pooling layer;
The third characteristic data is subjected to downsampling treatment through the second pooling layer, fourth characteristic data subjected to maximum pooling is generated, and the fourth characteristic data is input into the full-connection layer;
And processing the fourth characteristic data through the full connection layer to generate a prediction result.
In one possible embodiment, the pipe cleaning device based on a convolutional neural network is further used for:
Inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating an intermediate convolutional neural network model;
and evaluating and optimizing the middle convolutional neural network model according to the preset verification set to generate a preset convolutional neural network model.
In one possible implementation manner, the processing result includes an output value, and the control module is specifically configured to:
and if the output value meets the preset condition, controlling the pipeline cleaning robot to clean the pipeline according to the output value.
A third aspect of the present application proposes an electronic device comprising a processor and a memory, the memory storing at least one instruction, at least one program, a set of codes or a set of instructions, the at least one instruction, the at least one program, the set of codes or the set of instructions being loaded and executed by the processor to implement the convolutional neural network based pipe cleaning method as described in the first aspect.
A fourth aspect of the present application proposes a computer readable storage medium having stored therein at least one instruction, at least one program, a set of codes or a set of instructions, the at least one instruction, the at least one program, the set of codes or the set of instructions being loaded and executed by a processor to implement the convolutional neural network based pipe cleaning method as described in the first aspect.
The embodiment of the application has the following beneficial effects:
the pipeline cleaning method based on the convolutional neural network provided by the embodiment of the application comprises the following steps: the method comprises the steps of obtaining a current pipeline image collected by a pipeline cleaning robot, inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result, wherein the preset convolutional neural network model is generated based on preset training samples, the preset training samples comprise a preset number of pipeline image samples and marking information, each pipeline image sample is generated by carrying out data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample, and the pipeline cleaning robot is controlled to conduct pipeline cleaning according to the processing result. According to the scheme, the current pipeline image can be automatically processed through a pre-trained preset convolutional neural network model, so that the pipeline cleaning robot can automatically find a cleaning scene and automatically perform full-automatic cleaning operation; simultaneously, effectively combine the degree of depth learning technique and pipeline cleaning scene, can realize effectively discernment pipeline in need clean the scene, improve the automation level and the operating efficiency that pipeline maintained, reduce the risk of manual inspection simultaneously.
Drawings
FIG. 1 is a block diagram of a computer device according to an embodiment of the present application;
FIG. 2 is a flow chart of steps of a pipeline cleaning method based on a convolutional neural network according to an embodiment of the present application;
FIG. 3 is a flowchart illustrating steps for constructing a model of a convolutional neural network according to an embodiment of the present application;
FIG. 4 is a flowchart illustrating steps for obtaining a pipeline image sample and marking information according to an embodiment of the present application;
FIG. 5 is a flowchart illustrating steps for training to generate a predetermined convolutional neural network model according to an embodiment of the present application;
FIG. 6 is a schematic diagram of the overall structure of an initial convolutional neural network model according to an embodiment of the present application;
FIG. 7 is a flowchart illustrating steps for model optimization according to an embodiment of the present application;
Fig. 8 is a block diagram of a pipeline cleaning device based on a convolutional neural network according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The terms "first" and "second" are used below for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the embodiments of the present disclosure, unless otherwise indicated, the meaning of "a plurality" is two or more. In addition, the use of "based on" or "according to" is intended to be open and inclusive in that a process, step, calculation, or other action "based on" or "according to" one or more of the stated conditions or values may in practice be based on additional conditions or beyond the stated values.
The pipeline cleaning method based on the convolutional neural network can be applied to computer equipment (electronic equipment), wherein the computer equipment can be a server or a terminal, the server can be one server or a server cluster consisting of a plurality of servers, the embodiment of the application is not particularly limited to the embodiment, and the terminal can be but not limited to various personal computers, notebook computers, smart phones, tablet computers and portable formula wearable equipment.
Taking the example of a computer device being a server, FIG. 1 illustrates a block diagram of a server, as shown in FIG. 1, which may include a processor and memory connected by a system bus. Wherein the processor of the server is configured to provide computing and control capabilities. The memory of the server includes nonvolatile storage medium and internal memory. The nonvolatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The computer program when executed by a processor implements a pipeline cleaning method based on a convolutional neural network.
It will be appreciated by those skilled in the art that the structure shown in fig. 1 is merely a block diagram of a portion of the structure associated with the present inventive arrangements and does not constitute a limitation of the servers to which the present inventive arrangements are applied, alternatively the servers may include more or less components than those shown, or may combine certain components, or have different arrangements of components.
The execution subject of the embodiment of the present application may be a computer device, or may be a pipe cleaning device based on a convolutional neural network, and in the following method embodiment, the execution subject is described with reference to the computer device.
Fig. 2 is a flowchart of steps of a pipeline cleaning method based on a convolutional neural network according to an embodiment of the present application. As shown in fig. 2, the method comprises the steps of:
step 202, acquiring a current pipeline image acquired by the pipeline cleaning robot.
The pipeline related to the application can be a cable pipeline, and the pipeline cleaning robot can comprise a travelling mechanism, a visual mechanical arm device, a cleaning and sweeping disc device, a cleaning and dust collecting device and an image collecting module. Specifically, the walking mechanism can be used as a core, other modules can be assembled and disassembled, and the cleaning disc sweeping device and the cleaning dust collection device can share the same mounting seat.
The image acquisition module can be cameras, and in addition, the number, the installation position and the type of the cameras are not particularly limited, so long as the current pipeline image can be acquired.
Therefore, when the pipeline cleaning is performed, the current pipeline image can be automatically acquired through the pipeline cleaning robot, or the current pipeline image can be acquired according to the received acquisition instruction, and the embodiment of the application is not limited in particular. Alternatively, the pipe cleaning robot may be acquired in real time as the current pipe image is acquired.
And 204, inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result.
The preset convolutional neural network model is generated based on preset training samples, wherein the preset training samples can comprise a preset number of pipeline image samples and marking information, and each pipeline image sample is generated by performing data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample.
In some optional embodiments, the preset convolutional neural network model needs to be built in advance, and when the preset convolutional neural network model is built, as shown in fig. 3, fig. 3 is a flowchart of a step of building the preset convolutional neural network model according to an embodiment of the present application, where the step includes:
Step 302, obtaining a preset number of pipeline image samples and marking information corresponding to each pipeline image sample.
In some optional embodiments, as shown in fig. 4, fig. 4 is a flowchart of a step of obtaining a pipeline image sample and marking information according to an embodiment of the present application, including:
Step 402, obtaining a preset number of original pipeline images and preset marking information.
When the pipeline cleaning is performed, the pipeline cleaning robot can perform on-site image shooting, so that a preset number of original pipeline images are obtained, wherein the original pipeline images can comprise an image serving as a positive example and an image serving as a negative example. The image as the positive example is an original pipe image for which cleaning is confirmed, and the image as the negative example is an image for which cleaning is confirmed not to be required.
Then, the marking information corresponding to the image serving as the positive example can be preset to be 1, and the marking information corresponding to the image serving as the negative example can be preset to be 0, so that the marking information corresponding to all the original pipeline images can be determined.
Step 404, for each original pipeline image, performing data preprocessing on the original pipeline image to generate a preprocessed pipeline image.
The data preprocessing at least includes format conversion processing and normalization processing, and may also include other types of data preprocessing modes, which are not limited in particular in the embodiment of the present application.
Alternatively, for each original pipeline image, the image file may be read first, and then these pixel grids may be converted into floating point number tensors, that is, subjected to format conversion processing.
Then, the pixel value (within the range of 0 to 255) can be scaled to the [0, 1] interval, i.e., normalized. Alternatively, normalization processing may be performed by formula (1), so that the data may be scaled to the [0, 1] interval.
（1）
Wherein X refers to the actual value of a set of data; the data of subscripts min and max are respectively the maximum value and the minimum value in the dimension data; the data of the subscript norm are the data obtained after normalization processing.
And 406, performing data enhancement processing on the preprocessed pipeline image to generate a pipeline image sample.
In deep learning, when the number of training samples is too small, the model is often easy to be over-fitted, and in order to solve the over-fitting problem, data enhancement can be used to preprocess existing data. Data enhancement is the generation of more training data from existing training samples by increasing the number of samples using a variety of random transformations that can generate a trusted image. The data enhancement processing at least includes rotation processing, translation processing and scaling processing, and may also include other types of data enhancement processing modes, which are not limited in particular by the embodiment of the present application.
The rotation process is to rotate the image by a certain angle around its center point. For a two-dimensional image, the rotation process can be realized by a transformation matrix shown in formula (2).
（2）
Where θ is the rotation angle and R (θ) is the rotation matrix.
The translation process is to move the image in the horizontal or vertical direction by a certain distance, and for each pixel point (x, y) in the image, the translation process can be implemented by the transformation shown in formula (3).
（3）
Wherein T (tx, ty) represents a coordinate point after the translation process; tx is the translation distance in the x direction; and ty is the translation distance in the y-direction.
The scaling process is to change the size of the image, and for each pixel point (x, y) in the image, the scaling process can be implemented by the transformation shown in formula (4).
（4）
Wherein S (sx, sy) represents a coordinate point after the scaling process; sx and sy are scaling factors in the horizontal and vertical directions, respectively.
Step 304, inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a preset convolutional neural network model.
After a preset number of pipeline image samples and marking information corresponding to each pipeline image sample are obtained, the preset number of pipeline image samples and marking information corresponding to each pipeline image sample can be input into an initial convolutional neural network model for training, and a preset convolutional neural network model is generated.
In some alternative embodiments, as shown in fig. 5, fig. 5 is a flowchart of a step of training to generate a preset convolutional neural network model according to an embodiment of the present application, including:
Step 502, inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a prediction result.
And step 504, substituting the prediction result and the marking information corresponding to each pipeline image sample into a preset loss function, and calculating the value of the loss function.
And step 506, adjusting parameters of the initial convolutional neural network model based on the value of the loss function until a preset convergence condition is reached, and generating a preset convolutional neural network model.
In order to process the image data, a convolutional neural network may be used to design the model. The convolutional neural network may comprise a plurality of convolutional layers and a pooling layer, and the result is finally output through a full connection layer. Convolutional layers are a core component of convolutional neural networks in deep learning, and play an important role in image processing and visual recognition tasks, among other things. The main purpose of the convolution layer is to extract features of the input data (e.g., image) by a convolution operation. For image processing, a discrete convolution operation may be performed using equation (5).
（5）
Where I is the input image, K is the convolution kernel, (I, j) is the position coordinates of the output feature map, (m, n) is the position coordinates in the input image, (I x K) (I, j) is the output value at the (I, j) position after convolution.
The max-pooling layer is a downsampling operation commonly used in convolutional neural networks to reduce the spatial size of the data, thereby reducing the number of parameters and computation while maintaining important characteristic information. The max-pooling layer typically follows the convolution layer, helping to spatially invariance the extracted features, making the model more robust to small positional variations of the input data. The data after the convolution layer processing can be subjected to the maximum pooling processing by adopting the formula (6).
（6）
Wherein I is the characteristic diagram of the input,Representing the feature map output after maximum pooling, (i, j) is the position coordinates of the output feature map, m and n are coordinates sliding within the pooling window,/>Indicating that the maximum value is taken within a given window.
In some alternative embodiments, the design of the deep convolutional neural network model may be performed using multiple convolutional layers+pooled layers, including multiple convolutional layers, active layers, pooled layers, and fully-connected layers. The convolution layer is used for extracting local features in the image, the pooling layer is used for reducing the space dimension of the features, and the full connection layer is used for final classification judgment. Thus, the initial convolutional neural network model may include an input layer, a first convolutional layer, a first pooled layer, a second convolutional layer, a second pooled layer, and a fully-connected layer. As shown in fig. 6, fig. 6 is a schematic overall structure diagram of an initial convolutional neural network model according to an embodiment of the present application, where convolutional layer 1 is a first convolutional layer, pooled layer 1 is a first pooled layer, convolutional layer 2 is a second convolutional layer, and pooled layer 2 is a second pooled layer.
Wherein, optionally, the input layer may receive RGB images of 224x224x3 in size. The first convolution layer may use 32 filters of 3x3, with a step size of 1, and an activation function of ReLU. The first pooling layer may perform maximum pooling using a window of 2x2, with a stride of 2. The second convolution layer may use 64 3x3 filters, with a step size of 1, and an activation function of ReLU. The second pooling layer may perform maximum pooling using a window of 2x2, with a stride of 2. The fully connected layer can flatten the output of the last pooling layer and pass through a fully connected layer with two output neurons.
Therefore, when model training is performed, a preset number of pipeline image samples and marking information corresponding to each pipeline image sample can be input into an initial convolutional neural network model for training, and a prediction result is generated. Specifically, a preset number of pipeline image samples and marking information corresponding to each pipeline image sample may be received through an input layer, the received data is input into a first convolution layer, feature extraction is performed on the pipeline image samples through the first convolution layer, first feature data is generated, and the first feature data is input into a first pooling layer. And carrying out downsampling processing on the first characteristic data through the first pooling layer, generating second characteristic data subjected to maximum pooling, and inputting the second characteristic data into the second convolution layer. And performing feature extraction on the second feature data through the second convolution layer to generate third feature data, and inputting the third feature data into the second pooling layer. And carrying out downsampling treatment on the third characteristic data through the second pooling layer, generating fourth characteristic data subjected to maximum pooling, and inputting the fourth characteristic data into the full-connection layer. And processing the fourth characteristic data through the full connection layer to generate a prediction result.
Then, the prediction result and the marking information corresponding to each pipeline image sample can be substituted into a preset loss function, and the value of the loss function is calculated. Alternatively, a two-class cross entropy loss function may be employed as a preset loss function optimization target, whereby the value loss of the loss function is calculated by equation (7).
（7）
Wherein y is the marking information corresponding to each pipeline image sample,And (5) predicting results corresponding to the image samples of each pipeline.
Finally, parameters of the initial convolutional neural network model can be adjusted based on the value of the loss function until a preset convergence condition is reached, so that optimal model parameters are generated, and the preset convolutional neural network model is generated based on the optimal model parameters. Alternatively, a random gradient descent method or variants thereof, such as Adam optimizers, may be employed to update weights to adjust parameters of the initial convolutional neural network model.
In some alternative embodiments, as shown in fig. 7, fig. 7 is a flowchart of steps for model optimization according to an embodiment of the present application, including:
Step 702, inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating an intermediate convolutional neural network model.
And step 704, evaluating and optimizing the middle convolutional neural network model according to the preset verification set to generate a preset convolutional neural network model.
The method comprises the steps of inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating an intermediate convolutional neural network model. The model performance of the intermediate convolutional neural network model may then be evaluated on the validation set, and optionally the F1 value may be used to make a determination of the model performance, as shown in equation (8).
（8）
Therefore, the super parameters of the network structure can be adjusted according to the evaluation result on the verification set, the model is further trained, the model performance is optimized, and the preset convolutional neural network model is finally generated.
And 206, controlling the pipeline cleaning robot to clean the pipeline according to the processing result.
The trained preset convolutional neural network model can be connected into a control system of the pipeline cleaning robot, the pipeline cleaning robot can collect current pipeline images inside the pipeline in the process of moving in the pipeline, the current pipeline images are transmitted to the trained preset convolutional neural network model, and whether the pipeline cleaning robot needs cleaning or not is judged according to the generated processing result.
Alternatively, the processing result may include an output value, and the processing result is generated after the current pipeline image is input into the trained preset convolutional neural network model for processing. Thus, if the output value meets the preset condition, the pipeline cleaning robot is controlled to clean the pipeline according to the output value.
Illustratively, since in the training sample, a1 in the flag information indicates that cleaning is required, a0 indicates that cleaning is not required. Therefore, when the output value is 1, namely, the output value is determined to meet the preset condition, the pipeline cleaning robot can be controlled to clean the pipeline.
In the embodiment of the application, the method comprises the following steps: the method comprises the steps of obtaining a current pipeline image collected by a pipeline cleaning robot, inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result, wherein the preset convolutional neural network model is generated based on preset training samples, the preset training samples comprise a preset number of pipeline image samples and marking information, each pipeline image sample is generated by carrying out data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample, and the pipeline cleaning robot is controlled to conduct pipeline cleaning according to the processing result. According to the scheme, the current pipeline image can be automatically processed through a pre-trained preset convolutional neural network model, so that the pipeline cleaning robot can automatically find a cleaning scene and automatically perform full-automatic cleaning operation; simultaneously, effectively combine the degree of depth learning technique and pipeline cleaning scene, can realize effectively discernment pipeline in need clean the scene, improve the automation level and the operating efficiency that pipeline maintained, reduce the risk of manual inspection simultaneously.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Fig. 8 is a block diagram of a pipeline cleaning device based on a convolutional neural network according to an embodiment of the present application.
As shown in fig. 8, the pipe cleaning apparatus 800 based on the convolutional neural network includes:
An acquiring module 802 is configured to acquire a current pipeline image acquired by the pipeline cleaning robot.
The processing module 804 is configured to input the current pipeline image into a preset convolutional neural network model for processing, and generate a processing result; the method comprises the steps that a preset convolutional neural network model is generated based on preset training samples, wherein the preset training samples comprise a preset number of pipeline image samples and marking information, and each pipeline image sample is generated by carrying out data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample.
And a control module 806 for controlling the pipeline cleaning robot to perform pipeline cleaning according to the processing result.
The specific manner in which the various modules perform the operations in the apparatus of the above embodiments have been described in detail in connection with the embodiments of the method, and will not be described in detail herein. The above-mentioned construction method of the pipeline cleaning method based on the convolutional neural network may be implemented in whole or in part by software, hardware and a combination thereof. The above modules may be embedded in a hardware form or may be independent of a processor in the computer device, or may be stored in a memory in the computer device in a software form, so that the processor may call the operations of the above modules.
In one embodiment of the present application, there is provided a computer device including a memory and a processor, the memory having stored therein a computer program which when executed by the processor performs the steps of:
Acquiring a current pipeline image acquired by a pipeline cleaning robot;
Inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result; the method comprises the steps that a preset convolutional neural network model is generated based on preset training samples, wherein the preset training samples comprise a preset number of pipeline image samples and marking information, and each pipeline image sample is generated by carrying out data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample;
and controlling the pipeline cleaning robot to clean the pipeline according to the processing result.
In one embodiment of the application, the processor when executing the computer program further performs the steps of:
Acquiring a preset number of pipeline image samples and marking information corresponding to each pipeline image sample;
inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a preset convolutional neural network model.
In one embodiment of the application, the processor when executing the computer program further performs the steps of:
acquiring a preset number of original pipeline images and preset marking information;
For each original pipeline image, carrying out data preprocessing on the original pipeline image to generate a preprocessed pipeline image; the data preprocessing at least comprises format conversion processing and normalization processing;
Carrying out data enhancement processing on the preprocessed pipeline image to generate a pipeline image sample; the data enhancement processing at least comprises rotation processing, translation processing and scaling processing.
In one embodiment of the application, the processor when executing the computer program further performs the steps of:
Inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a prediction result;
substituting the prediction result and the marking information corresponding to each pipeline image sample into a preset loss function, and calculating the value of the loss function;
and adjusting parameters of the initial convolutional neural network model based on the value of the loss function until a preset convergence condition is reached, and generating a preset convolutional neural network model.
In one embodiment of the present application, the initial convolutional neural network model comprises an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, and a full-connection layer, and the processor when executing the computer program further implements the steps of:
Receiving a preset number of pipeline image samples and marking information corresponding to each pipeline image sample through an input layer, inputting the received data into a first convolution layer, performing feature extraction on the pipeline image samples through the first convolution layer, generating first feature data, and inputting the first feature data into a first pooling layer;
The first characteristic data is subjected to downsampling treatment through a first pooling layer, second characteristic data subjected to maximum pooling is generated, and the second characteristic data is input into a second convolution layer;
Performing feature extraction on the second feature data through the second convolution layer to generate third feature data, and inputting the third feature data into the second pooling layer;
The third characteristic data is subjected to downsampling treatment through the second pooling layer, fourth characteristic data subjected to maximum pooling is generated, and the fourth characteristic data is input into the full-connection layer;
And processing the fourth characteristic data through the full connection layer to generate a prediction result.
In one embodiment of the application, the processor when executing the computer program further performs the steps of:
Inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating an intermediate convolutional neural network model;
and evaluating and optimizing the middle convolutional neural network model according to the preset verification set to generate a preset convolutional neural network model.
In one embodiment of the application, the processing result comprises an output value, and the processor when executing the computer program further performs the steps of:
and if the output value meets the preset condition, controlling the pipeline cleaning robot to clean the pipeline according to the output value.
The implementation principle and technical effects of the computer device provided by the embodiment of the present application are similar to those of the above method embodiment, and are not described herein.
In one embodiment of the present application, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
Acquiring a current pipeline image acquired by a pipeline cleaning robot;
Inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result; the method comprises the steps that a preset convolutional neural network model is generated based on preset training samples, wherein the preset training samples comprise a preset number of pipeline image samples and marking information, and each pipeline image sample is generated by carrying out data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample;
and controlling the pipeline cleaning robot to clean the pipeline according to the processing result.
In one embodiment of the application, the computer program when executed by the processor further implements the steps of:
Acquiring a preset number of pipeline image samples and marking information corresponding to each pipeline image sample;
inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a preset convolutional neural network model.
In one embodiment of the application, the processor when executing the computer program further performs the steps of:
acquiring a preset number of original pipeline images and preset marking information;
For each original pipeline image, carrying out data preprocessing on the original pipeline image to generate a preprocessed pipeline image; the data preprocessing at least comprises format conversion processing and normalization processing;
Carrying out data enhancement processing on the preprocessed pipeline image to generate a pipeline image sample; the data enhancement processing at least comprises rotation processing, translation processing and scaling processing.
In one embodiment of the application, the processor when executing the computer program further performs the steps of:
Inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a prediction result;
substituting the prediction result and the marking information corresponding to each pipeline image sample into a preset loss function, and calculating the value of the loss function;
and adjusting parameters of the initial convolutional neural network model based on the value of the loss function until a preset convergence condition is reached, and generating a preset convolutional neural network model.
In one embodiment of the present application, the initial convolutional neural network model comprises an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, and a full-connection layer, and the processor when executing the computer program further implements the steps of:
Receiving a preset number of pipeline image samples and marking information corresponding to each pipeline image sample through an input layer, inputting the received data into a first convolution layer, performing feature extraction on the pipeline image samples through the first convolution layer, generating first feature data, and inputting the first feature data into a first pooling layer;
The first characteristic data is subjected to downsampling treatment through a first pooling layer, second characteristic data subjected to maximum pooling is generated, and the second characteristic data is input into a second convolution layer;
Performing feature extraction on the second feature data through the second convolution layer to generate third feature data, and inputting the third feature data into the second pooling layer;
The third characteristic data is subjected to downsampling treatment through the second pooling layer, fourth characteristic data subjected to maximum pooling is generated, and the fourth characteristic data is input into the full-connection layer;
And processing the fourth characteristic data through the full connection layer to generate a prediction result.
In one embodiment of the application, the processor when executing the computer program further performs the steps of:
Inputting a preset number of pipeline image samples and marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating an intermediate convolutional neural network model;
and evaluating and optimizing the middle convolutional neural network model according to the preset verification set to generate a preset convolutional neural network model.
In one embodiment of the application, the processing result comprises an output value, and the processor when executing the computer program further performs the steps of:
and if the output value meets the preset condition, controlling the pipeline cleaning robot to clean the pipeline according to the output value.
The computer readable storage medium provided in this embodiment has similar principles and technical effects to those of the above method embodiment, and will not be described herein.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (15)
1. A method for cleaning a pipeline based on a convolutional neural network, the method comprising:
Acquiring a current pipeline image acquired by a pipeline cleaning robot;
Inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result; the method comprises the steps that a preset convolutional neural network model is generated based on preset training samples, wherein the preset training samples comprise a preset number of pipeline image samples and marking information, and each pipeline image sample is generated by performing data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample;
and controlling the pipeline cleaning robot to clean the pipeline according to the processing result.
2. The method according to claim 1, wherein the construction process of the preset convolutional neural network model comprises:
acquiring a preset number of pipeline image samples and marking information corresponding to each pipeline image sample;
Inputting the preset number of pipeline image samples and the marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating the preset convolutional neural network model.
3. The method of claim 2, wherein the acquiring a predetermined number of pipeline image samples comprises:
acquiring a preset number of original pipeline images and preset marking information;
For each original pipeline image, carrying out data preprocessing on the original pipeline image to generate a preprocessed pipeline image; the data preprocessing at least comprises format conversion processing and normalization processing;
Performing data enhancement processing on the preprocessed pipeline image to generate the pipeline image sample; the data enhancement processing at least comprises rotation processing, translation processing and scaling processing.
4. The method according to claim 2 or 3, wherein inputting the preset number of pipeline image samples and the marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating the preset convolutional neural network model includes:
inputting the preset number of pipeline image samples and the marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a prediction result;
substituting the prediction result and the marking information corresponding to each pipeline image sample into a preset loss function, and calculating the value of the loss function;
And adjusting parameters of the initial convolutional neural network model based on the value of the loss function until a preset convergence condition is reached, and generating the preset convolutional neural network model.
5. The method of claim 4, wherein the initial convolutional neural network model includes an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, and a full-connection layer, the inputting the predetermined number of pipeline image samples and the marking information corresponding to each pipeline image sample into the initial convolutional neural network model for training, and generating the prediction result includes:
receiving the preset number of pipeline image samples and marking information corresponding to each pipeline image sample through the input layer, inputting the received data into the first convolution layer, performing feature extraction on the pipeline image samples through the first convolution layer, generating first feature data, and inputting the first feature data into the first pooling layer;
Performing downsampling processing on the first characteristic data through the first pooling layer to generate second characteristic data subjected to maximum pooling, and inputting the second characteristic data into the second convolution layer;
performing feature extraction on the second feature data through the second convolution layer to generate third feature data, and inputting the third feature data into the second pooling layer;
performing downsampling processing on the third characteristic data through the second pooling layer to generate fourth characteristic data subjected to maximum pooling, and inputting the fourth characteristic data into the full-connection layer;
And processing the fourth characteristic data through the full connection layer to generate the prediction result.
6. The method according to claim 2 or 3, wherein inputting the preset number of pipeline image samples and the marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating the preset convolutional neural network model, further comprises:
inputting the preset number of pipeline image samples and the marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating an intermediate convolutional neural network model;
And evaluating and optimizing the middle convolutional neural network model according to a preset verification set to generate the preset convolutional neural network model.
7. A method according to any one of claims 1-3, wherein the processing result includes an output value, and the controlling the pipe cleaning robot to perform pipe cleaning according to the processing result includes:
and if the output value meets the preset condition, controlling the pipeline cleaning robot to clean the pipeline according to the output value.
8. A pipe cleaning apparatus based on a convolutional neural network, the apparatus comprising:
the acquisition module is used for acquiring the current pipeline image acquired by the pipeline cleaning robot;
the processing module is used for inputting the current pipeline image into a preset convolutional neural network model for processing, and generating a processing result; the method comprises the steps that a preset convolutional neural network model is generated based on preset training samples, wherein the preset training samples comprise a preset number of pipeline image samples and marking information, and each pipeline image sample is generated by performing data preprocessing and data enhancement on an original pipeline image corresponding to the pipeline image sample;
And the control module is used for controlling the pipeline cleaning robot to clean the pipeline according to the processing result.
9. The apparatus of claim 8, wherein the convolutional neural network-based pipe cleaning apparatus is further configured to:
acquiring a preset number of pipeline image samples and marking information corresponding to each pipeline image sample;
Inputting the preset number of pipeline image samples and the marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating the preset convolutional neural network model.
10. The apparatus of claim 9, wherein the convolutional neural network-based pipe cleaning apparatus is further configured to:
acquiring a preset number of original pipeline images and preset marking information;
For each original pipeline image, carrying out data preprocessing on the original pipeline image to generate a preprocessed pipeline image; the data preprocessing at least comprises format conversion processing and normalization processing;
Performing data enhancement processing on the preprocessed pipeline image to generate the pipeline image sample; the data enhancement processing at least comprises rotation processing, translation processing and scaling processing.
11. The apparatus of claim 9 or 10, wherein the convolutional neural network-based pipe cleaning apparatus is further configured to:
inputting the preset number of pipeline image samples and the marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating a prediction result;
substituting the prediction result and the marking information corresponding to each pipeline image sample into a preset loss function, and calculating the value of the loss function;
And adjusting parameters of the initial convolutional neural network model based on the value of the loss function until a preset convergence condition is reached, and generating the preset convolutional neural network model.
12. The apparatus of claim 11, wherein the initial convolutional neural network model comprises an input layer, a first convolutional layer, a first pooled layer, a second convolutional layer, a second pooled layer, and a fully-connected layer, the convolutional neural network-based pipe cleaning apparatus further configured to:
receiving the preset number of pipeline image samples and marking information corresponding to each pipeline image sample through the input layer, inputting the received data into the first convolution layer, performing feature extraction on the pipeline image samples through the first convolution layer, generating first feature data, and inputting the first feature data into the first pooling layer;
Performing downsampling processing on the first characteristic data through the first pooling layer to generate second characteristic data subjected to maximum pooling, and inputting the second characteristic data into the second convolution layer;
performing feature extraction on the second feature data through the second convolution layer to generate third feature data, and inputting the third feature data into the second pooling layer;
performing downsampling processing on the third characteristic data through the second pooling layer to generate fourth characteristic data subjected to maximum pooling, and inputting the fourth characteristic data into the full-connection layer;
And processing the fourth characteristic data through the full connection layer to generate the prediction result.
13. The apparatus of claim 9 or 10, wherein the convolutional neural network-based pipe cleaning apparatus is further configured to:
inputting the preset number of pipeline image samples and the marking information corresponding to each pipeline image sample into an initial convolutional neural network model for training, and generating an intermediate convolutional neural network model;
And evaluating and optimizing the middle convolutional neural network model according to a preset verification set to generate the preset convolutional neural network model.
14. An electronic device comprising a processor and a memory having stored therein at least one instruction, at least one program, code set, or instruction set that is loaded and executed by the processor to implement the steps of the method of any of claims 1-7.
15. A computer readable storage medium having stored therein at least one instruction, at least one program, code set, or instruction set, the at least one instruction, the at least one program, code set, or instruction set being loaded and executed by a processor to implement the steps of the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410551474.9A CN118122723A (en) | 2024-05-07 | 2024-05-07 | Pipeline cleaning method, device and equipment based on convolutional neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410551474.9A CN118122723A (en) | 2024-05-07 | 2024-05-07 | Pipeline cleaning method, device and equipment based on convolutional neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118122723A true CN118122723A (en) | 2024-06-04 |
Family
ID=91246126
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410551474.9A Pending CN118122723A (en) | 2024-05-07 | 2024-05-07 | Pipeline cleaning method, device and equipment based on convolutional neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118122723A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108427953A (en) * | 2018-02-26 | 2018-08-21 | 北京易达图灵科技有限公司 | A kind of character recognition method and device |
| CN110962122A (en) * | 2019-11-11 | 2020-04-07 | 中广核工程有限公司 | Pipeline cleaning robot control system and method |
| US20210071897A1 (en) * | 2019-09-09 | 2021-03-11 | Alisea S.R.L. | Systems and Methods for Artificial Intelligence-Based Maintenance of an Air Conditioning System |
| CN112580581A (en) * | 2020-12-28 | 2021-03-30 | 英特灵达信息技术(深圳)有限公司 | Target detection method and device and electronic equipment |
| CN116871231A (en) * | 2023-06-29 | 2023-10-13 | 湖南哈工聚能科技有限公司 | Workpiece cleaning method and cleaning system |
| CN117809009A (en) * | 2022-09-23 | 2024-04-02 | 腾讯科技（深圳）有限公司 | Identity recognition method, equipment, device, computer equipment and storage medium |
-
2024
- 2024-05-07 CN CN202410551474.9A patent/CN118122723A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108427953A (en) * | 2018-02-26 | 2018-08-21 | 北京易达图灵科技有限公司 | A kind of character recognition method and device |
| US20210071897A1 (en) * | 2019-09-09 | 2021-03-11 | Alisea S.R.L. | Systems and Methods for Artificial Intelligence-Based Maintenance of an Air Conditioning System |
| CN110962122A (en) * | 2019-11-11 | 2020-04-07 | 中广核工程有限公司 | Pipeline cleaning robot control system and method |
| CN112580581A (en) * | 2020-12-28 | 2021-03-30 | 英特灵达信息技术(深圳)有限公司 | Target detection method and device and electronic equipment |
| CN117809009A (en) * | 2022-09-23 | 2024-04-02 | 腾讯科技（深圳）有限公司 | Identity recognition method, equipment, device, computer equipment and storage medium |
| CN116871231A (en) * | 2023-06-29 | 2023-10-13 | 湖南哈工聚能科技有限公司 | Workpiece cleaning method and cleaning system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6901802B2 (en) | Pedestrian detector learning method and learning device for robust surveillance based on image analysis, and test method and test device using it | |
| US10204299B2 (en) | Unsupervised matching in fine-grained datasets for single-view object reconstruction | |
| CN108537152B (en) | Method and apparatus for detecting living body | |
| CN110163344B (en) | Neural network training method, device, equipment and storage medium | |
| CN108416324B (en) | Method and apparatus for detecting living body | |
| EP3182334B1 (en) | License plate recognition using coarse-to-fine cascade adaptations of a convolutional neural network | |
| CN108305275B (en) | Active tracking method, device and system | |
| CN111079632A (en) | Training method and device of text detection model, computer equipment and storage medium | |
| CN110705684A (en) | Environment self-adaptive learning method and system based on end cloud cooperation | |
| CN111507333B (en) | Image correction method and device, electronic equipment and storage medium | |
| CN113705769A (en) | Neural network training method and device | |
| CN112052837A (en) | Target detection method and device based on artificial intelligence | |
| WO2022105197A1 (en) | Systems and methods for image detection | |
| CN111797992A (en) | Machine learning optimization method and device | |
| CN113838211B (en) | 3D point cloud classification attack defense method, device, equipment and storage medium | |
| CN113837942A (en) | Super-resolution image generation method, device, equipment and storage medium based on SRGAN | |
| CN111738403A (en) | Neural network optimization method and related equipment | |
| KR20230137991A (en) | Rendering new images of scenes using a geometry-aware neural network adjusted according to latent variables. | |
| CN114091554A (en) | Training set processing method and device | |
| CN117576724A (en) | Unmanned plane bird detection method, system, equipment and medium | |
| CN116863353A (en) | Electric power tower inclination degree detection method based on rotating target detection network | |
| CN215813842U (en) | Hydropower station gate control system based on deep learning | |
| CN117409083B (en) | Cable terminal identification method and device based on infrared image and improved YOLOV5 | |
| CN117217280A (en) | Neural network model optimization method and device and computing equipment | |
| CN118122723A (en) | Pipeline cleaning method, device and equipment based on convolutional neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |