CN108633411A - A kind of seedling case rice shoot amount real-time monitoring system and method based on machine vision - Google Patents
A kind of seedling case rice shoot amount real-time monitoring system and method based on machine vision Download PDFInfo
- Publication number
- CN108633411A CN108633411A CN201810463986.4A CN201810463986A CN108633411A CN 108633411 A CN108633411 A CN 108633411A CN 201810463986 A CN201810463986 A CN 201810463986A CN 108633411 A CN108633411 A CN 108633411A
- Authority
- CN
- China
- Prior art keywords
- seedling
- rice shoot
- seedling case
- rice
- camera
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01C—PLANTING; SOWING; FERTILISING
- A01C11/00—Transplanting machines
- A01C11/003—Transplanting machines for aquatic plants; for planting underwater, e.g. rice
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/182—Level alarms, e.g. alarms responsive to variables exceeding a threshold
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Soil Sciences (AREA)
- Environmental Sciences (AREA)
- Guiding Agricultural Machines (AREA)
Abstract
The invention discloses a kind of seedling case rice shoot amount real-time monitoring system and method based on machine vision, is made of seedling case, holder, bolt, guide rail, Rubber Conveyor Belt Scrap, camera mechanism for installing, partition board, rice shoot, camera and controller motherboard.By acquiring real-time monitoring and alarm of the realization of seedling case image information to rice shoot surplus in real time.Under controller action, distribution situation of the rice shoot in seedling case is monitored using image acquisition units, realizes the real-time monitoring to rice shoot surplus, in conjunction with each function module, is realized to monitoring that remaining rice shoot is less than the sound-light alarm of alarm line and the minimum rate of accumulation.Rice shoot amount real-time monitoring system of the present invention and method intelligence degree are high, at low cost, simple in structure, small change only, which need to be carried out, to existing rice transplanter to install and use, it is applied widely, it is applicable to seedling case rice shoot surplus in unmanned and manned rice transplanter operation process and carries out monitoring in real time and alarm.
Description
Technical field
The present invention relates to the seedling case rice shoot amount real-time monitoring systems and method of a kind of machine vision, belong to Agricultural Intelligent System machine
Tool field.
Background technology
Riding type high-speed transplanter is increasingly favored by Rice Cropping family, but existing riding type due to operating efficiency height
High-speed transplanter, which needs airborne 2-3 personnel to carry out rice transplanter driving and upper seedling, airborne personnel, causes rice transplanter in paddy-field-working
When the phenomenon that is short of power take place frequently；Rice transplanting is seasonal strong, it is necessary to concentrate rice transplanting by its prescribed period of time.In recent years, due to rural area
Labour to the transfer in city and the raising year by year of employment cost, concentrate between plantation, artificial shortage and employment cost by rice
Contradiction is more and more prominent.On the other hand, the pilot steering of rice transplanter is influenced by driver's technical merit and supervisor's factor, is often led
It causes rice shoot row winding, inserts not straight.Based on the above situation, scientific research institution and agricultural machinery enterprise are turned to unmanned rice transplanter
Research and development, the unmanned rice transplanter that in April, 2017 Japanese agriculture engineering research center is developed successfully list, and straight-line travelling is accurate
Exactness deviation<2cm realizes rice transplanting machine automatic drive, and 1 people can manipulate 1 rice transplanter in field, but due to seedling case
It is limited to carry seedling amount, needs to prejudge the area that the rice shoot amount for filling seedling case is capable of rice transplanting, it is often necessary in field into pedestrian
Seedling compares with seedling on the edge of a field in work, paddy field mud foot depth and needs rice shoot being transported to seedling on field from the edge of a field, greatly reduces
The efficiency of rice transplanter.There is an urgent need to research and develop the technology monitored in real time to seedling case residue rice shoot amount and device.
In the prior art, applicant not yet has found the real time monitoring and device to rice shoot surplus in seedling case.
In order to make up above-mentioned deficiency, the present invention is based on machine vision techniques, it is proposed that seedling case rice shoot amount real-time monitoring system
And method, by right over seedling case center and be approximately parallel to seedling case position install camera, make entire seedling case in camera
Horizon range in, have the green of rice shoot at seedling and the color characteristic difference without seedling case at seedling using seedling case, acquisition seedling case
Image information simultaneously carries out image procossing, calculates seedling case distribution shared by green rice shoot, realize real-time monitoring to seedling case rice shoot amount and
Alarm, bonding unit is apart from rice transplanting amount, it is ensured that seedling case rice shoot surplus can be planted at seedling on the edge of a field, to reach on the edge of a field
The purpose of the not upper seedling in seedling, field, promotes the operating efficiency of unmanned rice transplanter, has the advantages that intelligent, reliability is high.
Invention content
The seedling case rice shoot amount real-time monitoring system of machine vision of the present invention a kind of and method are intended to provide in a kind of seedling case
The real-time monitoring system and method for rice shoot surplus, to realize the real-time prison of rice shoot surplus in seedling case during rice transplanting transplanting
It surveys simultaneously and alarm is shut down.
In order to solve the above technical problems, the specific technical solution that the present invention uses is as follows：
A kind of seedling case rice shoot amount real-time monitoring system based on machine vision of the invention includes mainly：Seedling case, holder, spiral shell
Bolt, guide rail, Rubber Conveyor Belt Scrap, camera mechanism for installing, partition board, rice shoot, camera and controller motherboard；
The seedling case is separated into 6 seedling lattice by partition board, and each seedling lattice are equipped with Rubber Conveyor Belt Scrap close to the part of bottom；
The holder is process using stainless steel material, is inShape is mounted on bolt at left and right sides of seedling case,
Rack beam middle position is welded with camera mechanism for installing, and holder is located at right over seedling case and makes camera face seedling
Seedling adjusts cam lens focal length, makes seedling case in the horizon range of camera；
The camera mechanism for installing is welded on holder, and camera mechanism for installing uses thickness for 2mm
Stainless steel plate be process, open the hole there are two a diameter of 4mm thereon；
The guide rail is the guide rail on existing riding type high-speed transplanter, is used to support seedling case and is moved horizontally for seedling case
Guiding role is provided；
The Rubber Conveyor Belt Scrap is the rice shoot conveyer belt on existing riding type high-speed transplanter, and rice shoot is avoided to wrinkle；
The partition board is the seedling separator for container on existing riding type high-speed transplanter, is used for seedling case according to rice transplanting transplanting
Line number is separated into the seedling lattice of corresponding line number.
The camera is that woods cypress regards S908 industry wide-angle cameras, and image is powered and transmitted by USB interface, adjusts camera shooting
Head resolution ratio is 640*480, and camera is installed by the hole of two a diameter of 4mm of camera mechanism for installing with camera
Fixed mechanism is bolted；
The controller motherboard include controller, FIFO memory modules, WIFI signal transmitting module, SD card, button,
Buzzer, LED light, power voltage step down module and 12V DC power supplies are electrically connected with camera by USB interface bus；
The controller is S3C2410 embedded microprocessors, and the system on chip interior integrated system bus AHB is total
Lane controller, Power Management Unit, PLL clock generator, internal SRAM, external piloting control device, LCD control units, DMA controls are single
Member, interrupt control unit；The hardware component being connected on peripheral bus APB has the UART of triple channel, main I more than one2C bus marcos
Device, an I2S bus control units, Five-channel PWM timers and a timer internal, a WatchDog Timer, general purpose I/O
Mouth, the real-time clock RTC with calendar function, 10, eight channel ADC, two synchronous serial interface SIO interfaces and SDI/MMC
Deng.By the chip to the image collected carry out pretreatment and effective information extract, concurrently send warning message to cockpit and far
Journey receiving device；
The FIFO memory modules are FT2232H chips, which has USB2.0 interfaces and UART, FIFO, I2C、
The ability that the channel interfaces such as SPI, JPAG are connected and configure, for image data to be transmitted to controller；
The WIFI signal transmitting module be ESP8266EX chips, the chip height piece inner height be integrated with duplexer,
Radio frequency balun, power amplifier, low-noise reception amplifier, filter and power management manage module, for the remote of warning message
Journey transmits wirelessly；
The SD card can preserve image data in case subsequent analysis by the SDI Component drivers in controller；
The button is connected with S3C2410 chips, installs in cockpit, and the driver of cockpit can press when alarm
Button cancels alarm；
The buzzer and LED light and S3C2410 chip Is/O mouthfuls are connected, and acousto-optic report is carried out to the driver in cockpit
It is alert；
The voltage reduction module is LM2596S buck converters, is changed into 12V D/C voltages by power voltage step down module
5V D/C voltages be camera power supply, 12V D/C voltages are changed by 3.3V D/C voltages by DC voltage reduction modules, in order to control device,
FIFO memory modules and the power supply of WIFI signal transmitting module；
The 12V DC power supplies are rice transplanting ship battery.
A kind of seedling case rice shoot amount method of real-time based on machine vision, includes the following steps：
Step 1：The setting of job parameter
The rice shoot minimum rate of accumulation that rice shoot is alarmed is cured in program, after seedling piece is filled in rice seedling box of rice seedling transplanting machine, to control
Device powers on, and checks whether the system signal collecting unit is normal, and whether camera can persistently take pictures, and whether seedling case entirety, which is located at, is taken the photograph
As in range, if normal, system enters job state；If breaking down, failure is sent to cockpit and distant place receiving device
Information, notify operating personnel check camera whether failure, after trouble shooting, into job state.
Step 2：Image information collecting
It sets the image sampling period to 1s by controller motherboard, sets resolution ratio of camera head to 640*480, as
Plain depth is set as 8, then every image is sampled and quantization obtains the matrix of a 640*480, image coordinate f's (x, y)
Value is discrete magnitude, and the discrete magnitude of acquisition is carried out rounding, is embodied as
Since the color of each pixel is divided into tri- components of RGB, each coordinate of formula (1) includes three components, then obtains
Every photo has 640*480*3=921600 discrete data, each discrete data to be represented by
F (x, y, c)=F (f (x, y), c) (c=0,1,2) (2)
Picture is acquired from camera by FIFO memory modules, controller is then forwarded to and is pre-processed.
Step 3：Image procossing
(1) seedling case image segmentation
In addition to the picture of seedling case in the image of acquisition, some unwanted data also except seedling case, since seedling case is white
Color and rice shoot is green, rice shoot and seedling case aberration are very big, can be split to image according to the difference of color.
In view of lily rgb value is (255,255,255), tri- component sizes of linen RGB maintain an equal level, and all extremely
It is more than 200 less；The rgb value of green has apparent tendentiousness, and the values of G components is extremely much larger than the value of R and B component and the value of G components
It is more than 150 less；Inhibit R, B component by method of weighting, extract G components, specific formula is expressed as
P (x, y)=0.1*F (x, y, 0)+0.8*F (x, y, 1)+0,1*F (x, y, 2) (3)
In formula (3), F (x, y, 0), F (x, y, 1) and F (x, y, 2) correspond to R, G, B component respectively.Obtained p value is less than
170 cast out, and remaining coordinate points fit a rectangle in its surrounding, this rectangle is seedling case region.
(2) target area is extracted
The image obtained by image segmentation is handled again, each point is traversed, as the point and its point R, B of surrounding
The variance of component is larger, then it is assumed that the point is seedling piece edge, and the variance calculation formula of pixel is
Wherein,
For the pixel and its R value average values of surrounding totally 9 points；
F (x, y, 0) is the R values of the pixel；
F (x+i, y+j, 0) (i, j=± 1,0) is i distance on the directions pixel x, i range pixel on the directions y
R values；
V is the R value variances of 9 pixels of the pixel and its surrounding；
Obtained point is subjected to fitting a straight line, it is rice shoot boundary to fit the straight line come.
Step 4：The acquisition of seedling case rice shoot distribution situation and the calculating of rice shoot amount
With level to the right for zero angle, with the straight line for being 90 ° with horizontal direction angle, by six decile of seedling case, as six
Seedling lattice；By monitoring the distance in the frontier distance seedling bottom portion of the rice shoot in seedling lattice close to seedling case upper end, each seedling is obtained in real time
The quantity of lattice rice shoot.
Step 5：Alarm decision
If the rice shoot surplus of any seedling lattice is higher than the rice shoot minimum rate of accumulation of default in six seedling lattice, rice transplanter carries out just
Chang Zuoye, when the rice shoot surplus of any seedling lattice in six seedling lattice is less than the rice shoot minimum rate of accumulation of default, system passes through LED
Lamp and buzzer send sound and light alarm signal to cockpit and remote device, notify driver or the unmanned system of rice transplanter
System stops rice transplanting operation and carries out seedling, and cycle continues to check whether seedling piece is less than the minimum rate of accumulation, and after seedling box installed expires rice shoot, notice drives
Member or rice transplanter Unmanned Systems continue rice transplanting operation, and upper seedling is waited for conversely, continuing to shut down.
The present invention has advantageous effect
(1), the present invention is by acquiring seedling case image information in real time, and by image information analysis with processing to seedling case
Middle rice shoot distribution situation and rice shoot surplus are monitored on-line, and according to seedling case realize in seedling case rice shoot surplus it is real-time
Monitoring and alarm, have the advantages that intelligence degree is high, at low cost, real-time.
(2), in a kind of case to seedling of the present invention rice shoot surplus real-time monitoring system and method, usage range is wide, both may be used
The online of the step-by-step movement sold on existing market and the rice shoot amount of saddle type rice transplanter is can also be used for for automatic Pilot rice transplanter
Monitoring, it is only necessary to which slightly change is carried out to rice transplanter original structure to install and use.
Description of the drawings
Fig. 1 is rice shoot surplus real-time monitoring system structural schematic diagram in seedling case；
Fig. 2 is camera and controller motherboard attachment structure schematic diagram；
Fig. 3 is rice shoot surplus real-time monitoring system and the controller principle block diagram of method in seedling case；
Fig. 4 is the operational flow diagram of rice shoot surplus real-time monitoring system and method in seedling case；
Wherein：1- seedling casees；2- holders；3- bolts；4- guide rails；5- camera mechanism for installing；6- partition boards；7- rubber passes
Send band；8- rice shoots；9- cameras；10- controller motherboards；11- controllers, 12-FIFO memory modules；13-WIFI signals are sent out
Penetrate module；14- power voltage step down modules；15-12V DC power supplies；16-USB interfaces；17-I2C interface；18- alarm signals I/O connects
Mouthful；19-UART interfaces；20-SDI interfaces；21-SD cards；22- buttons；23-LED lamps；24- buzzers.
Specific implementation mode
Embodiment of the present invention is described in detail below in conjunction with the accompanying drawings.
As shown in Figure 1, a kind of seedling case rice shoot amount real-time monitoring system based on machine vision, mainly by seedling case 1, holder 2,
Bolt 3, guide rail 4, camera mechanism for installing 5, partition board 6, Rubber Conveyor Belt Scrap 7, rice shoot 8, camera 9 and controller motherboard
10 compositions.
The seedling case 1 is separated into 6 seedling lattice by partition board 6, and each seedling lattice are equipped with Rubber Conveyor Belt Scrap 7 close to the part of bottom；
The holder 2 is process using stainless steel material, is inShape is mounted on seedling case 1 or so two with bolt 3
Side, 1 crossbeam middle position of holder are welded with camera mechanism for installing 5, and holder 2 is located at right over seedling case 1 and makes camera 9
Face rice shoot 8 adjusts 9 lens focus of camera, makes seedling case 1 in the horizon range of camera 9；
The camera mechanism for installing 5 is welded on holder 2, camera mechanism for installing 5 use thickness for
The stainless steel plate of 2mm is process, and opens the hole there are two a diameter of 4mm thereon；
The guide rail 4 is the guide rail on existing riding type high-speed transplanter, is used to support seedling case 1 and is moved for seedling case level
It is dynamic that guiding role is provided；
The Rubber Conveyor Belt Scrap 7 is the rice shoot conveyer belt on existing riding type high-speed transplanter, and rice shoot is avoided to wrinkle；
The partition board 6 is the seedling separator for container on existing riding type high-speed transplanter, is used for seedling case according to the rice transplanting machine transplanting of rice
Seedling line number is separated into the seedling lattice of corresponding line number.
According to fig. 2, the camera shooting 9 regards S908 industry wide-angle cameras for woods cypress, and image is powered and transmitted by USB interface,
Adjusting resolution ratio of camera head is 640*480, camera 9 by the hole of two a diameter of 4mm of camera mechanism for installing 5 with
Camera mechanism for installing 5 is bolted；Controller motherboard 10 is electrically connected with camera 9；
As shown in figure 3, the controller motherboard 10 includes controller 11, FIFO memory modules 12, WIFI signal hair
Penetrate module 13, SD card 21, button 22, LED light 23, buzzer 24, power voltage step down module 14 and 12V DC power supplies 15；
The controller 11 is S3C2410 embedded microprocessors, which is integrated with many hardware, including even
It is connected to system bus controller on system bus AHB, Power Management Unit, PLL clock generator, internal SRAM, external main
Control device, LCD control units, DMA control units, interrupt control unit；The hardware component being connected on peripheral bus APB has triple channel
UART, main I more than one2C bus control units, an I2S bus control units, Five-channel PWM timers and an internal timing
Device, a WatchDog Timer, universaling I/O port, the real-time clock RTC with calendar function, 10, eight channel ADC, two it is same
Walk serial port SIO interfaces and SDI/MMC etc..Pretreatment is carried out to the image collected by the chip and effective information extracts,
Concurrently send warning message to cockpit and remote device；
The FIFO memory modules 12 are FT2232H chips, the chip have by USB2.0 interfaces and UART, FIFO,
I2Image data is transmitted to controller 11 by the C, ability that the channel interfaces such as SPI, JPAG are connected and configure by the chip；
The WIFI signal transmitting module 13 is ESP8266EX chips, the chip high height piece Embedded, including antenna
Switch, radio frequency balun, power amplifier, low-noise reception amplifier, filter and power management manage module, and warning message is logical
The chip is crossed to send；
The SD card 21 can preserve image data in case subsequent analysis by the SDI Component drivers in controller 11；
The button 22 is connected with controller 11, and the driver when alarm in cockpit can cancel report by lower button
It is alert；
The LED light 23 is connected with buzzer 24 with controller I/O mouths 18, and acousto-optic report is carried out to the driver in cockpit
It is alert；
The power voltage step down module 14 is LM2596S buck converters, by DC voltage reduction modules 14 by 12V D/C voltages
It is changed into 5V D/C voltages to power for camera 9,12V D/C voltages, which are changed into 3.3V D/C voltages, by DC voltage reduction modules 15 is
Controller 11, FIFO memory modules 12 and WIFI signal transmitting module 13 are powered；
The 12V DC power supplies 15 are rice transplanting ship battery.
According to Fig. 4, a kind of seedling case rice shoot amount method of real-time based on machine vision includes the following steps：
Step 1：The setting of job parameter
The rice shoot minimum rate of accumulation that rice shoot is alarmed is cured in program, after seedling piece is filled in rice seedling box of rice seedling transplanting machine, to control
Device powers on, and checks whether the system signal collecting unit is normal, and whether camera can persistently take pictures, and whether seedling case entirety, which is located at, is taken the photograph
As in range, if normal, system enters job state；If breaking down, failure is sent to cockpit and distant place receiving device
Information, notify operating personnel check camera whether failure, after trouble shooting, into job state.
Step 2：Image information collecting
It sets the image sampling period to 1s by controller motherboard, sets resolution ratio of camera head to 640*480, as
Plain depth is set as 8, then every image is sampled and quantization obtains the matrix of a 640*480, image coordinate f's (x, y)
Value is discrete magnitude, and the discrete magnitude of acquisition is carried out rounding, is embodied as
Since the color of each pixel is divided into tri- components of RGB, each coordinate of formula (1) includes three components, then obtains
Every photo has 640*480*3=921600 discrete data, each discrete data to be represented by
F (x, y, c)=F (f (x, y), c) (c=0,1,2) (2)
Picture is acquired from camera by FIFO memory modules, controller is then forwarded to and is pre-processed.
Step 3：Image procossing
(1) seedling case image segmentation
In addition to the picture of seedling case in the image of acquisition, some unwanted data also except seedling case, since seedling case is white
Color and rice shoot is green, rice shoot and seedling case aberration are very big, can be split to image according to the difference of color.
In view of lily rgb value is (255,255,255), tri- component sizes of linen RGB maintain an equal level, and all extremely
It is more than 200 less；The rgb value of green has apparent tendentiousness, and the values of G components is extremely much larger than the value of R and B component and the value of G components
It is more than 150 less；Inhibit R, B component by method of weighting, extract G components, specific formula is expressed as
P (x, y)=0.1*F (x, y, 0)+0.8*F (x, y, 1)+0,1*F (x, y, 2) (3)
In formula (3), F (x, y, 0), F (x, y, 1) and F (x, y, 2) correspond to R, G, B component respectively.Obtained p value is less than
170 cast out, and remaining coordinate points fit a rectangle in its surrounding, this rectangle is seedling case region.
(3) target area is extracted
The image obtained by image segmentation is handled again, each point is traversed, as the point and its point R, B of surrounding
The variance of component is larger, then it is assumed that the point is seedling piece edge, and the variance calculation formula of pixel is
Wherein,
For the pixel and its R value average values of surrounding totally 9 points；
F (x, y, 0) is the R values of the pixel；
F (x+i, y+j, 0) (i, j=± 1,0) is i distance on the directions pixel x, i range pixel on the directions y
R values；
V is the R value variances of 9 pixels of the pixel and its surrounding；
Obtained point is subjected to fitting a straight line, it is rice shoot boundary to fit the straight line come.
Step 4：The acquisition of seedling case rice shoot distribution situation and the calculating of rice shoot amount
With level to the right for zero angle, with the straight line for being 90 ° with horizontal direction angle, by six decile of seedling case, as six
Seedling lattice；By monitoring the distance in the frontier distance seedling bottom portion of the rice shoot in seedling lattice close to seedling case upper end, each seedling is obtained in real time
The quantity of lattice rice shoot.
Step 5：Alarm decision
If the rice shoot surplus of any seedling lattice is higher than the rice shoot minimum rate of accumulation of default in six seedling lattice, rice transplanter carries out just
Chang Zuoye, when the rice shoot surplus of any seedling lattice in six seedling lattice is less than the rice shoot minimum rate of accumulation of default, system passes through LED
Lamp and buzzer send sound and light alarm signal to cockpit and remote device, notify driver or the unmanned system of rice transplanter
System stops rice transplanting operation and carries out seedling, and cycle continues to check whether seedling piece is less than the minimum rate of accumulation, and after seedling box installed expires rice shoot, notice drives
Member or rice transplanter Unmanned Systems continue rice transplanting operation, and upper seedling is waited for conversely, continuing to shut down.
To sum up, a kind of seedling case rice shoot amount real-time monitoring system based on machine vision of the invention is by seedling case, holder, spiral shell
Bolt, guide rail, Rubber Conveyor Belt Scrap, camera mechanism for installing, partition board, rice shoot, camera and controller motherboard composition.It is a kind of
Seedling case rice shoot amount method of real-time based on machine vision includes the following steps：Step 1：The setting of job parameter；Step
Two：Image information collecting；Step 3：Image procossing；Step 4：The acquisition of seedling case rice shoot distribution situation and the calculating of rice shoot amount；
Step 5：Alarm decision.It is characterized in that：By acquiring real-time monitoring of the seedling case image information realization to rice shoot surplus in real time
With alarm.Under controller action, distribution situation of the rice shoot in seedling case is monitored using image acquisition units, realization pair
The real-time monitoring of rice shoot surplus is realized in conjunction with each function module to monitoring that remaining rice shoot is less than alarm line and the minimum rate of accumulation
Sound-light alarm.Rice shoot amount real-time monitoring system of the present invention and method intelligence degree are high, at low cost, simple in structure, only need pair
Existing rice transplanter, which carries out small change, to be installed and used, applied widely, be applicable to unmanned and manned rice transplanting
Seedling case rice shoot surplus carries out monitoring in real time and alarm during machine operation.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " illustrative examples ",
The description of " example ", " specific example " or " some examples " etc. means specific features described in conjunction with this embodiment or example, knot
Structure, material or feature are included at least one embodiment or example of the invention.In the present specification, to above-mentioned term
Schematic representation may not refer to the same embodiment or example.Moreover, specific features, structure, material or the spy of description
Point can be combined in any suitable manner in any one or more of the embodiments or examples.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that：Not
In the case of being detached from the principle of the present invention and objective a variety of change, modification, replacement and modification can be carried out to these embodiments, this
The range of invention is limited by claim and its equivalent.
Claims (8)
1. a kind of seedling case rice shoot amount real-time monitoring system based on machine vision, it is characterised in that：Mainly by seedling case (1), holder
(2), bolt (3), guide rail (4), camera mechanism for installing (5), partition board (6), Rubber Conveyor Belt Scrap (7), rice shoot (8), camera shooting
Head (9) and controller motherboard (10) composition；
The seedling case (1) is separated into multiple seedling lattice by partition board (6), and each seedling lattice are equipped with Rubber Conveyor Belt Scrap close to the part of bottom
(7)；The holder (2) is mounted at left and right sides of seedling case (1), and the crossbeam middle position of holder (2) is provided with camera peace
Fixed mechanism (5) is filled, the crossbeam of holder (2) is located at right over seedling case (1) and makes camera (9) face rice shoot (8), adjustment camera shooting
Head (9) lens focus, makes seedling case (1) in the horizon range of camera (9)；The guide rail (4) supports seedling case and is seedling case
Offer guiding is provided；
By online acquisition seedling case (1) image information, it is sent into controller motherboard (10), rice shoot is obtained through image processing and analysis
The rice shoot surplus in distribution situation and seedling case in seedling case (1), and by rice shoot surplus in the seedling case (1) monitored and setting
Value is compared, according to the difference of comparison result to judging whether to sound-light alarm.
2. a kind of seedling case rice shoot amount real-time monitoring system based on machine vision according to claim 1, it is characterised in that：
The holder (2) is process using stainless steel material, is inShape is mounted on seedling case (1) left and right two with bolt (3)
Side.
3. a kind of seedling case rice shoot amount real-time monitoring system based on machine vision according to claim 1, it is characterised in that：
The camera (9) is that woods cypress regards S908 industry wide-angle cameras, and image is powered and transmitted by USB interface (16), adjusts camera shooting
Head resolution ratio is 640*480, camera (9) by the hole of two a diameter of 4mm on camera mechanism for installing (5) with take the photograph
As head mechanism for installing (5) is bolted；Camera mechanism for installing (5) uses thickness for the stainless steel plate of 2mm
It is process.
4. a kind of seedling case rice shoot amount real-time monitoring system based on machine vision according to claim 1, it is characterised in that：
The controller motherboard (10) includes controller (11), FIFO memory modules (12), WIFI signal transmitting module (13), SD
Block (21), button (22), LED light (23), buzzer (24), power voltage step down module (14) and 12V DC power supplies (15)；
The controller (11) respectively with FIFO memory modules (12), WIFI signal transmitting module (13), SD card (21), button
(22), LED light (23), buzzer (24) are connected, and DC power supply (15) is respectively that FIFO stores mould by power voltage step down module (14)
Block (12), controller (11), WIFI signal transmitting module (13) power supply.
5. a kind of seedling case rice shoot amount real-time monitoring system based on machine vision according to claim 4, it is characterised in that：
The controller (11) is S3C2410 embedded microprocessors；The WIFI signal transmitting module (13) is
ESP8266EX chips；The FIFO memory modules (12) are FT2232H chips；The SD card (21) can be by the SDI in controller
Component driver；The power voltage step down module (14) is LM2596S buck converters, by DC voltage reduction modules by 12V D/C voltages
It is changed into 5V D/C voltages to power for camera (9), 12V D/C voltages, which are changed into 3.3V D/C voltages, by DC voltage reduction modules is
Controller (11), FIFO memory modules (12) and WIFI signal transmitting module (13) power supply.
6. the according to claim a kind of seedling case rice shoot amount real-time monitoring system based on machine vision, it is characterised in that：Institute
The 12V DC power supplies (15) stated are rice transplanting ship battery.
7. a kind of seedling case rice shoot amount method of real-time based on machine vision, which is characterized in that comprise the steps of：
Step 1：The setting of job parameter：The rice shoot minimum rate of accumulation that rice shoot is alarmed is cured in program, to rice seedling box of rice seedling transplanting machine
(1) it after filling seedling piece in, is powered on to controller, checks whether the system signal collecting unit is normal, and whether camera can continue
It takes pictures, whether seedling case is whole is located in image pickup scope, and if normal, system enters job state；If breaking down, to cockpit
Fault message is sent with distant place receiving device, notifies operating personnel to check whether camera (9) breaks down, trouble shooting
Afterwards, into job state；
Step 2：Image information collecting：The image sampling period is set by controller motherboard (10), and sets camera (9)
Resolution ratio, pixel depth, then every image is sampled and quantization obtains a matrix, and the value of image coordinate f (x, y) is discrete
The discrete magnitude of acquisition is carried out rounding, is embodied as by amount
Since the color of each pixel is divided into tri- components of RGB, each coordinate includes three components in formula (a), then what is obtained is every
Opening photo has multiple discrete datas, each discrete data to be represented by F (x, y, c)=F (f (x, y), c), c=0,1,2；
Picture is acquired from camera (9) by FIFO memory modules (12), controller (11) is then forwarded to and is pre-processed；
Step 3：Image procossing：Seedling case image segmentation and target area extraction；
Step 4：The acquisition of seedling case rice shoot distribution situation and the calculating of rice shoot amount：With level to the right for zero angle, with level side
The straight line for being 90 ° to angle, by (1) six decile of seedling case, as six seedling lattice；By monitoring in seedling lattice close to seedling case (1) upper end
Rice shoot (8) frontier distance seedling case (1) bottom distance, obtain the quantity of each seedling lattice rice shoot (8) in real time；
Step 5：Alarm decision：If the rice shoot surplus of any seedling lattice is higher than the rice shoot minimum rate of accumulation of default in multiple tracks seedling lattice,
Rice transplanter carries out normal operation, when the rice shoot surplus of any seedling lattice in multiple tracks seedling lattice is less than the rice shoot minimum rate of accumulation of default
When, system sends sound and light alarm signal to cockpit and remote device by LED light (23) and buzzer (24), and notice is driven
The person of sailing or rice transplanter Unmanned Systems stop rice transplanting operation and carry out seedling, and cycle continues to check whether seedling piece is less than the minimum rate of accumulation,
After seedling box installed expires rice shoot, driver or rice transplanter Unmanned Systems is notified to continue rice transplanting operation, is waited for conversely, continuing to shut down
Seedling.
8. a kind of seedling case rice shoot amount method of real-time based on machine vision according to claim 7, which is characterized in that
Step 3 detailed process is：
1) seedling case image segmentation：In addition to the picture of seedling case (1) in the image of acquisition, also some are unwanted except seedling case (1)
Data, due to seedling case (1) be white and rice shoot (8) is green, rice shoot (8) and seedling case (1) aberration are very big, according to color
Difference is split image；
In view of lily rgb value is (255,255,255), tri- component sizes of linen RGB maintain an equal level, and all at least big
In 200；The rgb value of green has apparent tendentiousness, and the value of G components is much larger than R and the value of B component and the value of G components is at least big
In 150；Inhibit R, B component by method of weighting, extract G components, specific formula is expressed as
P (x, y)=0.1*F (x, y, 0)+0.8*F (x, y, 1)+0,1*F (x, y, 2)
In formula, F (x, y, 0), F (x, y, 1) and F (x, y, 2) correspond to R, G, B component respectively, and obtained p value is less than to 170 house
It goes, remaining coordinate points fit a rectangle in its surrounding, this rectangle is seedling case (1) region；
2) target area is extracted：To again be handled by the image that image segmentation obtains, and traverse each point, when the point and its
The point R of surrounding, the variance of B component are larger, then it is assumed that the point is seedling piece edge, and the variance calculation formula of pixel is
Wherein,For the pixel and its R value average values of surrounding totally 9 points；F (x, y, 0) is the R values of the pixel；
F (x+i, y+j, 0) (i, j=± 1,0) is i distance on the directions pixel x, the R values of i range pixel on the directions y；
V is the R value variances of 9 pixels of the pixel and its surrounding；Obtained point is subjected to fitting a straight line, fitting the straight line come is
Rice shoot boundary.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810463986.4A CN108633411B (en) | 2018-05-15 | 2018-05-15 | Machine vision-based seedling box seedling quantity real-time monitoring system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810463986.4A CN108633411B (en) | 2018-05-15 | 2018-05-15 | Machine vision-based seedling box seedling quantity real-time monitoring system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108633411A true CN108633411A (en) | 2018-10-12 |
| CN108633411B CN108633411B (en) | 2020-07-31 |
Family
ID=63755930
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810463986.4A Active CN108633411B (en) | 2018-05-15 | 2018-05-15 | Machine vision-based seedling box seedling quantity real-time monitoring system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108633411B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110334750A (en) * | 2019-06-21 | 2019-10-15 | 西安工程大学 | Iron tower of power transmission line bolt corrosion degree image classification recognition methods |
| CN110809937A (en) * | 2019-12-03 | 2020-02-21 | 中国农业大学 | Method for quantitatively analyzing growth characteristics of root plants based on delayed photography technology |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101180928A (en) * | 2007-12-21 | 2008-05-21 | 浙江大学 | Seedling replanting system based on machine vision |
| CN103477770A (en) * | 2013-09-24 | 2014-01-01 | 浙江理工大学 | Automatic complementary planting device based on vision for seedlings for bowl plate |
| CN104322187A (en) * | 2014-10-23 | 2015-02-04 | 江苏大学 | Plug seedling image acquiring device and method |
| CN104704970A (en) * | 2015-03-26 | 2015-06-17 | 江苏大学 | Full-automatic plug seedling transplanting device and seedling selecting method thereof |
| CN204466226U (en) * | 2015-02-01 | 2015-07-15 | 浙江大学 | One is filled the gaps with seedlings mechanical work device |
| CN107464232A (en) * | 2017-07-13 | 2017-12-12 | 江苏大学 | A kind of unmanned rice transplanter plants the image detecting method of quality |
-
2018
- 2018-05-15 CN CN201810463986.4A patent/CN108633411B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101180928A (en) * | 2007-12-21 | 2008-05-21 | 浙江大学 | Seedling replanting system based on machine vision |
| CN103477770A (en) * | 2013-09-24 | 2014-01-01 | 浙江理工大学 | Automatic complementary planting device based on vision for seedlings for bowl plate |
| CN104322187A (en) * | 2014-10-23 | 2015-02-04 | 江苏大学 | Plug seedling image acquiring device and method |
| CN204466226U (en) * | 2015-02-01 | 2015-07-15 | 浙江大学 | One is filled the gaps with seedlings mechanical work device |
| CN104704970A (en) * | 2015-03-26 | 2015-06-17 | 江苏大学 | Full-automatic plug seedling transplanting device and seedling selecting method thereof |
| CN107464232A (en) * | 2017-07-13 | 2017-12-12 | 江苏大学 | A kind of unmanned rice transplanter plants the image detecting method of quality |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110334750A (en) * | 2019-06-21 | 2019-10-15 | 西安工程大学 | Iron tower of power transmission line bolt corrosion degree image classification recognition methods |
| CN110334750B (en) * | 2019-06-21 | 2021-06-08 | 西安工程大学 | Power transmission line iron tower bolt corrosion degree image classification and identification method |
| CN110809937A (en) * | 2019-12-03 | 2020-02-21 | 中国农业大学 | Method for quantitatively analyzing growth characteristics of root plants based on delayed photography technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108633411B (en) | 2020-07-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106774427B (en) | Unmanned aerial vehicle-based water area automatic inspection system and method | |
| CN106886225B (en) | Multifunctional unmanned aerial vehicle intelligent landing station system | |
| CN102136191B (en) | Multi-scene full-function all-in-one machine capable of capturing and monitoring | |
| CN110297498A (en) | A kind of rail polling method and system based on wireless charging unmanned plane | |
| CN107200130B (en) | Intelligent pesticide spraying aircraft and method | |
| CN105306796A (en) | Night vision equipment with regular infrared illumination function and global shutter CMOS (Complementary Metal Oxide Semiconductor) sensor | |
| DE212017000101U1 (en) | Camera-based headlight control system | |
| CN108633411A (en) | A kind of seedling case rice shoot amount real-time monitoring system and method based on machine vision | |
| EP3809808A1 (en) | Device of intelligent sensing system | |
| CN105957358A (en) | Intelligent traffic monitoring adjustment system | |
| CN110021158A (en) | A kind of camera shooting meter system and method based on deep learning | |
| CN106353717A (en) | Mining scraper intelligent positioning terminal | |
| CN208092975U (en) | The system for realizing semiclosed parking lot management based on long short focus camera | |
| CN203255039U (en) | Novel taxi intelligent screen display device | |
| CN110246347A (en) | A kind of intelligent traffic control system based on NB-IOT | |
| CN108510744A (en) | The system and method for semiclosed parking lot management is realized based on long short focus camera | |
| CN206077579U (en) | Hang automatic case number (CN) Train number recognition and casing check-fault system in a kind of bank | |
| CN204948211U (en) | A kind of airborne photographic equipment of taking photo by plane | |
| CN206075563U (en) | A kind of residential communities ground parking stall guide based on video image | |
| CN204406667U (en) | A kind of Intelligent license-plate of vehicle supervising device for parking lot | |
| CN209070097U (en) | A kind of generalization radar screen monitoring display system | |
| CN208384806U (en) | A kind of tracking floating electronic police | |
| CN208292520U (en) | Harbour gantry crane suspender based on machine vision it is automatic case apparatus | |
| JPS6117938A (en) | Flock monitor device | |
| CN207601491U (en) | A kind of energy-conserving intelligent projector of adjust automatically brightness |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |