DE102011120858A1 - Method and device for contactless determination of plant parameters and for processing this information - Google Patents

Method and device for contactless determination of plant parameters and for processing this information

Info

Publication number
DE102011120858A1
DE102011120858A1 DE102011120858A DE102011120858A DE102011120858A1 DE 102011120858 A1 DE102011120858 A1 DE 102011120858A1 DE 102011120858 A DE102011120858 A DE 102011120858A DE 102011120858 A DE102011120858 A DE 102011120858A DE 102011120858 A1 DE102011120858 A1 DE 102011120858A1
Authority
DE
Germany
Prior art keywords
image
plant
stock
stereo camera
images
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.)
Withdrawn
Application number
DE102011120858A
Other languages
German (de)
Inventor
Stefan Reusch
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yara International ASA
Original Assignee
Yara International ASA
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Yara International ASA filed Critical Yara International ASA
Priority to DE102011120858A priority Critical patent/DE102011120858A1/en
Publication of DE102011120858A1 publication Critical patent/DE102011120858A1/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • AHUMAN NECESSITIES
    • A01AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01MCATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
    • A01M7/00Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
    • A01M7/0089Regulating or controlling systems
    • AHUMAN NECESSITIES
    • A01AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01CPLANTING; SOWING; FERTILISING
    • A01C21/00Methods of fertilizing, sowing or planting
    • A01C21/007Determining fertilization requirements
    • AHUMAN NECESSITIES
    • A01AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01GHORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
    • A01G7/00Botany in general
    • A01G7/06Treatment of growing trees or plants, e.g. for preventing decay of wood, for tingeing flowers or wood, for prolonging the life of plants
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/04Analysing solids
    • G01N29/11Analysing solids by measuring attenuation of acoustic waves
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/22Details, e.g. general constructional or apparatus details
    • G01N29/30Arrangements for calibrating or comparing, e.g. with standard objects
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0012Biomedical image inspection
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/50Depth or shape recovery
    • G06T7/55Depth or shape recovery from multiple images
    • G06T7/593Depth or shape recovery from multiple images from stereo images
    • AHUMAN NECESSITIES
    • A01AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01GHORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
    • A01G7/00Botany in general
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C11/00Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
    • G01C11/04Interpretation of pictures
    • G01C11/06Interpretation of pictures by comparison of two or more pictures of the same area
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2291/00Indexing codes associated with group G01N29/00
    • G01N2291/02Indexing codes associated with the analysed material
    • G01N2291/024Mixtures
    • G01N2291/02466Biological material, e.g. blood
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2291/00Indexing codes associated with group G01N29/00
    • G01N2291/04Wave modes and trajectories
    • G01N2291/044Internal reflections (echoes), e.g. on walls or defects
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10004Still image; Photographic image
    • G06T2207/10012Stereo images
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30004Biomedical image processing
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30181Earth observation
    • G06T2207/30188Vegetation; Agriculture

Abstract

The invention relates to a method and a device for contactless determination of plant parameters of a plant population and for processing this information into a control quantity for fertilizing, watering and / or crop protection of the stock, in which a part of the plant population composed of a pixel digital image with recorded at least one image recording system and the image of the current plant parameters are determined by an image analysis. The invention has for its object to provide a method and apparatus for contactless determination of plant parameters of a plant stand and for processing this information into a control variable for fertilizing, watering the stock and / or applying pesticides to the stock to improve so that the Measuring accuracy for the plant parameters with simultaneous reduction of costs further increased and a significantly simplified handling for the farmer is achieved. This object is achieved in that by means of the image recording system simultaneously in slightly offset perspective in a vertical position to the plant stock at least one pair of images is taken, which is then subjected to the following steps: a) to bring the two images in such a way that an imaged object seemingly displaced in the horizontal direction only, the object being adjusted to exactly the same pixel position in the vertical direction on one and the other image, b) determining a disparity map which for each pixel shows the apparent disparity d of the object in the pixel contains an image compared to the other image, c) deriving a distance map from the disparity map containing the distance z between the image acquisition system and the object for each pixel, d) determining the plant parameters from the distance map, and e) converting the plant parameters determined in step d) into a control quantity for fertilising, watering and / or crop protection.

Description

  • The invention relates to a method for the non-contact determination of plant parameters of a plant population and for processing this information into a control quantity for fertilizing, watering and / or crop protection of the stock, in which a part of the plant population comprises a digital image composed of pixels with at least one image acquisition system recorded and from the image the current plant parameters are determined by an image analysis.
  • The invention further relates to a device for non-contact determination of current plant parameters of a plant population and for processing this information into a control variable for fertilizing, watering the stock and / or applying pesticides to the crop, with an image acquisition and evaluation system for image analysis, an image evaluation algorithm for image analysis, which is stored in the memory of a portable, local or central computer and is connected to the image recording system for transmitting the images, is provided as image evaluation system.
  • State of the art
  • From the DE 10 2005 050 302 A1 is a method for contactless determination of the current nutritional status of a crop stand and for processing this information, taking into account other parameters such as crop and / or variety and / or developmental stage and / or yield target in a fertilizer recommendation known in which of a part of the plant population at least one digital image recorded by means of an image recording system in at least two spectral channels, from the image of the current nutritional status determined by an image analysis and from the latter, the fertilizer recommendation is derived. This known prior art has the disadvantage that the method provides only a two-dimensional projection of the stock, in which height information is lost as well as the vertical distribution of the biomass and the position of the leaves, so that the results obtained have only a limited significance.
  • In the DE 10 2006 009 753 B3 Furthermore, a method for non-contact determination of the biomass and morphological parameters of plant stands is described, in which acted on the plants of the stock with an emitted from an ultrasonic source mounted on a mobile carrier sound field during the crossing acted by the plants and the ground reflected sound echoes by a detected on the carrier receiver and passed from this after conversion into digital signals to an evaluation and signal processing unit that currently evaluates the signals stored on a disk and displays on a monitor, optionally in an electronically controlled Ausbringeinheit the signals to control commands for Application of product agents are processed.
  • Although this prior art makes it possible to directly determine morphological parameters such as the number of leaf layers, the leaf position and the vertical distribution of the biomass, the known method requires specially adapted hardware.
  • From the DE 103 29 472 A1 Furthermore, a device for measuring the crop density, in particular the green masses of the plant population of an agricultural crop density for controlling and / or regulating an agricultural distribution machine is known, wherein the device has at least one signal to an on-board computer-supplying sensor with a transmitter and a receiver. The sensor is designed as a triangulation sensor. By means of the sensor, the inventory level of the plant population and from the signals by means of a stored in the on-board computer in a storage medium evaluation program, the crop density is determined. This known method works with a laser, which is focused on the measurement object. The disadvantage here is the very small point or line-shaped measuring field. A random, caused for example by drill errors streaking of the stock in the direction of travel can already lead here to unrepresentative measurement results.
  • The DE 40 42 672 C2 discloses an on-board computer system for agricultural machine and / or equipment combinations, consisting of a farm tractor and coupled to this and designed as fertilizer spreader and / or sprayer agricultural distribution machines, via the on-board computer system of the tractor and / or the coupled distribution over stored in the on-board computer system and / or entered setting values can be adjusted, controlled and / or regulated, wherein the on-board computer system has at least one sensor element cooperating with the on-board computer, which provides information about the area to be spread. The sensor element is designed as a non-contact sensor and currently detects information about the nature and condition of the plant growth by contactless scanning of the plants located on the area to be spread. This information is transmitted to the on-board computer, which currently determines the nutrient requirement and / or the nutrient supply of the plants by means of a stored evaluation program, wherein due to the nutrient requirement thus determined, the direct control of the metering is done by the on-board computer. The sensors used may be ultrasonic sensors or infrared sensors.
  • Although this known prior art describes a vehicle-based measuring system, however, the sensor attached to the farm tractor is guided at a small distance over the stock, so that such a system leads to erroneous results due to the small scanning surface.
  • task
  • In this prior art, the present invention seeks to improve the method for non-contact determination of plant parameters of a plant stand and for processing this information into a control variable for fertilizing, watering the stock and / or the application of pesticides to the stock so that the measurement accuracy for the plant parameters is further increased while at the same time reducing the costs and that the farmer is much more easily managed.
  • This object is achieved by a method of the type mentioned with the features of claim 1 and by a device having the features of claim 13.
  • Advantageous embodiments of the method and the device are the dependent claims.
  • The solution according to the invention is based on the knowledge that at least one image pair is simultaneously recorded in a slightly offset perspective in a vertical position relative to the plant population by means of the image recording system, which is then subjected to the following working steps:
    • a) to bring the two images in such a way that a mapped object is moved apparently only in the horizontal direction, wherein the object occupies exactly the same pixel position in the vertical direction on one and the other image,
    • b) determining a disparity map containing, for each pixel, the apparent displacement (disparity) d of the object in the one image compared to the other image,
    • c) deriving a distance map from the disparity map containing the distance z between the image acquisition system and the object for each pixel;
    • d) determining the plant parameters from the distance map and
    • e) converting the plant parameters determined in step d) into a control quantity for fertilizing, watering and / or crop protection.
  • In a further embodiment of the method according to the invention, to determine the disparity, a window formed by a small number of pixels is drawn around every nth pixel in the one image, the window in the other image being shifted horizontally by as many pixels as before a maximum match of the two window contents is achieved. This pixel number then corresponds to the disparity d (x, y).
  • A prerequisite for deriving the distance map from the disparity map is a calibration of the disparity map, preferably on a reference object with at least two different known distances or from the known camera parameters such as the focal point and the distance of the focal points.
  • The inventive method makes it possible in a very simple manner to determine from the distance map, the stock level of the stock and / or the soil cover and / or the vertical distribution of the biomass and / or the blade angle position and from a corresponding fertilizer and / or crop protection and / or to derive irrigation recommendation as the tax quantity for the stock.
  • It is particularly advantageous for the farmer if a hand-held mobile phone or smartphone with integrated stereo camera is used as the image recording system, the recorded images directly by a stored on the smartphone image analysis algorithm query of appropriate parameters such as crop and / or variety and / or development stage and / or yield target for fertilizer and / or crop protection and / or irrigation recommendation processed and displayed.
  • In another preferred embodiment of the method according to the invention, a hand-held mobile phone or smartphone with integrated stereo camera and Internet access is used as the image recording system, the recorded images being queried by appropriate parameters such as crop type and / or variety and / or development stage and / or yield target for fertilizer. and / or crop protection and / or irrigation recommendation to be sent to a central server, which processes the information to a fertilizer and / or crop protection and / or irrigation recommendation and sends this back to display on the smartphone.
  • The farmer is thus able to obtain with simple means up-to-date measurement results on the state of his crops and to initiate appropriate measures for fertilization and / or watering and / or for the application of pesticides.
  • As an image recording system, however, a commercially available hand-held stereo camera can be used, which stores the images in a camera-internal or external memory, of which the images after querying appropriate parameters such as crop and / or variety and / or development stage and / or income target directly to a local computer or transmitted via an Internet portal to the central server for evaluation.
  • In a preferred embodiment of the method according to the invention, at least one stationary camera mounted on a mast vertically above the crop is used as the image acquisition system with internet-enabled radio module which automatically picks up the stereo images of the crop in a preselectable time interval and sends them to the central server ( 8th ), which analyzes the stereo images, processes the results on fertilizer and / or crop protection and / or irrigation recommendations, and makes them available to the farmer via an internet-based platform. Preferably, at least two stationary stereo cameras can be used, of which one stereo camera monitors an optimally-guided reference stock and the other monitors a target stock so that the reference stock can serve to derive fertilizer, crop protection and irrigation measures on the target stock.
  • A further preferred embodiment of the method according to the invention provides that the image recording system used is at least one stationary camera mounted on a mast vertically above the plant stand with internet-capable radio module, which automatically records stereo images of the plant population in a preselectable time interval Server sends, which evaluates the images in terms of leaf position and the evaluation results as a control variable to an irrigation or liquid fertilizer system for discharging the required amount of water forwards.
  • In this case, the method according to the invention utilizes the knowledge that the plant leaves curl up or hang down under drought stress, as a result of which the blade angle or the blade angle distribution function changes. From the daily flow of measured data, it is possible, in conjunction with climate data such as temperature, dew point, irradiation, to determine and apply the required amount of water for watering.
  • In a preferred embodiment of the method according to the invention, at least one stereo camera mounted on a mobile carrier vehicle with processor unit is used as the image acquisition system, which continuously generates stereo images during the passage from the plant population, from which the processor unit continuously determines the biomass and with this size the output quantity of growth regulator the stock is controlled.
  • The device according to the invention is characterized in that the image recording system is a commercially available stereo camera integrated in an internet-capable mobile phone or smartphone or a stereo camera connected to the local or central computer, wherein either the smartphone or the local computer or central server uses the image evaluation algorithm Contains image analysis.
  • The device according to the invention can be applied in a simple manner in various ways, for example as manually operable by the farmer smartphone with built-in stereo camera, separately communicating with a smartphone stereo camera or stereo camera on a short boom next to the vehicle carrier, directly on the tractor roof or can be mounted on a standing mast with little effort.
  • Further advantages and details will become apparent from the following description with reference to the accompanying drawings.
  • embodiments
  • The invention will be explained in more detail below with reference to four exemplary embodiments.
  • Show it
  • 1a a schematic representation of the image recording and -auswertystems with integrated into an internet-enabled smartphone stereo camera,
  • 1b A typical image and evaluation system with an external image memory having stereo camera and local / central server,
  • 2 a schematic representation of the method according to the invention,
  • 3 an example of a stereo image for a rapeseed beat,
  • 4a and 4b an example of determining a Disparitätskarte and deriving a distance map from the stereo image according to 3 .
  • 5 one from the distance map after 4b derived histogram for the determination of the vertical biomass distribution in an unfertilized and optimally fertilized rapeseed stock,
  • 6 an example of the correlation between plant height and biomass determined by the method according to the invention for different fertilizer inputs and growth stages,
  • 7 a schematic representation of the device according to the invention on a stationary mast and
  • 8th a schematic representation of the device according to the invention on a mobile carrier.
  • The inventive method for the contactless determination of plant parameters of a plant population 4 and for processing this information into a control quantity for the fertilization, watering and / or plant protection of the stock should be applied manually on a rape cultivated field.
  • The farmer is putting - how 1a shows - for the manual process according to the invention as an image recording system a color stereo camera 1 in an internet enabled smartphone 3 is integrated, and as image evaluation system on in the memory 2 of the smartphone 3 implemented image evaluation algorithm for performing a 3D image analysis of color stereo images.
  • The from the stereo camera 1 taken pictures 5.1 and 5.2 be directly in the store 2 of the smartphone 3 by the image evaluation algorithm after retrieval of stock-specific data such as crop type and / or variety and / or development stage and / or yield target of the stock 5 processed to a fertilizer and / or crop protection and / or irrigation recommendation and the result on the display of the smartphone 3 brought to the display. The farmer can then, if necessary, immediately appropriate measures for an optimal supply of plant stock 4 take. But it is also possible to leave without the invention, the digital images 5.1 and 5.2 from the smartphone 3 via Internet I to a local computer 7 or central server 8th to send. This calculator 7 or server 8th , who has the image evaluation algorithm, processes the fertilizer and / or crop protection and / or irrigation recommendations, including the stock-specific data previously requested by the farmer, and sends them to the smartphone 3 back to the display.
  • However, the execution of the method according to the invention is not exclusive to a smartphone 3 but can also - like 1b shows - with a standard stereo camera 1 be performed.
  • The from the stereo camera 1 taken pictures 5.1 and 5.2 are preferably in a portable external memory 6 the stereo camera 1 and then this by the farmer in the equipped with the image evaluation algorithm local computer 7 loaded. The pictures 5.1 and 5.2 be in the local machine 7 processed to a fertilizer and / or crop protection and / or irrigation recommendation and the farmer on the display of the computer 7 so that the farmer can take appropriate action. Alternatively, it is also possible, the pictures 5.1. and 5.2 from the local computer to the central server 8th to send the image analysis and the result to the local computer 7 returns where it is displayed.
  • The color stereo camera 1 is about 1 to 3 m vertically above the plant 4 positioned and the plant stand 4 from the camera through at least one image pair (pic 5.1 and 5.2 ), preferably several image pairs (see 3 ) recorded simultaneously.
  • The stereo camera 1 has a resolution of z. B. 1.3 megapixels.
  • The inventive method is running, as 2 shows as follows.
  • Before the actual image analysis, camera-specific corrections to the double image must first be made in a pre-processing or calibration step S1.
  • The two images are made to coincide in such a way that a mapped object is apparently shifted only in the horizontal direction, wherein the object in the vertical direction on one and the other image exactly the same pixel position (x, y) occupies. For a given color stereo camera with given settings, the corrections to be made are always the same, so a one-time calibration step before the first use is sufficient.
  • In operation S2, a disparity map is created which for each pixel contains the apparent displacement (disparity) d (x, y) in one image compared to the other image.
  • The disparity d (x, y) can be determined by defining a small window, for example of a size of 7 × 7 pixels, around the relevant pixel of one image and moving this window horizontally in the other image until there is a maximum match of both window contents. This shift is detected as disparity d (x, y).
  • Since the focal points associated with the image planes of the color stereo camera are spatially separated from each other by a fixed amount b and the focal length f of the two lenses of the color stereo camera is the same, the distance z = b × f / d (x, y) so that the distance z to an object can be determined by measuring the disparity in the stereo image. The disparity map is thus a depth image. Alternatively, once a calibration of the camera system to two different distant objects with a known distance can be done. Then, the distance z between the image pickup system and the object generally increases z (x, y) = p / (d (x, y) + q), where p and q are camera-specific constants resulting from the calibration image and d (x, y) is the disparity.
  • Such a distance map derived according to step S3 shows 4b ,
  • Determining the disparity card can be quite time-consuming even with optimal implementation. This expenditure of time can be reduced if the disparity d (x, y) is determined for every nth pixel in the x and y direction instead of for each individual pixel of the output image. The prerequisite for this, however, is that no high spatial resolution is required or only the statistical distribution of the distances z (x, y) is of interest.
  • From the distance map obtained in step S3, the plant parameters such as the stock level, the soil cover level, the biomass and their vertical distribution and the leaf position can be derived in step S4 and as control variables for fertilizing, watering and / or spreading pesticides in step S5 use.
  • The stock height can at known distance z o the color stereo camera 1 from the ground simply by determining the smallest measured distance min (z) h = z o - min (z) respectively.
  • In order to exclude individual erroneously determined distance values, it has proven to be expedient, for example, to use a 5% or 10% percentile instead of the minimum.
  • If the distance between color stereo camera and ground is not known or is subject to fluctuations, for example when using a mobile carrier, then the stock height h can also be the difference between the highest and lowest distance value or, for example, from the 95% and 5% percentile determine. However, such a procedure presupposes that the plant population does not completely cover the soil and that the soil on the image area is still recognizable in some percent and also sufficiently illuminated. For many crops, the stock level is directly related to above-ground biomass. An example of the relationship between the inventively derived stock level and the actual biomass in a differently fertilized and measured at different times rape stock shows 6 ,
  • Soil coverage, ie the proportion of soil covered by parts of plants or leaves, can be determined simply by determining the number of pixels to which the condition applies: z (x, y) <z o , where z o is the distance of the color stereo camera 1 to the ground is. The prerequisite is that the lowest floor of the plant leaves does not rest directly on the ground and thus the distance is distinguishable.
  • In contrast to the determination of the soil cover level by a normal counting of the green pixels in a two-dimensional color image, the method according to the invention has the advantage that shallow weeds or ground algae and mosses do not distort the result. In addition, brown or discolored leaves and plant parts are reliably detected, since the process of the invention, the texture and not the color is evaluated.
  • The relative vertical distribution of the biomass results from the distribution of the distance values z (x, y) in a histogram. For this purpose, the statistical frequency of the individual distance values z (x, y) in the images is determined, so that the form of the distribution can be used to deduce the vertical distribution of the biomass.
  • An example of such a histogram shows 5 , which shows the frequency of pixel distances in unfertilized (N0) and optimally fertilized (N9) winter rape stock.
  • From the knowledge of the position of the leaves in three-dimensional space, the mean leaf position or a distribution function of the leaf positions can be derived. For this purpose, one level is adapted to the number of leaf pixels for each small leaf section and the angle between the normal vector of the plane and the vertical is determined. The procedure for determining this angle belongs to the general state of the art and therefore need not be described in detail.
  • example 1
  • The inventive method is intended in a portable device according to 1 be used. For the image capture is an internet enabled smartphone 3 with integrated commercial color stereo camera 1 used. After taking the pictures 5.1 and 5.2 These are after requesting further required for the evaluation parameters such. B. the type of fruit, the variety, the development status, etc. of the smartphone 3 via Internet I to a central server 8th sent on which the image evaluation algorithm is installed.
  • The image evaluation algorithm evaluates the images 5.1 and 5.2 determines the biomass and / or plant height and derives therefrom a fertilizer and / or plant protection recommendation for the stock. This fertilizer and / or crop protection recommendation is provided by the server 8th to the smartphone 3 returned via Internet I and displayed there for the user. Alternatively, the image evaluation algorithm can also be directly on the smartphone 3 installed, which when activated by the user asks them to record, for example, several color stereo images with the color stereo camera. After entering the information characteristic of the stock, the pictures are taken directly on the smartphone 3 evaluated by the image evaluation algorithm, determines the biomass and / or plant height, derived therefrom a fertilizer recommendation / crop protection recommendation and this displayed to the user.
  • Example 2
  • The method according to the invention is used for stationary monitoring of a plant population 4 used (see 7 ). The color stereo camera 1 with radio module 13 is in a weatherproof housing so on one in the plant stock 4 standing mast 9 that they are planted 4 can pick up vertically from above. This measuring station remains stationary at one point in the plant population throughout the growing season 4 ,
  • The color stereo camera 1 automatically takes pictures at an adjustable time interval, for example once a day, and sends them via mobile phone via the Internet directly to the central server 8th , The color stereo camera 1 and the radio module 13 be powered autonomously via a solar panel and possibly a battery backup with electricity.
  • On the central server 8th the sent pictures are with the on the central server 8th stored image evaluation algorithm processed and determines the plant parameters. The evaluation results with derived fertilization, irrigation and / or crop protection recommendations are made available to the user via an internet-based platform. The farmer is thus able to use his computer in the office to conveniently track the state of his crop and take appropriate action if the growth does not meet the expected value.
  • Optionally, several similar measuring stations can be used to simultaneously monitor several stocks or a stock at several locations.
  • Example 3
  • The inventive method is used as described in Example 2, but is directly connected to an irrigation system, the 7 shown in dashed lines and the server 8th receives its control commands. From the stereo images of the average blade angle is derived and used as a control variable for the irrigation system. This exploits the fact that the plant leaves curl up or hang down under drought stress, thereby changing the blade angle or the blade angle distribution function. The amount of water to be applied is determined from the daily course of the measurement data, possibly in conjunction with climate data such as temperature, dew point, sunshine duration, etc., and then applied to the stock accordingly.
  • Such an irrigation system is preferably suitable for use in the greenhouse in order to exclude the effects of wind movement.
  • Example 4
  • The process according to the invention is used for the partial area-specific application of growth regulators (see 8th ).
  • At least one color stereo camera 1 , but optionally also several color stereo cameras, is / are on a boom 10 a carrier vehicle 11 attached and capture the plant population 4 during the crossing. With the color stereo cameras 1 be continuous pictures 5.1 . 5.2 taken up by an aboard the carrier vehicle 11 located processor unit 12 processed and continuously determines the biomass. With the help of this size, the quantity of growth regulator is then continuously varied.
  • LIST OF REFERENCE NUMBERS
  • 1
    stereo camera
    2
    Memory in 3
    3
    Mobile phone / smartphone
    4
    plant stand
    5.1, 5.2
    stereo pictures
    6
    Memory in 1
    7
    Local calculator
    8th
    Central server
    9
    mast
    10
    boom
    11
    carrier vehicle
    12
    Processor unit in 11
    13
    radio module
    b
    Distance lenses
    I
    Internet
    S1, S2, S3, S4, S5
    steps
  • QUOTES INCLUDE IN THE DESCRIPTION
  • This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.
  • Cited patent literature
    • DE 102005050302 A1 [0003]
    • DE 102006009753 B3 [0004]
    • DE 10329472 A1 [0006]
    • DE 4042672 C2 [0007]

Claims (16)

  1. A method for the contactless determination of plant parameters of a plant population and for processing this information into a control quantity for fertilizing, watering and / or crop protection of the stock, in which a part of the plant population is a pixelized digital image taken with at least one image recording system and from the Image, the current plant parameters are determined by an image analysis, characterized in that at least one pair of images is simultaneously recorded by means of the image recording system in slightly offset perspective in a vertical position to the plant population, which is then subjected to the following steps: a) bring the two images in the way that an imaged object is apparently displaced only in the horizontal direction, wherein the object in the vertical direction on the one and the other image is set to exactly the same pixel position w b) determining a disparity map containing, for each pixel, the apparent displacement (disparity) d of the object in the one image compared to the other image, c) deriving a distance map containing the distance z between the image acquisition system and the object for each pixel Disparity map, d) determining the plant parameters from the distance map and e) converting the plant parameters determined in step d) into a control quantity for fertilising, watering and / or crop protection.
  2. A method according to claim 1, characterized in that, for determining the disparity, a window formed of a small number of pixels is drawn around each or every nth pixel in the one image, the window in the other image being shifted horizontally until a maximum match of the two window contents is achieved.
  3. A method according to claim 1, characterized in that the disparity card is previously calibrated once at a reference object with at least two different known distances.
  4. A method according to claim 1, characterized in that from the distance map, the stock level of the stock and / or the soil cover level and / or the vertical distribution of the biomass and / or the blade angle position is determined from which the fertilizer and / or crop protection and / or irrigation recommendation can be derived.
  5. A method according to claim 1, characterized in that as image recording system for the plant population, a hand-held mobile phone or smartphone with integrated stereo camera ( 1 ), wherein the captured images are processed and displayed directly by an image evaluation algorithm stored on the smartphone while interrogating corresponding parameters such as crop and / or variety and / or development stage and / or yield target.
  6. A method according to claim 1, characterized in that as image recording system for the plant population, a hand-held mobile phone or smartphone with integrated stereo camera ( 1 ) and accessing the Internet, wherein the recorded images are sent to a central server (interrogation of corresponding parameters such as crop type and / or variety and / or development stage and / or yield target). 8th ), which processes the information on a fertilizer and / or crop protection and / or irrigation recommendation and sends it back for display on the smartphone.
  7. A method according to claim 1, characterized in that the image recording system is a commercially available hand-held stereo camera ( 1 ), which stores the images in an in-camera or external memory, from which the images after query parameters such as crop and / or variety and / or development stage and / or income target directly to a local computer or via an Internet portal to the central server be transmitted for evaluation.
  8. A method according to claim 1, characterized in that as an image recording system at least one stationary during the growing season, on a mast ( 9 ) vertically above the crop ( 4 ) mounted stereo camera ( 1 ) is used with radio module and Internet access, which automatically generates stereo images of the plant 4 ) in a preselectable time interval and send this to the central server ( 8th ), which analyzes the stereo images, processes the results on fertilizer and / or crop protection and / or irrigation recommendations, and makes them available to the farmer via an internet-based platform.
  9. Method according to claim 8, characterized in that at least two stationary cameras fixed over the stock ( 1 ), of which one stereo camera monitors an optimally guided reference stock and the other monitors a target stock.
  10. A method according to claim 1, characterized in that at least one stationary during the growing season, as an image recording system a mast ( 9 ) vertically above the crop ( 4 ) mounted stereo camera ( 1 ) is used with radio module and Internet access, which automatically generates stereo images of the plant 4 ) in a preselectable time interval, this to the central server ( 8th ) which evaluates the images in terms of leaf position and forwards the evaluation results as a control quantity to an irrigation system for discharging the required amount of water.
  11. A method according to claim 1, characterized in that a hand-held stereo camera is used as the image recording system, with the first reference stock and then at least one target stock is measured.
  12. A method according to claim 1, characterized in that as an image recording system at least one on a mobile carrier vehicle ( 11 ) with processor unit ( 12 ) mounted stereo camera ( 1 ) used during the crossing of the plant stock ( 4 ) continuously generates stereo images from which the processor unit ( 12 ) determines the biomass continuously and with this size the quantity of growth regulator applied to the stock is controlled.
  13. Apparatus for the contactless determination of current plant parameters of a plant stock and for processing this information into a control variable for fertilizing, watering the stock and / or applying plant protection products to the plant stock, with an image recording and evaluation system for image analysis according to claim 1, wherein as image evaluation system one in the memory ( 2 ) of a portable, local or central computer ( 7 . 8th ) is provided for image analysis, which is connected to the image recording system for transmitting the images, characterized in that the image recording system in an internet-enabled mobile phone or smartphone ( 3 ) integrated commercial stereo camera ( 1 ) or one with the local or central computer ( 7 . 8th ) connected stereo camera ( 1 ), whereby either the smartphone ( 3 ) or the local computer ( 7 ) or central server ( 8th ) contains the image evaluation algorithm for image analysis.
  14. Device according to claim 13, characterized in that the stereo camera ( 1 ) stationary on a mast ( 9 ) is mounted in a vertical position to the plant population protected against water and with a radio module ( 13 ) is connected to the internet.
  15. Device according to claim 13, characterized in that the stereo camera ( 1 ) on a carrier vehicle ( 11 ) in a vertical position over the plant stock ( 4 ) is arranged guided.
  16. Device according to claim 13, characterized in that the stereo camera ( 1 ) is hand-carried in a vertical position directed to the plant population.
DE102011120858A 2011-12-13 2011-12-13 Method and device for contactless determination of plant parameters and for processing this information Withdrawn DE102011120858A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
DE102011120858A DE102011120858A1 (en) 2011-12-13 2011-12-13 Method and device for contactless determination of plant parameters and for processing this information

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE102011120858A DE102011120858A1 (en) 2011-12-13 2011-12-13 Method and device for contactless determination of plant parameters and for processing this information
PCT/DE2012/001150 WO2013087052A1 (en) 2011-12-13 2012-12-04 Method and apparatus for contactlessly determining plant parameters and for processing this information

Publications (1)

Publication Number Publication Date
DE102011120858A1 true DE102011120858A1 (en) 2013-06-13

Family

ID=47832834

Family Applications (1)

Application Number Title Priority Date Filing Date
DE102011120858A Withdrawn DE102011120858A1 (en) 2011-12-13 2011-12-13 Method and device for contactless determination of plant parameters and for processing this information

Country Status (2)

Country Link
DE (1) DE102011120858A1 (en)
WO (1) WO2013087052A1 (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2870859A1 (en) * 2013-11-07 2015-05-13 Heliospectra AB Method for controlling a growth cycle for growing plants using state oriented control
DE102014010079A1 (en) * 2014-07-05 2016-01-07 Rudolf Trepnau Procedure for cooperative monitoring of plant breeding
DE102014226291A1 (en) 2014-12-17 2016-06-23 Bayerische Motoren Werke Aktiengesellschaft Device and method for warning against surface damage to vehicles
DE102014226189A1 (en) * 2014-12-17 2016-06-23 Continental Automotive Gmbh Method for determining a weed component and agricultural technology control device
EP3165065A1 (en) * 2015-11-06 2017-05-10 Amazonen-Werke H. Dreyer GmbH & Co. KG Method for determining a distribution pattern of a distributor
WO2018073060A1 (en) * 2016-10-18 2018-04-26 Bayer Cropscience Aktiengesellschaft Planning and implementing agricultural measures
CN108471712A (en) * 2015-12-16 2018-08-31 富兰帝欧公司 Information processing unit, information processing method and program
DE102018251708A1 (en) * 2018-12-27 2020-07-02 Robert Bosch Gmbh Process for the destruction of at least one plant

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015193822A1 (en) 2014-06-17 2015-12-23 Casella Macchine Agricole S.R.L. Method and device for measuring vegetation cover on farmland
CN106595603B (en) * 2016-11-23 2019-06-28 华南农业大学 A method of the detection lodging cone of the canopy as caused by unmanned plane rotor air-flow
EP3343170A1 (en) 2016-12-27 2018-07-04 Yara International ASA Device and method for determining a height of an agricultural product
DE102017204650A1 (en) * 2017-03-21 2018-09-27 Robert Bosch Gmbh Method for classifying plants
DE102017124934A1 (en) * 2017-10-25 2019-04-25 M-Farms Gmbh System for semi-automatic and / or automatic weed removal
CN108074236B (en) * 2017-12-27 2020-05-19 Oppo广东移动通信有限公司 Plant watering reminding method, device, equipment and storage medium

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07115846A (en) * 1993-10-22 1995-05-09 Asia Kosoku Kk Method for preparing evaluation image of fluctuation in plant vitality
DE4042672C2 (en) 1990-05-23 2002-08-08 Amazonen Werke Dreyer H On-board computer system for agricultural machine and / or implement combinations
DE10329472A1 (en) 2003-07-01 2005-01-27 Amazonen-Werke H. Dreyer Gmbh & Co. Kg Crop density determination device, comprises an optoelectronic distance sensor for measuring crop height with the resultant measurement signal used by a control unit program to control fertilizer or muck spreading
DE10353212A1 (en) * 2003-11-13 2005-06-23 Db Netz Ag Method and device for detecting and measuring vegetation in the vicinity of traffic routes
DE102005050302A1 (en) 2005-10-17 2007-04-26 Yara International Asa Method and device for non-contact determination of the current nutritional status of a crop and for processing this information on fertilizer recommendations
DE102008009753B3 (en) * 2008-02-18 2009-09-10 Yara International Asa Method for contactless determination of biomass and morphological parameter of plant populations, involves treating plants of populations with ultrasound by releasing acoustic pulses from ultrasound source
WO2009153304A1 (en) * 2008-06-20 2009-12-23 Faculté Universitaire des Sciences Agronomiques de Gembloux Weed detection and/or destruction

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4132637C2 (en) * 1991-10-01 1995-04-27 Walter Prof Dr Kuehbauch Process for controlled weed control
DE19642439C1 (en) * 1996-10-15 1997-11-20 Heinz Dr Ing Ganzelmeier Treatment of crops based on fluorescence emission
AUPR457401A0 (en) * 2001-04-26 2001-05-24 Weed Control Australia Pty Ltd Selective weed discrimination
DE20221564U1 (en) * 2001-06-12 2006-06-29 Gebr. Pöttinger GmbH Agricultural vehicle and means for its remote control via an electronic control device with a Bluetooth(RTM) interface so that the tool, e.g. plough, being used can be accurately controlled by a remotely, comfortably, located operator
HU3056U (en) * 2005-04-29 2006-03-28 G & G Noevenyvedelmi Es Keresk Construction for making weed map
WO2007079805A1 (en) * 2006-01-08 2007-07-19 Hermann Tropf Creation of a distance image

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4042672C2 (en) 1990-05-23 2002-08-08 Amazonen Werke Dreyer H On-board computer system for agricultural machine and / or implement combinations
JPH07115846A (en) * 1993-10-22 1995-05-09 Asia Kosoku Kk Method for preparing evaluation image of fluctuation in plant vitality
DE10329472A1 (en) 2003-07-01 2005-01-27 Amazonen-Werke H. Dreyer Gmbh & Co. Kg Crop density determination device, comprises an optoelectronic distance sensor for measuring crop height with the resultant measurement signal used by a control unit program to control fertilizer or muck spreading
DE10353212A1 (en) * 2003-11-13 2005-06-23 Db Netz Ag Method and device for detecting and measuring vegetation in the vicinity of traffic routes
DE102005050302A1 (en) 2005-10-17 2007-04-26 Yara International Asa Method and device for non-contact determination of the current nutritional status of a crop and for processing this information on fertilizer recommendations
DE102008009753B3 (en) * 2008-02-18 2009-09-10 Yara International Asa Method for contactless determination of biomass and morphological parameter of plant populations, involves treating plants of populations with ultrasound by releasing acoustic pulses from ultrasound source
WO2009153304A1 (en) * 2008-06-20 2009-12-23 Faculté Universitaire des Sciences Agronomiques de Gembloux Weed detection and/or destruction

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
SONG, Y. et al.: Surface Modelling of Plants from Stereo Images. In: Sixth International Conference on 3-D Digital Imaging and Modeling (3DIM 2007), 2007, S.312-319 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2870859A1 (en) * 2013-11-07 2015-05-13 Heliospectra AB Method for controlling a growth cycle for growing plants using state oriented control
WO2015067691A1 (en) * 2013-11-07 2015-05-14 Heliospectra Ab Method for controlling a growth cycle for growing plants using state oriented control
DE102014010079A1 (en) * 2014-07-05 2016-01-07 Rudolf Trepnau Procedure for cooperative monitoring of plant breeding
DE102014226291A1 (en) 2014-12-17 2016-06-23 Bayerische Motoren Werke Aktiengesellschaft Device and method for warning against surface damage to vehicles
DE102014226189A1 (en) * 2014-12-17 2016-06-23 Continental Automotive Gmbh Method for determining a weed component and agricultural technology control device
DE102014226189B4 (en) * 2014-12-17 2017-08-24 Continental Automotive Gmbh Method for determining a weed component and agricultural technology control device
US10209179B2 (en) 2014-12-17 2019-02-19 Continental Automotive Gmbh Determining a weed percentage and agricultural control device
EP3165065A1 (en) * 2015-11-06 2017-05-10 Amazonen-Werke H. Dreyer GmbH & Co. KG Method for determining a distribution pattern of a distributor
CN108471712A (en) * 2015-12-16 2018-08-31 富兰帝欧公司 Information processing unit, information processing method and program
EP3387894A4 (en) * 2015-12-16 2018-12-12 Plantio, Inc. Information processing device, information processing method, and program
WO2018073060A1 (en) * 2016-10-18 2018-04-26 Bayer Cropscience Aktiengesellschaft Planning and implementing agricultural measures
DE102018251708A1 (en) * 2018-12-27 2020-07-02 Robert Bosch Gmbh Process for the destruction of at least one plant

Also Published As

Publication number Publication date
WO2013087052A8 (en) 2014-01-09
WO2013087052A1 (en) 2013-06-20

Similar Documents

Publication Publication Date Title
Virlet et al. Field Scanalyzer: An automated robotic field phenotyping platform for detailed crop monitoring
US10165722B2 (en) Scouting systems
JP2018160257A (en) System, machine, control method, and program
US9721181B2 (en) Cloud detection on remote sensing imagery
Bai et al. A multi-sensor system for high throughput field phenotyping in soybean and wheat breeding
US10761211B2 (en) Plant treatment based on morphological and physiological measurements
Santesteban et al. High-resolution UAV-based thermal imaging to estimate the instantaneous and seasonal variability of plant water status within a vineyard
US10192185B2 (en) Farmland management system and farmland management method
US10467472B2 (en) Ponding water detection on satellite imagery
US20180263171A1 (en) Generating an agriculture prescription
Das et al. Devices, systems, and methods for automated monitoring enabling precision agriculture
US20180129987A1 (en) Farming Data Collection and Exchange System
Underwood et al. Mapping almond orchard canopy volume, flowers, fruit and yield using lidar and vision sensors
US10028426B2 (en) Agronomic systems, methods and apparatuses
US20170090068A1 (en) Estimating soil properties within a field using hyperspectral remote sensing
Sanz et al. Relationship between tree row LIDAR-volume and leaf area density for fruit orchards and vineyards obtained with a LIDAR 3D Dynamic Measurement System
JPWO2016009688A6 (en) System, machine, control method, program
DE102009023896B4 (en) Apparatus and method for detecting a plant
Hall et al. Characterising and mapping vineyard canopy using high-spatial-resolution aerial multispectral images
EP3167698A1 (en) System for residue detection and implement control and method for such
US10614503B2 (en) Apparatus and method for surveying premises of a customer
US20180146612A1 (en) Agricultural data analysis
US8731836B2 (en) Wide-area agricultural monitoring and prediction
US10338602B2 (en) Multi-sensor, autonomous robotic vehicle with mapping capability
Andújar et al. Discriminating crop, weeds and soil surface with a terrestrial LIDAR sensor

Legal Events

Date Code Title Description
R012 Request for examination validly filed
R120 Application withdrawn or ip right abandoned
R120 Application withdrawn or ip right abandoned

Effective date: 20130425