CN117773405A - Method for detecting brazing quality of automobile radiator - Google Patents
Method for detecting brazing quality of automobile radiator Download PDFInfo
- Publication number
- CN117773405A CN117773405A CN202410217074.4A CN202410217074A CN117773405A CN 117773405 A CN117773405 A CN 117773405A CN 202410217074 A CN202410217074 A CN 202410217074A CN 117773405 A CN117773405 A CN 117773405A
- Authority
- CN
- China
- Prior art keywords
- image
- radiator
- brazing
- infrared
- visible light
- 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.)
- Granted
Links
- 238000005219 brazing Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000003466 welding Methods 0.000 claims abstract description 17
- 230000004927 fusion Effects 0.000 claims description 7
- 230000003993 interaction Effects 0.000 claims description 6
- 238000010330 laser marking Methods 0.000 claims description 3
- 238000006748 scratching Methods 0.000 claims description 3
- 230000002393 scratching effect Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 abstract description 8
- 238000012797 qualification Methods 0.000 abstract description 8
- 238000001931 thermography Methods 0.000 abstract description 6
- 238000003331 infrared imaging Methods 0.000 abstract description 2
- 239000013589 supplement Substances 0.000 abstract description 2
- 239000010410 layer Substances 0.000 description 21
- 238000001816 cooling Methods 0.000 description 11
- 230000005855 radiation Effects 0.000 description 9
- 238000012545 processing Methods 0.000 description 8
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 238000005476 soldering Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000005457 Black-body radiation Effects 0.000 description 1
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 description 1
- 238000010009 beating Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000002344 surface layer Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of detection and application of automobile radiators, and particularly relates to a detection method of brazing quality of an automobile radiator. The invention provides a method for detecting the brazing quality of an automobile radiator, which utilizes the characteristic that after brazing is finished, the temperature of a position with high brazing qualification rate is higher than that of a position with low brazing qualification rate, obtains a thermal imaging diagram of the radiator in an infrared imaging mode, supplements the details of the thermal imaging diagram by utilizing a visible light diagram to form a high-resolution infrared diagram, further conveniently and accurately determines the brazing qualification rate, and simultaneously utilizes experience to determine a threshold value of the brazing qualification rate, further conveniently determines the final welding quality and remedies.
Description
Technical Field
The invention belongs to the technical field of detection and application of automobile radiators, and particularly relates to a detection method of brazing quality of an automobile radiator.
Background
The pipe-strip radiator core adopts cooling pipes and radiating strips to be arranged at intervals along the longitudinal direction. In order to improve the heat radiation capability, the heat radiation belt is generally provided with blind-shaped slot holes for destroying the surface layer of the air flow on the surface of the heat radiation belt so as to improve the heat radiation capability. Compared with the duct type radiator core, the radiator core has the advantages of strong heat radiation capability, simple manufacturing process, small mass, low cost and the like, and is widely applied to automobiles at present.
The existing fixing mode of the pipe-strip radiator is mainly to fix by means of plug-in matching brazing, and low-concentration soldering flux is sprayed on the surface of the radiator during processing, and then the radiator after brazing is completed through a continuous brazing furnace can be obtained.
Aiming at the detection of welding quality, the existing detection method mainly detects through manual naked eye observation and beating, and the detection can observe the non-soldered point, but the situation of the welding rate in the area cannot be known, so that the use quality cannot be ensured, the requirements of the automobile radiator are continuously increased along with the increase of the automobile holding quantity, the detection of the soldering rate before leaving the factory of the radiator is more important in order to ensure the heat exchange performance of the radiator and avoid the damage of the engine caused by overheat during working. For this reason, how to effectively, quickly and accurately detect the brazing quality is a problem that needs to be solved at present.
Disclosure of Invention
Aiming at the problem that the welding quality of the pipe-strip radiator cannot be accurately determined, the invention provides the detection method for the brazing quality of the automobile radiator, which is reasonable in design, convenient to operate and capable of effectively ensuring the welding quality.
In order to achieve the above purpose, the invention adopts the following technical scheme: the invention provides a method for detecting brazing quality of an automobile radiator, which comprises the following steps:
a. firstly, acquiring an infrared image and a visible light image of a radiator welded in a continuous brazing furnace through a thermal camera embedded with a visible light RGB camera;
b. then, the images of the radiator parts in the infrared image and the visible light image are scratched out by utilizing a digital image scratching algorithm, and the scratched infrared radiator image and visible light radiator image are obtained;
c. then, obtaining low-frequency information in the visible light radiator image and high-frequency information of the infrared radiator image, and obtaining a high-resolution infrared radiator image after reconstruction and fusion;
d. obtaining an interested brazing area by using the obtained high-resolution infrared radiator image in a marking and labeling mode;
e. carrying out gray scale treatment on the interested brazing area, and adopting a binarization algorithm to treat the interested brazing area subjected to gray scale treatment to obtain a binarized image;
f. calculating the proportion of the whole binary image for the white part pixels in the binary image, wherein the proportion is the welding rate;
g. and then, determining whether the interested brazing area meets the brazing requirement or not by setting a welding rate threshold value, and marking the area by a laser marking machine if the interested brazing area does not meet the brazing requirement.
Preferably, the step c includes the steps of:
c1, amplifying an infrared radiator image by adopting a bicubic interpolation algorithm, respectively passing the amplified infrared radiator image and visible light radiator image through two convolution layers with convolution kernel sizes of 3 multiplied by 3, and performing downsampling by using downsampling convolution to respectively extract shallow features of the infrared radiator image and the visible light radiator image;
c2, inputting the shallow features of the visible light radiator image into a wavelet block layer to extract low-frequency information in the image; meanwhile, extracting high-frequency information from the shallow image features of the infrared radiator through a flow Fourier residual error module;
c3, splicing the obtained low-frequency information and high-frequency information in the channel dimension to form an input tensor;
c4, acquiring and generating a fine tensor by adopting the input tensor by adopting the fused triple attention module;
and c5, inputting the fine tensor into the 3X 3 convolution layers connected by three layers for feature enhancement, and obtaining the high-resolution infrared radiator image.
Preferably, the wavelet layer comprises a convolution module, a two-dimensional discrete wavelet transform module and a convolution output module.
Preferably, in the step c2, the flow fourier residual module performs deconvolution on the infrared radiator image shallow features, then performs downsampling to be consistent with the infrared radiator image in size through downsampling convolution, and then performs difference with the infrared radiator image shallow features to obtain the high-frequency information.
Preferably, in the step c4, the input tensor is output according to the interaction between the channel dimension and the height dimension, the interaction between the channel dimension and the width dimension, and the interaction between the height dimension and the width dimension, and the three outputs are summarized by using a simple average value, so as to generate the fine tensor.
Compared with the prior art, the invention has the advantages and positive effects that:
1. the invention provides a method for detecting the brazing quality of an automobile radiator, which utilizes the characteristic that after brazing is finished, the temperature of a position with high brazing qualification rate is higher than that of a position with low brazing qualification rate, obtains a thermal imaging diagram of the radiator in an infrared imaging mode, supplements the details of the thermal imaging diagram by utilizing a visible light diagram to form a high-resolution infrared diagram, further conveniently and accurately determines the brazing qualification rate, and simultaneously utilizes experience to determine a threshold value of the brazing qualification rate, further conveniently determines the final welding quality and remedies.
Detailed Description
In order that the above objects, features and advantages of the invention may be more clearly understood, a further description of the invention will be provided with reference to the following examples. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced otherwise than as described herein, and therefore the present invention is not limited to the specific embodiments of the disclosure that follow.
In an embodiment, the embodiment provides a method for detecting brazing quality of an automobile radiator.
The welding seam between the radiating belt and the cooling pipe and the welding seam between the cooling pipe and the water chamber in the pipe-strip radiator after welding are difficult to directly observe, so that the welding seam rate cannot be directly detected. However, according to the thermal imaging principle, although the materials of the radiating belt and the cooling pipe which are completely welded and the cooling pipe and the water chamber are different, the temperature difference at the output end of the continuous brazing furnace is not large, and obvious temperature difference exists between the radiating belt and the cooling pipe which are not completely welded and between the cooling pipe and the water chamber, so that any object higher than absolute zero (-273.15 ℃) can emit infrared radiation according to the Planck blackbody radiation law. The higher the object temperature under the same conditions, the more energy is radiated, and therefore the infrared radiation is also called thermal radiation. Therefore, the temperature change can be accurately observed by utilizing infrared thermal imaging, so that the brazing quality can be accurately judged.
In order to achieve the above object, according to the method for detecting soldering quality of an automotive radiator provided in this embodiment, an infrared image and a visible light image of a radiator welded from a continuous soldering furnace are obtained by a thermal camera embedded with a visible light RGB camera. The use of the visible light RGB camera is mainly because the problems of low resolution, blurred details, poor visual quality and the like of the infrared image generally exist in the acquisition process. The visible light image is used for guiding super-resolution reconstruction of the infrared image, and the super-resolution reconstruction method is a mode for effectively improving the resolution of the infrared image.
Considering that the brazing position of the radiator is not at the edge of the radiator, in order to reduce the calculation amount, in the embodiment, the images about the radiator part in the infrared image and the visible light image are scratched out by utilizing a digital image scratching algorithm, and the scratched infrared radiator image and the scratched visible light radiator image are obtained. Because the radiator is generally of a regular square structure, the drawing is simpler, the conventional common drawing algorithm is adopted for drawing, and in the embodiment, the radiator is not improved.
Because the imaging principles of the visible light image and the infrared image are different, the detail information between the two images is different, meanwhile, because the temperature difference of the radiator is large, the place where the temperature difference is highlighted is mainly in a high-frequency information section, in the visible light image, because the whole radiator is basically silver gray, the color difference is small, the visible light image belongs to low-frequency information, the texture and the detail of the whole radiator are required to be stored, and meanwhile, the heat information of the radiator is also reserved, so that the finally obtained high-resolution infrared image is more accurate. Therefore, the low-frequency information in the visible light radiator image and the high-frequency information of the infrared radiator image are acquired, and the high-resolution infrared radiator image is obtained after reconstruction and fusion.
The low frequency information and the high frequency information can be obtained directly by a filter, but the obtained information is easy to lose part of the information, therefore, in the embodiment, firstly, the infrared radiator image is amplified by adopting a bicubic interpolation algorithm, so that the amplified size of the infrared image is consistent with the overlapped part size of the visible light image. The purpose of this step is mainly to facilitate later fusion.
The basic idea of the bicubic interpolation algorithm is to correlate sixteen known pixel values in the 4×4 neighborhood of the floating point coordinates of the pixel points of the source image, then respectively perform third-order interpolation from the horizontal direction and the vertical direction, and finally calculate the gray value of the pixel points to be interpolated.
Then, the amplified infrared radiator image and visible light radiator image are respectively passed through two convolution layers with convolution kernel size of 3×3, and downsampled by using a downsampling convolution, so that the infrared radiator image and visible light radiator image shallow features are respectively extracted, and the aim of the processing is to preserve more pixel information while downsampling, so as to prepare for subsequent extraction of low-frequency information and high-frequency information.
For the extraction of the low-frequency information, in the present embodiment, the low-frequency information in the wavelet block layer extraction image is adopted by inputting the shallow features of the visible light radiator image. Specifically, the wavelet layer comprises a convolution module, a two-dimensional discrete wavelet transform module and a convolution output module. The convolution module is 3×3 convolution, and aims to perform further convolution processing on an image, distinguish high-frequency information and low-frequency information of a visible light radiator image, remove the high-frequency information through the discrete wavelet transformation module, and finally locate in the low-frequency information through the convolution output module, wherein the convolution output module comprises 3×3 convolution, a 3×3 filter kernel, a batch normalization processing unit and a correction linear unit.
The mathematical expression of the discrete wavelet transformation module is as follows:、. Thus, the wavelet transformation is operated by the functions of expansion, translation and the likeThe image is decomposed into a series of sub-band coefficients with different directional characteristics and space-time resolution, so that the multi-scale refinement and decomposition of the information are realized. Thereby better retaining details of the low frequency information of the visible light.
In order to better retain thermal information in the infrared radiator image, in the embodiment, the infrared radiator image shallow feature is extracted by a flow Fourier residual module, specifically speaking, the flow Fourier residual module carries out deconvolution on the infrared radiator image shallow feature, then downsamples to be consistent with the size of the infrared radiator image by downsampling convolution, and then makes a difference with the infrared radiator image shallow feature, so that the high-frequency information can be obtained.
And splicing the obtained low-frequency information and high-frequency information in the channel dimension to form an input tensor. And then the input tensor is acquired and generated by adopting a triple attention module, so that the loss in space can be effectively reduced, and the detail is ensured to be clear. In existing image processing, the image is often represented in the form of C, H, W, i.eThe method comprises the steps that C represents a channel dimension, three channels of the existing common picture mainly comprise RGB, H represents a height dimension, W represents a width dimension, a triple attention module mainly enables the three dimensions to interact with each other to form three branches, and finally, the cross dimensions obtained by the three branches are summarized by using a simple average value to generate a fine tensor.
And finally, inputting the fine tensor into a 3X 3 convolution layer connected by three layers for feature enhancement, and obtaining the high-resolution infrared radiator image. Specifically, first, the fused fine tensor is input to a convolution layer of 3×3 of the first layer for feature enhancement, and then the output is added to the fine tensor before fusion through addition operation, so as to obtain the output of the first layer. Next, the first layer output is passed into the second layer, again feature enhanced using a 3 x 3 convolutional layer, and the output is added to the output of the first layer to obtain a second output. And finally, transmitting the output of the second layer into a third layer, performing feature enhancement by using a 3×3 convolution layer, and adding the output and the output of the second layer to obtain a final fused high-resolution infrared radiator image. The obtained high-resolution infrared radiator image can meet the subsequent requirements.
In this embodiment, in order to correct the model error during the training of the fusion model, so that the model can acquire more source image information, the similarity between the fusion image and the source image is improved, and a loss function is also provided.
Specifically, the loss function is:wherein, the method comprises the steps of, wherein,in order to achieve a loss of structural similarity,is the pixel loss, wherein,wherein, the method comprises the steps of, wherein,representing the number of channels,Representing the high-level of the image,representing the width of the imageRepresenting the input image and,representing the output image of the image processing device,then the structural similarity of the input image and the output image is represented;wherein, the method comprises the steps of, wherein,representing an input imageAnd output imageEuclidean distance between them.
Since the entire radiator does not need to have a percentage of the total brazing area, for example, the brazing between the radiating strip and the cooling tube needs to be 60%, and the brazing between the water chamber and the cooling tube needs to be 100%, and since the radiator is fixed in structure, the welded area is also fixed, the obtained high-resolution infrared radiator image is marked by scribing to obtain the brazing area of interest. Therefore, the region of interest is obtained without complex deep learning and other methods, and the region of interest can be obtained only by a frame scribing mode because the region is fixed, so that the aim of simplifying the later operation is achieved.
In order to further simplify the operation, gray processing is carried out on the interested brazing area, and a binarization algorithm is adopted to process the interested brazing area subjected to gray processing, so that a binarized image is obtained.
In the embodiment, automatic threshold segmentation is adopted, the probability of each gray value, the distribution probability of a target and a background, the average gray value and the variance are calculated through statistics, and the gray value with the largest difference between classes is taken as a threshold. Then, the black-and-white region is divided according to the threshold value, and then the welding rate size is calculated according to the content of the region.
And then, determining whether the interested brazing area meets the brazing requirement by setting a welding rate threshold value, and if not, marking the area by a laser marking machine fixed on a cross moving platform. Namely, the brazing rate between the water chamber and the cooling pipe does not reach 100%, and the brazing between the heat radiation belt and the cooling pipe reaches 60%, namely, the brazing is qualified, and thus, after marking, the subsequent repair welding is finished.
Through the arrangement, the automatic identification and determination of the brazing quality are effectively realized, the working efficiency is improved, the labor intensity is reduced, and the method is more accurate and quicker.
The present invention is not limited to the above-mentioned embodiments, and any equivalent embodiments which can be changed or modified by the technical content disclosed above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above-mentioned embodiments according to the technical substance of the present invention without departing from the technical content of the present invention still belong to the protection scope of the technical solution of the present invention.
Claims (5)
1. The method for detecting the brazing quality of the automobile radiator is characterized by comprising the following steps of:
a. firstly, acquiring an infrared image and a visible light image of a radiator welded in a continuous brazing furnace through a thermal camera embedded with a visible light RGB camera;
b. then, the images of the radiator parts in the infrared image and the visible light image are scratched out by utilizing a digital image scratching algorithm, and the scratched infrared radiator image and visible light radiator image are obtained;
c. then, obtaining low-frequency information in the visible light radiator image and high-frequency information of the infrared radiator image, and obtaining a high-resolution infrared radiator image after reconstruction and fusion;
d. obtaining an interested brazing area by using the obtained high-resolution infrared radiator image in a marking and labeling mode;
e. carrying out gray scale treatment on the interested brazing area, and adopting a binarization algorithm to treat the interested brazing area subjected to gray scale treatment to obtain a binarized image;
f. calculating the proportion of the whole binary image for the white part pixels in the binary image, wherein the proportion is the welding rate;
g. and then, determining whether the interested brazing area meets the brazing requirement or not by setting a welding rate threshold value, and marking the area by a laser marking machine if the interested brazing area does not meet the brazing requirement.
2. The method for detecting brazing quality of an automotive radiator according to claim 1, wherein the step c includes the steps of:
c1, amplifying an infrared radiator image by adopting a bicubic interpolation algorithm, respectively passing the amplified infrared radiator image and visible light radiator image through two convolution layers with convolution kernel sizes of 3 multiplied by 3, and performing downsampling by using downsampling convolution to respectively extract shallow features of the infrared radiator image and the visible light radiator image;
c2, inputting the shallow features of the visible light radiator image into a wavelet block layer to extract low-frequency information in the image; meanwhile, extracting high-frequency information from the shallow image features of the infrared radiator through a flow Fourier residual error module;
c3, splicing the obtained low-frequency information and high-frequency information in the channel dimension to form an input tensor;
c4, acquiring and generating a fine tensor by adopting the input tensor by adopting the fused triple attention module;
and c5, inputting the fine tensor into the 3X 3 convolution layers connected by three layers for feature enhancement, and obtaining the high-resolution infrared radiator image.
3. The method for detecting brazing quality of an automotive radiator according to claim 2, wherein in the step c2, the wavelet layer includes a convolution module, a two-dimensional discrete wavelet transform module, and a convolution output module.
4. The method for detecting brazing quality of an automobile radiator according to claim 2, wherein in the step c2, the flow fourier residual error module performs deconvolution on the image shallow features of the infrared radiator, performs downsampling to be consistent with the size of the image of the infrared radiator through downsampling convolution, and performs difference with the image shallow features of the infrared radiator to obtain the high-frequency information.
5. The method for detecting brazing quality of an automotive radiator according to claim 2, wherein in the step c4, the input tensor is output according to channel dimension-to-height dimension interaction, channel dimension-to-width dimension interaction and height dimension-to-width dimension interaction, and the three outputs are summarized by using a simple average value to generate the fine tensor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410217074.4A CN117773405B (en) | 2024-02-28 | 2024-02-28 | Method for detecting brazing quality of automobile radiator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410217074.4A CN117773405B (en) | 2024-02-28 | 2024-02-28 | Method for detecting brazing quality of automobile radiator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117773405A true CN117773405A (en) | 2024-03-29 |
| CN117773405B CN117773405B (en) | 2024-05-14 |
Family
ID=90383745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410217074.4A Active CN117773405B (en) | 2024-02-28 | 2024-02-28 | Method for detecting brazing quality of automobile radiator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117773405B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120291172A1 (en) * | 2011-05-16 | 2012-11-22 | Lincoln Global, Inc. | Dual-spectrum digital imaging welding helmet |
| CN207642452U (en) * | 2017-09-29 | 2018-07-24 | 合肥埃科光电科技有限公司 | A kind of real-time thermal imaging infrared reflow weldering system |
| CN114359687A (en) * | 2021-12-07 | 2022-04-15 | 华南理工大学 | Target detection method, device, equipment and medium based on multi-mode data dual fusion |
| CN114429457A (en) * | 2022-01-19 | 2022-05-03 | 中国电建集团中南勘测设计研究院有限公司 | Intelligent fan blade defect detection method based on bimodal fusion |
| WO2023155324A1 (en) * | 2022-02-21 | 2023-08-24 | 烟台艾睿光电科技有限公司 | Image fusion method and apparatus, device and storage medium |
| CN116758389A (en) * | 2023-06-27 | 2023-09-15 | 太原理工大学 | Infrared and visible light image fusion method based on attention and threshold segmentation |
| CN116777808A (en) * | 2023-06-30 | 2023-09-19 | 武汉高德智感科技有限公司 | Infrared and visible light image fusion method, device and registration method |
| CN116921911A (en) * | 2023-08-10 | 2023-10-24 | 苏州探想光电科技有限公司 | Laser welding quality detection method |
| CN117237424A (en) * | 2023-09-13 | 2023-12-15 | 武汉高德智感科技有限公司 | Registering method, feature matching method and equipment for infrared and visible light images |
| CN117283133A (en) * | 2023-11-22 | 2023-12-26 | 海菲曼（天津）科技有限公司 | Earphone production line electrode laser welding method and laser welding equipment |
| CN117422858A (en) * | 2023-09-19 | 2024-01-19 | 国网浙江省电力有限公司电力科学研究院 | Dual-light image target detection method, system, equipment and medium |
-
2024
- 2024-02-28 CN CN202410217074.4A patent/CN117773405B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120291172A1 (en) * | 2011-05-16 | 2012-11-22 | Lincoln Global, Inc. | Dual-spectrum digital imaging welding helmet |
| CN207642452U (en) * | 2017-09-29 | 2018-07-24 | 合肥埃科光电科技有限公司 | A kind of real-time thermal imaging infrared reflow weldering system |
| CN114359687A (en) * | 2021-12-07 | 2022-04-15 | 华南理工大学 | Target detection method, device, equipment and medium based on multi-mode data dual fusion |
| CN114429457A (en) * | 2022-01-19 | 2022-05-03 | 中国电建集团中南勘测设计研究院有限公司 | Intelligent fan blade defect detection method based on bimodal fusion |
| WO2023155324A1 (en) * | 2022-02-21 | 2023-08-24 | 烟台艾睿光电科技有限公司 | Image fusion method and apparatus, device and storage medium |
| CN116758389A (en) * | 2023-06-27 | 2023-09-15 | 太原理工大学 | Infrared and visible light image fusion method based on attention and threshold segmentation |
| CN116777808A (en) * | 2023-06-30 | 2023-09-19 | 武汉高德智感科技有限公司 | Infrared and visible light image fusion method, device and registration method |
| CN116921911A (en) * | 2023-08-10 | 2023-10-24 | 苏州探想光电科技有限公司 | Laser welding quality detection method |
| CN117237424A (en) * | 2023-09-13 | 2023-12-15 | 武汉高德智感科技有限公司 | Registering method, feature matching method and equipment for infrared and visible light images |
| CN117422858A (en) * | 2023-09-19 | 2024-01-19 | 国网浙江省电力有限公司电力科学研究院 | Dual-light image target detection method, system, equipment and medium |
| CN117283133A (en) * | 2023-11-22 | 2023-12-26 | 海菲曼（天津）科技有限公司 | Earphone production line electrode laser welding method and laser welding equipment |
Non-Patent Citations (1)
| Title |
|---|
| 赵立昌;张宝辉;吴杰;吴旭东;吉莉;: "基于灰度能量差异性的红外与可见光图像融合", 红外技术, no. 08, 20 August 2020 (2020-08-20), pages 75 - 82 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117773405B (en) | 2024-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110866924B (en) | Line structured light center line extraction method and storage medium | |
| CN110717872B (en) | Method and system for extracting characteristic points of V-shaped welding seam image under laser-assisted positioning | |
| CN112837233B (en) | Polarization image defogging method for acquiring transmissivity based on differential polarization | |
| CN110596116B (en) | Vehicle surface flaw detection method and system | |
| CN112304960B (en) | High-resolution image object surface defect detection method based on deep learning | |
| CN111369605A (en) | Infrared and visible light image registration method and system based on edge features | |
| CN107169947A (en) | A kind of image co-registration experimental method of feature based point location and rim detection | |
| CN113139900B (en) | Method for acquiring complete surface image of bar | |
| CN111476750B (en) | Method, device, system and storage medium for detecting stain of imaging module | |
| CN109544577B (en) | Improved straight line extraction method based on edge point grouping | |
| Kong et al. | Infrared and visible image fusion using structure-transferring fusion method | |
| CN102129670B (en) | Method for detecting and repairing movie scratch damage | |
| CN117773405B (en) | Method for detecting brazing quality of automobile radiator | |
| Lee et al. | Improving focus measurement via variable window shape on surface radiance distribution for 3D shape reconstruction | |
| CN111931537B (en) | Granular QR two-dimensional code positioning method | |
| CN112163996B (en) | Flat angle video fusion method based on image processing | |
| CN110852947B (en) | Infrared image super-resolution method based on edge sharpening | |
| CN116681879A (en) | Intelligent interpretation method for transition position of optical image boundary layer | |
| CN114858301B (en) | Object surface temperature measurement system, measurement method, intelligent terminal and storage medium | |
| CN114882122A (en) | Image local automatic calibration method and device and related equipment | |
| CN114897712A (en) | Infrared image blind pixel compensation method based on two wave bands | |
| CN112634228A (en) | Infrared image edge high-precision detection positioning method based on sub-pixel positioning algorithm | |
| Zhu et al. | Experimental study on image filtering algorithm | |
| CN111539967A (en) | Method and system for identifying and processing interference fringe region in terahertz imaging of focal plane | |
| CN111968049B (en) | Light field image hot pixel point removing method based on side window guide filtering |
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 |