CN114138458A - Intelligent vision processing system - Google Patents
Intelligent vision processing system Download PDFInfo
- Publication number
- CN114138458A CN114138458A CN202111270775.7A CN202111270775A CN114138458A CN 114138458 A CN114138458 A CN 114138458A CN 202111270775 A CN202111270775 A CN 202111270775A CN 114138458 A CN114138458 A CN 114138458A
- Authority
- CN
- China
- Prior art keywords
- vision processing
- processing server
- output end
- processing system
- intelligent
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
- G06F3/041—Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/485—Task life-cycle, e.g. stopping, restarting, resuming execution
- G06F9/4856—Task life-cycle, e.g. stopping, restarting, resuming execution resumption being on a different machine, e.g. task migration, virtual machine migration
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an intelligent vision processing system, and relates to the technical field of intelligent processing. The intelligent vision processing system comprises a vision processing server, an imaging acquisition device, a comprehensive application client and an input trigger controller, wherein the vision processing server is respectively in electric signal connection with the imaging acquisition device and the comprehensive application client, and the vision processing server is in I/0 signal connection with the input trigger controller. According to the invention, by improving the calculation performance of the algorithm, including the modes of adjusting algorithm logic, optimizing the algorithm and the like and performing instruction set optimization according to the characteristics of the CPU, AVX mixed operation instructions are provided for the CPU of the recent generations, so that the AVX mixed operation instructions are greatly improved in deep learning, the overall calculation performance of the system is improved by adopting a distributed calculation mode, the problems of low efficiency, high error rate and the like caused by manual detection are reduced, the labor intensity of production personnel is reduced, and the production efficiency is improved.
Description
Technical Field
The invention relates to the technical field of intelligent processing, in particular to an intelligent vision processing system.
Background
The intelligent visual processing system is based on an image processing technology, carries a high-efficiency and stable visual algorithm library, has the processing function of realizing high-speed image reading and stabilization, is mostly managed by adopting equipment and algorithm plug-in modules, and has good expansibility and compatibility.
Traditional intelligent vision processing system is when using, and it is limited not to solve its inside algorithm, and the analysis of the production process each item detected data in the very difficult quick processing industry manufacturing has increased the production of the inefficiency that artifical detection caused, the high scheduling problem of error rate, has improved producers' intensity of labour, has reduced production efficiency, provides a novel device for this and in order to solve the problem that above-mentioned exists.
Disclosure of Invention
It is an object of the present invention to provide an intelligent vision processing system to solve the problems set forth in the background art.
In order to achieve the purpose, the invention provides the following technical scheme: the intelligent vision processing system comprises a vision processing server, an imaging acquisition device, a comprehensive application client and an input trigger controller, wherein the vision processing server is respectively in electric signal connection with the imaging acquisition device and the comprehensive application client, and is in I/0 signal connection with the input trigger controller;
the vision processing server comprises a central server, a management server and a processing engine which are in electric signal connection.
The imaging acquisition equipment comprises a camera, the top of the output end of the camera is rotatably connected with an auxiliary light source, and the auxiliary light source is electrically connected with the camera;
the comprehensive application client comprises an image display screen, the image display screen is an electronic touch screen, and a receiving end of the image display screen is electrically connected with an output end of the processing engine;
the input trigger controller comprises an I/O controller, and a receiving end of the I/O controller is electrically connected with an output end of the processing engine.
Furthermore, an upper computer is arranged at the output end of the visual processing server, an 6/7-generation Intel processor with a core display card is assembled in the upper computer, and the receiving end of the upper computer is electrically connected with the output end of the processing engine.
Furthermore, the output end of the vision processing server is provided with a subsequent actuating mechanism, the subsequent actuating mechanism comprises an acoustic-optical-electrical lamp feedback device and a mechanical linkage device, and the input end of the subsequent actuating mechanism is in I/0 signal connection with the output end of the vision processing server.
Furthermore, the processor inside the camera supports algorithm plug-in management, and the hardware architecture of the camera conforms to the standards of GigeVision and usb3.0 vision.
Furthermore, the motherboard inside the I/O controller is equipped with two chipsets respectively, wherein one chipset is a south bridge, the other chipset is a north bridge, the south bridge controls the release of the low-speed I/0 signal, and the north bridge controls the reading of the high-speed feedback signal.
Compared with the prior art, the invention has the beneficial effects that:
(1) the intelligent vision processing system improves the calculation performance of the algorithm, comprises modes of adjusting algorithm logic, optimizing the algorithm and the like and optimizes an instruction set according to the characteristics of the CPU, provides AVX mixed operation instructions for the CPU of the recent generations, greatly improves the deep learning, realizes the mode of adopting distributed calculation, improves the whole calculation performance of the system, reduces the problems of low efficiency, high error rate and the like caused by manual detection, reduces the labor intensity of production personnel and improves the production efficiency.
(2) This intelligence vision processing system calculates through a part calculation task, the part that migrates outside CPU to the task of will decoding is whole to be migrated and is gone on in the core display card, has reduced because core and CPU share memory can bring the overhead, thereby has promoted the whole calculation performance of this system, and the cost is reduced and whole energy consumption has reduced the production of the inefficiency that manual detection caused, error rate height etc. problem, the indirect reduction production personnel's intensity of labour and improvement production efficiency.
(3) This intelligence vision processing system, through the application to the sub-pixel technique, make the camera in this system under the limited condition of resolution ratio, the analysis that goes on the edge and the setting of drawing still can be according to calculation and the statistics of edge color transition through this system, the mode that adopts the sub-pixel is calculated and is used for promoting the setting of computational accuracy, the best recognition effect of acquireing that has realized can be better, and then the whole calculation performance of this system has been promoted, the inefficiency that manual detection caused has been reduced, the production of problems such as error rate height, the indirect reduction production personnel's intensity of labour and improvement production efficiency.
(4) This intelligence vision processing system, through the adaptation to different trades, even the different production lines of same trade, the different product models of same production line have very frequent switching phenomenon, adopt MPEG-7 coding mode's setting, have promoted the overall computation performance of this system, have reduced the production of the inefficiency that manual detection caused, error rate height problem, indirect reduction producers' intensity of labour and improved production efficiency.
Drawings
FIG. 1 is a block diagram of the system components of the present invention;
FIG. 2 is an internal assembly view of the vision processing server of the present invention;
FIG. 3 is a system flow diagram of the present invention;
FIG. 4 is a first analytical view of a system feature of the present invention;
FIG. 5 is a second analysis of a system feature of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that in the description of the present invention, the terms "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, which are only for convenience of description and simplification of description, and do not indicate or imply that the referred device or element must have a specific orientation, be configured in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
Further, it will be appreciated that the dimensions of the various elements shown in the figures are not drawn to scale, for ease of description, and that the thickness or width of some layers may be exaggerated relative to other layers, for example.
It should be noted that like reference numerals and letters refer to like items in the following figures, and thus, once an item is defined or illustrated in one figure, it will not need to be further discussed or illustrated in detail in the description of the following figure.
As shown in fig. 1 to 5, the present invention provides a technical solution: the intelligent vision processing system comprises a vision processing server, an imaging acquisition device, a comprehensive application client and an input trigger controller.
It should be noted that, in the embodiments provided in the present application, the vision processing server is respectively in electrical signal connection with the imaging acquisition device and the integrated application client, and the vision processing server is in I/0 signal connection with the input trigger controller.
It is further noted that in the embodiments provided in the present application, the functions of the devices are different.
For example: the visual processing server is the core of the whole system, can complete a plurality of visual tasks such as image processing, analysis and detection, and the like, and also supports the integration and deployment with a codesys control linkage system, thereby playing a role in realizing the integration of calculation, analysis and control linkage.
The imaging acquisition equipment is a sensing part of the whole system, and can quickly and accurately finish the acquisition of images according to different application scenes.
The comprehensive application client is an image application interface of the whole visual system, programming is not needed, and the simple and streamlined design of the operation interface of the device can help a user to operate by hands more quickly.
In addition, as shown in fig. 2, in the embodiment provided in the present application, the vision processing server includes a central server, a management server and a processing engine, and the three are electrically connected to each other.
In the embodiments provided in the present application, the apparatus in the vision processing server has the following functions.
The central server is responsible for data acquisition, mainly for acquiring video data.
The processing engine is responsible for video decoding, feature extraction and contrast debugging.
The management server is responsible for presenting and feeding back processed data results, including processes of man-machine interaction, data arrangement and the like, and selects and controls the linkage state of a mechanical device at the rear end of the system according to data analysis results.
It should be further noted that, in the embodiments provided in the present application, the imaging acquisition device includes a camera, wherein the top of the camera output end is rotatably connected with an auxiliary light source, and the auxiliary light source is electrically connected with the camera.
It is further noted that the embodiments provided herein may be implemented in a variety of mechanical arrangements within the system.
For example: in some embodiments, the camera may be a normal digital camera, and in the embodiments provided herein, the processor inside the camera supports algorithmic plug-in management, and the hardware architecture of the camera conforms to GigeVision, usb3.0vision standards.
In addition, it should be noted that, in the embodiments provided in the present application, the integrated application client includes an image display screen, where the image display screen is an electronic touch screen, and a receiving end of the image display screen is electrically connected to an output end of the processing engine.
It should be further noted that, in the embodiments provided in the present application, the input trigger controller includes an I/O controller, and a receiving end of the I/O controller is electrically connected to an output end of the processing engine.
It should be further noted that, in the embodiments provided in the present application, the motherboard inside the I/O controller is respectively equipped with two chipsets, wherein one chipset is a south bridge, the other chipset is a north bridge, the south bridge controls the release of the low-speed I/0 signal, and the north bridge controls the reading of the high-speed feedback signal.
As shown in fig. 1, in the embodiment provided in the present application, the output end of the vision processing server is provided with an upper computer, wherein it should be noted that, the upper computer in the system is internally equipped with an 6/7 generation intel processor with a core graphics card, and the receiving end of the upper computer is electrically connected with the output end of the processing engine.
In addition, as shown in fig. 1, in the embodiment provided by the present application, the output end of the vision processing server is provided with a subsequent actuator, the subsequent actuator includes an acousto-optic and electric light feedback device and a mechanical linkage device, and the input end of the subsequent actuator is in I/0 signal connection with the output end of the vision processing server.
As shown in fig. 4 and 5, it should be noted that the system has the following features:
firstly, the method comprises the following steps: and (4) high-performance calculation.
In order to solve the problem of low computing performance of the traditional device and enable the traditional device to have high-performance capability, the system provides the following two methods:
1. by improving the calculation performance of the algorithm, including adjusting algorithm logic, optimizing the algorithm and the like, and optimizing the instruction set according to the characteristics of the CPU, AVX mixed operation instructions are provided for the CPU of the recent generations, so that the deep learning of the CPU is greatly improved, the distributed calculation mode is adopted, the overall calculation performance is improved, the problems of low efficiency, high error rate and the like caused by manual detection are reduced, the labor intensity of production personnel is reduced, and the production efficiency is improved.
2. A part of calculation tasks are migrated to a component outside the CPU for calculation, and the decoding task is wholly migrated to the core display card for calculation, so that the system overhead caused by memory sharing of the core and the CPU is reduced, the overall performance is improved, the cost and the overall energy consumption are reduced, the problems of low efficiency, high error rate and the like caused by manual detection are reduced, the labor intensity of production personnel is indirectly reduced, and the production efficiency is improved.
Secondly, the method comprises the following steps: and (5) high-precision algorithm.
Besides the calculation performance, the accuracy and the false alarm rate of the algorithm are important indexes, and for machine vision, special algorithm detail optimization is required besides the conventional algorithm process, wherein the optimization mode comprises the following points:
1. the system also adopts the sub-pixel mode to calculate according to the calculation and statistics of the edge color transition to improve the calculation precision.
In addition, the camera has different recognition accuracy in different scenes, and the system also adopts a dual-template recognition mode which adopts different templates according to different feature dimensions and weights, so that the optimal recognition effect is obtained, and the calculation performance of the system is improved.
2. In the aspect of planning and using characteristics of a traditional device, because the traditional device has adaptation of different industries, even different production lines of the same industry have very frequent switching phenomena of different product models of the same production line, under the phenomenon, the requirement on the expansibility of an algorithm frame is very high, so that the system forms a characteristic tree by adopting an MPEG-7 coding mode, and the overall calculation performance of the system is improved in the aspect of establishing a model.
In addition, the system has an algorithm library, a feedback mechanism and a framework, namely the algorithm 'self-learning' capability frequently taught by people, and a background model is adjusted through the feedback of results by people to form negative feedback, so that the system can feed back and process information in time, and the computing capability of the device is enhanced.
According to the two modes, the embodiment provided by the application also provides a novel mode.
For example: in some embodiments, computational control within the system may be integrated.
The background of the embodiment is that the trend of IT is quite obvious in the industrial field at present, most of systems utilize a CPU to provide calculation analysis service, and simultaneously, an RTOS real-time operating system is deployed to run soft PLC control, so that image acquisition, data analysis and linkage control are all completed on one computer, and calculation control integration is formed.
In the embodiment, the system is based on the Intel series CPU platform, most of motion control performance requirements can be met, the scheme brings more paths, more complexity and more flexibility of motion control, and compared with the traditional PLC control mode, the mode is more convenient and simpler in development.
In addition, as shown in fig. 3, in the embodiment provided by the present application, the system works differently in each link.
1. And data acquisition mainly refers to the acquisition of video data.
2. After the video data is acquired, the uncompressed video baseband data is obtained through video decoding.
3. After decoding is completed, feature extraction needs to be performed on the video data of the baseband.
4. After the target extraction is completed, the standard modules need to be compared to determine whether the target is in compliance, and a decision of good products and defective products is made.
5. And after the detection of the algorithm is completed, the result is presented and fed back, and the processes comprise human-computer interaction, data arrangement and the like. Further, according to the results of good products and defective products, the mechanism is linked at the rear end.
It should be noted that, in the embodiments provided in the present application, the functions of the system have the following points.
1. Engineering management, wherein engineering management has the following advantages:
a. can systematize centralized management.
b. The method can support engineering replication, and is applicable to similar model products only by finely adjusting algorithm parameters.
c. And the complete backup and migration of the engineering data can be supported.
d. And multiple machine positions can be supported to simultaneously run the project aiming at the same equipment, so that the performance is maximized.
e. The alarm system can support the setting of a qualification rate alarm line to give an alarm and can automatically stop the operation of the project.
2. Operation monitoring, wherein the operation monitoring has the following advantages:
a. the engineering detection result can be checked in real time.
b. The data structure can be checked under the condition that multiple machine positions run simultaneously.
c. The historical data can be viewed and printed and retained in the form of a picture.
3. Statistical analysis, wherein the statistical analysis has the following advantages:
a. the engineering detection data can be counted, summarized and viewed.
b. And the data statistics of the multi-dimensional chart of the qualification rate, the NG rate and the real-time detection result are supported.
c. And supporting the tracking of engineering historical results.
d. And supporting the statistics of the overall detection result of the project and the individual result of each algorithm.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (5)
1. Intelligent vision processing system, including vision processing server, formation of image collection equipment, synthesize application client and input trigger control ware, its characterized in that: the vision processing server is respectively in electric signal connection with the imaging acquisition equipment and the comprehensive application client, and the vision processing server is in I/0 signal connection with the input trigger controller;
the vision processing server comprises a central server, a management server and a processing engine which are in electric signal connection with each other;
the imaging acquisition equipment comprises a camera, the top of the output end of the camera is rotatably connected with an auxiliary light source, and the auxiliary light source is electrically connected with the camera;
the comprehensive application client comprises an image display screen, the image display screen is an electronic touch screen, and a receiving end of the image display screen is electrically connected with an output end of the processing engine;
the input trigger controller comprises an I/O controller, and a receiving end of the I/O controller is electrically connected with an output end of the processing engine.
2. The intelligent vision processing system of claim 1, wherein: the output end of the visual processing server is provided with an upper computer, an 6/7-generation Intel processor with a core display card is assembled in the upper computer, and the receiving end of the upper computer is electrically connected with the output end of the processing engine.
3. The intelligent vision processing system of claim 1, wherein: the output end of the vision processing server is provided with a follow-up actuating mechanism, the follow-up actuating mechanism comprises an acoustic-optical-electric lamp feedback device and a mechanical linkage device, and the input end of the follow-up actuating mechanism is in I/0 signal connection with the output end of the vision processing server.
4. The intelligent vision processing system of claim 1, wherein: the processor inside the camera supports algorithm plug-in management, and the hardware architecture of the camera conforms to the standards of GigeVision and USB3.0Vision.
5. The intelligent vision processing system of claim 1, wherein: the mainboard in the I/O controller is respectively provided with two chip sets, wherein one chip set is a south bridge, the other chip set is a north bridge, the south bridge controls the release of low-speed I/0 signals, and the north bridge controls the reading of high-speed feedback signals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111270775.7A CN114138458A (en) | 2021-10-29 | 2021-10-29 | Intelligent vision processing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111270775.7A CN114138458A (en) | 2021-10-29 | 2021-10-29 | Intelligent vision processing system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114138458A true CN114138458A (en) | 2022-03-04 |
Family
ID=80394889
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111270775.7A Pending CN114138458A (en) | 2021-10-29 | 2021-10-29 | Intelligent vision processing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114138458A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024007206A1 (en) * | 2022-07-06 | 2024-01-11 | 宁德时代新能源科技股份有限公司 | Debugging method and apparatus for production line devices, and production line system |
-
2021
- 2021-10-29 CN CN202111270775.7A patent/CN114138458A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024007206A1 (en) * | 2022-07-06 | 2024-01-11 | 宁德时代新能源科技股份有限公司 | Debugging method and apparatus for production line devices, and production line system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3836077B1 (en) | Product defect detection method and apparatus, electronic device, storage medium and program | |
| US11954904B2 (en) | Real-time gesture recognition method and apparatus | |
| CN108197547B (en) | Face pose estimation method, device, terminal and storage medium | |
| Li et al. | A deep learning approach for real-time rebar counting on the construction site based on YOLOv3 detector | |
| CN111192293B (en) | Moving target pose tracking method and device | |
| US10521264B2 (en) | Data processing architecture for improved data flow | |
| US20210073973A1 (en) | Method and apparatus for component fault detection based on image | |
| Lin et al. | Capacitor detection in PCB using YOLO algorithm | |
| CN113435431B (en) | Posture detection method, training device and training equipment of neural network model | |
| DE102018124211A1 (en) | Learning-based camera pose estimation of images of an environment | |
| CN103440667A (en) | Automatic device for stably tracing moving targets under shielding states | |
| CN112040198A (en) | Intelligent water meter reading identification system and method based on image processing | |
| CN114138458A (en) | Intelligent vision processing system | |
| CN113838015B (en) | Electrical product appearance defect detection method based on network cooperation | |
| Zhang et al. | Gesture recognition system based on improved YOLO v3 | |
| CN116027743A (en) | Intelligent monitoring method, device and system for production line | |
| Wang et al. | SE-COTR: A novel fruit segmentation model for green apples application in complex orchard | |
| Peng et al. | RGB-D human matting: A real-world benchmark dataset and a baseline method | |
| CN111797862A (en) | Task processing method and device, storage medium and electronic equipment | |
| CN112270760A (en) | Positioning method, positioning device, electronic equipment and storage medium | |
| CN106908455A (en) | IMAQ terminal, glass surface defects on-line detecting system and method | |
| CN115619698A (en) | Method and device for detecting defects of circuit board and model training method | |
| Qian et al. | Multi-Scale tiny region gesture recognition towards 3D object manipulation in industrial design | |
| CN116740721B (en) | Finger sentence searching method, device, electronic equipment and computer storage medium | |
| CN111857478B (en) | Image layout method and device, electronic equipment and storage medium |
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 |