CN208367733U - Embedded A I machine vision hardware configuration - Google Patents
Embedded A I machine vision hardware configuration Download PDFInfo
- Publication number
- CN208367733U CN208367733U CN201821034832.5U CN201821034832U CN208367733U CN 208367733 U CN208367733 U CN 208367733U CN 201821034832 U CN201821034832 U CN 201821034832U CN 208367733 U CN208367733 U CN 208367733U
- Authority
- CN
- China
- Prior art keywords
- interface
- cpu
- circuit
- embedded
- hardware configuration
- 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 application provides a kind of Embedded A I machine vision hardware configuration, it include: top plate and lower plywood, in lower plywood, multiple power supplies respectively with CPU, storage device, debugging interface, camera interface, network PHY chip, clock and support circuit connection, clock and support circuit are connect with CPU, clock and the selection of CPU start-up mode are provided for CPU, CPU also with storage device, external interface, debugging interface, the connection of network PHY chip, CPU is for realizing high speed image identifying processing and operation, in top plate, power circuit is the power supply mode of POE mode, POE rectification decompression isolation exports stable voltage.Power circuit is camera model power supply by camera power supply circuit, and HDMI conversion chip is connect with CPU, and POE network module is connect with network PHY chip.Embedded A I is able to carry out complicated operation on end, while meeting requirement of real-time.
Description
Technical field
The invention relates to field of artificial intelligence more particularly to a kind of embedded artificial intelligence
(Artificial Intelligence, AI) machine vision hardware configuration.
Background technique
Artificial intelligence (Artificial Intelligence, AI), AI are a branches of computer science, it attempts
Understand the essence of intelligence, and produces a kind of new intelligence machine that can be made a response in such a way that human intelligence is similar, the neck
The research in domain includes robot, language identification, image recognition, natural language processing and expert system etc..
Scene not high in requirement of real time at this stage, but when needing to handle mass data or complicated algorithm, usually handle
The information of acquisition is transferred on cloud, and cloud carries out model training or deduction, on the terminal device that result is transmitted, then by end
The result that end equipment display needs.
However, the occasion high in real-time, terminal device handles AI far from satisfaction in computing capability and power consumption
The ability of algorithm uploads to cloud and calculating feedback is gone then to take a long time, is not able to satisfy the requirement of real-time.
Utility model content
The embodiment of the present application provides a kind of Embedded A I machine vision hardware configuration, sets for solving terminal in above scheme
The standby ability for handling AI algorithm far from satisfaction in computing capability and power consumption uploads to cloud and calculating feedback is gone then to need
Longer time, the problem of not being able to satisfy the requirement of real-time.
The application provides a kind of Embedded A I machine vision hardware configuration, comprising: top plate and lower plywood;
The lower plywood includes: that processor CPU, multiple power supplies, storage device, external interface, debugging interface, camera connect
Mouth, network PHY chip, clock and support circuit;
The multiple power supplies respectively with the CPU, storage device, the debugging interface, the camera interface, described
Network PHY chip, the clock and support circuit connection, the device power supply that the multiple power supplies are used to connect to it;
The clock and support circuit are connect with the CPU, provide clock and the selection of CPU start-up mode for the CPU;
The CPU is also connect with storage device, the external interface, the debugging interface, the network PHY chip respectively;The CPU
For realizing high speed image identifying processing and Multiple Models Algorithm operation;
The top plate includes: power circuit, camera power supply circuit, camera model, high-definition multimedia interface HDMI
Conversion chip, active Ethernet POE network module;
The power circuit is connect with the camera power supply circuit, and the camera power supply circuit is connect with camera model, institute
It is camera model power supply that power circuit, which is stated, by the camera power supply circuit;
The power circuit is connect with the HDMI conversion chip, is powered for the HDMI conversion chip;
The power circuit is also connect with the multiple power supplies, provides power supply for the lower plywood;
The camera model is connect by the camera interface with the CPU;The HDMI conversion chip and the CPU connect
It connects;The POE network module is connect with the network PHY chip.
Embedded A I machine vision hardware configuration one kind in the specific implementation, the storage device includes EMMC chip
And TF card.
Embedded A I machine vision hardware configuration one kind in the specific implementation, the debugging interface be jtag interface.
Embedded A I machine vision hardware configuration one kind in the specific implementation, the external interface include it is following at least
One:
USB2.0 interface, USB3.0 interface and USB type C interface.
Embedded A I machine vision hardware configuration one kind in the specific implementation, the multiple power supplies be integrated power supply pipe
Manage circuit PMIC.
Embedded A I machine vision hardware configuration one kind in the specific implementation, the CPU be graphics processor
MA2450。
Embedded A I machine vision hardware configuration one kind in the specific implementation, the top plate further includes secure digital
The Wireless Fidelity WIFI circuit of input-output card SDIO interface;The WIFI circuit is connect with the CPU;The power circuit is
The WIFI circuit power supply.
Embedded A I machine vision hardware configuration one kind in the specific implementation, the top plate further includes expansible
I2C bus and/or universal input/output GPIO.
Embedded A I machine vision hardware configuration one kind in the specific implementation, the top plate further includes at least one
Interface for external connection.
In the another in the specific implementation, the top plate and the lower plywood of Embedded A I machine vision hardware configuration
Between by connection plug-in unit connection.
In the another in the specific implementation, the top plate is located at the lower layer of Embedded A I machine vision hardware configuration
Plate top.
In the another of Embedded A I machine vision hardware configuration in the specific implementation, the power circuit includes POE electricity
Source and DC-DC circuit.
Embedded A I machine vision hardware configuration provided by the embodiments of the present application, comprising: top plate and lower plywood, lower plywood
In, multiple power supplies connect with CPU, storage device, debugging interface, camera interface, network PHY chip, clock and support circuit respectively
Connect, clock and support circuit are connect with CPU, provide clock and CPU start-up mode for CPU and select, CPU also with storage device,
External interface, debugging interface, the connection of network PHY chip, CPU transport for realizing high speed image identifying processing and Multiple Models Algorithm
It calculates, in top plate, power circuit is camera model power supply by camera power supply circuit, and power circuit is also that HDMI conversion chip supplies
Electricity, power circuit are also connect with multiple power supplies, provide power supply for lower plywood, camera model is connect by camera interface with CPU;
HDMI conversion chip is connect with CPU, and POE network module is connect with network PHY chip.Embedded A I is able to carry out complexity on end
Operation, while meeting requirement of real-time.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this Shen
Some embodiments please for those of ordinary skill in the art without any creative labor, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the schematic diagram of Embedded A I machine vision hardware configuration provided by the embodiments of the present application;
Fig. 2 is a kind of specific implementation of the lower plywood of Embedded A I machine vision hardware configuration provided by the embodiments of the present application
Schematic diagram;
Fig. 3 is a kind of specific implementation of the top plate of Embedded A I machine vision hardware configuration provided by the embodiments of the present application
Schematic diagram;
Fig. 4 is the schematic diagram of one example of Embedded A I machine vision hardware configuration provided by the embodiments of the present application.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is
Some embodiments of the present application, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art
All other embodiment obtained without creative efforts, shall fall in the protection scope of this application.
Artificial intelligence is a branch of computer science, it attempts to understand essence of intelligence, and is produced a kind of new
The intelligence machine that can be made a response in such a way that human intelligence is similar, the research in the field include robot, language identification, image
Identification, natural language processing and expert system etc..
It is so-called embedded, just refer to a kind of dedicated computer system that can be built in device.Usually, have
There is the equipment of digital interface that all there is embedded system, such as mobile phone, vehicle-mounted computer, smartwatch.And Embedded A I, then it is one
The technological concept that AI algorithm can be run on the terminal device for kind.In other words, its effect can exactly allow speaker, mobile phone,
The Intelligent hardwares such as robot, intelligent automobile, security device completed in real time in the case where not networking environment sensing, human-computer interaction,
The functions such as Decision Control.
The application of embedded artificial intelligence (Artificial Intelligence, AI) is universal all thus relatively uncomplicated, mainly
In terms of being following two:
(1) scene not high in requirement of real time, but when needing to handle mass data or complicated algorithm, usually acquisition
Information be transferred on cloud, cloud carries out model training or deduction, sets by the terminal device of result transmission, then by terminal
The result that standby display needs.
(2) in requirement of real-time height, but when data volume is little and algorithm is uncomplicated, the ARM of performance well can be used,
Element programmable logic array (Field Programmable Gate Array, PFGA), Digital Signal Processing (Digital
Signal Processing, DSP) or dedicated ASIC realize.
Powerful computing capability and quick real-time are not combined, limit system by two class above-mentioned
Usage scenario and commercialized popularization.The disadvantage of the first kind is exactly real-time, and similar unmanned aspect is needed
The scene quickly made a response is not suitable for.The shortcomings that second class is exactly that computing capability is not powerful, can only calculate uncomplicated algorithm,
It is unable to satisfy the AI model algorithm of larger data, and power consumption is big, it is at high cost.
The problem of for above scheme, this programme provide a kind of Embedded A I machine vision hardware configuration, this is embedding
Entering formula AI machine vision hardware system structure is a kind of hardware structure that can allow AI algorithm that can run on the terminal device, tool
Have that computing capability is strong, real-time is high, feature low in energy consumption, at low cost.It may be implemented to run a variety of AI algorithms, real-time processing data
Embedded A I system, and can satisfy low-power consumption, inexpensive requirement.
Using the product or project of Embedded A I machine vision hardware configuration: gate, security protection, it is unmanned, nobody zero
It sells, with no restrictions to this this programme.
Fig. 1 is the schematic diagram of Embedded A I machine vision hardware configuration provided by the embodiments of the present application, as shown in Figure 1, should
Embedded A I machine vision hardware configuration, comprising: top plate and lower plywood;I.e. the system is realized by upper layer and lower layer plate, under
Laminate is mainly minimum system, for the data storage of optimized integration, treatment process.
The lower plywood includes: that processor CPU, multiple power supplies, storage device, external interface, debugging interface, camera connect
Mouth, network PHY chip, clock and support circuit;
In the program, CPU is for realizing supercomputing, and generally strong using computing capability, the processor of low-power consumption is real
It is existing, it can specifically be selected according to the product of application.
The multiple power supplies respectively with the CPU, storage device, the debugging interface, the camera interface, described
Network PHY chip, the clock and support circuit connection, it is CPU that the multiple power supplies are used to connect to it, storage device, each
The devices such as a interface and network chip, clock circuit and other support circuits are powered.
The clock and support circuit are connect with the CPU, provide clock and the selection of CPU start-up mode for the CPU;
The CPU is also connect with storage device, the external interface, the debugging interface, the network PHY chip respectively;The CPU
For realizing high speed image identifying processing and Multiple Models Algorithm operation.
Top plate can be extended according to different product functions, such as applied in AI machine vision in this programme
In processing, then the top plate may include:
Power circuit, camera power supply circuit, camera model, high-definition multimedia interface (High Definition
Multimedia Interface, HDMI) conversion chip, active Ethernet (Power Over Ethernet, POE) network mould
Block;
The power circuit is connect with the camera power supply circuit, and the camera power supply circuit is connect with camera model, institute
It is camera model power supply that power circuit, which is stated, by the camera power supply circuit;Here camera model refers to obtain
The device of image and video, such as camera etc..
The power circuit is also connect with the HDMI conversion chip respectively, is powered for the HDMI conversion chip;The party
HDMI conversion chip in case is used to convert image, can connect with external display device, image is shown.
HDMI is a kind of digitized video/audio interface technology, is suitble to the tailored version digital interface of image transmission, can transmit sound simultaneously
Frequency and video signal, while without carrying out D/A or analog/digital conversion before signal transmits.HDMI supports EDID, DDC2B, because
This has the characteristics that the equipment of HDMI has " plug and play ", can carry out automatically " negotiation " between signal source and display equipment, from
The dynamic most suitable video/audio format of selection.
The power circuit is also connect with the multiple power supplies, provides power supply for the lower plywood;
The camera model is connect by the camera interface with the CPU;The HDMI conversion chip and the CPU connect
It connects;The POE network module is connect with the network PHY chip.
It is to be electrically connected between the device on device and lower plywood on top plate, needs to carry out the interaction of data, on
Laminate and lower plywood itself can carry out physical connection by plug-in unit or bracket etc..
Embedded A I machine vision hardware configuration provided in this embodiment, real-time height, power consumption strong using computing capability
Low, at low cost processor chips cooperate other function device, realize the calculation process of mass data, are able to carry out complexity
Operation, while meeting requirement of real-time, and power consumption is lower.
Fig. 2 is a kind of specific implementation of the lower plywood of Embedded A I machine vision hardware configuration provided by the embodiments of the present application
Schematic diagram, as shown in Fig. 2, on the basis of the above embodiments, one kind of Embedded A I machine vision hardware configuration is specific
In realization, the storage device includes EMMC chip and TF card.
The EMMC chip is that MMC association concludes, mainly for the embedded memory mark of the products such as mobile phone or tablet computer
The chip of quasi- specification.EMMC is integrated with a controller in a package, provides standard interface and manages flash memory, embedded by one
Storage solution composition, has MMC (multimedia card) interface, flash memory device and master controller.It is all at one
Small-sized welded ball array encapsulation (Ball Grid Array, BGA) encapsulation.
TF card, also referred to as Micro SD Card are a kind of memory devices of new generation based on semiconductor fast-flash memory device,
Since it is small in size, data transmission bauds is fast, the excellent characteristic such as hot-swappable, it is widely used in portable device such as electricity
Sub- dictionary, mobile phone, digital camera, auto-navigation system, and have in miniaturization, large capacity, high data rate storage
Whether there is or not alternative effect, can be used to store a large amount of data in this programme.
The EMMC chip and TF card are required to multiple power supplies and are powered, since EMMC chip and TF card are to carry out data
Storage, therefore it is required to connect with CPU, so as to can be with the realization according to system to data progress calculation process and transmission.
Optionally, in a kind of specific implementation, the debugging interface is jtag interface.
Optionally, in another concrete scheme, it should be understood that external interface can be one or more, the type of interface
It can also be selected according to the realization of product, which includes at least USB2.0 interface, USB3.0 interface and USB
One or more in type C interface, is also possible to other kinds of interface, this programme is with no restrictions.
Optionally, one kind of the program is in the specific implementation, the multiple power supplies are that integrated power supply manages circuit (Power
Management IC, PMIC) it is used to the power-supply device in managing main frame system, it is usually used in mobile phone and various mobile terminals is set
Standby, the power supply in the program for accessing the power circuit of top plate is managed, and is connected respectively to each different device
In be powered.
The program one kind in the specific implementation, the CPU be graphics processor MA2450.It can select model
The chip of MA2450 realizes quick data operation, also can choose the better CPU of performance in this scenario, or according to reality
Product selects the CPU of different processing capacity advantages, and this programme is with no restrictions.
According to shown in Fig. 2, showing the minimum system block diagram (i.e. lower plywood) of Embedded A I, function strong by computing capability, low
The main process task MA2450 of consumption, the PMIC of multiple-output electric power, the TYPE_C interface of high-speed transfer, massive store EMMC or TF card
Memory, JTAG debugging interface, all the way MIPI interface and gigabit networking and clock, reset circuit.The connection of upper and lower level
Interface.TYPEC power supply individually can be used in lower plywood, can also be powered, can individually be debugged with USB2.0 interface, surveyed to performance
It is convenient that examination and malfunction elimination provide.Flexible configuration can be carried out to the input clock of sensor by also having on plate, to meet difference
sensor。
Fig. 3 is a kind of specific implementation of the top plate of Embedded A I machine vision hardware configuration provided by the embodiments of the present application
Schematic diagram, as shown in figure 3, top plate can be extended according to different needs in the realization of above-mentioned two scheme, this
Scheme mainly introduces a kind of hardware configuration applied in machine vision, therefore is provided with camera model, the top plate it is specific
It further include the nothing of secure digital input-output card (Secure Digital Input and Output, SDIO) interface in realization
Line fidelity (WIreless-Fidelity, WIFI) circuit;The WIFI circuit is connect with the CPU;The power circuit is institute
State the power supply of WIFI circuit.
WIFI circuit in the program is used to be connected to the network with external network, carries out data transmission as needed, should
WIFI circuit is the alternative construction during this programme is realized.
Another in the program realize, the top plate further include expansible I2C bus and/or universal input/
It exports (General Purpose Input Output, GPIO), bus extender can also be known as, when microcontroller or chip
The no enough ports I/O of group, or when system is needed using distal end serial communication or control, GPIO product is capable of providing additionally
Control and function for monitoring etc..
In the another in the specific implementation, the top plate further includes at least one of Embedded A I machine vision hardware configuration
A interface for external connection, external connecter type this programme is with no restrictions.
As shown in figure 3, the power circuit includes POE power supply and DC-DC circuit.The POE power supply is referred in ether
In the case that screen cloth line architecture does not make any changes, while transmitting data-signal for some terminal devices, moreover it is possible to be
Such equipment provides direct current supply.DC-DC circuit refers to DC/DC conversion circuit.Main purpose is to carry out the change of voltage
It changes.In Embedded A I machine vision hardware configuration in the specific implementation, POE power supply and DC-DC circuit can be arranged simultaneously,
A setting can be selected, power supply can be provided for entire Embedded A I machine vision hardware configuration.
Top plate structural block diagram as shown in Figure 3, mainly have MIPI interface, POE interface, HDMI interface, WIFi interface,
I2C interface, GPIO interface, top plate are mainly that POE provides power supply, and power supply can be provided to whole system, have MIPI on plate
Interface can directly paste camera, can also be different demand and flexibly change;Have on plate I2C and GPIO for extend use.
In the implementation that any of the above-described scheme provides, the top plate of Embedded A I machine vision hardware configuration and described
Pass through connection plug-in unit connection between lower plywood.Optionally, the top plate is located at the lower plywood top.According to different functions
Demand, multiple top plates can be extended in each Embedded A I hardware configuration, with no restrictions to this this programme.
The design of this upper and lower level double-layer structure, can satisfy different demands, client can according to demand to top plate into
Row independent design achievees the purpose that want realization, it might even be possible to delete some functions on the basis of existing, realize more inexpensive
Requirement.This system architecture design increases flexibility, can satisfy a variety of usage scenarios in commercial applications.
Fig. 4 is the schematic diagram of one example of Embedded A I machine vision hardware configuration provided by the embodiments of the present application, such as Fig. 4 institute
Show, is realized in the specific implementation, holding and going up Embedded A I using powerful CPU any of the above-described, such as can choose: vision
Processor SOC Myriad 2, be integrated with 4Gb DDR3,12SHAVE, 2LEON4 32-BIT RISC processor,
Over 1TFLOPs Performance, dedicated PMIC module can support multiple-output electric power, interface MIPI interface abundant, TF
Card interface, SDIO interface, 3.0 interface of usb, I2C interface, HDMI interface, gigabit networking interface, multichannel GPIO interface etc. can
Meet existing most of embedded system peripheral hardwares.
System realizes that lower plywood is mainly minimum system using upper layer and lower layer plate, and top plate meets the expansion of different demands
Zhan Yong.Physical connection and electric connection can be carried out by connector J15 between top plate and lower plywood.
Whole system is mainly used for machine vision, and the image data of sensor acquisition is transmitted directly to by MIPI interface
CPU (Myrid 2) is internal, using CPU (Myrid 2) powerful calculating ability, calculates various AI algorithms such as: recognition of face, human body
Gesture recognition, commodity identify scheduling algorithm.Algorithm model is placed in EMMC or TF card, system can realize different function
Demand, system also increase the EEPROM that can identify hardware and software version, thus can in systems can be straight
It connects and finds out whether hardware and software and algorithm arrange in pairs or groups, the production of failure is likely to occur to later batch production shipment verification and later period
Product carry out investigation and provide convenience.System supports TYPE C power supply also to support POE power supply, and WIFI is supported to communicate.
The Embedded A I machine vision hardware configuration that this programme provides, using the design of upper and lower level double-layer structure, Ke Yiman
Sufficient different demands, client can carry out independent design to top plate according to demand, achieve the purpose that want realization, it might even be possible to
Some functions are deleted on the basis of existing, realize more inexpensive requirement.This system architecture design increases flexibility, in quotient
It can satisfy a variety of usage scenarios in industryization application.Select processing speed fast, operational capability is strong, and the lesser CPU of power consumption, so that
Embedded A I system framework solves those and needs to run complicated algorithm on end, and can satisfy the scene of requirement of real-time, because
To be at low cost, the scene low in energy consumption that also can satisfy conventional machines vision, and also it is small in size, it is flexible for installation changeable, it is easier to quotient
Industry.
In above-mentioned Embedded A I machine vision hardware configuration in the specific implementation, it should be understood that the function of CPU can also pass through
Other general processors, digital signal processor (English: Digital Signal Processor, abbreviation: DSP), dedicated collection
It is realized at circuit (English: Application Specific Integrated Circuit, abbreviation: ASIC) etc..General place
Reason device can be microprocessor or the processor is also possible to any conventional processor etc..In conjunction with the embodiment of the present application, institute is public
The step of method opened, can be embodied directly in hardware processor and execute completion, or with the hardware and software module in processor
Combination executes completion.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence;And storage medium above-mentioned includes: read-only memory (English
Text: read-only memory, abbreviation: ROM), RAM, flash memory, hard disk, solid state hard disk, tape (English: magnetic
Tape), floppy disk (English: floppy disk), CD (English: optical disc) and any combination thereof.
Finally, it should be noted that the above various embodiments is only to illustrate the technical solution of the application, rather than its limitations;To the greatest extent
Pipe is described in detail the application referring to foregoing embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement;And these are modified or replaceed, each embodiment technology of the application that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (12)
1. a kind of Embedded A I machine vision hardware configuration characterized by comprising top plate and lower plywood;
The lower plywood includes: processor CPU, multiple power supplies, storage device, external interface, debugging interface, camera interface, net
Network PHY chip, clock and support circuit;
The multiple power supplies respectively with the CPU, the storage device, the debugging interface, the camera interface, the network
PHY chip, the clock and support circuit connection, the device power supply that the multiple power supplies are used to connect to it;
The clock and support circuit are connect with the CPU, provide clock and the selection of CPU start-up mode for the CPU;It is described
CPU is also connect with storage device, the external interface, the debugging interface, the network PHY chip respectively;The CPU is used for
Realize high speed image identifying processing and Multiple Models Algorithm operation;
The top plate includes: power circuit, camera power supply circuit, camera model, high-definition multimedia interface HDMI conversion
Chip, active Ethernet POE network module;
The power circuit is connect with the camera power supply circuit, and the camera power supply circuit is connect with camera model, the electricity
Source circuit is camera model power supply by the camera power supply circuit;
The power circuit is connect with the HDMI conversion chip, is powered for the HDMI conversion chip;
The power circuit is also connect with the multiple power supplies, provides power supply for the lower plywood;
The camera model is connect by the camera interface with the CPU;The HDMI conversion chip is connect with the CPU;
The POE network module is connect with the network PHY chip.
2. Embedded A I machine vision hardware configuration according to claim 1, which is characterized in that the storage device includes
EMMC chip and TF card.
3. Embedded A I machine vision hardware configuration according to claim 1, which is characterized in that the debugging interface is
Jtag interface.
4. Embedded A I machine vision hardware configuration according to claim 1, which is characterized in that the external interface includes
Below at least one:
USB2.0 interface, USB3.0 interface and USB type C interface.
5. Embedded A I machine vision hardware configuration according to claim 1, which is characterized in that the multiple power supplies are collection
At electric power management circuit PMIC.
6. Embedded A I machine vision hardware configuration according to claim 1, which is characterized in that the CPU is at figure
Manage device MA2450.
7. Embedded A I machine vision hardware configuration according to claim 1, which is characterized in that the top plate further includes
The Wireless Fidelity WIFI circuit of secure digital input-output card SDIO interface;The WIFI circuit is connect with the CPU;The electricity
Source circuit is WIFI circuit power supply.
8. Embedded A I machine vision hardware configuration according to any one of claims 1 to 7, which is characterized in that on described
Laminate further includes expansible I2C bus and/or universal input/output GPIO.
9. Embedded A I machine vision hardware configuration according to any one of claims 1 to 7, which is characterized in that on described
Laminate further includes the interface that at least one is used for external connection.
10. Embedded A I machine vision hardware configuration according to any one of claims 1 to 7, which is characterized in that on described
Pass through connection plug-in unit connection between laminate and the lower plywood.
11. Embedded A I machine vision hardware configuration according to claim 10, which is characterized in that the top plate is located at
The lower plywood top.
12. Embedded A I machine vision hardware configuration according to any one of claims 1 to 7, which is characterized in that the electricity
Source circuit includes POE power supply and direct current DC-DC circuit.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201821034832.5U CN208367733U (en) | 2018-07-02 | 2018-07-02 | Embedded A I machine vision hardware configuration |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201821034832.5U CN208367733U (en) | 2018-07-02 | 2018-07-02 | Embedded A I machine vision hardware configuration |
Publications (1)
Publication Number | Publication Date |
---|---|
CN208367733U true CN208367733U (en) | 2019-01-11 |
Family
ID=65032938
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201821034832.5U Active CN208367733U (en) | 2018-07-02 | 2018-07-02 | Embedded A I machine vision hardware configuration |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN208367733U (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113360145A (en) * | 2021-07-14 | 2021-09-07 | 深圳思悦创新有限公司 | Customizable embedded AI model floor software architecture system |
CN113407482A (en) * | 2021-07-14 | 2021-09-17 | 深圳思悦创新有限公司 | Customizable embedded type AI model floor hardware architecture |
-
2018
- 2018-07-02 CN CN201821034832.5U patent/CN208367733U/en active Active
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113360145A (en) * | 2021-07-14 | 2021-09-07 | 深圳思悦创新有限公司 | Customizable embedded AI model floor software architecture system |
CN113407482A (en) * | 2021-07-14 | 2021-09-17 | 深圳思悦创新有限公司 | Customizable embedded type AI model floor hardware architecture |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104063290B (en) | Handle system, the method and apparatus of time-out | |
CN104956347B (en) | By a kind of interconnection agreement enumerate and/or configuration mechanism be used for different interconnection agreements | |
CN105874682B (en) | Charging method for electronic equipment and charging unit | |
CN107409056A (en) | Use the communication path of repeater selective enabling first and the second communication path | |
CN109388595A (en) | High-bandwidth memory systems and logic dice | |
CN103870429A (en) | High-speed-signal processing board based on embedded GPU | |
CN103685890A (en) | Intelligent camera with dynamic reconfiguration image processing function and dynamic reconfiguration method thereof | |
CN208367733U (en) | Embedded A I machine vision hardware configuration | |
CN103200199A (en) | Out of band (OOB) data collection system | |
CN104067200A (en) | Communication and monitoring of a battery via a single wire | |
CN103729319A (en) | Equipment system based on serial bus and data transmission method | |
CN104850516B (en) | A kind of DDR Frequency Conversion Designs method and apparatus | |
CN113609045A (en) | Intelligent network card BMC communication structure and method with strong universality | |
CN204217080U (en) | A kind of audio-video collection plate based on PPC processor | |
CN112000545A (en) | Graphics processor board card and graphics processor management method | |
CN201909987U (en) | Constant-current supply networking system on basis of serial bus | |
CN207124680U (en) | A kind of external camera assembly | |
CN110569173A (en) | Server health management chip based on Loongson IP core and implementation method | |
CN107911610A (en) | A kind of data handling system applied to image capture module | |
CN105005246A (en) | Portable data acquisition terminal | |
CN108965368A (en) | Data management circuit with network function and data management method based on network | |
CN202975894U (en) | Mobile logging terminal on based on cloud computing service | |
CN115237849B (en) | Dynamically reconfigurable and heterogeneous computing module and computing device | |
CN207542806U (en) | A kind of battery management system based on FPGA | |
CN206312135U (en) | A kind of Android intelligent kernel plate |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant |