CN116408647B - Automatic assembly system and method for automobile radiator core - Google Patents
Automatic assembly system and method for automobile radiator core Download PDFInfo
- Publication number
- CN116408647B CN116408647B CN202310445030.2A CN202310445030A CN116408647B CN 116408647 B CN116408647 B CN 116408647B CN 202310445030 A CN202310445030 A CN 202310445030A CN 116408647 B CN116408647 B CN 116408647B
- Authority
- CN
- China
- Prior art keywords
- feature
- radiator core
- classification
- feature map
- neural network
- 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
- 238000000034 method Methods 0.000 title claims abstract description 58
- 230000007547 defect Effects 0.000 claims abstract description 78
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 46
- 238000013527 convolutional neural network Methods 0.000 claims description 82
- 238000001514 detection method Methods 0.000 claims description 68
- 230000004927 fusion Effects 0.000 claims description 60
- 238000009826 distribution Methods 0.000 claims description 37
- 238000012545 processing Methods 0.000 claims description 37
- 238000010586 diagram Methods 0.000 claims description 29
- 239000013598 vector Substances 0.000 claims description 26
- 238000000605 extraction Methods 0.000 claims description 24
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000005457 optimization Methods 0.000 claims description 19
- 238000004519 manufacturing process Methods 0.000 claims description 16
- 230000002950 deficient Effects 0.000 claims description 13
- 230000000903 blocking effect Effects 0.000 claims description 11
- 238000011176 pooling Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 7
- 238000012546 transfer Methods 0.000 claims description 6
- 238000009434 installation Methods 0.000 claims description 4
- 230000017525 heat dissipation Effects 0.000 abstract description 8
- 230000006870 function Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 11
- 238000004590 computer program Methods 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 7
- 230000001788 irregular Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000013461 design Methods 0.000 description 5
- 239000000110 cooling liquid Substances 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000005065 mining Methods 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 239000008358 core component Substances 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000005094 computer simulation Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000000306 component Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23P—METAL-WORKING NOT OTHERWISE PROVIDED FOR; COMBINED OPERATIONS; UNIVERSAL MACHINE TOOLS
- B23P21/00—Machines for assembling a multiplicity of different parts to compose units, with or without preceding or subsequent working of such parts, e.g. with programme control
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
An automatic assembly system of an automobile radiator core and a method thereof are provided, wherein radiating fins meeting preset requirements are manufactured; mounting a water tank and a water pipe on the radiating fin to obtain a radiating core; mounting a bracket on the radiating core to obtain an automobile radiator core; and detecting surface defects of the assembled automobile radiator core. Therefore, the quality and the size suitability of the automobile radiator core body can be ensured, and further the automobile radiator core body can be ensured to effectively perform engine heat dissipation, and the normal operation of an engine is ensured.
Description
Technical Field
The application relates to the technical field of intelligent assembly, in particular to an automatic assembly system and method for an automobile radiator core.
Background
The automobile radiator core is a core component of an automobile radiating system and mainly comprises a water tank, a water pipe, radiating fins, a bracket and the like. When an automobile engine runs, a large amount of heat is generated, if the heat cannot be timely emitted, the engine can be overheated, and therefore safety and reliability of the automobile are affected. Therefore, the automobile radiator core plays a very important role, and can radiate heat generated by the engine into the air, so that the engine is ensured not to overheat. However, the conventional automotive radiator core assembly method mainly adopts manual operation, and is low in efficiency and easy to cause quality problems.
Accordingly, an optimized automotive radiator core automatic assembly scheme is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides an automatic assembly system and a method for an automobile radiator core, wherein radiating fins meeting preset requirements are manufactured; mounting a water tank and a water pipe on the radiating fin to obtain a radiating core; mounting a bracket on the radiating core to obtain an automobile radiator core; and detecting surface defects of the assembled automobile radiator core. Therefore, the quality and the size suitability of the automobile radiator core body can be ensured, and further the automobile radiator core body can be ensured to effectively perform engine heat dissipation, and the normal operation of an engine is ensured.
In a first aspect, an automotive radiator core automatic assembly method is provided, comprising:
manufacturing a radiating fin meeting preset requirements;
mounting a water tank and a water pipe on the radiating fin to obtain a radiating core;
mounting a bracket on the radiating core to obtain an automobile radiator core;
and detecting surface defects of the assembled automobile radiator core.
In the above-mentioned auto radiator core automatic assembly method, the surface defect detection is performed on the assembled auto radiator core, including: acquiring a surface detection image of the assembled automobile radiator core; performing image blocking processing on the surface detection image to obtain a sequence of image blocks; each image block in the sequence of image blocks is passed through a convolutional neural network model comprising a depth feature fusion module to obtain a plurality of image block feature matrixes; arranging the image block feature matrixes into two-dimensional feature matrixes according to the positions of the image blocks, and then obtaining a classification feature map through a convolutional neural network model serving as a feature extractor; performing feature distribution optimization on the classification feature map to obtain an optimized classification feature map; and passing the optimized classification characteristic diagram through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the surface of the assembled automobile radiator core has defects.
In the above method for automatically assembling an automotive radiator core, passing each image block in the sequence of image blocks through a convolutional neural network model including a depth feature fusion module to obtain a plurality of image block feature matrices, including: extracting a shallow feature map from a shallow layer of the convolutional neural network model comprising the deep and shallow feature fusion module; extracting a deep feature map from the deep layer of the convolutional neural network model comprising the deep-shallow feature fusion module; using a depth feature fusion module of the convolutional neural network model comprising a depth feature fusion module to fuse the shallow feature map and the deep feature map to obtain a fused feature map; and carrying out global mean pooling on the fusion feature map along the channel dimension to obtain the image block feature matrixes.
In the above automatic assembling method for an automotive radiator core, the steps of arranging the plurality of image block feature matrices into a two-dimensional feature matrix according to the positions of the image blocks, and obtaining a classification feature map by using a convolutional neural network model as a feature extractor include: and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transfer of layers by using each layer of the convolutional neural network model serving as the feature extractor, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the two-dimensional feature matrix.
In the above automatic assembling method for an automotive radiator core, performing feature distribution optimization on the classification feature map to obtain an optimized classification feature map, including: carrying out secondary regularization on Gaussian probability density parameters of the manifold curved surface on the classification feature map by using the following optimization formula to obtain the optimized classification feature map; wherein, the optimization formula is:
f′ i，j，k ＝(μσ)f i，j，k 2 +f i，j，k μ+(f i，j，k -σ)μ 2
wherein f i，j，k Is the characteristic value of the (i, j, k) th position of the classification characteristic diagram, mu and sigma are the mean value and standard deviation of the characteristic value sets of the respective positions of the classification characteristic diagram, and f' i，j，k Is the eigenvalue of the (i, j, k) th position of the optimized classification characteristic graph.
In the above automatic assembling method of an automobile radiator core, the classifying result obtained by passing the optimized classifying feature map through a classifier is used for indicating whether the surface of the assembled automobile radiator core has defects, and the method comprises the following steps: expanding the optimized classification 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.
In a second aspect, an automotive radiator core automatic assembly system is provided, comprising:
the radiating fin manufacturing module is used for manufacturing radiating fins meeting preset requirements;
the first installation module is used for installing the water tank and the water pipe on the radiating fin to obtain a radiating core body;
the second installation module is used for installing a bracket on the radiating core body to obtain an automobile radiator core body;
and the surface defect detection module is used for detecting the surface defect of the assembled automobile radiator core.
In the above-mentioned auto radiator core assembly system, the surface defect detection module includes: an image acquisition unit for acquiring a surface detection image of the assembled automobile radiator core; the image blocking processing unit is used for carrying out image blocking processing on the surface detection image to obtain a sequence of image blocks; the depth feature fusion unit is used for enabling each image block in the sequence of the image blocks to pass through a convolutional neural network model comprising a depth feature fusion module so as to obtain a plurality of image block feature matrixes; the feature extraction unit is used for arranging the feature matrixes of the image blocks into a two-dimensional feature matrix according to the positions of the image blocks and obtaining a classification feature map through a convolutional neural network model serving as a feature extractor; the feature distribution optimizing unit is used for carrying out feature distribution optimization on the classification feature map so as to obtain an optimized classification feature map; and a surface result generating unit for passing the optimized classification characteristic map through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the surface of the assembled automobile radiator core has defects.
In the above-mentioned auto-assembly system for automotive radiator core, the depth feature fusion unit includes: a shallow layer extraction subunit, configured to extract a shallow layer feature map from a shallow layer of the convolutional neural network model including the depth feature fusion module; a deep layer extraction subunit, configured to extract a deep layer feature map from a deep layer of the convolutional neural network model that includes a deep-shallow feature fusion module; the fusion subunit is used for fusing the shallow feature map and the deep feature map by using a depth feature fusion module of the convolutional neural network model comprising a depth feature fusion module so as to obtain a fusion feature map; and a mean Chi Huazi unit configured to pool the global mean of the fused feature graphs along a channel dimension to obtain the plurality of image block feature matrices.
In the above-mentioned auto radiator core assembly system, the feature extraction unit is configured to: and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transfer of layers by using each layer of the convolutional neural network model serving as the feature extractor, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the two-dimensional feature matrix.
Compared with the prior art, the automatic assembly system and the method for the automobile radiator core, provided by the application, have the advantages that the radiating fins meeting the preset requirements are manufactured; mounting a water tank and a water pipe on the radiating fin to obtain a radiating core; mounting a bracket on the radiating core to obtain an automobile radiator core; and detecting surface defects of the assembled automobile radiator core. Therefore, the quality and the size suitability of the automobile radiator core body can be ensured, and further the automobile radiator core body can be ensured to effectively perform engine heat dissipation, and the normal operation of an engine is ensured.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic view of a scene of an automatic assembly method of an automotive radiator core according to an embodiment of the application.
Fig. 2 is a flowchart of an auto radiator core assembly method according to an embodiment of the present application.
Fig. 3 is a flowchart of the sub-steps of step 140 of an auto radiator core assembly method according to an embodiment of the present application.
Fig. 4 is a schematic diagram of the structure of step 140 in the auto-assembling method of the core of the radiator for an automobile according to the embodiment of the application.
Fig. 5 is a flow chart of the sub-steps of step 143 of the auto radiator core assembly method of an automotive vehicle according to an embodiment of the application.
Fig. 6 is a flowchart of the substep of step 146 in an automotive radiator core automatic assembly method according to an embodiment of the application.
Fig. 7 is a block diagram of an automotive radiator core automatic assembly system according to an embodiment of the application.
Fig. 8 is a block diagram of the surface defect detection module in an automotive radiator core automatic assembly system according to an embodiment of the application.
Detailed Description
The following description of the technical solutions according to the embodiments of the present application will be given with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Unless defined otherwise, all technical and scientific terms used in the embodiments of the application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In describing embodiments of the present application, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.
It should be noted that, the term "first\second\third" related to the embodiment of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the application described herein may be practiced in sequences other than those illustrated or described herein.
As described above, a large amount of heat is generated when the automobile engine is operated, and if the heat cannot be timely dissipated, the engine is overheated, thereby affecting the safety and reliability of the automobile. Therefore, the automobile radiator core plays a very important role, and can radiate heat generated by the engine into the air, so that the engine is ensured not to overheat. However, the conventional automotive radiator core assembly method mainly adopts manual operation, and is low in efficiency and easy to cause quality problems. Accordingly, an optimized automotive radiator core automatic assembly scheme is desired.
It should be understood that the automotive radiator core is a core component of an automotive radiator system, and is mainly composed of a water tank, a water pipe, a radiator fin, a bracket and the like. Specifically, the water tank is two end parts of the radiator core, which are connected with the water inlet and the water outlet and used for storing cooling liquid. The water pipe is a main body portion of the radiator core, which penetrates the water tank to form a flow passage of the cooling liquid. The fin is a peripheral portion of the radiator core, which is in close contact with the water pipe for increasing a radiating area and transferring heat. The bracket is an auxiliary part of the radiator core, and is used for fixing and supporting the radiator core to prevent the radiator core from being deformed and damaged.
Based on the above, in the technical scheme of the application, an automatic assembly method of an automobile radiator core is provided, which comprises the following steps: manufacturing a radiating fin: and manufacturing the radiating fin meeting the preset design requirements. Installing a water tank and a water pipe: and installing the water tank and the water pipe on the radiating fin, and accurately positioning and connecting to obtain the radiating core. And (2) mounting a bracket: and installing the bracket on the radiating core body to ensure the stability of the whole radiator core body, thereby obtaining the automobile radiator core body. Defect detection: and detecting surface defects of the assembled automobile radiator core.
Accordingly, considering that it is particularly important to detect defects of the automobile radiator core in the actual assembly process of the automobile radiator core, whether the size and the shape of the radiator core meet the design requirements or not can be timely found, whether the radiator core has defects of deformation, cracks, damage and the like or not is facilitated, and the follow-up radiator core can effectively play a role of the radiator core so as to ensure normal operation of an engine. In view of this, in the technical solution of the present application, it is desirable to perform defect detection of the automobile radiator core by analyzing the surface detection image of the assembled automobile radiator core. However, since a large amount of characteristic information about the automobile radiator core exists in the surface detection image, it is difficult to capture and delineate the quality characteristic information and the dimensional defect characteristic information of the automobile radiator core, resulting in difficulty in ensuring the manufacturing quality of the automobile radiator core. Therefore, in this process, the difficulty is how to fully express the quality implicit characteristic information about the automobile radiator core in the surface detection image, so as to ensure the quality and the size suitability of the automobile radiator core, further ensure that the automobile radiator core can effectively perform engine heat dissipation, and ensure the normal operation of the engine.
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. The development of deep learning and neural networks provides new solutions and solutions for mining quality implicit characteristic information about the core of an automotive radiator in the surface detection image.
Specifically, in the technical scheme of the application, first, a surface detection image of an assembled automobile radiator core is acquired. Next, considering that the implicit characteristic of the surface detection image regarding the defect of the automobile radiator core body is a fine characteristic of a small scale, in order to improve the expression capability of the surface detection image regarding the defect of the automobile radiator core body, the defect detection accuracy of the automobile radiator core body is improved. It will be appreciated that the dimensions of the individual image blocks in the sequence of image blocks are reduced compared to the original image, and therefore that the underlying features of the automotive radiator core mass defect in relation to small dimensions in the automotive radiator core are no longer small-sized objects in the individual image blocks for subsequent defect detection.
Then, feature mining of each image block in the sequence of image blocks is performed using a convolutional neural network model having excellent performance in implicit feature extraction of images, and in particular, in consideration of shallow features in each image block, such as shape, contour, texture, etc., regarding the automobile radiator core, which have important significance for surface defect detection of the automobile radiator core, in order to enable more accurate detection of surface defects and sizes of the automobile radiator core when extracting the hidden features of each image block in the sequence of image blocks. While convolutional neural networks are coded, as their depth deepens, shallow features become blurred and even buried in noise. Therefore, in the technical scheme of the application, each image block in the sequence of image blocks is processed through a convolutional neural network model comprising a depth feature fusion module to obtain a plurality of image block feature matrixes. It should be appreciated that the convolutional neural network model according to the present application can preserve shallow features and deep features of the respective image blocks with respect to the automobile radiator core, so that not only feature information is more abundant, but also features of different depths can be preserved to improve the accuracy of surface defect detection of the automobile radiator core.
Further, in order to accurately detect defects by considering that there is a correlation between depth fusion feature information about surface quality defects of an automobile radiator core in each image block, in order to enable sufficient expression of the surface quality defect features of the automobile radiator core, in the technical solution of the present application, the plurality of image block feature matrices are further arranged as two-dimensional feature matrices according to positions of the image blocks and then processed in a convolutional neural network model serving as a feature extractor, so as to extract semantic correlation feature information about the surface quality defect features of the automobile radiator core in each image block, thereby obtaining a classification feature map.
And then, further classifying the classification characteristic map through a classifier to classify by utilizing semantic relevance characteristic distribution information about the surface quality defect characteristics of the automobile radiator core based on the whole surface detection image in each image block, thereby realizing defect detection of the automobile radiator core and obtaining a classification result for indicating whether defects exist on the surface of the assembled automobile radiator core. That is, in the technical solution of the present application, the tag of the classifier includes that the surface of the assembled automobile radiator core is defective (first tag) and that the surface of the assembled automobile radiator core is not defective (second tag), wherein the classifier determines to which classification tag the classification feature map belongs through a soft maximum function. It is noted that the first tag p1 and the second tag p2 do not include the concept of artificial setting, and in fact, during the training process, the computer model does not have the concept of "whether there are defects on the surface of the assembled radiator core of the automobile", which is simply two kinds of classification tags, and the probability that the output characteristics are under the two classification tags, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether the surface of the assembled automobile radiator core is defective is actually that the classification label is converted into a classification probability distribution conforming to the natural law, and the physical meaning of the natural probability distribution of the label is essentially used instead of the language text meaning of whether the surface of the assembled automobile radiator core is defective. It should be understood that, in the technical solution of the present application, the classification label of the classifier is a detection and evaluation label for detecting whether a defect exists on the surface of the assembled automobile radiator core, so after the classification result is obtained, the surface defect detection of the automobile radiator core can be performed based on the classification result, so as to ensure the quality and the size suitability of the automobile radiator core.
Particularly, in the technical scheme of the application, when a plurality of image block feature matrixes are obtained by passing each image block in the sequence of the image blocks through the convolutional neural network model comprising the depth feature fusion module, the convolutional neural network model comprising the depth feature fusion module can extract shallow and deep image semantic features of a single image block, but the feature extraction can be converged in different directions due to different source image semantic contents of each image block, so that the feature distribution of a two-dimensional feature matrix obtained by arranging the plurality of image block feature matrixes is irregular. Moreover, such irregular feature distribution may further cause the problem of irregularity in the feature distribution of the obtained classification feature map to affect the classification accuracy of the classification feature map after feature extraction of the local associated features via the convolutional neural network model as the feature extractor.
Based on the above, the applicant of the present application performs gaussian probability density parameter quadratic regularization on the manifold curved surface of the classification feature map F, specifically expressed as:
f′ i，j，k ＝(μσ)f i，j，k 2 +f i，j，k μ+(f i，j，k -σ)μ 2
wherein μ and σ are the feature value set f i，j，k E means and standard deviation of F, and F i， ′ j，k Is the eigenvalue of the (i, j, k) th position of the classification characteristic diagram F' after optimization.
Specifically, in order to solve the problem of irregular distribution of high-dimensional feature distribution of the feature set of the classification feature map F in a high-dimensional feature space, secondary regularization of each feature value of the classification feature map F is performed by using feature values for likelihood of gaussian probability density parameters of class probability distribution of a classifier, so that equidistant distribution in a parameter space of the gaussian probability density parameters based on target class probability is subjected to smooth constraint of the feature values, and regular reformation of an original probability density likelihood function expressed by a manifold curved surface of the high-dimensional feature in the parameter space is obtained, so that the regularity of the feature distribution of the optimized classification feature map F 'is improved, and the classification accuracy of the optimized classification feature map F' passing through the classifier is improved. Therefore, the surface defect detection of the automobile radiator core can be accurately performed, so that the quality and the size suitability of the automobile radiator core are guaranteed, the automobile radiator core can be further guaranteed to effectively perform engine heat dissipation, and the normal operation of an engine is guaranteed.
Fig. 1 is a schematic view of a scene of an automatic assembly method of an automotive radiator core according to an embodiment of the application. As shown in fig. 1, in this application scenario, first, a surface detection image (e.g., C as illustrated in fig. 1) of the assembled automotive radiator core (e.g., M as illustrated in fig. 1) is acquired; the acquired surface detection image is then input into a server (e.g., S as illustrated in fig. 1) that deploys an auto radiator core assembly algorithm, wherein the server is capable of processing the surface detection image based on the auto radiator core assembly algorithm to generate a classification result that indicates whether the surface of the assembled auto radiator core is defective.
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.
In one embodiment of the present application, FIG. 2 is a flow chart of a method for automatically assembling an automotive radiator core in accordance with an embodiment of the present application. As shown in fig. 2, an automatic assembly method 100 for a radiator core of an automobile according to an embodiment of the present application includes: 110, manufacturing a radiating fin meeting preset requirements; 120, mounting a water tank and a water pipe on the radiating fin to obtain a radiating core; 130, mounting a bracket on the heat-dissipating core to obtain an automotive radiator core; 140, performing surface defect detection on the assembled automobile radiator core.
Fig. 3 is a flowchart of the sub-steps of step 140 of an auto radiator core assembly method according to an embodiment of the present application. As shown in fig. 3, the surface defect detection of the assembled automobile radiator core includes: 141, acquiring a surface detection image of the assembled automobile radiator core; 142, performing image blocking processing on the surface detection image to obtain a sequence of image blocks; 143, passing each image block in the sequence of image blocks through a convolutional neural network model comprising a depth feature fusion module to obtain a plurality of image block feature matrices; 144, arranging the image block feature matrixes into two-dimensional feature matrixes according to the positions of the image blocks, and obtaining a classification feature map through a convolutional neural network model serving as a feature extractor; 145, optimizing the characteristic distribution of the classification characteristic map to obtain an optimized classification characteristic map; and, 146, passing the optimized classification characteristic map through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the surface of the assembled automobile radiator core has defects.
As described above, a large amount of heat is generated when the automobile engine is operated, and if the heat cannot be timely dissipated, the engine is overheated, thereby affecting the safety and reliability of the automobile. Therefore, the automobile radiator core plays a very important role, and can radiate heat generated by the engine into the air, so that the engine is ensured not to overheat. However, the conventional automotive radiator core assembly method mainly adopts manual operation, and is low in efficiency and easy to cause quality problems. Accordingly, an optimized automotive radiator core automatic assembly scheme is desired.
It should be understood that the automotive radiator core is a core component of an automotive radiator system, and is mainly composed of a water tank, a water pipe, a radiator fin, a bracket and the like. Specifically, the water tank is two end parts of the radiator core, which are connected with the water inlet and the water outlet and used for storing cooling liquid. The water pipe is a main body portion of the radiator core, which penetrates the water tank to form a flow passage of the cooling liquid. The fin is a peripheral portion of the radiator core, which is in close contact with the water pipe for increasing a radiating area and transferring heat. The bracket is an auxiliary part of the radiator core, and is used for fixing and supporting the radiator core to prevent the radiator core from being deformed and damaged.
Based on the above, in the technical scheme of the application, an automatic assembly method of an automobile radiator core is provided, which comprises the following steps: manufacturing a radiating fin: and manufacturing the radiating fin meeting the preset design requirements. Installing a water tank and a water pipe: and installing the water tank and the water pipe on the radiating fin, and accurately positioning and connecting to obtain the radiating core. And (2) mounting a bracket: and installing the bracket on the radiating core body to ensure the stability of the whole radiator core body, thereby obtaining the automobile radiator core body. Defect detection: and detecting surface defects of the assembled automobile radiator core.
Fig. 4 is a schematic diagram of the structure of step 140 in the auto-assembling method of the core of the radiator for an automobile according to the embodiment of the application. As shown in fig. 4, in the network architecture, first, a surface detection image of the assembled automobile radiator core is acquired; then, carrying out image blocking processing on the surface detection image to obtain a sequence of image blocks; then, each image block in the sequence of image blocks is passed through a convolutional neural network model comprising a depth feature fusion module to obtain a plurality of image block feature matrixes; then, arranging the image block feature matrixes into two-dimensional feature matrixes according to the positions of the image blocks, and obtaining a classification feature map through a convolutional neural network model serving as a feature extractor; then, carrying out feature distribution optimization on the classification feature map to obtain an optimized classification feature map; and finally, the optimized classification characteristic diagram is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the surface of the assembled automobile radiator core has defects.
Specifically, in step 141, a surface inspection image of the assembled automotive radiator core is acquired. Accordingly, considering that it is particularly important to detect defects of the automobile radiator core in the actual assembly process of the automobile radiator core, whether the size and the shape of the radiator core meet the design requirements or not can be timely found, whether the radiator core has defects of deformation, cracks, damage and the like or not is facilitated, and the follow-up radiator core can effectively play a role of the radiator core so as to ensure normal operation of an engine.
In view of this, in the technical solution of the present application, it is desirable to perform defect detection of the automobile radiator core by analyzing the surface detection image of the assembled automobile radiator core. However, since a large amount of characteristic information about the automobile radiator core exists in the surface detection image, it is difficult to capture and delineate the quality characteristic information and the dimensional defect characteristic information of the automobile radiator core, resulting in difficulty in ensuring the manufacturing quality of the automobile radiator core. Therefore, in this process, the difficulty is how to fully express the quality implicit characteristic information about the automobile radiator core in the surface detection image, so as to ensure the quality and the size suitability of the automobile radiator core, further ensure that the automobile radiator core can effectively perform engine heat dissipation, and ensure the normal operation of the engine.
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. The development of deep learning and neural networks provides new solutions and solutions for mining quality implicit characteristic information about the core of an automotive radiator in the surface detection image.
Specifically, in the technical scheme of the application, first, a surface detection image of an assembled automobile radiator core is acquired.
Specifically, in step 142, the surface detection image is subjected to image blocking processing to obtain a sequence of image blocks. Next, considering that the implicit characteristic of the surface detection image regarding the defect of the automobile radiator core body is a fine characteristic of a small scale, in order to improve the expression capability of the surface detection image regarding the defect of the automobile radiator core body, the defect detection accuracy of the automobile radiator core body is improved. It will be appreciated that the dimensions of the individual image blocks in the sequence of image blocks are reduced compared to the original image, and therefore that the underlying features of the automotive radiator core mass defect in relation to small dimensions in the automotive radiator core are no longer small-sized objects in the individual image blocks for subsequent defect detection.
Specifically, in step 143, each image block in the sequence of image blocks is passed through a convolutional neural network model including a depth feature fusion module to obtain a plurality of image block feature matrices. Then, feature mining of each image block in the sequence of image blocks is performed using a convolutional neural network model having excellent performance in implicit feature extraction of images, and in particular, in consideration of shallow features in each image block, such as shape, contour, texture, etc., regarding the automobile radiator core, which have important significance for surface defect detection of the automobile radiator core, in order to enable more accurate detection of surface defects and sizes of the automobile radiator core when extracting the hidden features of each image block in the sequence of image blocks. While convolutional neural networks are coded, as their depth deepens, shallow features become blurred and even buried in noise.
Therefore, in the technical scheme of the application, each image block in the sequence of image blocks is processed through a convolutional neural network model comprising a depth feature fusion module to obtain a plurality of image block feature matrixes. It should be appreciated that the convolutional neural network model according to the present application can preserve shallow features and deep features of the respective image blocks with respect to the automobile radiator core, so that not only feature information is more abundant, but also features of different depths can be preserved to improve the accuracy of surface defect detection of the automobile radiator core.
Fig. 5 is a flowchart of the substeps of step 143 in the auto radiator core assembly method according to an embodiment of the present application, as shown in fig. 5, the step of passing each image block in the sequence of image blocks through a convolutional neural network model including a depth feature fusion module to obtain a plurality of image block feature matrices includes: 1431, extracting a shallow feature map from a shallow layer of the convolutional neural network model comprising the depth feature fusion module; 1432, extracting a deep feature map from the deep layer of the convolutional neural network model comprising the deep-shallow feature fusion module; 1433, fusing the shallow feature map and the deep feature map by using a depth feature fusion module of the convolutional neural network model comprising a depth feature fusion module to obtain a fused feature map; and 1434, performing global averaging on the fusion feature map along a channel dimension to obtain the feature matrices of the plurality of image blocks.
It should be appreciated that, compared to a standard convolutional neural network model, the convolutional neural network model according to the present application can retain the shallow features and deep features of each image block in the sequence of image blocks, so that not only feature information is more abundant, but also features of different depths can be retained to improve the accuracy of classification results. Meanwhile, the structure of the deep neural network is complex, a large amount of sample data is needed for training and adjusting, the training time of the deep neural network is long, and fitting is easy. Therefore, in the design of the neural network model, the combination of the shallow network and the deep network is generally adopted, and through depth feature fusion, the complexity of the network and the risk of overfitting can be reduced to a certain extent, and meanwhile, the feature extraction capability and the generalization capability of the model are improved.
In encoding each image block in the sequence of image blocks using a convolutional neural network model, a shallow feature map is first extracted from a shallow layer of the convolutional neural network model (e.g., the shallow layer refers to a first layer through a sixth layer), and a deep feature map is extracted from a deep layer of the convolutional neural network model (e.g., a last layer of the convolutional neural network model), and then feature representations including shallow features and deep features are obtained by fusing the shallow feature map and the deep feature map. In a specific encoding process, the extraction position of the shallow feature map is determined by the overall network depth of the convolutional neural network model, for example, from layer 3 of the convolutional neural network model when the network depth is 30, and from layer 4 of the convolutional neural network model when the network depth is 40, which is not limited by the present application. Likewise, the extraction position of the deep feature map is not limited by the present application, and may be the last layer, the last but one layer, or the last but one layer and the last but one layer.
Specifically, in step 144, the plurality of image block feature matrices are arranged into two-dimensional feature matrices according to the positions of the image blocks, and then a classification feature map is obtained through a convolutional neural network model serving as a feature extractor. Further, in order to accurately detect defects by considering that there is a correlation between depth fusion feature information about surface quality defects of an automobile radiator core in each image block, in order to enable sufficient expression of the surface quality defect features of the automobile radiator core, in the technical solution of the present application, the plurality of image block feature matrices are further arranged as two-dimensional feature matrices according to positions of the image blocks and then processed in a convolutional neural network model serving as a feature extractor, so as to extract semantic correlation feature information about the surface quality defect features of the automobile radiator core in each image block, thereby obtaining a classification feature map.
The method for obtaining the classification feature map by using the convolutional neural network model as a feature extractor after arranging the feature matrixes of the image blocks into two-dimensional feature matrixes according to the positions of the image blocks comprises the following steps: and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transfer of layers by using each layer of the convolutional neural network model serving as the feature extractor, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the two-dimensional feature matrix.
The convolutional neural network (Convolutional Neural Network, CNN) is an artificial neural network and has wide application in the fields of image recognition and the like. The convolutional neural network may include an input layer, a hidden layer, and an output layer, where the hidden layer may include a convolutional layer, a pooling layer, an activation layer, a full connection layer, etc., where the previous layer performs a corresponding operation according to input data, outputs an operation result to the next layer, and obtains a final result after the input initial data is subjected to a multi-layer operation.
The convolutional neural network model has excellent performance in the aspect of image local feature extraction by taking a convolutional kernel as a feature filtering factor, and has stronger feature extraction generalization capability and fitting capability compared with the traditional image feature extraction algorithm based on statistics or feature engineering.
Specifically, in step 145, the classification feature map is feature distribution optimized to obtain an optimized classification feature map. Particularly, in the technical scheme of the application, when a plurality of image block feature matrixes are obtained by passing each image block in the sequence of the image blocks through the convolutional neural network model comprising the depth feature fusion module, the convolutional neural network model comprising the depth feature fusion module can extract shallow and deep image semantic features of a single image block, but the feature extraction can be converged in different directions due to different source image semantic contents of each image block, so that the feature distribution of a two-dimensional feature matrix obtained by arranging the plurality of image block feature matrixes is irregular. Moreover, such irregular feature distribution may further cause the problem of irregularity in the feature distribution of the obtained classification feature map to affect the classification accuracy of the classification feature map after feature extraction of the local associated features via the convolutional neural network model as the feature extractor.
Based on the above, the applicant of the present application performs gaussian probability density parameter quadratic regularization on the manifold curved surface of the classification feature map F, specifically expressed as: carrying out secondary regularization on Gaussian probability density parameters of the manifold curved surface on the classification feature map by using the following optimization formula to obtain the optimized classification feature map; wherein, the optimization formula is:
f′ i，j，k ＝(μσ)f i，j，k 2 +f i，j，k μ+(f i，j，k -σ)μ 2
wherein f i，j，k Is the characteristic value of the (i, j, k) th position of the classification characteristic diagram, mu and sigma are the mean value and standard deviation of the characteristic value sets of the respective positions of the classification characteristic diagram, and f' i，j，k Is the eigenvalue of the (i, j, k) th position of the optimized classification characteristic graph.
Specifically, in order to solve the problem of irregular distribution of high-dimensional feature distribution of the feature set of the classification feature map F in a high-dimensional feature space, secondary regularization of each feature value of the classification feature map F is performed by using feature values for likelihood of gaussian probability density parameters of class probability distribution of a classifier, so that equidistant distribution in a parameter space of the gaussian probability density parameters based on target class probability is subjected to smooth constraint of the feature values, and regular reformation of an original probability density likelihood function expressed by a manifold curved surface of the high-dimensional feature in the parameter space is obtained, so that the regularity of the feature distribution of the optimized classification feature map F 'is improved, and the classification accuracy of the optimized classification feature map F' passing through the classifier is improved. Therefore, the surface defect detection of the automobile radiator core can be accurately performed, so that the quality and the size suitability of the automobile radiator core are guaranteed, the automobile radiator core can be further guaranteed to effectively perform engine heat dissipation, and the normal operation of an engine is guaranteed.
Specifically, in step 146, the optimized classification feature map is passed through a classifier to obtain a classification result that is used to indicate whether the surface of the assembled radiator core is defective. And then, further classifying the classification characteristic map through a classifier to classify by utilizing semantic relevance characteristic distribution information about the surface quality defect characteristics of the automobile radiator core based on the whole surface detection image in each image block, thereby realizing defect detection of the automobile radiator core and obtaining a classification result for indicating whether defects exist on the surface of the assembled automobile radiator core.
That is, in the technical solution of the present application, the tag of the classifier includes that the surface of the assembled automobile radiator core is defective (first tag) and that the surface of the assembled automobile radiator core is not defective (second tag), wherein the classifier determines to which classification tag the classification feature map belongs through a soft maximum function. It is noted that the first tag p1 and the second tag p2 do not include the concept of artificial setting, and in fact, during the training process, the computer model does not have the concept of "whether there are defects on the surface of the assembled radiator core of the automobile", which is simply two kinds of classification tags, and the probability that the output characteristics are under the two classification tags, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether the surface of the assembled automobile radiator core is defective is actually that the classification label is converted into a classification probability distribution conforming to the natural law, and the physical meaning of the natural probability distribution of the label is essentially used instead of the language text meaning of whether the surface of the assembled automobile radiator core is defective.
It should be understood that, in the technical solution of the present application, the classification label of the classifier is a detection and evaluation label for detecting whether a defect exists on the surface of the assembled automobile radiator core, so after the classification result is obtained, the surface defect detection of the automobile radiator core can be performed based on the classification result, so as to ensure the quality and the size suitability of the automobile radiator core.
Fig. 6 is a flowchart of the substeps of step 146 in the auto-assembling method of a radiator core for an automobile according to an embodiment of the present application, as shown in fig. 6, the optimized classification feature map is passed through a classifier to obtain a classification result, where the classification result is used to indicate whether a surface of the assembled radiator core for an automobile is defective, and the method includes: 1461, expanding the optimized classification feature map into classification feature vectors according to row vectors or column vectors; 1462, 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 1463, passing the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
Processing the optimized classification characteristic diagram by using the classifier according to the following formula to obtain the classification result; wherein, the formula is: o=softmax { (W) n ，B n )：...：(W 1 ，B 1 ) Project (F), where W 1 To W n Is a weight matrix, B 1 To B n For bias vectors Project (F) is to Project the optimized classification feature map as a vector.
In summary, an auto radiator core assembly method 100 according to an embodiment of the present application is illustrated, which produces a heat sink that meets predetermined requirements; mounting a water tank and a water pipe on the radiating fin to obtain a radiating core; mounting a bracket on the radiating core to obtain an automobile radiator core; and detecting surface defects of the assembled automobile radiator core. Therefore, the quality and the size suitability of the automobile radiator core body can be ensured, and further the automobile radiator core body can be ensured to effectively perform engine heat dissipation, and the normal operation of an engine is ensured.
In one embodiment of the present application, FIG. 7 is a block diagram of an automotive radiator core automatic assembly system according to an embodiment of the present application. As shown in fig. 7, an automotive radiator core automatic assembly system 200 according to an embodiment of the present application includes: a fin manufacturing module 210 for manufacturing a fin meeting a predetermined requirement; a first mounting module 220 for mounting a water tank and a water pipe on the radiating fin to obtain a radiating core; a second mounting module 230 for mounting a bracket on the heat radiating core to obtain an automotive radiator core; and the surface defect detection module 240 is used for detecting the surface defect of the assembled automobile radiator core.
Fig. 8 is a block diagram of the surface defect detection module in an automotive radiator core automatic assembly system according to an embodiment of the application. As shown in fig. 8, the surface defect detection module 240 includes: an image acquisition unit 241 for acquiring a surface detection image of the assembled automobile radiator core; an image blocking processing unit 242, configured to perform image blocking processing on the surface detection image to obtain a sequence of image blocks; the depth feature fusion unit 243 is configured to obtain a plurality of image block feature matrices by passing each image block in the sequence of image blocks through a convolutional neural network model including a depth feature fusion module; a feature extraction unit 244, configured to arrange the feature matrices of the plurality of image blocks into two-dimensional feature matrices according to the positions of the image blocks, and obtain a classification feature map through a convolutional neural network model serving as a feature extractor; the feature distribution optimizing unit 245 is configured to perform feature distribution optimization on the classification feature map to obtain an optimized classification feature map; and a surface result generating unit 246 for passing the optimized classification characteristic map through a classifier to obtain a classification result indicating whether or not the surface of the assembled radiator core is defective.
In a specific example, in the above automotive radiator core automatic assembly system, the depth feature fusion unit includes: a shallow layer extraction subunit, configured to extract a shallow layer feature map from a shallow layer of the convolutional neural network model including the depth feature fusion module; a deep layer extraction subunit, configured to extract a deep layer feature map from a deep layer of the convolutional neural network model that includes a deep-shallow feature fusion module; the fusion subunit is used for fusing the shallow feature map and the deep feature map by using a depth feature fusion module of the convolutional neural network model comprising a depth feature fusion module so as to obtain a fusion feature map; and a mean Chi Huazi unit configured to pool the global mean of the fused feature graphs along a channel dimension to obtain the plurality of image block feature matrices.
In a specific example, in the above-described auto radiator core assembling system, the feature extraction unit is configured to: and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transfer of layers by using each layer of the convolutional neural network model serving as the feature extractor, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the two-dimensional feature matrix.
In a specific example, in the above-described automotive radiator core automatic assembly system, the feature distribution optimizing unit is configured to: carrying out secondary regularization on Gaussian probability density parameters of the manifold curved surface on the classification feature map by using the following optimization formula to obtain the optimized classification feature map; wherein, the optimization formula is:
f′ i，j，k ＝(μσ)f i，j，k 2 +f i，j，k μ+(f i，j，k -σ)μ 2
wherein f i，j，k Is the characteristic value of the (u, j, k) th position of the classification characteristic diagram, mu and sigma are the mean value and standard deviation of the characteristic value sets of the respective positions of the classification characteristic diagram, and f' i，j，k Is the eigenvalue of the (i, j, k) th position of the optimized classification characteristic graph.
In a specific example, in the above-described automotive radiator core automatic assembly system, the surface result generation unit includes: a development subunit, configured to develop the optimized classification feature map into a classification feature vector according to a row vector or a column vector; a full-connection coding subunit, configured to perform full-connection coding on the classification feature vector by using multiple full-connection layers of the classifier to obtain a coded classification feature vector; and a classification result subunit, configured to pass the encoded 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 functions and operations of the respective units and modules in the above-described auto radiator core assembling system have been described in detail in the above description of the auto radiator core assembling method with reference to fig. 1 to 6, and thus, repetitive descriptions thereof will be omitted.
As described above, the auto radiator core assembling system 200 according to the embodiment of the present application can be implemented in various terminal devices, such as a server for auto radiator core assembling, and the like. In one example, automotive radiator core automatic assembly system 200 according to an embodiment of the present application may be integrated into a terminal device as a software module and/or hardware module. For example, the auto radiator core automatic assembly system 200 may be a software module in the operating system of the terminal device or may be an application developed for the terminal device; of course, the automotive radiator core automatic assembly system 200 may also be one of a number of hardware modules of the terminal equipment.
Alternatively, in another example, the auto radiator core mounting system 200 and the terminal device may be separate devices, and the auto radiator core mounting system 200 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a agreed data format.
The present application also provides a computer program product comprising instructions which, when executed, cause an apparatus to perform operations corresponding to the above-described method.
In one embodiment of the present application, there is also provided a computer-readable storage medium storing a computer program for executing the above-described method.
It should be appreciated that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the forms of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects may be utilized. Furthermore, the computer program product may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Methods, systems, and computer program products of embodiments of the present application are described in the flow diagrams and/or block diagrams. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The basic principles of the present application have been described above in connection with specific embodiments, however, 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 considered as essential to the various embodiments of the present 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.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
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 (7)
1. An automatic assembly method for an automotive radiator core, comprising:
manufacturing a radiating fin meeting preset requirements;
mounting a water tank and a water pipe on the radiating fin to obtain a radiating core;
mounting a bracket on the radiating core to obtain an automobile radiator core;
detecting surface defects of the assembled automobile radiator core;
wherein, carry out surface defect detection to the car radiator core that the assembly is accomplished, include:
acquiring a surface detection image of the assembled automobile radiator core;
performing image blocking processing on the surface detection image to obtain a sequence of image blocks;
each image block in the sequence of image blocks is passed through a convolutional neural network model comprising a depth feature fusion module to obtain a plurality of image block feature matrixes;
arranging the image block feature matrixes into two-dimensional feature matrixes according to the positions of the image blocks, and then obtaining a classification feature map through a convolutional neural network model serving as a feature extractor;
performing feature distribution optimization on the classification feature map to obtain an optimized classification feature map; and
the optimized classification characteristic diagram is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the surface of the assembled automobile radiator core has defects or not;
The feature distribution optimization is performed on the classification feature map to obtain an optimized classification feature map, which comprises the following steps:
carrying out secondary regularization on Gaussian probability density parameters of the manifold curved surface on the classification feature map by using the following optimization formula to obtain the optimized classification feature map;
wherein, the optimization formula is:
wherein,is the +.o of the classification characteristic diagram>Characteristic value of the location->And->Is the mean and standard deviation of the respective sets of position feature values of the classification feature map, and +.>Is the +.f. of the optimized classification characteristic graph>Characteristic values of the location.
2. The auto-assembling method of an automotive radiator core according to claim 1, wherein passing each image block in the sequence of image blocks through a convolutional neural network model including a depth feature fusion module to obtain a plurality of image block feature matrices, comprises:
extracting a shallow feature map from a shallow layer of the convolutional neural network model comprising the deep and shallow feature fusion module;
extracting a deep feature map from the deep layer of the convolutional neural network model comprising the deep-shallow feature fusion module;
using a depth feature fusion module of the convolutional neural network model comprising a depth feature fusion module to fuse the shallow feature map and the deep feature map to obtain a fused feature map; and
And carrying out global mean pooling on the fusion feature map along the channel dimension to obtain the image block feature matrixes.
3. The automatic assembling method of an automotive radiator core according to claim 2, wherein the step of arranging the plurality of image block feature matrices into a two-dimensional feature matrix in accordance with the positions of the image blocks and then obtaining a classification feature map by a convolutional neural network model as a feature extractor includes: and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transfer of layers by using each layer of the convolutional neural network model serving as the feature extractor, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the two-dimensional feature matrix.
4. The auto-assembling method of an automotive radiator core according to claim 3, characterized in that the optimizing classification feature map is passed through a classifier to obtain a classification result indicating whether or not a surface of an assembled automotive radiator core is defective, comprising:
Expanding the optimized classification 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
and the coding classification feature vector is passed through a Softmax classification function of the classifier to obtain the classification result.
5. An automotive radiator core automatic assembly system, comprising:
the radiating fin manufacturing module is used for manufacturing radiating fins meeting preset requirements;
the first installation module is used for installing the water tank and the water pipe on the radiating fin to obtain a radiating core body;
the second installation module is used for installing a bracket on the radiating core body to obtain an automobile radiator core body;
the surface defect detection module is used for detecting surface defects of the assembled automobile radiator core;
wherein the surface defect detection module comprises:
an image acquisition unit for acquiring a surface detection image of the assembled automobile radiator core;
the image blocking processing unit is used for carrying out image blocking processing on the surface detection image to obtain a sequence of image blocks;
The depth feature fusion unit is used for enabling each image block in the sequence of the image blocks to pass through a convolutional neural network model comprising a depth feature fusion module so as to obtain a plurality of image block feature matrixes;
the feature extraction unit is used for arranging the feature matrixes of the image blocks into a two-dimensional feature matrix according to the positions of the image blocks and obtaining a classification feature map through a convolutional neural network model serving as a feature extractor;
the feature distribution optimizing unit is used for carrying out feature distribution optimization on the classification feature map so as to obtain an optimized classification feature map; and
the surface result generating unit is used for passing the optimized classification characteristic diagram through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the surface of the assembled automobile radiator core has defects or not;
wherein the feature distribution optimizing unit includes:
carrying out secondary regularization on Gaussian probability density parameters of the manifold curved surface on the classification feature map by using the following optimization formula to obtain the optimized classification feature map;
wherein, the optimization formula is:
wherein,is the +.o of the classification characteristic diagram>Characteristic value of the location->And->Is the mean and standard deviation of the respective sets of position feature values of the classification feature map, and +. >Is the +.f. of the optimized classification characteristic graph>Characteristic values of the location.
6. The automotive radiator core automatic assembling system according to claim 5, wherein the depth feature fusion unit includes:
a shallow layer extraction subunit, configured to extract a shallow layer feature map from a shallow layer of the convolutional neural network model including the depth feature fusion module;
a deep layer extraction subunit, configured to extract a deep layer feature map from a deep layer of the convolutional neural network model that includes a deep-shallow feature fusion module;
the fusion subunit is used for fusing the shallow feature map and the deep feature map by using a depth feature fusion module of the convolutional neural network model comprising a depth feature fusion module so as to obtain a fusion feature map; and
and the average Chi Huazi unit is used for carrying out global average pooling on the fusion feature map along the channel dimension so as to obtain the image block feature matrixes.
7. The automotive radiator core automatic assembling system according to claim 6, wherein the feature extraction unit is configured to: and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transfer of layers by using each layer of the convolutional neural network model serving as the feature extractor, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the two-dimensional feature matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310445030.2A CN116408647B (en) | 2023-04-23 | 2023-04-23 | Automatic assembly system and method for automobile radiator core |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310445030.2A CN116408647B (en) | 2023-04-23 | 2023-04-23 | Automatic assembly system and method for automobile radiator core |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116408647A CN116408647A (en) | 2023-07-11 |
| CN116408647B true CN116408647B (en) | 2023-11-14 |
Family
ID=87054462
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310445030.2A Active CN116408647B (en) | 2023-04-23 | 2023-04-23 | Automatic assembly system and method for automobile radiator core |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116408647B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102218653A (en) * | 2011-05-27 | 2011-10-19 | 重庆迅良机电设备有限公司 | Parallel flow technology-based system for manufacturing heat exchanger |
| CN103170807A (en) * | 2013-04-12 | 2013-06-26 | 惠州东风易进工业有限公司 | Production process and system for automotive radiators |
| CN103419022A (en) * | 2013-08-20 | 2013-12-04 | 天津市亚星散热器有限公司 | Automatic core assembling machine for core body of warm wind radiator and working method thereof |
| CN115564766A (en) * | 2022-11-09 | 2023-01-03 | 浙江振兴阿祥集团有限公司 | Method and system for preparing volute casing seat ring of water turbine |
| CN115965217A (en) * | 2023-01-10 | 2023-04-14 | 吉安吉之模型有限公司 | Intelligent production monitoring method and system for plastic model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112414164A (en) * | 2020-12-07 | 2021-02-26 | 惠州汉旭五金塑胶科技有限公司 | Multi-runner type efficient radiating water-cooling radiator |
-
2023
- 2023-04-23 CN CN202310445030.2A patent/CN116408647B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102218653A (en) * | 2011-05-27 | 2011-10-19 | 重庆迅良机电设备有限公司 | Parallel flow technology-based system for manufacturing heat exchanger |
| CN103170807A (en) * | 2013-04-12 | 2013-06-26 | 惠州东风易进工业有限公司 | Production process and system for automotive radiators |
| CN103419022A (en) * | 2013-08-20 | 2013-12-04 | 天津市亚星散热器有限公司 | Automatic core assembling machine for core body of warm wind radiator and working method thereof |
| CN115564766A (en) * | 2022-11-09 | 2023-01-03 | 浙江振兴阿祥集团有限公司 | Method and system for preparing volute casing seat ring of water turbine |
| CN115965217A (en) * | 2023-01-10 | 2023-04-14 | 吉安吉之模型有限公司 | Intelligent production monitoring method and system for plastic model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116408647A (en) | 2023-07-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108734210B (en) | Object detection method based on cross-modal multi-scale feature fusion | |
| Kang et al. | Deep learning-based weather image recognition | |
| CN107085731B (en) | Image classification method based on RGB-D fusion features and sparse coding | |
| CN116152267A (en) | Point cloud instance segmentation method based on contrast language image pre-training technology | |
| CN110188827B (en) | Scene recognition method based on convolutional neural network and recursive automatic encoder model | |
| CN112163114B (en) | Image retrieval method based on feature fusion | |
| CN116167989A (en) | Intelligent production method and system for aluminum cup | |
| CN112115806B (en) | Remote sensing image scene accurate classification method based on Dual-ResNet small sample learning | |
| CN112949510A (en) | Human detection method based on fast R-CNN thermal infrared image | |
| CN111126155B (en) | Pedestrian re-identification method for generating countermeasure network based on semantic constraint | |
| CN113222072A (en) | Lung X-ray image classification method based on K-means clustering and GAN | |
| CN116844126A (en) | YOLOv7 improved complex road scene target detection method | |
| US20150086118A1 (en) | Method for recognizing a visual context of an image and corresponding device | |
| CN112967296B (en) | Point cloud dynamic region graph convolution method, classification method and segmentation method | |
| CN116408647B (en) | Automatic assembly system and method for automobile radiator core | |
| CN116797533B (en) | Appearance defect detection method and system for power adapter | |
| CN111062409A (en) | Unsupervised image feature extraction and classification method | |
| CN117011274A (en) | Automatic glass bottle detection system and method thereof | |
| CN115861229A (en) | YOLOv5 s-based X-ray detection method for packaging defects of components | |
| Hong et al. | Neural matching fields: Implicit representation of matching fields for visual correspondence | |
| CN115049611A (en) | Continuous casting billet crack defect identification method based on improved yolov5 | |
| CN115063831A (en) | High-performance pedestrian retrieval and re-identification method and device | |
| CN114612707A (en) | Image automatic labeling method and device based on deep learning | |
| CN113177599A (en) | Enhanced sample generation method based on GAN | |
| CN115424250A (en) | License plate recognition method and device |
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 |