CN116045104B - Air conditioner soft and hard pipe connection sealing device and method thereof - Google Patents
Air conditioner soft and hard pipe connection sealing device and method thereof Download PDFInfo
- Publication number
- CN116045104B CN116045104B CN202310036644.5A CN202310036644A CN116045104B CN 116045104 B CN116045104 B CN 116045104B CN 202310036644 A CN202310036644 A CN 202310036644A CN 116045104 B CN116045104 B CN 116045104B
- Authority
- CN
- China
- Prior art keywords
- feature map
- image
- hose
- detection
- feature
- 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
- 238000007789 sealing Methods 0.000 title claims abstract description 75
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000012544 monitoring process Methods 0.000 claims abstract description 80
- 230000007704 transition Effects 0.000 claims abstract description 37
- 238000005336 cracking Methods 0.000 claims abstract description 28
- 230000008878 coupling Effects 0.000 claims abstract description 6
- 238000010168 coupling process Methods 0.000 claims abstract description 6
- 238000005859 coupling reaction Methods 0.000 claims abstract description 6
- 238000001514 detection method Methods 0.000 claims description 96
- 238000010586 diagram Methods 0.000 claims description 67
- 238000009826 distribution Methods 0.000 claims description 32
- 239000013598 vector Substances 0.000 claims description 26
- 238000012937 correction Methods 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 13
- 230000004927 fusion Effects 0.000 claims description 11
- 238000012512 characterization method Methods 0.000 claims description 10
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 4
- 238000012423 maintenance Methods 0.000 abstract description 10
- 238000010438 heat treatment Methods 0.000 description 9
- 230000007246 mechanism Effects 0.000 description 6
- 239000002184 metal Substances 0.000 description 6
- 238000003062 neural network model Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000006872 improvement Effects 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 238000004378 air conditioning Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005485 electric heating Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 229910001566 austenite Inorganic materials 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 229910000734 martensite Inorganic materials 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 229910001220 stainless steel Inorganic materials 0.000 description 1
- 239000010935 stainless steel Substances 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/16—Measuring arrangements characterised by the use of optical techniques for measuring the deformation in a solid, e.g. optical strain gauge
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60H—ARRANGEMENTS OF HEATING, COOLING, VENTILATING OR OTHER AIR-TREATING DEVICES SPECIALLY ADAPTED FOR PASSENGER OR GOODS SPACES OF VEHICLES
- B60H1/00—Heating, cooling or ventilating [HVAC] devices
- B60H1/00507—Details, e.g. mounting arrangements, desaeration devices
- B60H1/00557—Details of ducts or cables
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16L—PIPES; JOINTS OR FITTINGS FOR PIPES; SUPPORTS FOR PIPES, CABLES OR PROTECTIVE TUBING; MEANS FOR THERMAL INSULATION IN GENERAL
- F16L33/00—Arrangements for connecting hoses to rigid members; Rigid hose connectors, i.e. single members engaging both hoses
- F16L33/26—Arrangements for connecting hoses to rigid members; Rigid hose connectors, i.e. single members engaging both hoses specially adapted for hoses of metal
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
- F17D5/02—Preventing, monitoring, or locating loss
- F17D5/06—Preventing, monitoring, or locating loss using electric or acoustic means
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
- Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Acoustics & Sound (AREA)
- Thermal Sciences (AREA)
- Image Processing (AREA)
Abstract
The application discloses a connecting and sealing device and a connecting and sealing method for soft and hard pipes of an air conditioner. Wherein, the soft or hard union coupling sealing device of air conditioner includes: a transition joint having opposed first and second ends; a hose connected to a first end of the transition joint; a hard tube connected to the second end of the transition joint; a buckle sleeve which is buckled on the transition joint in a sealing way; the camera is arranged on the transition joint and faces the hose; and a controller communicatively coupled to the camera. The controller may construct a condition monitoring scheme for the hose based on machine vision to determine whether the risk of cracking of the hose exceeds a predetermined threshold, such that upon detecting that the risk of cracking of the hose exceeds the predetermined threshold, an early warning prompt is generated to prompt a service person for maintenance. In this way, the potential safety hazard can be reduced.
Description
Technical Field
The application relates to the technical field of intelligent monitoring, in particular to an air conditioner soft and hard pipe connection sealing device and an air conditioner soft and hard pipe connection sealing method.
Background
With the continuous improvement of consumer demand for comfort experience, the state is increasingly paying attention to environmental protection, and new energy electric traffic has become a trend. The improvement of the new energy electric automobile on the requirements of the endurance mileage plays a key role in the popularization of urban electric traffic, the extension of the service life of batteries, the improvement of the endurance capacity and the improvement of the performance of a driving system.
At present, a heat pump air conditioner is an effective solution for heating a pure electric automobile. Under the condition that the power battery does not break through, low-energy consumption heating is ensured, the heat pump air conditioner is a few feasible technologies, the efficiency coefficient is much higher than that of PTC heating, and the endurance mileage can be effectively prolonged. When the heating wire is used in a new energy automobile, the heating wire can only be used for heating the air conditioner, and the heat pump technology can not be used for heating. The electric heating wire is adopted for heating, so that the endurance mileage of the automobile is greatly reduced; if all the pipes are connected by hard pipes, the problem of sealing the pipes can be solved, but the hard pipes cannot be adopted due to the limitation of the internal space of the automobile, and a pipe assembly formed by combining the hose and the hard pipes is necessary, but the hose and the hard pipes have different properties, so that the sealing of the joint between the hose and the hard pipes becomes an important technical problem when the pipe assembly is carried out by using the hose and the hard pipes.
Therefore, an air conditioner soft and hard tube connection sealing device is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a device and a method for connecting and sealing soft and hard pipes of an air conditioner. Wherein, the soft or hard union coupling sealing device of air conditioner includes: a transition joint having opposed first and second ends; a hose connected to a first end of the transition joint; a hard tube connected to the second end of the transition joint; a buckle sleeve which is buckled on the transition joint in a sealing way; the camera is arranged on the transition joint and faces the hose; and a controller communicatively coupled to the camera. The controller may construct a condition monitoring scheme for the hose based on machine vision to determine whether the risk of cracking of the hose exceeds a predetermined threshold, such that upon detecting that the risk of cracking of the hose exceeds the predetermined threshold, an early warning prompt is generated to prompt a service person for maintenance. In this way, the potential safety hazard can be reduced.
According to an aspect of the present application, there is provided an air conditioner soft and hard pipe connection sealing device, comprising: a transition joint having opposed first and second ends; a hose connected to a first end of the transition joint; a hard tube connected to the second end of the transition joint; and a buckle sleeve which is buckled on the transition joint in a sealing way.
In the above-mentioned soft or hard union coupling sealing device of air conditioner, the soft or hard union coupling sealing device of air conditioner still includes: the camera is arranged on the transition joint and faces the hose; and a controller communicatively coupled to the camera.
In the above-mentioned air conditioner soft and hard tube connection sealing device, the controller includes: the monitoring image receiving module is used for receiving the monitoring image of the hose from the camera; the reference image scheduling module is used for acquiring a reference image of the hose, wherein the reference image is an image of the hose which is not deformed; the twin detection module is used for enabling the monitoring image and the reference image to pass through a twin network model comprising a first image encoder and a second image encoder to obtain a detection characteristic diagram and a reference characteristic diagram, and the first image encoder and the second image encoder have the same network structure; the spatial enhancement module is used for enabling the detection feature map and the reference feature map to pass through the spatial attention module so as to obtain a spatial enhancement detection feature map and a spatial enhancement reference feature map; the difference characterization module is used for calculating a difference characteristic diagram between the space enhancement detection characteristic diagram and the space enhancement reference characteristic diagram; the characteristic distribution correction module is used for carrying out characteristic distribution correction on the differential characteristic map based on the scale of the differential characteristic map so as to obtain an optimized differential characteristic map; and the monitoring result generation module is used for enabling the optimized differential feature map to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the risk of hose cracking exceeds a preset threshold value.
In the air conditioner soft and hard tube connection sealing device, the first image encoder and the second image encoder are convolution neural network models comprising depth feature fusion modules.
In the above-mentioned air conditioner soft and hard tube connection sealing device, the space enhancement module is further used for: performing depth convolution coding on the detection feature map and the reference feature map by using a convolution coding part of the spatial attention module to obtain a detection convolution feature map and a reference convolution feature map; inputting the detected convolution feature map and the reference convolution feature map into a spatial attention portion of the spatial attention module to obtain a detected spatial attention map and a reference spatial attention map, respectively; -activating the detection spatial attention map and the reference spatial attention map by Softmax activation functions, respectively, to obtain a detection spatial attention profile and a reference spatial attention profile; and calculating the detection space attention characteristic diagram and the detection characteristic diagram according to the position point multiplication to obtain the space enhancement detection characteristic diagram, and calculating the reference space attention characteristic diagram and the reference characteristic diagram according to the position point multiplication to obtain the space enhancement reference characteristic diagram.
In the above-mentioned air conditioner soft and hard tube connection sealing device, the difference characterization module is further used for: calculating a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map with the following formula; wherein, the formula is: Wherein F 1 is the spatially enhanced detection feature map, F 2 is the spatially enhanced reference feature map,/> Representing the difference by location.
In the above-mentioned air conditioner soft and hard tube connection sealing device, the characteristic distribution correction module is further configured to: performing feature distribution correction on the differential feature map based on the scale of the differential feature map by using the following formula to obtain an optimized differential feature map;
wherein, the formula is:
Wherein f i' is the optimized differential feature map, f i is a predetermined feature value of the differential feature map, f j is other feature values except the predetermined feature value of the differential feature map, f is a mean value of all feature values of the differential feature map, and N is a scale of the differential feature map, that is, a width multiplied by a height multiplied by a channel number, exp (·) represents an exponential operation of the feature map, and performing the exponential operation on the feature map represents a natural exponential function value with the feature value of each position in the feature map as a power.
In the above-mentioned air conditioner soft and hard tube connection sealing device, the monitoring result generation module includes: the unfolding unit is used for unfolding the optimized differential feature map into a classification feature vector according to a row vector or a column vector; the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and the classification unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
According to another aspect of the present application, there is provided a method for monitoring a soft and hard pipe connection sealing device of an air conditioner, comprising: receiving a monitoring image of a hose of the air conditioner soft and hard tube connecting sealing device from a camera; acquiring a reference image of the hose, wherein the reference image is an image of the hose which is not deformed; passing the monitored image and the reference image through a twin network model comprising a first image encoder and a second image encoder to obtain a detected feature map and a reference feature map, the first image encoder and the second image encoder having the same network structure; the detection feature map and the reference feature map pass through a spatial attention module to obtain a spatial enhancement detection feature map and a spatial enhancement reference feature map; calculating a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map; based on the scale of the differential feature map, carrying out feature distribution correction on the differential feature map to obtain an optimized differential feature map; and passing the optimized differential feature map through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the risk of hose cracking exceeds a preset threshold value.
In the method for monitoring the air conditioner soft and hard tube connection sealing device, the first image encoder and the second image encoder are convolution neural network models comprising depth feature fusion modules.
Compared with the prior art, the air conditioner soft and hard tube connection sealing device and the method thereof provided by the application, wherein the air conditioner soft and hard tube connection sealing device comprises: a transition joint having opposed first and second ends; a hose connected to a first end of the transition joint; a hard tube connected to the second end of the transition joint; a buckle sleeve which is buckled on the transition joint in a sealing way; the camera is arranged on the transition joint and faces the hose; and a controller communicatively coupled to the camera. The controller may construct a condition monitoring scheme for the hose based on machine vision to determine whether the risk of cracking of the hose exceeds a predetermined threshold, such that upon detecting that the risk of cracking of the hose exceeds the predetermined threshold, an early warning prompt is generated to prompt a service person for maintenance. In this way, the potential safety hazard can be reduced.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
Fig. 1 is a schematic structural diagram of a soft and hard tube connection sealing device of an air conditioner according to an embodiment of the application.
Fig. 2 is an application scenario diagram of an air conditioner soft and hard tube connection sealing device according to an embodiment of the application.
Fig. 3 is a schematic diagram of a controller of an air conditioner soft and hard tube connection sealing device according to an embodiment of the application.
Fig. 4 is a schematic block diagram of the monitoring result generating module in the controller of the air conditioner soft and hard tube connection sealing device according to the embodiment of the application.
Fig. 5 is a flowchart of a method for monitoring a soft and hard tube connection sealing device of an air conditioner according to an embodiment of the application.
Fig. 6 is a schematic diagram of a system architecture of a method for monitoring a soft and hard tube connection sealing device of an air conditioner according to an embodiment of the application.
Detailed Description
Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Scene overview
As described above, the adoption of the electric heating wire for heating greatly reduces the endurance mileage of the automobile; if all the pipes are connected by hard pipes, the problem of sealing the pipes can be solved, but the hard pipes cannot be adopted due to the limitation of the internal space of the automobile, and a pipe assembly formed by combining the hose and the hard pipes is necessary, but the hose and the hard pipes have different properties, so that the sealing of the joint between the hose and the hard pipes becomes an important technical problem when the pipe assembly is carried out by using the hose and the hard pipes. Therefore, an air conditioner soft and hard tube connection sealing device is desired.
In view of the above technical problems, as shown in fig. 1, the present application provides an air conditioner soft and hard pipe connection sealing device 10, which includes: a transition fitting 11 having opposite first and second ends, a hose 12 connected to the first end of the transition fitting 11, a rigid tube 13 connected to the second end of the transition fitting 11, and a grommet 14 sealingly crimped to the transition fitting 11. That is, the pipe assembly is realized by receiving the hose 12 and the hard pipe 13 through the transition joint 11, but the joint between the transition joint 11 and the hose 12 and the hard pipe 13 may cause leakage problems. To improve the sealing, the sleeve 14 is further pressed against the outer surface of the transition joint 11.
Specifically, in the technical scheme of the application, the hose is a metal corrugated pipe, and the metal corrugated pipe is a metal corrugated pipe made of stainless steel. During production, the metal corrugated pipe is cut into a specified length and then connected with the transition joint, so that the situation that the metal corrugated pipe is coated with rubber firstly, then the pipe is cut and the outer side of the joint is not coated with a rubber layer after the transition joint is welded is avoided, and the buckling position of the buckling sleeve cannot be positioned at the outer side of the transition joint. If the rubber layer is buckled on the outer side of the metal corrugated pipe, the corrugated pipe can deform in the buckling stress area, the material structure can deform martensite or austenite twin crystals, and the deformation area of the corrugated pipe is easy to crack under high pressure to cause leakage. To solve the above problems, the conventional method is to check whether the hose is cracked or not through regular maintenance, which not only consumes time, but also causes additional safety hazards because the hose cannot be judged in real time, so that in an actual use scene, the hose is cracked when in use.
According to the technical problem, in the technical scheme, a state monitoring scheme for the hose is constructed based on machine vision to determine whether the risk of hose cracking exceeds a preset threshold, so that after the risk of hose cracking is detected to exceed the preset threshold, an early warning prompt is generated to prompt maintenance personnel to carry out maintenance so as to avoid potential safety hazards.
With continued reference to fig. 1, the air conditioner soft and hard tube connection sealing device 10 further includes: a camera 15 disposed at the transition joint 11 and facing the hose 12, and a controller 16 communicatively connected to the camera 15. Specifically, a camera 15 facing the hose 12 is first provided at the transition joint 11, and a real-time status image of the hose 12 is acquired by the camera 15. It should be understood that the hose is deformed due to stress, and the surface state of the hose is changed due to continuous deformation, for example, twisted stripes, tiny cracks, etc., so that whether the risk of cracking the hose exceeds a predetermined threshold can be determined by comparing the real-time state image with the reference image. However, the relationship between the difference between the implementation state image of the hose and the reference image of the hose and whether the risk of the hose cracking exceeds a predetermined threshold is a complex nonlinear relationship, and it is difficult for a conventional statistical model or a feature engineering model to establish an accurate function mapping relationship. In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks have also shown levels approaching and even exceeding humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like. The development of deep learning and neural networks provides new solutions and solutions for hose condition monitoring.
Specifically, firstly, the monitoring image of the hose 12 is received from the camera 15, and meanwhile, a reference image of the hose is obtained by calling from a background database, wherein the reference image is an image of the hose which is not deformed. It should be understood that in some embodiments of the present application, the reference image of the hose may also be stored directly in the controller 16, without being called up by a background server, which is not a limitation of the present application.
Considering that when the monitoring image of the hose is collected by the camera, a plurality of interferences and noises are introduced due to the orientation, resolution and distortion of the camera, if the monitoring image and the reference image are directly compared pixel by pixel, the interferences can cause erroneous judgment. Meanwhile, the reference image and the monitoring image are not completely aligned, and various expression visualization errors exist in the difference expression generated by comparing the reference image with the monitoring image pixel by pixel. Therefore, in the technical scheme of the application, a feature extractor is constructed by using a deep neural network model to extract high-dimensional implicit features in the monitoring image and the reference image, and the difference between the monitoring image and the reference image is represented by the difference between the monitoring image and the reference image in a high-dimensional feature space. That is, the monitoring image and the reference image are mapped from an image source domain to a high-dimensional feature domain with a deep neural network model as a domain mapper, and substantial differences of the two are represented based on feature distribution differences of the two in the high-dimensional feature domain.
Specifically, the monitoring image and the reference image are passed through a twin network model including a first image encoder and a second image encoder to obtain a detection feature map and a reference feature map, the first image encoder and the second image encoder having the same network structure. In a specific example of the present application, the first image encoder and the second image encoder are convolutional neural network models that include a depth feature fusion module. That is, in the technical solution of the present application, a convolutional neural network model with a depth feature fusion mechanism is used as a feature extractor, so that the depth feature fusion mechanism is used to preserve the difference between the shallow features of the hoses in the reference image and the monitoring image, where the shallow features include textures, lines, edges, and the like.
After the detection feature map and the reference feature map are obtained, in order to enable the feature expression of each pixel to have better space discriminativity, the detection feature map and the reference feature map are further subjected to a space attention module to obtain a space enhancement detection feature map and a space enhancement reference feature map in consideration of different confidence levels of feature values of each pixel position in the detection feature map and the reference feature map for final classification judgment. Then, a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map is calculated to represent the difference between the monitored image and the reference image of the hose in the high-dimensional feature space. Further, the differential feature map is passed through a classifier to obtain a classification result indicating whether the risk of hose cracking exceeds a predetermined threshold.
In particular, in the technical solution of the present application, since the detection feature map and the reference feature map are obtained by a spatial attention module respectively, differences of feature distributions in different image semantic encoding directions of the detection image and the reference image, which are introduced by the first image encoder and the second image encoder of the twin network model, of the detection feature map and the reference feature map are amplified by the spatial attention mechanism, so that the differential feature map obtained by calculating the position-by-position difference between the spatial enhancement detection feature map and the spatial enhancement reference feature map needs to express associated differential feature distributions in different image semantic encoding directions, so that after the differential feature map is expanded into feature vectors in a classifier, fitting burden between the feature distributions and a weight matrix of the classifier is heavy, thereby affecting training speed of the classifier and accuracy of classification results.
Thus, the differential feature map is preferably class-characterization flattened, specifically expressed as:
f i is a predetermined feature value of the differential feature map, f j is other feature values than the predetermined feature value of the differential feature map, Is the average of all feature values of the differential feature map, and N is the scale of the differential feature map, i.e. width times height times channel number.
Here, the class representation flattening of the differential feature map flattens a finite polyhedral manifold for class representation of feature distribution in a high-dimensional feature space while maintaining the inherent distance between planes of the manifold and intuitively avoiding intersection based on space, which essentially decomposes the finite polyhedral manifold into a cube lattice based on right-angle plane intersection and with vertices, thereby obtaining flat "slice" continuity of the class planes to enhance the fitting performance of the differential feature map to the weight matrix of the classifier. Therefore, the training speed of classifying the differential feature map through the classifier and the accuracy of the classification result are improved.
Based on the above, the application provides a controller of an air conditioner soft and hard pipe connection sealing device, which comprises: the monitoring image receiving module is used for receiving the monitoring image of the hose from the camera; the reference image scheduling module is used for acquiring a reference image of the hose, wherein the reference image is an image of the hose which is not deformed; the twin detection module is used for enabling the monitoring image and the reference image to pass through a twin network model comprising a first image encoder and a second image encoder to obtain a detection characteristic diagram and a reference characteristic diagram, and the first image encoder and the second image encoder have the same network structure; the spatial enhancement module is used for enabling the detection feature map and the reference feature map to pass through the spatial attention module so as to obtain a spatial enhancement detection feature map and a spatial enhancement reference feature map; the difference characterization module is used for calculating a difference characteristic diagram between the space enhancement detection characteristic diagram and the space enhancement reference characteristic diagram; the characteristic distribution correction module is used for carrying out characteristic distribution correction on the differential characteristic map based on the scale of the differential characteristic map so as to obtain an optimized differential characteristic map; and the monitoring result generation module is used for enabling the optimized differential feature map to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the risk of hose cracking exceeds a preset threshold value.
Fig. 2 is an application scenario diagram of an air conditioner soft and hard tube connection sealing device according to an embodiment of the application. As shown in fig. 2, in this application scenario, a camera is acquired to receive a monitoring image (e.g., D1 as illustrated in fig. 2) of the hose and a reference image (e.g., D2 as illustrated in fig. 2) of the hose, the reference image being an image of the hose that is not deformed, and then the monitoring image and the reference image are input to a server (e.g., S as illustrated in fig. 2) in which a monitoring algorithm of an air-conditioning hose-to-hose connection sealing device is deployed, wherein the server is able to process the monitoring image and the reference image using the monitoring algorithm of the air-conditioning hose-to-hose connection sealing device to generate a classification result indicating whether a risk of hose cracking exceeds a predetermined threshold.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Exemplary controller
Fig. 3 is a schematic diagram of a controller of an air conditioner soft and hard tube connection sealing device according to an embodiment of the application. As shown in fig. 3, a controller 100 of an air conditioner soft and hard tube connection sealing device according to an embodiment of the present application includes: a monitoring image receiving module 110 for receiving a monitoring image of the hose from the camera; the reference image scheduling module 120 is configured to obtain a reference image of the hose, where the reference image is an image of the hose that is not deformed; a twin detection module 130 for passing the monitoring image and the reference image through a twin network model including a first image encoder and a second image encoder to obtain a detection feature map and a reference feature map, the first image encoder and the second image encoder having the same network structure; a spatial enhancement module 140, configured to pass the detection feature map and the reference feature map through a spatial attention module to obtain a spatial enhanced detection feature map and a spatial enhanced reference feature map; a difference characterization module 150 for calculating a difference signature between the spatially enhanced detection signature and the spatially enhanced reference signature; a feature distribution correction module 160, configured to perform feature distribution correction on the differential feature map based on the scale of the differential feature map to obtain an optimized differential feature map; and a monitoring result generating module 170, configured to pass the optimized differential feature map through a classifier to obtain a classification result, where the classification result is used to indicate whether the risk of hose cracking exceeds a predetermined threshold.
More specifically, in an embodiment of the present application, the monitoring image receiving module 110 is configured to receive a monitoring image of the hose from the camera. In the prior art, the hose is regularly maintained to see whether cracking occurs, which not only consumes time, but also can not judge the state of the hose in real time, so that in an actual use scene, the hose can be cracked when in use, and additional potential safety hazards are caused. In this regard, the application constructs a state monitoring scheme for the hose based on machine vision to determine whether the risk of the hose cracking exceeds a predetermined threshold, so that after detecting that the risk of the hose cracking exceeds the predetermined threshold, an early warning prompt is generated to prompt maintenance personnel to carry out maintenance so as to avoid potential safety hazards. It should be understood that the hose is deformed due to stress, and the surface state of the hose is changed due to continuous deformation, for example, twisted stripes, tiny cracks, etc., so that whether the risk of cracking the hose exceeds a predetermined threshold can be determined by comparing the real-time state image with the reference image.
More specifically, in the embodiment of the present application, the reference image scheduling module 120 is configured to obtain a reference image of the hose, where the reference image is an image of the hose that is not deformed. It should be understood that in some embodiments of the present application, the reference image of the hose may also be stored directly in the controller, without being called up by a background server, which is not a limitation of the present application.
More specifically, in an embodiment of the present application, the twin detection module 130 is configured to pass the monitored image and the reference image through a twin network model including a first image encoder and a second image encoder to obtain a detection feature map and a reference feature map, where the first image encoder and the second image encoder have the same network structure. Considering that when the monitoring image of the hose is collected by the camera, a plurality of interferences and noises are introduced due to the orientation, resolution and distortion of the camera, if the monitoring image and the reference image are directly compared pixel by pixel, the interferences can cause erroneous judgment. Meanwhile, the reference image and the monitoring image are not completely aligned, and various expression visualization errors exist in the difference expression generated by comparing the reference image with the monitoring image pixel by pixel. Therefore, in the technical scheme of the application, a feature extractor is constructed by using a deep neural network model to extract high-dimensional implicit features in the monitoring image and the reference image, and the difference between the monitoring image and the reference image is represented by the difference between the monitoring image and the reference image in a high-dimensional feature space. That is, the monitoring image and the reference image are mapped from an image source domain to a high-dimensional feature domain with a deep neural network model as a domain mapper, and substantial differences of the two are represented based on feature distribution differences of the two in the high-dimensional feature domain.
Accordingly, in one specific example, the first image encoder and the second image encoder are convolutional neural network models that include a depth feature fusion module. That is, in the technical solution of the present application, a convolutional neural network model with a depth feature fusion mechanism is used as a feature extractor, so that the depth feature fusion mechanism is used to preserve the difference between the shallow features of the hoses in the reference image and the monitoring image, where the shallow features include textures, lines, edges, and the like.
More specifically, in the embodiment of the present application, the spatial enhancement module 140 is configured to pass the detection feature map and the reference feature map through a spatial attention module to obtain a spatial enhanced detection feature map and a spatial enhanced reference feature map. After the detection feature map and the reference feature map are obtained, in order to enable the feature expression of each pixel to have better space discriminativity, the detection feature map and the reference feature map are further subjected to a space attention module to obtain a space enhancement detection feature map and a space enhancement reference feature map in consideration of different confidence levels of feature values of each pixel position in the detection feature map and the reference feature map for final classification judgment.
Accordingly, in one specific example, the spatial enhancement module 140 is further configured to: performing depth convolution coding on the detection feature map and the reference feature map by using a convolution coding part of the spatial attention module to obtain a detection convolution feature map and a reference convolution feature map; inputting the detected convolution feature map and the reference convolution feature map into a spatial attention portion of the spatial attention module to obtain a detected spatial attention map and a reference spatial attention map, respectively; -activating the detection spatial attention map and the reference spatial attention map by Softmax activation functions, respectively, to obtain a detection spatial attention profile and a reference spatial attention profile; and calculating the detection space attention characteristic diagram and the detection characteristic diagram according to the position point multiplication to obtain the space enhancement detection characteristic diagram, and calculating the reference space attention characteristic diagram and the reference characteristic diagram according to the position point multiplication to obtain the space enhancement reference characteristic diagram.
More specifically, in an embodiment of the present application, the difference characterization module 150 is configured to calculate a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map. And calculating a difference characteristic diagram between the space enhancement detection characteristic diagram and the space enhancement reference characteristic diagram so as to represent the difference between the monitoring image and the reference image of the hose in a high-dimensional characteristic space.
Accordingly, in one specific example, the variance characterization module 150 is further configured to: calculating a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map with the following formula; wherein, the formula is: Wherein F 1 is the spatially enhanced detection feature map, F 2 is the spatially enhanced reference feature map,/> Representing the difference by location.
More specifically, in the embodiment of the present application, the feature distribution correction module 160 is configured to perform feature distribution correction on the differential feature map based on the scale of the differential feature map to obtain an optimized differential feature map.
In particular, in the technical solution of the present application, since the detection feature map and the reference feature map are obtained by a spatial attention module respectively, differences of feature distributions in different image semantic encoding directions of the detection image and the reference image, which are introduced by the first image encoder and the second image encoder of the twin network model, of the detection feature map and the reference feature map are amplified by the spatial attention mechanism, so that the differential feature map obtained by calculating the position-by-position difference between the spatial enhancement detection feature map and the spatial enhancement reference feature map needs to express associated differential feature distributions in different image semantic encoding directions, so that after the differential feature map is expanded into feature vectors in a classifier, fitting burden between the feature distributions and a weight matrix of the classifier is heavy, thereby affecting training speed of the classifier and accuracy of classification results. Therefore, the differential feature map is preferably class-characterization flattened.
Accordingly, in one specific example, the feature distribution correction module 160 is further configured to: performing feature distribution correction on the differential feature map based on the scale of the differential feature map by using the following formula to obtain an optimized differential feature map; wherein, the formula is:
Wherein f i' is the optimized differential feature map, f i is a predetermined feature value of the differential feature map, f j is other feature values of the differential feature map than the predetermined feature value, The method is characterized in that the method is the average value of all characteristic values of the differential characteristic diagram, N is the scale of the differential characteristic diagram, namely the width multiplied by the height multiplied by the channel number, exp (·) represents the exponential operation of the characteristic diagram, and the exponential operation of the characteristic diagram represents a natural exponential function value which takes the characteristic value of each position in the characteristic diagram as a power.
Here, the class representation flattening of the differential feature map flattens a finite polyhedral manifold for class representation of feature distribution in a high-dimensional feature space while maintaining the inherent distance between planes of the manifold and intuitively avoiding intersection based on space, which essentially decomposes the finite polyhedral manifold into a cube lattice based on right-angle plane intersection and with vertices, thereby obtaining flat "slice" continuity of the class planes to enhance the fitting performance of the differential feature map to the weight matrix of the classifier. Therefore, the training speed of classifying the differential feature map through the classifier and the accuracy of the classification result are improved.
More specifically, in the embodiment of the present application, the monitoring result generating module 170 is configured to pass the optimized differential feature map through a classifier to obtain a classification result, where the classification result is used to indicate whether the risk of hose cracking exceeds a predetermined threshold.
Accordingly, in one specific example, as shown in fig. 4, the monitoring result generating module 170 includes: a developing unit 171 for developing the optimized differential feature map into a classification feature vector according to a row vector or a column vector; a full-connection encoding unit 172, configured to perform full-connection encoding on the classification feature vector by using multiple full-connection layers of the classifier to obtain an encoded classification feature vector; and a classification unit 173, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
In summary, the controller 100 of the air conditioner soft and hard pipe connection sealing device according to the embodiment of the present application is illustrated, which may construct a state monitoring scheme for the hose based on machine vision to determine whether the risk of cracking of the hose exceeds a predetermined threshold, so that after detecting that the risk of cracking of the hose exceeds the predetermined threshold, an early warning prompt is generated to prompt a maintenance person to perform maintenance. In this way, the potential safety hazard can be reduced.
As described above, the controller 100 of the air conditioner soft and hard tube connection sealing device according to the embodiment of the present application may be implemented in various terminal devices, for example, a server based on a monitoring algorithm of the air conditioner soft and hard tube connection sealing device, and the like. In one example, the controller 100 of the air conditioner soft and hard tube connection sealing device may be integrated into the terminal device as a software module and/or a hardware module. For example, the controller 100 of the air conditioner soft and hard pipe connection sealing device may be a software module in the operating system of the terminal device, or may be an application program developed for the terminal device; of course, the controller 100 of the air conditioner soft and hard tube connection sealing device can be one of a plurality of hardware modules of the terminal device.
Alternatively, in another example, the controller 100 of the air conditioner soft and hard tube connection sealing device and the terminal device may be separate devices, and the controller 100 of the air conditioner soft and hard tube connection sealing device may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information according to an agreed data format.
Exemplary method
Fig. 5 is a flowchart of a method for monitoring a soft and hard tube connection sealing device of an air conditioner according to an embodiment of the application. As shown in fig. 5, a method for monitoring a soft and hard pipe connection sealing device of an air conditioner according to an embodiment of the application includes: s110, receiving a monitoring image of a hose of the air conditioner soft and hard tube connection sealing device from a camera; s120, acquiring a reference image of the hose, wherein the reference image is an image of the hose which is not deformed; s130, passing the monitoring image and the reference image through a twin network model comprising a first image encoder and a second image encoder to obtain a detection characteristic diagram and a reference characteristic diagram, wherein the first image encoder and the second image encoder have the same network structure; s140, the detection feature map and the reference feature map pass through a spatial attention module to obtain a spatial enhancement detection feature map and a spatial enhancement reference feature map; s150, calculating a difference characteristic diagram between the space enhancement detection characteristic diagram and the space enhancement reference characteristic diagram; s160, carrying out feature distribution correction on the differential feature map based on the scale of the differential feature map to obtain an optimized differential feature map; and S170, passing the optimized differential feature map through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the risk of hose cracking exceeds a preset threshold value.
Fig. 6 is a schematic diagram of a system architecture of a method for monitoring a soft and hard tube connection sealing device of an air conditioner according to an embodiment of the application. As shown in fig. 6, in the system architecture of the method for monitoring the air conditioner soft and hard tube connection sealing device, firstly, a monitoring image of a hose of the air conditioner soft and hard tube connection sealing device is received from a camera; then, acquiring a reference image of the hose, wherein the reference image is an image of the hose which is not deformed; then, the monitoring image and the reference image are passed through a twin network model comprising a first image encoder and a second image encoder to obtain a detection feature map and a reference feature map, wherein the first image encoder and the second image encoder have the same network structure; then, the detection feature map and the reference feature map pass through a spatial attention module to obtain a spatial enhancement detection feature map and a spatial enhancement reference feature map; then, calculating a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map; then, based on the scale of the differential feature map, carrying out feature distribution correction on the differential feature map to obtain an optimized differential feature map; and finally, the optimized differential feature map is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the risk of hose cracking exceeds a preset threshold value.
In a specific example, in the method for monitoring the air conditioner soft and hard tube connection sealing device, the first image encoder and the second image encoder are convolutional neural network models including depth feature fusion modules.
In a specific example, in the method for monitoring a soft and hard pipe connection sealing device of an air conditioner, the detecting feature map and the reference feature map are passed through a spatial attention module to obtain a spatial enhancement detecting feature map and a spatial enhancement reference feature map, and the method further includes: performing depth convolution coding on the detection feature map and the reference feature map by using a convolution coding part of the spatial attention module to obtain a detection convolution feature map and a reference convolution feature map; inputting the detected convolution feature map and the reference convolution feature map into a spatial attention portion of the spatial attention module to obtain a detected spatial attention map and a reference spatial attention map, respectively; -activating the detection spatial attention map and the reference spatial attention map by Softmax activation functions, respectively, to obtain a detection spatial attention profile and a reference spatial attention profile; and calculating the detection space attention characteristic diagram and the detection characteristic diagram according to the position point multiplication to obtain the space enhancement detection characteristic diagram, and calculating the reference space attention characteristic diagram and the reference characteristic diagram according to the position point multiplication to obtain the space enhancement reference characteristic diagram.
In a specific example, in the method for monitoring a soft and hard pipe connection sealing device of an air conditioner, the calculating a difference feature map between the space enhancement detection feature map and the space enhancement reference feature map further includes: calculating a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map with the following formula; wherein, the formula is: Wherein F 1 is the spatially enhanced detection feature map, F 2 is the spatially enhanced reference feature map,/> Representing the difference by location.
In a specific example, in the method for monitoring a soft and hard tube connection sealing device of an air conditioner, the performing feature distribution correction on the differential feature map based on the scale of the differential feature map to obtain an optimized differential feature map further includes: performing feature distribution correction on the differential feature map based on the scale of the differential feature map by using the following formula to obtain an optimized differential feature map; wherein, the formula is:
Wherein f i' is the optimized differential feature map, f i is a predetermined feature value of the differential feature map, f j is other feature values of the differential feature map than the predetermined feature value, The method is characterized in that the method is the average value of all characteristic values of the differential characteristic diagram, N is the scale of the differential characteristic diagram, namely the width multiplied by the height multiplied by the channel number, exp (·) represents the exponential operation of the characteristic diagram, and the exponential operation of the characteristic diagram represents a natural exponential function value which takes the characteristic value of each position in the characteristic diagram as a power.
In a specific example, in the method for monitoring the soft and hard pipe connection sealing device of the air conditioner, the step of passing the optimized differential feature map through a classifier to obtain a classification result, where the classification result is used to indicate whether the risk of hose cracking exceeds a predetermined threshold, includes: expanding the optimized differential feature map into classification feature vectors according to row vectors or column vectors; performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described monitoring method of the air conditioner soft and hard tube connection sealing device have been described in detail in the above description of the controller of the air conditioner soft and hard tube connection sealing device with reference to fig. 2 to 4, and thus, repetitive descriptions thereof will be omitted.
The basic principles of the present application have been described above in connection with specific embodiments, but it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be construed as necessarily possessed by the various embodiments of the application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.
The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (3)
1. The utility model provides a soft or hard union coupling sealing device of air conditioner which characterized in that includes:
A transition joint having opposed first and second ends;
a hose connected to a first end of the transition joint;
a hard tube connected to the second end of the transition joint; and
A buckle sleeve which is buckled on the transition joint in a sealing way;
Wherein, the soft or hard union coupling sealing device of air conditioner still includes:
the camera is arranged on the transition joint and faces the hose; and
A controller communicatively coupled to the camera;
Wherein, the controller includes:
The monitoring image receiving module is used for receiving the monitoring image of the hose from the camera;
The reference image scheduling module is used for acquiring a reference image of the hose, wherein the reference image is an image of the hose which is not deformed;
The twin detection module is used for enabling the monitoring image and the reference image to pass through a twin network model comprising a first image encoder and a second image encoder to obtain a detection characteristic diagram and a reference characteristic diagram, and the first image encoder and the second image encoder have the same network structure;
The spatial enhancement module is used for enabling the detection feature map and the reference feature map to pass through the spatial attention module so as to obtain a spatial enhancement detection feature map and a spatial enhancement reference feature map;
the difference characterization module is used for calculating a difference characteristic diagram between the space enhancement detection characteristic diagram and the space enhancement reference characteristic diagram;
The characteristic distribution correction module is used for carrying out characteristic distribution correction on the differential characteristic map based on the scale of the differential characteristic map so as to obtain an optimized differential characteristic map; and
The monitoring result generation module is used for enabling the optimized differential feature map to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the risk of hose cracking exceeds a preset threshold value or not;
the first image encoder and the second image encoder are convolutional neural network models comprising a depth feature fusion module;
wherein the spatial enhancement module is further configured to:
Performing depth convolution coding on the detection feature map and the reference feature map by using a convolution coding part of the spatial attention module to obtain a detection convolution feature map and a reference convolution feature map;
Inputting the detected convolution feature map and the reference convolution feature map into a spatial attention portion of the spatial attention module to obtain a detected spatial attention map and a reference spatial attention map, respectively;
-activating the detection spatial attention map and the reference spatial attention map by Softmax activation functions, respectively, to obtain a detection spatial attention profile and a reference spatial attention profile; and
Calculating the detection space attention characteristic diagram and the detection characteristic diagram according to the position point multiplication to obtain the space enhancement detection characteristic diagram, and calculating the reference space attention characteristic diagram and the reference characteristic diagram according to the position point multiplication to obtain the space enhancement reference characteristic diagram;
wherein, the difference characterization module is further configured to: calculating a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map with the following formula;
wherein, the formula is: wherein/> For the spatially enhanced detection feature map,/>For the spatial enhancement reference feature map,/>Representing the difference by location;
Wherein the feature distribution correction module is further configured to: performing feature distribution correction on the differential feature map based on the scale of the differential feature map by using the following formula to obtain an optimized differential feature map;
wherein, the formula is:
，
Wherein, Is the optimized differential feature map,/>Is a predetermined eigenvalue of the differential eigenvector,/>Is a feature value other than the predetermined feature value of the differential feature map,/>Is the average of all eigenvalues of the differential eigenvector and/>Is the scale of the differential feature map, i.e. width times height times channel number,/>Performing exponential operation on the feature map to represent natural exponential function values with the feature values of all positions in the feature map as powers;
Wherein, the monitoring result generation module includes:
The unfolding unit is used for unfolding the optimized differential feature map into a classification feature vector according to a row vector or a column vector;
the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and
And the classification unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
2. A method for monitoring the soft and hard tube connection sealing device of the air conditioner according to claim 1, comprising:
receiving a monitoring image of a hose of the air conditioner soft and hard tube connecting sealing device from a camera;
acquiring a reference image of the hose, wherein the reference image is an image of the hose which is not deformed;
Passing the monitored image and the reference image through a twin network model comprising a first image encoder and a second image encoder to obtain a detected feature map and a reference feature map, the first image encoder and the second image encoder having the same network structure;
The detection feature map and the reference feature map pass through a spatial attention module to obtain a spatial enhancement detection feature map and a spatial enhancement reference feature map;
calculating a difference feature map between the spatially enhanced detection feature map and the spatially enhanced reference feature map;
Based on the scale of the differential feature map, carrying out feature distribution correction on the differential feature map to obtain an optimized differential feature map; and
And (3) passing the optimized differential feature map through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the risk of hose cracking exceeds a preset threshold value.
3. The method for monitoring the soft and hard air conditioner pipe connection sealing device according to claim 2, wherein the first image encoder and the second image encoder are convolutional neural network models comprising a depth feature fusion module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310036644.5A CN116045104B (en) | 2023-01-10 | 2023-01-10 | Air conditioner soft and hard pipe connection sealing device and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310036644.5A CN116045104B (en) | 2023-01-10 | 2023-01-10 | Air conditioner soft and hard pipe connection sealing device and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116045104A CN116045104A (en) | 2023-05-02 |
| CN116045104B true CN116045104B (en) | 2024-05-10 |
Family
ID=86119719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310036644.5A Active CN116045104B (en) | 2023-01-10 | 2023-01-10 | Air conditioner soft and hard pipe connection sealing device and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116045104B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0556019A1 (en) * | 1992-02-13 | 1993-08-18 | Titeflex Corporation | Means for and methods of attaching metal hoses to end fittings |
| CN201496698U (en) * | 2009-07-29 | 2010-06-02 | 常州市盛士达汽车空调有限公司 | Buckle structure for automotive air conditioning pipeline |
| CN109296857A (en) * | 2018-10-11 | 2019-02-01 | 南京利德东方橡塑科技有限公司 | A kind of hose assembling for automobile air-conditioner and preparation method thereof |
| KR102008973B1 (en) * | 2019-01-25 | 2019-08-08 | (주)나스텍이앤씨 | Apparatus and Method for Detection defect of sewer pipe based on Deep Learning |
| CN209622278U (en) * | 2019-02-12 | 2019-11-12 | 南京协众汽车空调集团有限公司 | The soft or hard pipe connecting structure of pipelines of automobile air conditioner |
| WO2020210004A1 (en) * | 2019-04-12 | 2020-10-15 | University Of Iowa Research Foundation | System and method to predict, prevent, and mitigate workplace injuries |
| CN113374962A (en) * | 2021-07-06 | 2021-09-10 | 大洋流体技术(上海)有限公司 | R744 refrigerant air conditioner pipeline connecting technology |
| US11521379B1 (en) * | 2021-09-16 | 2022-12-06 | Nanjing University Of Information Sci. & Tech. | Method for flood disaster monitoring and disaster analysis based on vision transformer |
| CN115456047A (en) * | 2022-08-23 | 2022-12-09 | 华能新能源股份有限公司 | Offshore wind turbine structure state monitoring system and method based on artificial intelligence |
| CN115527134A (en) * | 2022-10-27 | 2022-12-27 | 浙江九烁光电工程技术有限公司 | Urban garden landscape lighting monitoring system and method based on big data |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7387012B2 (en) * | 2006-07-14 | 2008-06-17 | Veyance Technologies, Inc. | Leak detection sensor system and method for double carcass hose |
| US7966889B2 (en) * | 2008-06-20 | 2011-06-28 | General Electric Company | Method of measuring strain in a flex hose or other flexible member |
| US8888139B2 (en) * | 2011-10-04 | 2014-11-18 | American Boa Incorporated | Metal hose end fitting |
| BR112014026342B1 (en) * | 2012-04-23 | 2021-08-24 | Eaton Intelligent Power Limited | HOSE DEGRADATION MONITORING SYSTEM AND METHOD FOR MONITORING THE DEGRADATION OF A HOSE ARRANGEMENT |
| AU2019249431A1 (en) * | 2018-04-05 | 2020-10-29 | Dynamic Infrastructure Ltd. | System and method for early identification and monitoring of defects in transportation infrastructure |
| EP3936224A4 (en) * | 2019-03-08 | 2022-04-27 | FUJIFILM Corporation | Data generation device, data generation method, learning device, and learning method |
| JP7396832B2 (en) * | 2019-06-19 | 2023-12-12 | 株式会社ブリヂストン | Hose remaining life prediction method and hose remaining life prediction system |
| LU102103B1 (en) * | 2020-09-30 | 2022-03-30 | Wurth Paul Sa | Computer System and Method Providing Operating Instructions for Thermal Control of a Blast Furnace |
| US20240062530A1 (en) * | 2020-12-17 | 2024-02-22 | Trustees Of Tufts College | Deep perceptual image enhancement |
-
2023
- 2023-01-10 CN CN202310036644.5A patent/CN116045104B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0556019A1 (en) * | 1992-02-13 | 1993-08-18 | Titeflex Corporation | Means for and methods of attaching metal hoses to end fittings |
| CN201496698U (en) * | 2009-07-29 | 2010-06-02 | 常州市盛士达汽车空调有限公司 | Buckle structure for automotive air conditioning pipeline |
| CN109296857A (en) * | 2018-10-11 | 2019-02-01 | 南京利德东方橡塑科技有限公司 | A kind of hose assembling for automobile air-conditioner and preparation method thereof |
| KR102008973B1 (en) * | 2019-01-25 | 2019-08-08 | (주)나스텍이앤씨 | Apparatus and Method for Detection defect of sewer pipe based on Deep Learning |
| CN209622278U (en) * | 2019-02-12 | 2019-11-12 | 南京协众汽车空调集团有限公司 | The soft or hard pipe connecting structure of pipelines of automobile air conditioner |
| WO2020210004A1 (en) * | 2019-04-12 | 2020-10-15 | University Of Iowa Research Foundation | System and method to predict, prevent, and mitigate workplace injuries |
| CN113374962A (en) * | 2021-07-06 | 2021-09-10 | 大洋流体技术(上海)有限公司 | R744 refrigerant air conditioner pipeline connecting technology |
| US11521379B1 (en) * | 2021-09-16 | 2022-12-06 | Nanjing University Of Information Sci. & Tech. | Method for flood disaster monitoring and disaster analysis based on vision transformer |
| CN115456047A (en) * | 2022-08-23 | 2022-12-09 | 华能新能源股份有限公司 | Offshore wind turbine structure state monitoring system and method based on artificial intelligence |
| CN115527134A (en) * | 2022-10-27 | 2022-12-27 | 浙江九烁光电工程技术有限公司 | Urban garden landscape lighting monitoring system and method based on big data |
Non-Patent Citations (2)
| Title |
|---|
| Variable block-size double predictor DPCM (VBDP-DPCM) image transmission system over wireless fading channel;Wu, JC等;《 2003 IEEE 58TH VEHICULAR TECHNOLOGY CONFERENCE》;20031231;第S1-5卷;第3183-3187页 * |
| 图像去雾算法综述;谢勇等;《计算机与数字工程》;20221220;第50卷(第12期);第2765-2774页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116045104A (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104595170B (en) | A kind of air compressor machine monitoring and diagnosis system and method for self-adaptive kernel gauss hybrid models | |
| CN107230015A (en) | A kind of power distribution network toughness appraisal procedure based on system information entropy | |
| CN115099285A (en) | Intelligent detection method and system based on neural network model | |
| CN116045104B (en) | Air conditioner soft and hard pipe connection sealing device and method thereof | |
| CN115013298B (en) | Real-time performance online monitoring system and monitoring method of sewage pump | |
| CN103279033B (en) | State identification method during crude oil pipeline blending process | |
| CN115346345B (en) | Intelligent toxic and harmful gas alarm system for preparing hexafluorobutadiene | |
| CN115147655A (en) | Oil gas gathering and transportation monitoring system and method thereof | |
| CN113111543A (en) | Internet of things service system | |
| CN115410069A (en) | Fault detection method and system based on multiple attention mechanism | |
| CN105809182A (en) | Image classification method and device | |
| CN115774014A (en) | Welding seam defect detection system and method based on vision and ultrasound | |
| CN114985826A (en) | Maximum over-cutting depth limiting device and limiting method of ratchet type cable shears | |
| CN105786763B (en) | A kind of generation method in the fault propagation path of device integration system network | |
| CN116843656B (en) | Plastic coating control method and system for steel belt pipe | |
| CN102194232B (en) | Layering-guided video image target segmenting method | |
| CN115762558B (en) | Performance detection system and method for escalator production | |
| CN116001579A (en) | Emergency power-off method and system for new energy vehicle | |
| CN115810007A (en) | Energy-saving production method and system for aluminum composite plate | |
| CN116167994A (en) | Mineral cable and intelligent preparation method thereof | |
| CN115222798A (en) | Vehicle processing state monitoring system and method thereof | |
| CN116091987A (en) | Industrial scene-oriented multi-strategy image anomaly sample generation method | |
| CN116030014B (en) | Intelligent processing method and system for soft and hard air conditioner pipes | |
| CN115130520A (en) | Circuit fault diagnosis method and system based on multi-line association | |
| CN111694342A (en) | Industrial boiler fault diagnosis method based on similarity |
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 |