CN112070721B - Antenna parameter measurement method, device and storage medium based on instance division network - Google Patents
Antenna parameter measurement method, device and storage medium based on instance division network Download PDFInfo
- Publication number
- CN112070721B CN112070721B CN202010811969.2A CN202010811969A CN112070721B CN 112070721 B CN112070721 B CN 112070721B CN 202010811969 A CN202010811969 A CN 202010811969A CN 112070721 B CN112070721 B CN 112070721B
- Authority
- CN
- China
- Prior art keywords
- antenna
- network
- inputting
- module
- instance
- 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.)
- Active
Links
- 238000000691 measurement method Methods 0.000 title claims abstract description 18
- 230000011218 segmentation Effects 0.000 claims abstract description 47
- 230000007246 mechanism Effects 0.000 claims abstract description 41
- 238000005259 measurement Methods 0.000 claims abstract description 40
- 238000000605 extraction Methods 0.000 claims abstract description 19
- 238000001914 filtration Methods 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 33
- 238000012549 training Methods 0.000 claims description 18
- 238000005070 sampling Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 7
- 238000013526 transfer learning Methods 0.000 claims description 7
- 238000012512 characterization method Methods 0.000 claims description 6
- 230000009977 dual effect Effects 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000006870 function Effects 0.000 description 7
- 238000013508 migration Methods 0.000 description 7
- 230000005012 migration Effects 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 6
- 238000011176 pooling Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000003044 adaptive effect 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
- 238000010276 construction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- 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
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an antenna parameter measurement method, device and storage medium based on an example segmentation network, which comprises the steps of obtaining antenna image information; inputting the antenna image information into a network for feature extraction to obtain a first feature quantity; inputting the first characteristic quantity to a dual-attention mechanism module to obtain an accurate boundary box; inputting the first feature quantity into a feature pyramid network to obtain a prototype mask map, inputting the prototype mask map into a PointRend module, and obtaining a smooth instance segmentation mask map by combining an accurate boundary box and threshold filtering; and performing data fitting on the smooth instance segmentation mask map to obtain antenna parameters, and realizing safe, rapid and high-precision antenna parameter measurement.
Description
Technical Field
The present invention relates to the field of antenna tilt angle measurement, and in particular, to a method and apparatus for measuring antenna parameters based on an example division network, and a storage medium.
Background
The antenna downward inclination angle is the included angle between the antenna on the signal tower and the vertical direction, and also is the antenna mechanical downward inclination angle. The antenna downtilt angle is generally determined by comprehensively analyzing coverage area, topography, site distribution condition and hanging height of an area where the antenna is located, and site spacing in combination with wireless propagation environments such as population density of the area where the antenna is located. Along with the change of factors such as urban population density, environment, social development, network optimization and the like, the antenna downtilt angle needs to be timely detected to determine whether adjustment is needed. With the gradual perfection of network construction, the antenna downtilt angle may need to be optimally adjusted in the later network optimization. At this time, the mechanical downtilt angle information of the antenna is particularly important, and the acquisition of the downtilt angle information of the antenna is often difficult for network optimization personnel, while the traditional measurement method requires the professional tower crane to perform tower crane measurement, which consumes manpower and time, and is difficult to ensure safety.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art.
Therefore, the invention provides an antenna parameter measurement method based on an example division network, which can realize safe, rapid and high-precision antenna parameter measurement.
The invention also provides an antenna parameter measuring device based on the example division network, which applies the antenna parameter measuring method based on the example division network.
The invention also provides a computer readable storage medium applying the antenna parameter measurement method based on the example division network.
An embodiment of the present invention provides a method for measuring antenna parameters based on an example split network, including:
acquiring antenna image information;
inputting the antenna image information into a network for feature extraction to obtain a first feature quantity;
inputting the first characteristic quantity to a dual-attention mechanism module to obtain an accurate boundary box;
inputting the first feature quantity into a feature pyramid network to obtain a prototype mask graph;
inputting the prototype mask map to a PointRend module, and combining the accurate bounding box and threshold filtering to obtain a smooth instance segmentation mask map;
and performing data fitting on the smooth instance segmentation mask map to obtain antenna parameters.
The method for measuring the antenna parameters based on the example division network has the following advantages: firstly, acquiring antenna image information, and then extracting characteristics of the antenna image information by utilizing a network; the migration learning can well relieve the problems of great training difficulty and over-fitting caused by the problem of small samples; the dual-attention mechanism module can well improve the accuracy of the network; the PointRend module can well smooth the edges of the prototype mask map in the pyramid network, and the smooth instance mask map is fitted by a data fitting method, so that antenna parameters are obtained, quick and high-precision antenna parameter measurement is realized, manual measurement is not needed, and the antenna parameter measurement can be more labor-saving and safer.
According to some embodiments of the invention, the acquiring antenna image information includes:
and controlling the camera to shoot, and collecting the antenna image information.
According to some embodiments of the present invention, the inputting the antenna image information into a network for feature extraction, to obtain a first feature quantity, includes:
obtaining the weight of a pre-training model through transfer learning;
training the network by loading the weights;
and inputting the antenna image information into the network to perform feature extraction to obtain a first feature quantity.
According to some embodiments of the invention, the dual attention mechanism module includes a channel attention mechanism module and a spatial attention mechanism module.
According to some embodiments of the invention, the inputting the first feature quantity into the dual-attention mechanism module, to obtain an accurate bounding box, includes:
inputting the first characteristic quantity into the channel attention module to obtain an intermediate characteristic quantity;
and inputting the intermediate feature quantity into the spatial attention module to obtain an accurate boundary box.
According to some embodiments of the invention, the inputting the prototype mask map to the pointrand module, in combination with the exact bounding box and threshold filtering, results in a smoothed instance segmentation mask map, comprises:
selecting a plurality of sampling points from the prototype mask graph;
extracting features of the sampling points to obtain a characterization vector;
and predicting the characterization vector by using a multi-layer perceptron, and filtering by combining the accurate boundary box and a threshold value to obtain a smooth instance segmentation mask map.
According to some embodiments of the invention, the performing data fitting on the smoothed instance segmentation mask map to obtain antenna parameters includes:
and performing data fitting on the smooth instance segmentation mask map by using a least square method to obtain antenna parameters.
An example split network-based antenna parameter measurement apparatus according to an embodiment of the second aspect of the present invention includes:
the acquisition unit is used for acquiring antenna image information;
the extraction unit inputs the antenna image information into a network to perform feature extraction to obtain a first feature quantity;
the first processing unit is used for inputting the first characteristic quantity into the dual-attention mechanism module to obtain an accurate bounding box;
the second processing unit is used for inputting the first characteristic quantity into a characteristic pyramid network to obtain a prototype mask graph; the segmentation unit is used for inputting the prototype mask graph to a PointRend module, and combining the accurate bounding box and threshold filtering to obtain a smooth instance segmentation mask graph;
and the calculating unit is used for carrying out data fitting on the smooth instance segmentation mask diagram to obtain antenna parameters.
According to some embodiments of the invention, the first processing unit comprises:
the first operation unit is used for inputting the first characteristic quantity into the channel attention module to obtain an intermediate characteristic quantity;
and the second operation unit is used for inputting the intermediate characteristic quantity into the space attention module to obtain an accurate boundary box.
The antenna parameter measuring device based on the example division network has at least the following beneficial effects: firstly, acquiring antenna image information, and then extracting characteristics of the antenna image information by utilizing a network; the migration learning can well relieve the problems of great training difficulty and over-fitting caused by the problem of small samples; the dual-attention mechanism module can well improve the accuracy of the network; the PointRend module can well smooth the edges of the prototype mask map in the pyramid network, and the smooth instance mask map is fitted by a data fitting method, so that antenna parameters are obtained, quick and high-precision antenna parameter measurement is realized, manual measurement is not needed, and the antenna parameter measurement can be more labor-saving and safer.
The computer-readable storage medium according to the embodiment of the third aspect of the present invention can be applied to the example-division-network-based antenna parameter measurement method according to the above-described first aspect of the present invention.
The computer-readable storage medium according to the embodiment of the invention has at least the following advantageous effects: firstly, acquiring antenna image information, and then extracting characteristics of the antenna image information by utilizing a network; the migration learning can well relieve the problems of great training difficulty and over-fitting caused by the problem of small samples; the dual-attention mechanism module can well improve the accuracy of the network; the PointRend module can well smooth the edges of the prototype mask map in the pyramid network, and the smooth instance mask map is fitted by a data fitting method, so that antenna parameters are obtained, quick and high-precision antenna parameter measurement is realized, manual measurement is not needed, and the antenna parameter measurement can be more labor-saving and safer.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the invention will become apparent and may be better understood from the following description of embodiments taken in conjunction with the accompanying drawings in which:
fig. 1 is a flowchart of an example split network-based antenna parameter measurement method according to a first embodiment of the present invention;
fig. 2 is a flowchart of feature extraction in an example split network-based antenna parameter measurement method according to an embodiment of the present invention;
FIG. 3 is a flowchart of the operation of bounding box validation in an example split network based antenna parameter measurement method according to an embodiment of the present invention;
fig. 4 is a flowchart of an example division in an example division network-based antenna parameter measurement method according to the first embodiment of the present invention;
fig. 5 is a schematic diagram of an antenna parameter measurement device based on an example split network according to a second embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
In the description of the present invention, unless explicitly defined otherwise, terms such as arrangement, connection, etc. should be construed broadly and the specific meaning of the terms in the present invention can be reasonably determined by a person skilled in the art in combination with the specific contents of the technical solution.
Example 1
Referring to fig. 1, a first embodiment of the present invention provides an antenna parameter measurement method based on an example split network, wherein one embodiment includes, but is not limited to, the following steps:
step S100, antenna image information is acquired.
In this embodiment, the step collects the antenna image information, and prepares for the measurement of the antenna parameters, which is the basis of the measurement of the antenna parameters.
Step S200, inputting the antenna image information into a network for feature extraction to obtain a first feature quantity.
In this embodiment, the step inputs the antenna image information into the network to perform feature extraction, thereby obtaining a first feature quantity; the pre-training model can well overcome the problems of over fitting, difficult training and the like caused by less sample data.
Step S300, inputting the first feature quantity into a dual-attention mechanism module to obtain an accurate bounding box.
In this embodiment, the first feature quantity is input to the dual-attention mechanism module in this step, so as to obtain an accurate bounding box, where the dual-attention mechanism module can well improve accuracy of instance segmentation of the network.
Step S400, the first feature quantity is input into a feature pyramid network to obtain a prototype mask graph.
In this embodiment, the step inputs the first feature quantity into the feature pyramid network to obtain a prototype mask map, ready for antenna parameter measurement.
Step S500, inputting the prototype mask graph into the PointRend module, and combining the accurate bounding box and the threshold filtering to obtain the smooth instance segmentation mask graph.
In this embodiment, the mask map is input to the PointRend module in this step, so that the edges of the mask map can be smoothed, and the segmentation accuracy can be improved well.
Step S600, data fitting is carried out on the smooth instance segmentation mask diagram, and antenna parameters are obtained.
In this embodiment, the data fitting is performed on the smoothed instance segmentation mask map in this step, and the antenna parameters are calculated, so that the method is simple and fast.
In step S100 of the present embodiment, the following steps may be included, but are not limited to:
and controlling the camera to shoot, and collecting the antenna image information.
In this embodiment, the step controls the camera to shoot, for example, by setting the camera on the unmanned aerial vehicle, shooting the antenna, collecting the antenna image information, and preparing for the measurement of the antenna parameters.
Referring to fig. 2, in step S200 of the present embodiment, the following steps may be included, but are not limited to:
step S210, obtaining the weight of the pre-training model through transfer learning.
In the embodiment, the pre-training model weight parameters are obtained through transfer learning, and the transfer learning method can well relieve the problems of large training difficulty, overfitting and the like caused by the problem of small samples.
Step S220, training the network by loading the weights.
In this embodiment, this step trains the network by loading the weights, ready for feature extraction.
In step S230, the antenna image information is input into the network to perform feature extraction, so as to obtain a first feature quantity.
In this embodiment, the step inputs the antenna image information into the network to perform feature extraction, and a first feature value is obtained, so as to prepare for measurement of the antenna parameters.
In some embodiments of the invention, the dual attention mechanism module includes a channel attention mechanism module and a spatial attention mechanism module. The channel attention mechanism module uses a maximum pooling and an average pooling output over the shared network; the spatial attention mechanism module combines two similar outputs along the channel axis and maps them to the convolutional layers.
Referring to fig. 3, in step S300 of the present embodiment, the following steps may be included, but are not limited to:
in step S310, the first feature quantity is input to the channel attention module, and an intermediate feature quantity is obtained.
In this embodiment, the step inputs the first feature quantity into the channel attention module, thereby obtaining an intermediate feature quantity, ready for the subsequent network segmentation.
Step S320, inputting the intermediate feature quantity into the spatial attention module to obtain an accurate boundary box.
In this embodiment, this step inputs the intermediate feature quantity into the spatial attention module, so that an accurate bounding box can be obtained.
Referring to fig. 4, in step S500 of the present embodiment, the following steps may be included, but are not limited to:
step S510, selecting a plurality of sampling points from the prototype mask map.
In this embodiment, this step selects a number of sampling points from the mask map, in particular points at edge positions in the mask map.
And step S520, extracting features of the sampling points to obtain a characterization vector.
In this embodiment, the feature extraction is performed on the selected sampling points in this step, so as to obtain the characterization vector.
And step S530, predicting the feature vector by using a multi-layer perceptron, and combining the accurate bounding box and threshold filtering to obtain a smooth instance segmentation mask diagram.
In this embodiment, the present step predicts the representative vector by using the multi-layer perceptron, so that a smooth instance segmentation mask map can be obtained, and the edge of the prototype mask map is well smoothed, thereby improving the accuracy of network segmentation.
In step S600 of the present embodiment, the following steps may be included, but are not limited to:
and performing data fitting on the smooth instance segmentation mask map by using a least square method to obtain antenna parameters.
In this embodiment, the present step uses a least square method to perform linear fitting, and obtains the antenna parameters from the smoothed instance division mask map, so that the measurement of the antenna parameters can be more accurate.
A specific embodiment is provided below to further describe an antenna parameter measurement method based on an example split network:
the embodiment of the invention mainly provides an antenna parameter measurement method based on an instance segmentation network, which combines an attention mechanism and a PointRend module to improve segmentation accuracy, and adopts a single-step instance segmentation network to improve segmentation speed. Wherein a RetinaNet based on feature pyramid network and resnet101 is used to build the basic network framework, and a dual-attention mechanism module is used to channel and space attention on each convolution block of the depth network to perfect the feature map of the middle layer, and a PointRend module is added after the mask map to generate a finer instance mask.
Because marked data are difficult to acquire, marked data are difficult to be obtained, and the problem that fitting is performed on small samples exists, the embodiment of the invention utilizes transfer learning to acquire parameters from a large database model trained by a deep convolutional neural network to train an antenna segmentation network so as to achieve the aim of high-precision segmentation.
The dual-attention mechanism module can deduce attention weights along two dimensions of a channel and a space sequentially through the intermediate feature diagram and then multiply the attention weights by the original feature diagram to adaptively adjust the features, and the dual-attention mechanism module can be seamlessly integrated into any convolutional neural network architecture and has negligible additional cost due to the very light design process of the dual-attention mechanism module, and can perform end-to-end training together with any basic CNN framework. The dual-attention mechanism has two consecutive sub-modules: channels and spatial sub-modules. The channel attention mechanism module uses maximum pooling and average pooling output through a shared network; the spatial attention mechanism module combines two similar outputs along the channel axis and maps them to the convolutional layer. In the channel attention module, each channel in the feature map represents a detector, so it is meaningful to say that the channel attention mechanism focuses on the feature. To collect spatial features, a global average pool and a maximum pool are used to obtain different feature information, which can be described as follows:
the characteristic F is input as h×w×c. Spatial global averaging pooling and maximum pooling are first performed, respectively, to obtain two 1 x C channel information. They are then sent to a two-layer neural network, the number of neurons in the first layer being C/r, the activation function being ReLU, and the number of neurons in the second layer being C. The two-layer neural network is feature-shared. The two features obtained are then added and the function is activated by sigmoid to obtain the weight coefficients. Finally, multiplying the weight coefficient with the original feature F to obtain a new scaled feature.
After passing through the attention mechanism module, the spatial attention module, which will be described in detail below, is used where the attention features are of interest.
The spatial attention mechanism is similar to the channel attention mechanism, with feature F being hxw x C, and first, the channel dimensions are averaged pooled and maximally pooled to obtain two hxw x 1 channel information, and the two information are combined together into a channel. Subsequently, after a 7×7 convolutional layer and a sigmoid activation function, the weight coefficient M can be obtained s . Finally, the weight coefficient M s Multiplying the new feature by the feature F to obtain a new scaled feature. The two modules of channel attention and spatial attention can be combined in parallel or sequentially with any basic CNN, and experiments show that sequential combination and placing the channel attention mechanism first can achieve better results. The channel attention module and the space attention module are taken as two independent parts, so that parameters and calculation capacity can be saved, and the channel attention module and the space attention module can be integrated into the existing network architecture as a plug-and-play module.
Because of the over-sampling and under-sampling problems in the pixel labeling task, the masking branch cannot add a very fine and accurate mask for each case, which will greatly affect the measurement accuracy of the antenna parameter measurement. Thus, an efficient and high quality PointRend module is combined to smooth the segmented instance mask edges and minimize the measurement error of the antenna parameters. PointRend consists essentially of three parts: point selection strategies, point feature representations and point predictions. Implementation of the point selection strategy relies primarily on flexible adaptive selection of suitable sampling points. For example, more points are sampled in the boundary region, and fewer points are sampled in the non-boundary region. After selecting the point strategy, the most uncertain points are selected to re-predict the segmentation results for these points. After the punctiform feature representation module calculates the feature of each particular point (the most uncertain point), the feature of each point is processed by the point prediction to obtain a new prediction result. The above steps can be summarized as special treatment of points on boundary positions to make the results more accurate. Therefore, the mask edge of the mobile communication base station antenna can be smoothed subtly by the points to improve fitting accuracy and parameter measurement accuracy.
Downtilt fitting structures in PointRend modules. The least squares method is the most basic and important data fitting method for revealing the relationship between pixels in the mask to quantify the antenna parameters. The least squares method has wide application in data modeling. The use of this theory to solve problems has a very important role and role, especially in the study of extensive data analysis. To reduce measurement errors and computational costs, the edge 3/5 pixel value coordinates on both sides of the mask (top and bottom 1/5 removed mask) are chosen to accommodate downtilt angle FIG. 5 illustrates the PointRend module function with a parameter fitting structure.
First, it is assumed that a parameter estimation equation is as follows, where x and y represent an abscissa point and an ordinate point of a pixel value coordinate point selected in a mask, β represents a correlation coefficient, and μ represents a deviation, respectively.
y＝β 0 +β 1 x+β 2 x+…+β k x+μ
Substituting n groups of pixel value coordinates into the above formula can obtain:
the matrix representation of the above formula is as follows:
y＝BX+μ
thus, the sum of squares of the difference between the observed value and the estimated value of the interpreted variable is:
according to the principle of the least square method, the method comprises the following steps:
therefore, the least squares estimation of the parameters is:
the deep convolutional neural network has the capability of learning rich image representations, so that the deep convolutional neural network is widely used in various large application scenes. However, to learn these functions, a large amount of data is required. For the datasets used in the present invention, the amount of data is insufficient, resulting in overfitting and training difficulties. Transfer learning is a method of transferring knowledge from the relevant domain to a new problem. The migration learning strategy learns low-level and medium-level features from the migrated domain, so that a small amount of data can be obtained from the new domain to obtain higher performance. Thus, application migration learning will be trained on massive amounts
The identification model on the image is transferred to the mobile communication base station antenna segmentation task, so that the trouble caused by a small sample can be effectively reduced, and the performance the same as that of the massive data task can be obtained.
According to the scheme, the antenna image information is firstly obtained, and then the network is utilized to extract the characteristics of the antenna image information; the migration learning can well relieve the problems of great training difficulty and over-fitting caused by the problem of small samples; the dual-attention mechanism module can well improve the accuracy of the network; the PointRend module can well smooth the edges of the prototype mask map in the pyramid network, and the smooth instance segmentation mask map is fitted by a data fitting method, so that antenna parameters are obtained, quick and high-precision antenna parameter measurement is realized, manual measurement is not needed, and the antenna parameter measurement can be more labor-saving and safer.
Example two
Referring to fig. 5, a second embodiment of the present invention provides an antenna parameter measurement apparatus 1000 based on an example split network, including:
an acquisition unit 1100 for acquiring antenna image information;
an extracting unit 1200, configured to input the antenna image information into a pre-training model to perform feature extraction, so as to obtain a first feature quantity;
a first processing unit 1300, configured to input the first feature quantity to a dual-attention mechanism module, to obtain an accurate bounding box;
a second processing unit 1400, configured to input the first feature quantity to a feature pyramid network to obtain a prototype mask map;
a segmentation unit 1500, configured to input the prototype mask map to a poinrand module, to obtain a smoothed instance segmentation mask map;
a calculating unit 1600, configured to perform data fitting on the smoothed instance segmentation mask map to obtain antenna parameters.
In the present embodiment, the first processing unit 1300 includes:
a first operation unit 1310, configured to input the first feature quantity into the channel attention module, to obtain an intermediate feature quantity;
a second operation unit 1320 for inputting the intermediate feature quantity into the spatial attention module, resulting in an accurate bounding box.
It should be noted that, since the antenna parameter measurement apparatus based on the instance-division network in the present embodiment is based on the same inventive concept as the antenna parameter measurement method based on the instance-division network in the first embodiment, the corresponding content in the first method embodiment is also applicable to the present system embodiment, and will not be described in detail herein.
According to the scheme, the antenna image information is firstly obtained, and then the network is utilized to extract the characteristics of the antenna image information; the migration learning can well relieve the problems of great training difficulty and over-fitting caused by the problem of small samples; the dual-attention mechanism module can well improve the accuracy of the network; the mask pattern edge segmented by the dual-attention mechanism model can be well smoothed by the PointRend module, and the smooth instance segmentation mask pattern is fitted by a data fitting method, so that antenna parameters are obtained, quick and high-precision antenna parameter measurement is realized, manual measurement is not needed, and the antenna parameter measurement can be more labor-saving and safer.
Example III
A third embodiment of the present invention also provides a computer-readable storage medium storing an antenna parameter measurement apparatus executable instruction based on an instance-split network, for causing the antenna parameter measurement apparatus based on the instance-split network to perform the above-described antenna parameter measurement method based on the instance-split network, for example, performing the above-described method steps S100 to S600 in fig. 1, to implement the functions of units 1000 to 1500 in fig. 5.
In the description of the present specification, reference to the terms "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.
Claims (9)
1. An antenna parameter measurement method based on an example division network is characterized by comprising the following steps:
acquiring antenna image information;
inputting the antenna image information into a network for feature extraction to obtain a first feature quantity;
inputting the first characteristic quantity to a dual-attention mechanism module to obtain an accurate boundary box;
inputting the first feature quantity into a feature pyramid network to obtain a prototype mask graph;
inputting the prototype mask map to a PointRend module, and combining the accurate bounding box and threshold filtering to obtain a smooth instance segmentation mask map;
performing data fitting on the smooth instance segmentation mask map to obtain antenna parameters;
the performing data fitting on the smoothed instance segmentation mask map to obtain antenna parameters includes:
performing data fitting on the smooth instance segmentation mask map by using a least square method to obtain antenna parameters;
wherein, the antenna parameter estimation equation is as follows:
y＝β 0 +β 1 x+β 2 x+…+β k x+μ
wherein x and y respectively represent an abscissa point and an ordinate point of a pixel value coordinate point selected in the smoothed instance division mask map, β represents a correlation coefficient, and μ represents a deviation;
substituting n groups of pixel value coordinates into the above formula can obtain:
the matrix representation of the above formula is as follows:
y＝BX+μ
the sum of squares of the difference between the observed value and the estimated value of the interpreted variable is:
according to the principle of the least square method, the method comprises the following steps:
the least squares estimation of the antenna parameters is:
2. the method for measuring antenna parameters based on an instance-partitioned network according to claim 1, wherein the acquiring antenna image information comprises:
and controlling the camera to shoot, and collecting the antenna image information.
3. The method for measuring antenna parameters based on an example split network according to claim 1, wherein the inputting the antenna image information into a network for feature extraction to obtain a first feature value comprises: obtaining the weight of a pre-training model through transfer learning;
training the network by loading the weights;
and inputting the antenna image information into the network to perform feature extraction to obtain a first feature quantity.
4. The method for measuring antenna parameters based on an instance-partitioned network according to claim 1, wherein:
the dual attention mechanism module includes a channel attention mechanism module and a spatial attention mechanism module.
5. The method for measuring antenna parameters based on an instance-partitioned network according to claim 4, wherein said inputting the first feature quantity into a dual-attention mechanism module results in an accurate bounding box, comprising:
inputting the first characteristic quantity into the channel attention module to obtain an intermediate characteristic quantity;
and inputting the intermediate feature quantity into the spatial attention module to obtain an accurate boundary box.
6. The method for measuring antenna parameters based on an instance segmentation network according to claim 1, wherein the inputting the prototype mask graph into a poinrand module, in combination with the accurate bounding box and threshold filtering, obtains a smoothed instance segmentation mask graph, comprises:
selecting a plurality of sampling points from the prototype mask graph;
extracting features of the sampling points to obtain a characterization vector;
and predicting the characterization vector by using a multi-layer perceptron, and combining the accurate bounding box and threshold filtering to obtain a smooth instance segmentation mask diagram.
7. An example-split network-based antenna parameter measurement apparatus, comprising:
the acquisition unit is used for acquiring antenna image information;
the extraction unit inputs the antenna image information into a network to perform feature extraction to obtain a first feature quantity;
the first processing unit is used for inputting the first characteristic quantity into the dual-attention mechanism module to obtain an accurate bounding box;
the second processing unit is used for inputting the first characteristic quantity into a characteristic pyramid network to obtain a prototype mask graph; the segmentation unit is used for inputting the prototype mask graph to a PointRend module, and combining the accurate bounding box and threshold filtering to obtain a smooth instance segmentation mask graph;
the computing unit is used for carrying out data fitting on the smooth instance segmentation mask graph to obtain antenna parameters;
the performing data fitting on the smoothed instance segmentation mask map to obtain antenna parameters includes:
performing data fitting on the smooth instance segmentation mask map by using a least square method to obtain antenna parameters;
wherein, the antenna parameter estimation equation is as follows:
y＝β 0 +β 1 x+β 2 x+…+β k x+μ
wherein x and y respectively represent an abscissa point and an ordinate point of a pixel value coordinate point selected in the smoothed instance division mask map, β represents a correlation coefficient, and μ represents a deviation;
substituting n groups of pixel value coordinates into the above formula can obtain:
the matrix representation of the above formula is as follows:
y＝BX+μ
the sum of squares of the difference between the observed value and the estimated value of the interpreted variable is:
according to the principle of the least square method, the method comprises the following steps:
the least squares estimation of the antenna parameters is:
8. the instance-partitioned network-based antenna parameter measurement apparatus of claim 7, wherein the first processing unit comprises:
the first operation unit is used for inputting the first characteristic quantity into the channel attention module to obtain an intermediate characteristic quantity;
and the second operation unit is used for inputting the intermediate characteristic quantity into the space attention module to obtain an accurate boundary box.
9. A computer-readable storage medium, characterized by: the computer-readable storage medium stores an example-divided-network-based antenna parameter measurement apparatus executable instruction for causing the example-divided-network-based antenna parameter measurement apparatus to perform the example-divided-network-based antenna parameter measurement method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010811969.2A CN112070721B (en) | 2020-08-13 | 2020-08-13 | Antenna parameter measurement method, device and storage medium based on instance division network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010811969.2A CN112070721B (en) | 2020-08-13 | 2020-08-13 | Antenna parameter measurement method, device and storage medium based on instance division network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112070721A CN112070721A (en) | 2020-12-11 |
| CN112070721B true CN112070721B (en) | 2024-01-12 |
Family
ID=73661585
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010811969.2A Active CN112070721B (en) | 2020-08-13 | 2020-08-13 | Antenna parameter measurement method, device and storage medium based on instance division network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112070721B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109458980A (en) * | 2018-11-06 | 2019-03-12 | 五邑大学 | A kind of Downtilt measurement method based on linear regression fit |
| CN110532955A (en) * | 2019-08-30 | 2019-12-03 | 中国科学院宁波材料技术与工程研究所 | Example dividing method and device based on feature attention and son up-sampling |
| CN111062466A (en) * | 2019-12-11 | 2020-04-24 | 南京华苏科技有限公司 | Method for predicting field intensity distribution of cell after antenna adjustment based on parameters and neural network |
| CN111445474A (en) * | 2020-05-25 | 2020-07-24 | 南京信息工程大学 | Kidney CT image segmentation method based on bidirectional complex attention depth network |
-
2020
- 2020-08-13 CN CN202010811969.2A patent/CN112070721B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109458980A (en) * | 2018-11-06 | 2019-03-12 | 五邑大学 | A kind of Downtilt measurement method based on linear regression fit |
| CN110532955A (en) * | 2019-08-30 | 2019-12-03 | 中国科学院宁波材料技术与工程研究所 | Example dividing method and device based on feature attention and son up-sampling |
| CN111062466A (en) * | 2019-12-11 | 2020-04-24 | 南京华苏科技有限公司 | Method for predicting field intensity distribution of cell after antenna adjustment based on parameters and neural network |
| CN111445474A (en) * | 2020-05-25 | 2020-07-24 | 南京信息工程大学 | Kidney CT image segmentation method based on bidirectional complex attention depth network |
Non-Patent Citations (2)
| Title |
|---|
| 基于数据拟合的补全模型研究;王国民;谷晓鹏;路广勋;;军事通信技术(第01期);第1-3页 * |
| 基于最小二乘法的误差分析与误差修正;陈林斌;孙赐恩;;中国新通信(第07期);第1-2页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112070721A (en) | 2020-12-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108961235B (en) | Defective insulator identification method based on YOLOv3 network and particle filter algorithm | |
| CN109711288B (en) | Remote sensing ship detection method based on characteristic pyramid and distance constraint FCN | |
| CN108122247B (en) | A kind of video object detection method based on saliency and feature prior model | |
| CN107369166B (en) | Target tracking method and system based on multi-resolution neural network | |
| CN108960135B (en) | Dense ship target accurate detection method based on high-resolution remote sensing image | |
| CN110298387A (en) | Incorporate the deep neural network object detection method of Pixel-level attention mechanism | |
| CN111626176B (en) | Remote sensing target rapid detection method and system based on dynamic attention mechanism | |
| CN110232471B (en) | Rainfall sensor network node layout optimization method and device | |
| CN109389102A (en) | The system of method for detecting lane lines and its application based on deep learning | |
| CN108230354A (en) | Target following, network training method, device, electronic equipment and storage medium | |
| CN115457006B (en) | Unmanned aerial vehicle inspection defect classification method and device based on similarity consistency self-distillation | |
| CN112862090A (en) | Air temperature forecasting method based on deep space-time neural network | |
| CN114896561B (en) | Wetland carbon reserve calculation method based on remote sensing algorithm | |
| CN114596500A (en) | Remote sensing image semantic segmentation method based on channel-space attention and DeeplabV3plus | |
| CN113920262A (en) | Mining area FVC calculation method and system for enhancing edge sampling and improving Unet model | |
| CN107610156A (en) | Infrared small object tracking based on guiding filtering and core correlation filtering | |
| CN116824396A (en) | Multi-satellite data fusion automatic interpretation method | |
| CN102202216A (en) | Mouse and insect monitoring device and method for performing image tracking and identification by adopting same | |
| CN114998251A (en) | Air multi-vision platform ground anomaly detection method based on federal learning | |
| CN112070721B (en) | Antenna parameter measurement method, device and storage medium based on instance division network | |
| CN111428191B (en) | Antenna downtilt angle calculation method and device based on knowledge distillation and storage medium | |
| CN116403071A (en) | Method and device for detecting few-sample concrete defects based on feature reconstruction | |
| US20220215619A1 (en) | Geospatial modeling system providing 3d geospatial model update based upon iterative predictive image registration and related methods | |
| CN114782709A (en) | Image small target detection method and system based on Gaussian distribution strategy | |
| CN113591740A (en) | Deep learning-based method and device for identifying silt particles in complex river environment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |