CN104568971B - A kind of image detection device based on data statistics data mining - Google Patents
A kind of image detection device based on data statistics data mining Download PDFInfo
- Publication number
- CN104568971B CN104568971B CN201510059368.XA CN201510059368A CN104568971B CN 104568971 B CN104568971 B CN 104568971B CN 201510059368 A CN201510059368 A CN 201510059368A CN 104568971 B CN104568971 B CN 104568971B
- Authority
- CN
- China
- Prior art keywords
- data
- product
- image
- database
- register
- 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 present invention discloses a kind of image detection device based on data statistics data mining, its data statistics data digging method.Data statistics first:Database is set, some product data are write in database, databases have product register when initial luminance data and AE completion object brightnesses.Data mining afterwards, in follow-up test product, obtain product current light source initial luminance data to be measured, search for database, do Data Matching, data mining processing, the product register to match is found in database according to the initial luminance data of product to be measured, this product register is write into product to be measured, then finely tune AE.AE actions so can be completed in two steps:Database, then the data mining in database are formed by data statistics.
Description
Technical field
The technical field of image detection of the present invention and product, more particularly to a kind of image based on data statistics data mining
Detection means.
Background technology
Applied to mobile phone, vehicle-mounted, safety-security area camera product image detection device, for MTF, SFR, dirty point, wound
The scheduling algorithms such as point, stain (Blemish), ColorUniform, SNR, photocentre detection are based on different PC cpu instruction collection
SSE4.1, SSE4.2, SSE2, SSE3 dynamics use the research emphasis that instruction set optimization is those skilled in the art.
The content of the invention
For overcome the deficiencies in the prior art, the present invention provides a kind of image detection dress based on data statistics data mining
Put, it uses the software architecture based on QT/C++, for Product checking type:Auto-focusing (AF) embedding data statistics is dug
Function is dug, for MTF, SFR, dirty point, hinders the scheduling algorithms such as point, stain (Blemish), ColorUniform, SNR, photocentre detection
Based on different PC cpu instruction collection SSE4.1, SSE4.2, SSE2, SSE3 dynamics are optimized using instruction set.
The technical solution adopted for the present invention to solve the technical problems is:A kind of image based on data statistics data mining
Detection means, it includes data statistics data mining device, the data statistics data mining device include data statistics module and
Data-mining module;
The data statistics module is used to form database by data statistics, and it includes initial luminance data acquiring unit
First, product register acquiring unit two, recording unit;The initial luminance data acquiring unit one is used to extract test product in light
Initial luminance data under source;The product register acquiring unit two is used to control test product to do automatic exposure, reaches target
Brightness, product register when extracting the target of test product;The recording unit is used for initial luminance data and automatic exposure is complete
Into product register information during object brightness, the database is charged to;N test product is counted repeatedly, and n is natural number;
The data mining that the data-mining module is used in the database goes out the product register that product to be measured needs, its
Including initial luminance data acquiring unit two, search unit, product register acquiring unit two, writing unit, fine-adjusting unit;Should
Initial luminance data acquiring unit two is used for the current light source initial luminance data for obtaining product to be measured;The search unit is used to search
Rope database, do Data Matching, data mining processing;The product register acquiring unit two is used for according to the initial of product to be measured
Monochrome information finds the product register to match in database;The writing unit is used to treat the write-in of this product register
Survey product;The fine-adjusting unit is used to start corresponding product fine setting automatic exposure to be measured according to the product register of write-in;
Wherein, described image detection means also includes Image-capturing platform, parallel port transmission board;The data statistics data are dug
Device is dug to establish on the basis of PC, the primary control program interacted with parallel port transmission board, the Image-capturing platform;The parallel port
Transmission board card is connected by I2C, double camera parallel port, GPIO mouths with the Image-capturing platform;Parallel port transmission board includes adopting
With the control unit of fpga chip one, the signal conversion unit using two MC20901 chips, camera driver element;The camera
Driver element is assigned instruction and be passed to the MIPI signals of the camera of the Image-capturing platform by the GPIO mouths by the I2C to be somebody's turn to do
Two MC20901 chips;The MIPI signals are converted into one receivable LVDS of fpga chip letters by two MC20901 chips
Number;The fpga chip one possesses the view data of line synchronising signal by decoding IP kernel and being decoded into the LVDS signals, then should
View data Synchronous fluorimetry is transmitted to the image to the double camera parallel port and adopted for the simultaneously view data of port transmission 16 or 8 and output
Collect platform.
As the further improvement of such scheme, n is not less than 1000.
The present invention also provides another image detection device based on data statistics data mining, and it uses data statistics number
According to method for digging, the data statistics data digging method comprises the following steps:
Database is formed by data statistics:Extract initial luminance data of the test product under light source;Control test production
Product do automatic exposure, reach object brightness, product register when extracting the target of test product;By initial luminance data and automatically
Product register information during object brightness is completed in exposure, charges to the database;N test product is counted repeatedly, and n is nature
Number;
Data mining in the database goes out the product register that product to be measured needs:Obtain the current light of product to be measured
Source initial luminance data;Database is searched for, does Data Matching, data mining processing;According to the initial luminance data of product to be measured
The product register to match is found in database;This product register is write into product to be measured;According to the product of write-in
Register starts corresponding product fine setting automatic exposure to be measured;
Wherein, described image detection means also includes Image-capturing platform, parallel port transmission board;The data statistics data are dug
Device is dug to establish on the basis of PC, the primary control program interacted with parallel port transmission board, the Image-capturing platform;The parallel port
Transmission board card is connected by I2C, double camera parallel port, GPIO mouths with the Image-capturing platform;Parallel port transmission board includes adopting
With the control unit of fpga chip one, the signal conversion unit using two MC20901 chips, camera driver element;The camera
Driver element is assigned instruction and be passed to the MIPI signals of the camera of the Image-capturing platform by the GPIO mouths by the I2C to be somebody's turn to do
Two MC20901 chips;The MIPI signals are converted into one receivable LVDS of fpga chip letters by two MC20901 chips
Number;The fpga chip one possesses the view data of line synchronising signal by decoding IP kernel and being decoded into the LVDS signals, then should
View data Synchronous fluorimetry is transmitted to the image to the double camera parallel port and adopted for the simultaneously view data of port transmission 16 or 8 and output
Collect platform.
As the further improvement of such scheme, n is not less than 1000.
The present invention for establish on the basis of PC, with parallel port transmission board, current detection module, open short-circuit detecting module,
The primary control program of Image-capturing platform interaction, is connected by USB3.0 line with Part IV.PC is responsible for reception processing picture number
According to, extract characteristics of image, judge the work such as properties of product;Possess preferable human-computer interaction interface, Control card and production simultaneously
The sequential of product, control parallel port transmit board, current detection module, open short-circuit detecting module, Image-capturing platform (respectively first
Partly, Part II, Part III, Part IV) work etc., and embedded in Database Unit, possess to historical product
Characteristic extraction, analysis, data mining duty.
Brief description of the drawings
Fig. 1 is the modular structure schematic diagram of the image detection device of the present invention.
Fig. 2 is the electrical connection schematic diagram of fpga chip one in Fig. 1.
Fig. 3 is the electrical connection schematic diagram of MC20901 chips one in Fig. 1.
Fig. 4 is the electrical connection schematic diagram of MC20901 chips two in Fig. 1.
Fig. 5 is the current detection module structural representation of image detection device of the present invention.
The electrical connection that Fig. 6 is four Current Voltage detection chip IC IN220A of current detection module in Fig. 5 shows
It is intended to.
Fig. 7 is the electricity of the optional switch chip IC ADG811-YRU of four independences built in four of current detection module in Fig. 5
Property connection relationship diagram.
Fig. 8 is the electrical connection signal of the critical piece for opening short-circuit detecting module of image detection device of the present invention
Figure.
Fig. 9 is the modular structure schematic diagram of the Image-capturing platform of image detection device of the present invention.
Figure 10 is the software architecture schematic diagram that image detection device of the present invention uses.
Figure 11 is the data statistics flow signal for the data statistics data digging method that image detection device of the present invention uses
Figure.
Figure 12 is the data digging flow signal for the data statistics data digging method that image detection device of the present invention uses
Figure.
Figure 13 is the schematic perspective view of the board housing of image detection device of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, not
For limiting the present invention.
The 8lane MIPI decoding USB3.0 image detection devices of embedding data statistical Data Mining are applied to hand for one kind
Machine, vehicle-mounted, safety-security area camera product image detection device, it is mainly made up of following six part.Each section is set
All there is respective characteristic in meter scheme, while also can each form independent product, maintenance, safeguard, using being all very convenient.
1) Part I, 1lane mipi, 2lane are supported based on xilinx Spartan FPGA exploitation designs
MIPI, 4lane MIPI, 8LANE MIPI (double 4LANE mipi synchronous) MIPI decodings and synchronously into 8/16 and oral instructions
Input board card.
2) Part II, four tunnels for the product shooting that can focus for possessing 4 ranges of high accuracy are examined for electrically independent electric current
Survey module.
3) Part III, short-circuit detecting module is opened.Short-circuit OPEN SHAORT detection functions are opened for camera product,
And hot drawing plugging function is added by software optimization, four-way power supply (AF power supplies, AVDD, DOVDD, DVDD) is from dynamic circuit breaker during plug product
Electricity Functional on electricity.
4) Part IV, developed based on Cypress 3014 and xilinx Spartan FPGA, valid data transmission efficiency
Reach 3.7Gbps USB3.0 Image-capturing platforms.The platform be directed to picture frame before with postamble be embedded in frame before postamble synchronizing signal,
Host computer constantly detects the frame head and postamble of view data, avoids the processing of wrong frame, the wrong data processing of view data, and carries
For two-way, per road, a variety of Sensor reference frequencies provide, the two-way a variety of supply of I2C communication patterns, a variety of Sensor power supplys and controls
Signal.
5) Part V, the software architecture based on QT/C++, auto-focusing (AF), automatic is included for Product checking type
The embedding data statistical Data Mining function such as focusing, the burning of AF positions, automatic exposure (AE), automatic white flat horizontal (AWB).For
MTF, SFR, dirty point, hinder the scheduling algorithms such as point, stain (Blemish), ColorUniform, SNR, photocentre detection based on different PC
Cpu instruction collection SSE4.1, SSE4.2, SSE2, SSE3 dynamics are optimized using instruction set.
6) Part VI, board housing, using fan-free heat dissipation technology, device case towards high-speed motion apparatus, for
Designed outside USB3.0 interface hulls using non-threaded.
Referring to Fig. 1, parallel port transmission board mainly includes fpga chip one (such as FPGA XC6SLX45), 2 MC20901
Chip (such as MC20901TQLMP-48), camera (Camera) driver element.Parallel port transmission board uses xilinx
Spartan6XC6SLX45FPGA exploitation designs, using I2C (Inter-Integrated Circuit, bus), by I2C,
Double camera parallel port, GPIO mouths are connected with Image-capturing platform.Camera driver element assigns instruction and by this by the I2C
The MIPI signals of the camera of the Image-capturing platform are passed to two MC20901 chips by GPIO mouths, and MC20901 chips are by signal
The receivable LVDS signals of fpga chip one are converted into, fpga chip one possesses row by decoding IP kernel and being decoded into LVDS signals
The view data of synchronizing signal, then the view data Synchronous fluorimetry for the simultaneously view data of port transmission 16 or 8 and is passed through
Level conversion IC is exported to double Camera parallel ports and is transmitted to Image-capturing platform.The GPIO mouths can be double 40 pins (PIN) insert rows.Figure
As acquisition platform uses fpga chip two (can also use xilinx Spartan6XC6SLX45FPGA exploitations design) to realize and control
System.
Incorporated by reference to Fig. 2, Fig. 3 and Fig. 4, wherein Fig. 2 is the electrical connection schematic diagram of fpga chip one;2 MC20901
Chip is MC20901 chips one and MC20901 chips two, and Fig. 3, Fig. 4 are respectively MC20901 chips one and MC20901 chips two
Electrical connection schematic diagram.The annexation of fpga chip one and 2 MC20901 chips is as shown in table 1, table 2, table 3.
The pin and its network label of the fpga chip one of table 1
The pin and its network label of the MC20901 chips one of table 2
The pin and its network label of the MC20901 chips two of table 3
In summary, the electrical connection between fpga chip one and 2 MC20901 chips is as shown in table 4 below.
Transmit the electrical connection of board in the parallel port of table 4
In MC20901 chips one, pin 45 connects power Vcc one, is also connect via electric capacity C85 ground connection, via electric capacity C86
Ground;Pin 15 connects power Vcc one, is also grounded via electric capacity C87 ground connection, via electric capacity C88;The power Vcc two of pin 32, is also passed through
It is grounded by electric capacity C67 ground connection, via electric capacity C68.
In MC20901 chips two, pin 45 connects power Vcc one, is also connect via electric capacity C62 ground connection, via electric capacity C63
Ground;Pin 15 connects power Vcc one, is also grounded via electric capacity C64 ground connection, via electric capacity C65;The power Vcc two of pin 32, is also passed through
It is grounded by electric capacity C60 ground connection, via electric capacity C61.
Image detection device of the present invention, its parallel port used transmission board are based on xilinx Spartan FPGA exploitations and set
Meter, 1lane mipi, 2lane MIPI, 4lane MIPI, 8LANE MIPI (double 4LANE mipi are synchronous) MIPI solutions can be supported
Code and synchronously into 8/16.
Referring to Fig. 5, current detection module (for Part II) is using four IN220A points of Current Voltage detection chip IC
It Pei He not realize that four tunnels are independently-powered by the optional switch chip IC ADG811-YRU of four independences built in four.Current detection module
Four road power supply circuits be directed to respectively Image-capturing platform (Part IV) camera (such as camera product) need four tunnels
Power supply AF, AVDD, DOVDD, DVDD do the measurement of four ranges, and precision is up to 1uA.Current detection module uses four electric current electricity
Pressure detection chip IC IN220A realize that four tunnels are independently-powered.Four Current Voltage detection chip IC IN220A by I2C (such as
SDA and SCL) it is connected with the fpga chip two (i.e. xilinx Spartan6 XC6SLX45FPGA) of Image-capturing platform.IC
IN220A refers to IN220A chips.Therefore current detection module possesses 4 ranges of high accuracy, four tunnels of the product that can focus shooting independently supply
Electricity.
Incorporated by reference to Fig. 6, Fig. 7, it is that four Current Voltage detection chip IC IN220A and four independences built in four are optional
Switch chip IC ADG811-YRU electrical connection schematic diagram, IC ADG811-YRU refer to ADG811-YRU chips.Four
Current Voltage detection chip IC IN220A are respectively U20, U22, U24, U27, the optional switch chip IC of four independences built in four
ADG811-YRU is respectively U21, U23, U25, U26, and the electrical connection between them is as shown in table 4.
The electrical connection of the current detection module of table 4
Wherein, according to popular understanding, V33 is expressed as 33V power supply, and GND is expressed as ground, and IN1~IN16 represents defeated respectively
The corresponding output for entering to connect fpga chip two is used to control U21, U23, U25, U26.
In U20, pin 6 is via electric capacity C164 electrical grounds, and pin 9 is via resistance R43 connection power supply DVDD1, pin
10 via resistance R33 connection power supply DVDD, and pin 10 is also via electric capacity C161 connections pin 9.
In U22, pin 6 is via electric capacity C165 electrical grounds, and pin 9 is via resistance R44 connection power supply DODD1, pin
10 via resistance R38 connection power supply DODD, and pin 10 is also via electric capacity C162 connections pin 9.
In U24, pin 6 is via electric capacity C171 electrical grounds, and pin 9 is via resistance R55 connection power supply AVDD1, pin
10 via resistance R47 connection power supply AVDD, and pin 10 is also via electric capacity C168 connections pin 9.
In U27, pin 6 is via electric capacity C172 electrical grounds, and pin 9 is via resistance R56 connection power supply AF1, pin 10
Via resistance R48 connection power supply AF, and pin 10 is also via electric capacity C167 connections pin 9.
In U21, pin 2 via resistance R35 connection power supply DVDD1, pin 7 via resistance R37 connection power supply DVDD1,
Pin 10 connects power supply DVDD1 via resistance 34, and pin 13 connects power supply V33 and via electric capacity C164 electrical grounds, pin 15
Via resistance R36 connection power supplys DVDD1.
In U23, pin 2 via resistance R42 connection power supply DODD1, pin 7 via resistance R39 connection power supply DODD1,
Pin 10 connects power supply DODD1 via resistance 40, and pin 13 connects power supply V33 and via electric capacity C166 electrical grounds, pin 15
Via resistance R41 connection power supplys DODD1.
In U25, pin 2 via resistance R51 connection power supply AVDD1, pin 7 via resistance R49 connection power supply AVDD1,
Pin 10 connects power supply AVDD1 via resistance 50, and pin 13 connects power supply V33 and via electric capacity C169 electrical grounds, pin 15
Via resistance R45 connection power supplys AVDD1.
In U26, pin 2 is via resistance R54 connection power supply AF1, and pin 7 is via resistance R53 connection power supply AF1, pin
10 connect power supply AF1 via resistance 52, and pin 13 connects power supply V33 and via electric capacity C170 electrical grounds, and pin 15 is via electricity
Hinder R46 connection power supplys AF1.
Core is detected in the state of being not powered on referring to Fig. 8, opening short-circuit detecting module and can be directed to the camera product inserted
Piece pin opens short circuit, avoids under product short-circuit conditions electricity, product and mainboard are caused to damage, at the same can also intercept it is such not
Good product.Open short-circuit detecting module with the fpga chip one in parallel port transmission board to be connected, pass through one split short circuit of fpga chip
AD data be acquired and analyze, while fpga chip one with controlling when opening short-circuit detecting camera is isolated with mainboard ground
With short circuit, and the disconnection and connection of the Signal Products such as PWDN, RESET, I2C and receiving terminal.
Open short-circuit detecting module and combine analog-digital converter using 2 optional switch chips (ADG706 chips)
(AD7477AAKSZ-500RL7 chips) design is realized.2 ADG706 chips are U6, U7, AD7477AAKSZ-500RL7 chips
For U5, the annexation between them is as shown in table 5.
Table 5 opens the electrical connection of short-circuit detecting module
Wherein, in U5, pin 3 is via resistance R90 electrical grounds.
In U6, pin 1 distinguishes electrical ground via electric capacity C40, electric capacity C41, and pin 28 connects power supply via resistance R9
VCC。
In U7, pin 1 distinguishes electrical ground via electric capacity C38, electric capacity C39.
Referring to Fig. 9, Image-capturing platform as whole platform PC and parallel port transmission board, current detection module and
The maincenter plate of the data transfer of short-circuit detecting module is opened, not only provides voltage module (its of the different cameras of polymorphic type
Middle AVDD supports 3.3V, 2.8V, 2.5V, 1.8V, 1.5V, 0V;DOVDD supports 3.3V, 2.8V, 2.5V, 1.8V, 1.5V, 0V;
DVDD provides 3.3V, 2.8V, 1.8V, 1.5V, 1.2V, 0V) and support per road independence power down function, provide duplexing polytype
I2C communication patterns (place value of 8 bit address 8, the place value of 8 bit address 16, the place value of 16 bit address 8, the place value of 16 bit address 16,16 ground
The place value of location 32, the place value of 32 address 32 etc.);Provide multi-frequency camera product supply when kind (6.25M, 12.5M, 25M,
50M);Support PWDN and RESET software exchange.The major function of the Image-capturing platform by control signal and picture signal with
The interaction of PC main programs.
Referring to Fig. 10, software architecture is transmitted board, current detection module to establish on the basis of PC, with parallel port, opened
Short-circuit detecting module, the primary control program of Image-capturing platform interaction, are connected by USB3.0 line with Part IV.PC is responsible for connecing
Processing view data is received, characteristics of image is extracted, judges the work such as properties of product;Possess preferable human-computer interaction interface simultaneously, control
The sequential of making sheet card and product, control parallel port transmission board, current detection module, open short-circuit detecting module, IMAQ is put down
Work of platform (being respectively Part I, Part II, Part III, Part IV) etc., and Database Unit is embedded in, gather around
There is the characteristic extraction to historical product, analyze, data mining duty.
The data statistics data digging method that software architecture uses is as shown in FIG. 11 and 12.First:Conventional 6 steps are completed certainly
Dynamic exposure (AE).Figure 11 is referred to, database is set, if dryed product is write in database ((refers to the image detection dress of the present invention
Put)) data, such as record 1000 product data.What the data of each product were such to:1st, A products are tested;2nd, A productions are extracted
Initial luminance data of the product under light source;3rd, control A products do automatic exposure (as completed automatic exposure using conventional 6 steps), reach
To object brightness, product register when extracting the target of A products, that is to say, that the register of extraction A products now product;4th, will
Initial luminance data and AE complete product register information data-in storehouse during object brightness.
Afterwards, in follow-up test product, Figure 12 is referred to, such as product B to be measured, you can 2 steps complete AE, obtain product B
Current light source initial luminance data, database is searched for, do Data Matching, data mining processing, believed according to product B original intensity
Breath finds the product register to match in database, by this product register (the product register of product in database)
Product B to be measured is write, then finely tunes AE.AE actions so can be completed in two steps:Database is formed by data statistics, then in data
Data mining in storehouse.
The data statistics data digging method can be designed to that following data statistics data are dug in Software Module Design
Dig device:The data statistics data mining device includes data statistics module and data-mining module.
Data statistics module be used for by data statistics formation database, it include initial luminance data acquiring unit one,
Product register acquiring unit two, recording unit.Initial luminance data acquiring unit one is used to extract test product under light source
Initial luminance data;Product register acquiring unit two is used to control test product to do automatic exposure (as used conventional 6 steps complete
Into automatic exposure), reach object brightness, product register when extracting the target of test product, that is to say, that extraction test product
The now register of product;Recording unit is used to complete initial luminance data and AE product register information during object brightness
Data-in storehouse.At least 1000 test products are counted repeatedly.
The data mining that data-mining module is used in database goes out the product register that product to be measured needs, and it includes
Initial luminance data acquiring unit two, search unit, product register acquiring unit two, writing unit, fine-adjusting unit.It is initial bright
Degree information acquisition unit two is used for the current light source initial luminance data for obtaining product to be measured;Search unit is used to search for data
Storehouse, does Data Matching, data mining processing, and product register acquiring unit two is used for the initial luminance data according to product to be measured
The product register to match is found in database;Writing unit is used for this product register (product in database
Product register) write-in product to be measured, fine-adjusting unit according to the product register of write-in for starting product fine setting AE to be measured.
The data statistics data digging method that software architecture uses, when the test product in early stage, according to original intensity
Information is established after the database of product register, and just the test of convenient follow-up trial product to be measured, improves testing efficiency.
Board housing, as shown in figure 13, using aluminum alloy casing, the design of category mechanism, especially set to protect overall board
Meter.Shell includes the upper casing 18 being fastened togather up and down and lower casing 17.Transmit the caloric value of board and Image-capturing platform in parallel port
Higher fpga chip is directly connected by radiating silica gel sheet with housing, and USB3.0 lines are to avoid and equipment height with board interface
Produce and rock during motion, transmitting board with parallel port by an accessory 15 is connected, and accessory 15 is used to fix USB3.0 lines and simultaneously oral instructions
The interface section of input board card.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement made within refreshing and principle etc., should be included in the scope of the protection.
Claims (4)
- A kind of 1. image detection device based on data statistics data mining, it is characterised in that:It includes the digging of data statistics data Device is dug, the data statistics data mining device includes data statistics module and data-mining module;The data statistics module is used to form database by data statistics, and it includes initial luminance data acquiring unit one, production Product register acquiring unit two, recording unit;The initial luminance data acquiring unit one is used to extract test product under light source Initial luminance data;The product register acquiring unit two is used to control test product to do automatic exposure, reaches object brightness, Product register when extracting the target of test product;The recording unit is used to initial luminance data and automatic exposure completing target Product register information during brightness, charges to the database;N test product is counted repeatedly, and n is natural number;The data mining that the data-mining module is used in the database goes out the product register that product to be measured needs, and it includes Initial luminance data acquiring unit two, search unit, product register acquiring unit two, writing unit, fine-adjusting unit;This is initial Monochrome information acquiring unit two is used for the current light source initial luminance data for obtaining product to be measured;The search unit is used to search for number According to storehouse, Data Matching, data mining processing are done;The product register acquiring unit two is used for the original intensity according to product to be measured Information finds the product register to match in database;The writing unit is used to this product register writing production to be measured Product;The fine-adjusting unit is used to start corresponding product fine setting automatic exposure to be measured according to the product register of write-in;Wherein, described image detection means also includes Image-capturing platform, parallel port transmission board;The data statistics data mining fills Foundation is set on the basis of PC, the primary control program interacted with parallel port transmission board, the Image-capturing platform;It is somebody's turn to do simultaneously port transmission Board is connected by I2C, double camera parallel port, GPIO mouths with the Image-capturing platform;Parallel port transmission board includes using The control unit of fpga chip one, the signal conversion unit using two MC20901 chips, camera driver element;The camera drives Moving cell by the I2C assign instruction and by the GPIO mouths by the MIPI signals of the camera of the Image-capturing platform be passed to this two Individual MC20901 chips;The MIPI signals are converted into one receivable LVDS of fpga chip letters by two MC20901 chips Number;The fpga chip one possesses the view data of line synchronising signal by decoding IP kernel and being decoded into the LVDS signals, then should View data Synchronous fluorimetry is transmitted to the image to the double camera parallel port and adopted for the simultaneously view data of port transmission 16 or 8 and output Collect platform.
- 2. the image detection device according to claim 1 based on data statistics data mining, it is characterised in that:N is not small In 1000.
- A kind of 3. image detection device based on data statistics data mining, it is characterised in that:It uses data statistics data to dig Pick method, the data statistics data digging method comprise the following steps:Database is formed by data statistics:Extract initial luminance data of the test product under light source;Control test product is done Automatic exposure, reach object brightness, product register when extracting the target of test product;By initial luminance data and automatic exposure Product register information during object brightness is completed, charges to the database;N test product is counted repeatedly, and n is natural number;Data mining in the database goes out the product register that product to be measured needs:At the beginning of the current light source for obtaining product to be measured Beginning monochrome information;Database is searched for, does Data Matching, data mining processing;According to the initial luminance data of product to be measured in number The product register to match according to being found in storehouse;This product register is write into product to be measured;Deposited according to the product of write-in Device starts corresponding product fine setting automatic exposure to be measured;Wherein, described image detection means also includes Image-capturing platform, parallel port transmission board;The data statistics data mining fills Foundation is set on the basis of PC, the primary control program interacted with parallel port transmission board, the Image-capturing platform;It is somebody's turn to do simultaneously port transmission Board is connected by I2C, double camera parallel port, GPIO mouths with the Image-capturing platform;Parallel port transmission board includes using The control unit of fpga chip one, the signal conversion unit using two MC20901 chips, camera driver element;The camera drives Moving cell by the I2C assign instruction and by the GPIO mouths by the MIPI signals of the camera of the Image-capturing platform be passed to this two Individual MC20901 chips;The MIPI signals are converted into one receivable LVDS of fpga chip letters by two MC20901 chips Number;The fpga chip one possesses the view data of line synchronising signal by decoding IP kernel and being decoded into the LVDS signals, then should View data Synchronous fluorimetry is transmitted to the image to the double camera parallel port and adopted for the simultaneously view data of port transmission 16 or 8 and output Collect platform.
- 4. the image detection device according to claim 3 based on data statistics data mining, it is characterised in that:N is not small In 1000.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510059368.XA CN104568971B (en) | 2015-02-05 | 2015-02-05 | A kind of image detection device based on data statistics data mining |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510059368.XA CN104568971B (en) | 2015-02-05 | 2015-02-05 | A kind of image detection device based on data statistics data mining |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104568971A CN104568971A (en) | 2015-04-29 |
CN104568971B true CN104568971B (en) | 2018-02-02 |
Family
ID=53085538
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510059368.XA Active CN104568971B (en) | 2015-02-05 | 2015-02-05 | A kind of image detection device based on data statistics data mining |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104568971B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110519527A (en) * | 2019-09-19 | 2019-11-29 | 天津英田视讯科技有限公司 | A kind of camera auto exposure parameter generation method based on image big data |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101916429A (en) * | 2010-07-09 | 2010-12-15 | 浙江大学 | Geometric correction and disparity extraction device of binocular camera |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4415674B2 (en) * | 2002-01-29 | 2010-02-17 | 株式会社ニコン | Image forming state adjusting system, exposure method, exposure apparatus, program, and information recording medium |
CN101975357B (en) * | 2010-09-25 | 2012-07-04 | 东华大学 | Self-adaptive light source lighting system and method for high-precision image measuring device |
CN102846301B (en) * | 2011-07-08 | 2014-07-02 | 上海交通大学 | Intelligent adjusting device of endoscope illumination |
-
2015
- 2015-02-05 CN CN201510059368.XA patent/CN104568971B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101916429A (en) * | 2010-07-09 | 2010-12-15 | 浙江大学 | Geometric correction and disparity extraction device of binocular camera |
Also Published As
Publication number | Publication date |
---|---|
CN104568971A (en) | 2015-04-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104581151B (en) | A kind of image detection device that board is transmitted based on parallel port | |
CN104539945B (en) | It is a kind of that there is the image detection device for opening function of short circuit detection | |
CN105427320B (en) | A kind of image division and extracting method | |
CN108141549A (en) | Imaging sensor and the electronic equipment with imaging sensor | |
WO2020105863A1 (en) | Electronic device including camera module in display and method for compensating for image around camera module | |
US11314672B2 (en) | Configurable input / output connector in a camera | |
CN103037242B (en) | A kind of camera module test system | |
CN205961333U (en) | Cell -phone camera module multimode while testing of group device | |
CN102625056A (en) | FPGA (Field-Programmable Gate Array)-based CIS (Contact Image Sensor) image acquisition system and acquisition method thereof | |
CN107872394A (en) | Fault self-checking type wireless router | |
CN104580919B (en) | A kind of image detection device | |
CN104568971B (en) | A kind of image detection device based on data statistics data mining | |
CN101738827A (en) | usb digital industrial camera | |
CN101778219A (en) | Device and method for ensuring synchronous working of plurality of camera heads | |
CN109951617A (en) | A kind of double spectrum high speed cameras based on fpga | |
CN102967531A (en) | Novel gel-imaging system and method | |
CN106331452A (en) | Device and method for performing image acquisition by utilizing synchronous slave SLAVEFIFO manner | |
CN105046192A (en) | Recognition control method and device for fingerprint recognition module | |
CN104581024B (en) | The compatible color high-definition CCD front-end video collection modules of multi-mode and implementation method | |
CN105930293A (en) | Serial transceiving interface module and USB (universal serial bus) interface module shared port circuit and working method | |
CN102866165A (en) | Surface defect detecting device of hardware | |
CN104539946A (en) | Four power supply independent image detection device | |
CN111010524A (en) | VGA video signal screen capture device and screen capture method | |
CN104867201A (en) | Mobile terminal equipment for operation and maintenance | |
CN206042135U (en) | Image sensor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |