CN104483325B - On-line measuring device and detection method based on multimodal registration - Google Patents
On-line measuring device and detection method based on multimodal registration Download PDFInfo
- Publication number
- CN104483325B CN104483325B CN201410790951.3A CN201410790951A CN104483325B CN 104483325 B CN104483325 B CN 104483325B CN 201410790951 A CN201410790951 A CN 201410790951A CN 104483325 B CN104483325 B CN 104483325B
- Authority
- CN
- China
- Prior art keywords
- registration
- pattern registration
- pattern
- subelement
- unit
- 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
Abstract
The invention discloses a kind of on-line measuring device and detection method based on multimodal registration, including it is installed on pattern registration unit and intelligent lighting assembly above mechanically supported platform successively from top to bottom；Manual intervention unit is installed on the right side of pattern registration unit, and pattern registration unit is made up of multiple pattern registration subelements, and each pattern registration subelement and manual intervention unit are connected with CPU respectively by interaction data communication bus.The present invention carries out registering analysis to the multiple characteristic informations for detecting target simultaneously, realizes the classification of multiple characteristic informations, collects and integrate, and realizes that efficiently, accurately on-line checking identifies；Combine the result of multiple characteristic information registrations, effectively realize the compression of multicharacteristic information, reduce redundancy feature information, mitigate the pressure of hardware system processing image information significantly, system detectio precision is improved, the result of comprehensive multimodal registration information provides effective support for the analysis and decision of the Automated condtrol of production line.
Description
Technical field
The invention belongs to a kind of on-line measuring device and detection method, and in particular to a kind of based on the online of multimodal registration
Detection means and detection method.
Background technology
During modern industrial and agricultural production, online measuring technique is in production, quality restriction and quality control etc.
Field is widely used, and is increasingly becoming one of basic technology of modern industry.In such applications, by various
Some feature of product component carries out registering identification, the defects of to realize product, cut, spot and shape, the inspection of installation site
Survey, and then eliminate or reduce defect ware.Comparatively, this is a kind of traditional image matching method, it is necessary to detecting target
The information such as gradation of image carry out more strict processing, and by a series of optimization processing of vision algorithms, can be only achieved satisfaction
Accuracy of identification requirement.
But inevitably, there is phenomena such as various tension fluctuations, noise jamming in most production line, actually obtain
The problems such as more or less stretcher strain that different situations all be present of the detection target image got, background complexity.With online
The complication of application scenario and the continuous improvement to testing requirements are detected, is difficult to description detection scene by single characteristic information
In full detail, and traditional on-line detecting system carries out detection identification using single characteristics of image pattern, specific
Under the conditions of light source etc., a characteristic information of detection target can only be obtained every time, and is often used as detection identification after final process
This feature information of matching does not simultaneously meet expected requirement.
Meanwhile the uncertainty at scene is detected to the operational efficiency of on-line detecting system, hardware cost, algorithm complex band
Many difficulties are carried out, among being in the occasion of change due to conditions such as detection target location, light sources, many times using single
Feature carry out recognition detection target, be extremely difficult to preferable effect.
At many scenes for needing on-line checking, many times need to detect multiple features of product, such as produce
The surface condition of engineering goods, internal structure damage, installation site etc. are detected on line, now matched somebody with somebody using single pattern
Its efficiency of quasi- method and the degree of accuracy are obviously difficult to meet to require, are the on-line checking for quickly and accurately realizing detection target,
It is necessary to design a kind of simple, reliable, low cost technical solution.
The content of the invention
The present invention proposes that the purpose is to provide one kind to be based on multimode in order to overcome shortcoming present in prior art
The on-line measuring device and detection method of formula registration.
The technical scheme is that：
A kind of on-line measuring device based on multimodal registration, the detection means includes mechanically supported platform, intelligence is shone
Bright assembly, pattern registration unit, manual intervention unit, interaction data communication bus and CPU；Pattern registration unit
It is installed on successively from top to bottom with intelligent lighting assembly above mechanically supported platform；It is single that manual intervention unit is installed on pattern registration
The right side of member, pattern registration unit are made up of multiple pattern registration subelements, each pattern registration subelement and manual intervention list
Member is connected with CPU respectively by interaction data communication bus.
The pattern registration subelement by camera lens, sensor, drive circuit, STM32 processor groups into；Drive circuit is by believing
Number acquisition module and high-speed pulse module composition, STM32 processors produce signal acquisition module and high-speed pulse module work institute
The time sequential pulse needed, high-speed pulse module produce the high speed time series pulse signals needed for working sensor, and signal acquisition module is adopted
Collect the signal of sensor, and signal is transmitted into the processing of STM32 processors.
The intelligent lighting assembly is made up of multiple lighting sources and light source communication module, and each lighting source is with detection mesh
The target direction of motion is installed successively, and each lighting source subelement registering with each pattern is corresponded and set, each illumination
Light source is formed by lamp body, multiple light emitting sources, optics auxiliary accessories and multiple drive control modules, and lamp body is oppositely arranged by two
Arc composition, light emitting source is distributed in lamp body inner periphery, and optics auxiliary accessories are installed on below lamp body, each light emitting source with it is each
Individual drive control module is connected one to one, and each drive control module is connected with light source communication module, light source communication module
It is connected with CPU.
The CPU is by FPGA, DSP, SDRAM and data memory cell group into FPGA connection interaction datas
Communication bus, connected between FPGA and DSP, DSP connects with data storage cell, and DSP connects with SDRAM.
A kind of online test method based on multimodal registration, comprises the following steps：
（ⅰ）Initialization
According to the testing requirements of detection target, sample is detected using artificial collect, suitable registration pattern is primarily determined that and matches somebody with somebody
Quasi-regular；
（ⅱ）Quasi-regular is matched somebody with somebody in optimization
Set comprehensive expert to match somebody with somebody quasi-regular, with reference to field working conditions, determine each pattern registration in pattern registration unit successively
Subelement is matched somebody with somebody quasi-regular and optimized, in the data storage cell with quasi-regular input CPU after optimization；
（ⅲ）Pattern registration unit configures
According to step（ⅱ）In rule and detection target specific testing requirements, to each pattern in pattern registration unit
Registering subelement is configured, and includes the selection of sensor and camera lens；
（ⅳ）Lighting source configures
With reference to step（ⅲ）The selection of middle respective sensor and camera lens, from suitable illumination sources, pass through central processing list
Member is controlled to each lighting source in intelligent lighting assembly, according to pattern registration subelement corresponding to each lighting source,
Configure suitable light level scheme；
（ⅴ）Start registration procedure
CPU being issued to each pattern registration subelement with quasi-regular by interaction data communication bus
The pattern registration subelement, when detection target is successively by i-th of pattern registration subelement, i-th i pattern registration subelement
During the subelements such as 2 ... n-th pattern registration subelement, associative mode registration subelement brings into operation；
（ⅵ）Registration result judges
Registration result is delivered to CPU, centre by each pattern registration subelement respectively in pattern registration unit
Reason unit carries out integrating calculation process to each registration result；
（ⅶ）Change Expert Rules
If inconsistent with the expected detection target of setting, comprehensive expert is directly changed by manual intervention and matches somebody with somebody quasi-regular,
Restart step（ⅰ）Perform detection, until testing result meets expection；
（ⅷ）Terminate
If being consistent with the expected detection target of setting, this detection terminates.
The beneficial effects of the invention are as follows：
The present invention realizes multiple pattern registration units in a detecting system, can realize simultaneously to detection target
Multiple characteristic informations carry out registering analysis；The final output result of detecting system carries out comprehensive analysis through all registration unit results
Drawn after processing, at the same in detection process manual intervention unit real-time participation, have modified because field working conditions complicated band come inspection
Dendrometry is imitated or mistake, is advantageous to the intelligent perfect of Expert Rules；On-line checking based on multimodal registration realizes multiple spies
The classification of reference breath, collect and integrate, help to realize the sample classification statistics of detection sample, realize efficient, accurately online
Detection identification, improve the automated production level of the entire production line；The result of multiple characteristic information registrations is combined, is effectively realized
The compression of multicharacteristic information, reduce redundancy feature information, mitigate the pressure of hardware system processing feature information, improve and be
System accuracy of detection；The result for combining multimodal registration information provides effectively for the analysis and decision that production line automation controls
Support.
Brief description of the drawings
Fig. 1 is the structural representation of the on-line measuring device of the invention based on multimodal registration；
Fig. 2 is the structural representation of pattern registration unit in the present invention；
Fig. 3 is the structural representation of pattern registration subelement in the present invention；
Fig. 4 is the structural representation of the lighting source of intelligent lighting assembly in the present invention；
Fig. 5 is the structural representation of CPU in the present invention；
Fig. 6 is the flow chart of the online test method of the invention based on multimodal registration.
Wherein：
1 mechanically supported platform 2 detects target
The pattern registration unit of 3 intelligent lighting assembly 4
The interaction data communication bus of 5 manual intervention unit 6
The pattern registration subelement of 7 CPU 8
The drive circuit of 9 STM32 processors 10
The camera lens of 11 sensor 12
The high-speed pulse module of 13 signal acquisition module 14
The lamp body of 15 optics auxiliary accessories 16
The drive control module of 17 light emitting source 18
The FPGA of 19 light source communication module 20
The data storage cells of 21 DSP 22
23 SDRAM。
Embodiment
With reference to Figure of description and embodiment to on-line measuring device of the present invention based on multimodal registration and detection
Method is described in detail：
As shown in figure 1, a kind of on-line measuring device based on multimodal registration, including mechanically supported platform 1, intelligent lighting
Assembly 3, pattern registration unit 4, manual intervention unit 5, interaction data communication bus 6 and CPU 7；Pattern registration is single
Member 4 and intelligent lighting assembly 3 are installed on the top of mechanically supported platform 1 successively from top to bottom；Manual intervention unit 5 is installed on pattern
The right side of registration unit 4, pattern registration unit 4 are made up of multiple pattern registration subelements 8, each He of pattern registration subelement 8
Manual intervention unit 5 is connected with CPU 7 respectively by interaction data communication bus 6.
As shown in Figure 2,3, the pattern registration subelement 8 is by camera lens 12, sensor 11, drive circuit 10, STM32 processing
Device 9 forms；Drive circuit 10 realizes the driving and control to sensor 11, and it is by signal acquisition module 13 and high-speed pulse mould
Block 14 forms, and produces signal acquisition module 13 by STM32 processors 9 and high-speed pulse module 14 works required time sequential pulse.
High-speed pulse module 14 is responsible for producing the high speed time series pulse signals that sensor 11 works required, and signal acquisition module 13 is mainly born
The collection of the signal of sensor 11 is blamed, the signal collected is ultimately delivered to STM32 processors 9 and handled.According to detection target
2 difference, the sensor 11 in each pattern registration subelement 8 can select Visible-light CCD, infrared CCD etc. different types of
Sensor, and the pattern registration subelement 8 that can configure repetition as needed repeats the purpose of detection with realization.Each pattern is matched somebody with somebody
Camera lens 12, drive circuit 10 in quasi- subelement 8 is different and specific all in accordance with the sensor 11 in the pattern registration subelement 8
Designed with alignment request.STM32 processors 9 be responsible for the execution of the registration action of pattern registration subelement 8 and with CPU 7
Real-time, interactive is carried out, STM32F103 Series MCU of the STM32 processors 9 from ST companies in the present embodiment.
The manual intervention unit 5 is made up of input equipments such as keyboard or touch-screens, to directly input intervention rule, its
Last right side of pattern registration subelement 8 of pattern registration unit 4 is installed on, when all pattern registration subelements 8 fail
Export effective detection result or testing result is undesirable, intervened by manual intervention unit 5 according to actual conditions, there is provided be artificial
Court verdict.Manual intervention unit 5 formed with quasi-regular can be with real-time update to Expert Rules among.
As shown in figure 4, the intelligent lighting assembly 3 is made up of multiple lighting sources and light source communication module 19, Ge Gezhao
Mingguang City source is with the direction of motion for detecting target 2（As shown by the arrows in Figure 1）Install successively, and each lighting source and each pattern
Registering subelement 8, which corresponds, to be set, and each lighting source is by lamp body 16, multiple light emitting sources 17, the and of optics auxiliary accessories 15
Multiple drive control modules 18 are formed, and lamp body 16 is made up of two arcs being oppositely arranged, and light emitting source 17 is distributed in lamp body 16
Inner periphery, optics auxiliary accessories 15 are installed on the lower section of lamp body 16, a pair of each light emitting source 17 and each drive control module 18 1
It should connect, each drive control module 18 is connected with light source communication module 19, light source communication module 19 and CPU 7
Connection.Each lighting source in intelligent lighting assembly 3 is all in accordance with each pattern registration subelement 8 in pattern registration unit 4
It is not all that the subelement provides different lighting sources, each lighting source can provide different illumination sides as needed
To, irradiating angle, light source colour and different light source light spectrums.As detection chip surface print character when, light source can be controlled
It is illuminated with low angle, so as to prominent character.In each lighting source drive control module 18 with light source communication module 19
Connect to receive light source configuration order.Lamp body 16 can need to be designed as vaulted or annular structure according to detection, light
Source 17 can select the light sources such as LED, infrared Halogen lamp LED, and drive control module 18 mainly realizes the driving and control of light source, rationally
The lighting quality that detecting system in the present invention can be effectively improved from optics auxiliary accessories 15, according to field working conditions and reality
The requirement optics auxiliary accessories 15 of detection can typically select optical element, the drive control modules 18 such as polarizer or diffusing panel negative
The drive control of light source is blamed, and CPU 6 can be realized to some independent lighting source by light source communication module 19
Real-time control.
As shown in figure 5, the CPU 7 is responsible for the coordination and control of whole detecting system, it passes through interaction number
Connected according to the subelement 8 registering with each pattern of pattern registration unit 4 of communication bus 6 and manual intervention unit 5, realize registration fortune
With the real-time, interactive of each registering sub-unit of data during row, it is realized single to each registering son by the connection with intelligent lighting assembly 3
The configuration of lighting source corresponding to member.CPU 6 is mainly by on-site programmable gate array FPGA（Field
Programmable Gate Array）20th, digital signal processor DSP (Digital Signal Processor) 21, same
Walk dynamic RAM SDRAM (Synchronous Dynamic Random Access Memory) 23 and data storage list
Member 22 forms.Wherein coprocessors of the FPGA 20 as DSP signal processors, mainly complete light source control, driving and pre- place
The logic function of reason, bus management and some low orders, and DSP is mainly responsible for completing regular sample learning, result of decision processing
Deng high-order Processing Algorithm.SDRAM 23 and data storage cell 22 are respectively used to storage during 6 high speed operation of CPU
With reading, intelligence learning and the real-time update of such as Expert Rules.This platform architecture, system operating efficiency is improved, lifted
The overall performance of system.
As shown in fig. 6, a kind of online test method based on multimodal registration, comprises the following steps：
（ⅰ）Initialization
According to the testing requirements of detection target 2, sample is detected using artificial collect, suitable registration pattern is primarily determined that and matches somebody with somebody
Quasi-regular（S1）；
（ⅱ）Quasi-regular is matched somebody with somebody in optimization
Set comprehensive expert to match somebody with somebody quasi-regular, with reference to field working conditions, determine each pattern registration in pattern registration unit 4 successively
Subelement 8 is matched somebody with somebody quasi-regular and optimized, the data storage cell 22 with quasi-regular input CPU 7 after optimization
In（S2）；
（ⅲ）Pattern registration unit configures
According to step（ⅱ）In rule and detection target 2 specific testing requirements, to each mould in pattern registration unit 4
Formula registration subelement 8 is configured, and includes the selection of sensor 11 and camera lens 12（S3）；
（ⅳ）Lighting source configures
With reference to step（ⅲ）The selection of middle respective sensor 11 and camera lens 12, from suitable illumination sources, passes through centre
Reason unit 7 is controlled to each lighting source in intelligent lighting assembly 3, according to pattern registration corresponding to each lighting source
Unit 8, configure suitable light level scheme（S4）；
（ⅴ）Start registration procedure
Being issued each pattern registration subelement 8 with quasi-regular by interaction data communication bus 6 of CPU 7
To the pattern registration subelement 8, when detection target 2 is single by i-th of pattern registration subelement, i-th i pattern registration successively
During the subelements such as n-th pattern registration subelement of member 2 ..., associative mode registration subelement 8 brings into operation S5；
（ⅵ）Registration result judges
Registration result is delivered to CPU 7 by each pattern registration subelement 8 respectively in pattern registration unit 4, in
Central Processing Unit 7 carries out integrating calculation process to each registration result（S6）；
（ⅶ）Change Expert Rules
If inconsistent with the expected detection target of setting, comprehensive expert's registration is directly changed by manual intervention unit 5
Rule（S7）, restart step（ⅰ）Perform detection, until testing result meets expection；
（ⅷ）Terminate
If being consistent with the expected detection target of setting, this detection terminates（S8）.
Main advantages of the present invention have：
(i) multiple pattern registration units are realized in a detecting system, can be realized simultaneously to the more of detection target
Individual characteristic information carries out registering analysis.
(ii) the final output result of detecting system is drawn after all registration unit results carry out comprehensive analysis processing, together
When detection process in manual intervention unit real-time participation, have modified because field working conditions complicated band come detection failure or mistake,
Be advantageous to the intelligent perfect of Expert Rules.
(iii) the on-line checking based on multimodal registration realizes the classification of multiple characteristic informations, collects and integrate, and contributes to
The sample classification statistics of detection sample is realized, realizes that efficiently, accurately on-line checking identifies, improves the automation of the entire production line
The level of production.
(iv) the result of multiple characteristic information registrations is combined, the compression of multicharacteristic information is effectively realized, reduces redundancy
Characteristic information, mitigate the pressure of hardware system processing feature information, improve system detectio precision.
(v) the result for combining multimodal registration information provides effectively for the analysis and decision that production line automation controls
Support.
Claims (1)
- A kind of 1. on-line measuring device based on multimodal registration, it is characterised in that：The detection means is put down including mechanically supported Platform（1）, intelligent lighting assembly（3）, pattern registration unit（4）, manual intervention unit（5）, interaction data communication bus（6）With in Central Processing Unit（7）；Pattern registration unit（4）With intelligent lighting assembly（3）It is installed on mechanically supported platform successively from top to bottom（1）Top；Pattern registration unit（4）By multiple pattern registration subelements（8）Composition, each pattern registration subelement（8）Done with artificial Pre- unit（5）Pass through interaction data communication bus（6）Respectively with CPU（7）Connection, when being run with realization registration Subelement registering with each pattern（8）Data real-time, interactive, CPU（7）By with intelligent lighting assembly（3）Even Realization is connect to each pattern registration subelement（8）The configuration of corresponding lighting source；The pattern registration subelement（8）By camera lens（12）, sensor（11）, drive circuit（10）, STM32 processors（9）Group Into；Drive circuit（10）By signal acquisition module（13）With high-speed pulse module（14）Composition, is realized to sensor（11）Drive Dynamic and control, STM32 processors（9）Produce signal acquisition module（13）With high-speed pulse module（14）When serial vein needed for work Punching, high-speed pulse module（14）Produce sensor（11）High speed time series pulse signals needed for work, signal acquisition module（13） Gather sensor（11）Signal, and signal is transmitted into STM32 processors（9）Processing；According to detection target（2）Difference, Each pattern registration subelement（8）In sensor（11）For Visible-light CCD or infrared CCD, and configuration repeats as needed Pattern registration subelement（8）To realize the purpose for repeating detection；Each pattern registration subelement（8）In camera lens（12）, drive Dynamic circuit（10）All in accordance with the pattern registration subelement（8）In sensor（11）Designed with alignment request is specifically matched somebody with somebody；The intelligent lighting assembly（3）By multiple lighting sources and light source communication module（19）Composition, each lighting source is with inspection Survey target（2）The direction of motion install successively, and each lighting source subelement registering with each pattern（8）One-to-one corresponding is set Put, each lighting source is by lamp body（16）, multiple light emitting sources（17）, optics auxiliary accessories（15）With multiple drive control modules （18）Composition, lamp body（16）It is made up of two arcs being oppositely arranged；Light emitting source（17）It is distributed in lamp body（16）Inner periphery, light Learn auxiliary accessories（15）It is installed on lamp body（16）Lower section, each light emitting source（17）With each drive control module（18）Correspond Connection, each drive control module（18）With light source communication module（19）Connection, light source communication module（19）With central processing Unit（7）Connection；Intelligent lighting assembly（3）In each lighting source all in accordance with needing to provide different illumination directions, irradiation Angle, light source colour and different light source light spectrums；Lamp body（16）For vaulted or annular structure；Light emitting source（17）For LED or red Outer Halogen lamp LED；Optics auxiliary accessories（15）For polarizer or diffusing panel；The manual intervention unit（5）It is installed on pattern registration unit（4）Last pattern registration subelement（8）Right side, Manual intervention unit（5）For keyboard or touch-screen；When all pattern registration subelements（8）Fail export effective detection result or Testing result is undesirable, by manual intervention unit（5）Intervened according to actual conditions, there is provided artificial court verdict, it is artificial dry Pre- unit（5）What is formed matches somebody with somebody quasi-regular real-time update into Expert Rules；The CPU（7）By FPGA（20）、DSP（21）、SDRAM（23）And data storage cell（22）Composition, FPGA（20）Connect interaction data communication bus（6）, FPGA（20）And DSP（21）Between connect, DSP（21）With data storage list Member（22）Connection, DSP（21）And SDRAM（23）Connection；FPGA（20）For DSP（21）Coprocessor, complete light source control, drive The dynamic logic function with pretreatment, bus management and some low orders, and DSP（21）It is responsible for completing regular sample learning, decision-making Result treatment class high-order Processing Algorithm；SDRAM（23）And data storage cell（22）It is respectively used to CPU（6）At a high speed Storage and reading during work, the intelligence learning of Expert Rules and real-time update；The detection method of the on-line measuring device based on multimodal registration, comprises the following steps：（ⅰ）InitializationAccording to detection target（2）Testing requirements, detect sample using artificial collect, primarily determine that suitable registration pattern and registration Rule（S1）；（ⅱ）Quasi-regular is matched somebody with somebody in optimizationSet comprehensive expert to match somebody with somebody quasi-regular, with reference to field working conditions, determine pattern registration unit successively（4）In each pattern registration Unit（8）With quasi-regular and being optimized, after optimization with quasi-regular input CPU（7）Data storage cell （22）In（S2）；（ⅲ）Pattern registration unit configuresAccording to step（ⅱ）In rule and detection target（2）Specific testing requirements, to pattern registration unit（4）In each mould Formula registration subelement（8）Configured, including sensor（11）And camera lens（12）Selection（S3）；（ⅳ）Lighting source configuresWith reference to step（ⅲ）Middle respective sensor（11）And camera lens（12）Selection, from suitable illumination sources, pass through centre Manage unit（7）To intelligent lighting assembly（3）In each lighting source be controlled, matched somebody with somebody according to pattern corresponding to each lighting source Quasi- subelement（8）, configure suitable light level scheme（S4）；（ⅴ）Start registration procedureCPU（7）Pass through interaction data communication bus（6）By each pattern registration subelement（8）Match somebody with somebody quasi-regular under It is sent to the pattern registration subelement（8）, when detection target（2）Match somebody with somebody successively by i-th of pattern registration subelement, i-th i pattern During the subelements such as quasi- n-th of pattern registration subelement of subelement 2 ..., associative mode registration subelement（8）Bring into operation（S5）；（ⅵ）Registration result judgesPattern registration unit（4）In each pattern registration subelement（8）Registration result is delivered into CPU respectively（7）, CPU（7）Each registration result is carried out to integrate calculation process（S6）；（ⅶ）Change Expert RulesIf the expected detection target with setting（2）It is inconsistent, pass through manual intervention unit（5）Directly change comprehensive expert's registration Rule（S7）, restart step（ⅰ）Perform detection, until testing result meets expection；Manual intervention unit（5）What is formed matches somebody with somebody Quasi-regular real-time update is into Expert Rules；（ⅷ）TerminateIf the expected detection target with setting（2）It is consistent, then this detection terminates（S8）.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410790951.3A CN104483325B (en) | 2014-12-19 | 2014-12-19 | On-line measuring device and detection method based on multimodal registration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410790951.3A CN104483325B (en) | 2014-12-19 | 2014-12-19 | On-line measuring device and detection method based on multimodal registration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104483325A CN104483325A (en) | 2015-04-01 |
| CN104483325B true CN104483325B (en) | 2017-12-15 |
Family
ID=52757894
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410790951.3A Active CN104483325B (en) | 2014-12-19 | 2014-12-19 | On-line measuring device and detection method based on multimodal registration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104483325B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108345476A (en) * | 2018-02-02 | 2018-07-31 | 曙光信息产业(北京)有限公司 | A kind of method and apparatus of batch processing hardware configuration information |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1436301A (en) * | 2000-06-14 | 2003-08-13 | 泰拉丁公司 | Optical inspection system |
| CN101210886A (en) * | 2006-12-29 | 2008-07-02 | 财团法人工业技术研究院 | Optical detector |
| CN204330614U (en) * | 2014-12-19 | 2015-05-13 | 核工业理化工程研究院华核新技术开发公司 | A kind of on-line measuring device based on multimodal registration |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0949805A (en) * | 1995-08-07 | 1997-02-18 | Asahi Optical Co Ltd | Method and device for inspecting optical member |
| JP3980149B2 (en) * | 1998-02-13 | 2007-09-26 | 大日本印刷株式会社 | Method and apparatus for correcting misalignment of multiple cameras |
-
2014
- 2014-12-19 CN CN201410790951.3A patent/CN104483325B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1436301A (en) * | 2000-06-14 | 2003-08-13 | 泰拉丁公司 | Optical inspection system |
| CN101210886A (en) * | 2006-12-29 | 2008-07-02 | 财团法人工业技术研究院 | Optical detector |
| CN204330614U (en) * | 2014-12-19 | 2015-05-13 | 核工业理化工程研究院华核新技术开发公司 | A kind of on-line measuring device based on multimodal registration |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104483325A (en) | 2015-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102218406B (en) | Intelligent detection device of defects of mobile phone outer shell based on machine vision | |
| Gamanayake et al. | Cluster pruning: An efficient filter pruning method for edge ai vision applications | |
| CN100562761C (en) | Intelligent light-emitting diode (LED) car lamp lighting quality on-line detecting system | |
| CN109840900A (en) | A kind of line detection system for failure and detection method applied to intelligence manufacture workshop | |
| CN112100865A (en) | Method for predicting remaining life of aircraft engine based on parallel CNN model | |
| CN109063728A (en) | A kind of fire image deep learning mode identification method | |
| CN108932711A (en) | Image analysis nerve network system | |
| CN104992449A (en) | Information identification and surface defect on-line detection method based on machine visual sense | |
| CN102854194A (en) | Object surface defect detection method and apparatus based on linear array CCD | |
| CN108761237A (en) | Unmanned plane electric inspection process image vital electrical component diagnoses automatically and labeling system | |
| CN113219917B (en) | Bioremediation strengthening process for non-uniform discrimination guided twin model | |
| CN103279763A (en) | Method for automatically identifying wheel hub types based on structural features | |
| CN109693140A (en) | A kind of intelligent flexible production line and its working method | |
| JP2021515885A (en) | Methods, devices, systems and programs for setting lighting conditions and storage media | |
| CN204330614U (en) | A kind of on-line measuring device based on multimodal registration | |
| CN104483325B (en) | On-line measuring device and detection method based on multimodal registration | |
| CN113145492A (en) | Visual grading method and grading production line for pear appearance quality | |
| CN104483324B (en) | On-line measuring device based on multimodal registration | |
| CN116434777A (en) | Transformer fault diagnosis method and system based on multistage attention and feature fusion | |
| CN110412055A (en) | A kind of lens white haze defect inspection method based on multiple light courcess dark-ground illumination | |
| CN112150416A (en) | Coating defect detection system based on deep learning | |
| CN108226175B (en) | Plastic container detection system and detection method | |
| TW202326114A (en) | Fault or defect detection system for long-code zipper capable of meeting the requirement of continuous detection of the production assembly line having transmission speed at 300 m/min or lower, and has high detection precision | |
| CN216525503U (en) | Carbon fiber prepreg surface defect on-line measuring device based on machine vision | |
| CN103148783B (en) | A kind of automatic testing method of valve rocker installation site |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 300384 Tianjin hi tech District of Xiqing City Huayuan Industrial Zone Development six road No. 6 Haitai green industry base building J room 104 Applicant after: China core (Tianjin) New Technology Development Co., Ltd. Address before: 300384 Tianjin hi tech District of Xiqing City Huayuan Industrial Zone Development six road No. 6 Haitai green industry base building J room 104 Applicant before: Huahe New Technology Development Company, Research Institute of Physical and Chemical Engineering of Nuclear Industry |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200310 Address after: 300180, No. 168 Jintang Road, Hedong District, Tianjin Patentee after: Inst. of Physics and Chemistry Engineering Address before: 300384 Tianjin hi tech District of Xiqing City Huayuan Industrial Zone Development six road No. 6 Haitai green industry base building J room 104 Patentee before: China core (Tianjin) New Technology Development Co., Ltd. |
|
| TR01 | Transfer of patent right |