EP3528609A1 - Prévisions de rendement pour un champ de blé - Google Patents

Prévisions de rendement pour un champ de blé

Info

Publication number
EP3528609A1
EP3528609A1 EP17790724.3A EP17790724A EP3528609A1 EP 3528609 A1 EP3528609 A1 EP 3528609A1 EP 17790724 A EP17790724 A EP 17790724A EP 3528609 A1 EP3528609 A1 EP 3528609A1
Authority
EP
European Patent Office
Prior art keywords
ear
ears
image
cornfield
determining
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Withdrawn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
EP17790724.3A
Other languages
German (de)
English (en)
Inventor
Klaus Ruelberg
Gregor Fischer
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.)
BASF Agro Trademarks GmbH
Original Assignee
BASF Agro Trademarks GmbH
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 BASF Agro Trademarks GmbH filed Critical BASF Agro Trademarks GmbH
Publication of EP3528609A1 publication Critical patent/EP3528609A1/fr
Withdrawn legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02—Agriculture; Fishing; Forestry; Mining
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
    • A01G22/00—Cultivation of specific crops or plants not otherwise provided for
    • A01G22/20—Cereals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30181—Earth observation
    • G06T2207/30188—Vegetation; Agriculture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume

Definitions

  • the invention relates generally to a crop yield prediction, and more particularly to a grain crop yield prediction method.
  • the invention further relates to a corresponding system for yield prediction of a corn field and to a computer system relating thereto.
  • a method for yield prediction for a cornfield may include positioning a digital camera at a defined distance above a central plane of ears of a cornfield and capturing a digital image of a section of the cornfield with the positioned digital camera.
  • the middle level of the ears may be parallel to an image plane of the digital camera.
  • the method may include determining an area of the picked cornfield section from the defined distance and an angle of view of the digital camera, and determining a total area of the ears in the digital image as compared to a total area of the digital image by an algorithm for differentiating between image pixels Have ears unlike other picture pixels that are not associated with ears.
  • the method may include determining an output of the field from the total area of the ears in the digital image. This can be done in comparison to a total area of the digital image, the particular area of the recorded cornfield section, an average grain weight of an ear, a field total area, and a first calibration factor.
  • a system for yield forecasting a cornfield may include a digital camera positioned at a defined distance above a central plane of ears of a cornfield.
  • the digital camera can be adapted for taking a digital image of a section of the cornfield with the positioned digital camera.
  • the middle level of the ears and one image plane of the digital camera can be parallel to each other.
  • the system may include a partial area determining unit for determining an area of the picked cornfield section from the defined distance and an angle of view of the digital camera and a ⁇ hren vombeéessaku, the Determining a total area of the ears is adapted in the digital image. This can be done as compared to a total area of the digital image by an algorithm for differentiating between image pixels of the ears as opposed to other image pixels that are not associated with ears.
  • a yield determination module for determining an output of the field from the total area of the ears in the digital image compared to a total area of the digital image, the particular area of the picked cornfield section, an average grain weight of an ear, a field total area and a Calibration factor be present.
  • the presented system can be implemented as part of a smartphone.
  • the method presented here by a powerful form of a smartphone can be completely or partially performed by this smartphone.
  • the determination of the total number of grains of an ear of corn may also be carried out on a dedicated computer specially adapted for this purpose, a server computer or any other computer system.
  • embodiments have the form of a corresponding computer program product.
  • This may include instructions that, when executed on a computer system, perform steps of the described method.
  • the alternative possible positioning of the digital camera above the cornfield allows on the one hand to produce the digital image from an elevated location above the surface of the cornfield (eg from an agricultural machine); On the other hand, it is also possible to produce in the cornfield standing by the digital camera, which is mounted on a rod, the required digital recording.
  • the proposed method for yield prediction can be combined with another elegant method for determining the grain weight of an ear of corn.
  • an estimate of the yield of a cornfield can be made directly from images of the ears of the cornfield.
  • the same digital camera can be used for the procedure for the yield analysis and the partial method of determining the grain weight.
  • the yield prediction calculations as well as the computation for the determination of the grain weight can be performed on the same computer system, so that it becomes possible to provide the results of one calculation (grain weight) of the second calculation (yield prediction) as input values.
  • an upstream determination of the grain weight of an ear can - as already mentioned - as the front end of the same digital camera -. from a smartphone - can be used for improved yield prediction.
  • Using a mobile device is enough to enable farmers to make an improved yield prediction of their cornfield.
  • One or two digital images of an ear of corn are already enough to enable the farmer to significantly improve his field yield forecast.
  • Another simple everyday item in the form of a reference card does not complicate the handling and acceptance of the method.
  • the ear can either be picked or cut and placed on the reference card, or the ear can remain on the stalk and the reference card can be easily placed behind the ear.
  • the scale of the reference card provides a clear, unadulterated scale with the digital image.
  • the required computing power for the automated measurement of the ear and the grain weight of the ear can be made available in a data center. This data center can - as well as the data center for the calculation of the field yield - be operated at any point. It can either be a computer used by a farmer, several farmers can operate the computer together or a service provider takes over the analysis work and offers the required computing power.
  • the latter would have the further advantage that the service could be operated in the form of a cloud computing service for a large number of farmers in different regions or even transnationally. It would also be easier to take into account parallels between different regions, global and even local weather conditions or even regionally known pest infestation, use of fertilizers or use of pesticides, etc.
  • the digital image can be transmitted via a mobile network to an evaluation computer.
  • the analysis can be carried out and the result transmitted wirelessly back to the farmer or the mobile device.
  • An extrapolation, based on the grain weight of an ear on the entire field, could be made by further methods.
  • the analysis could also be carried out directly on site.
  • the necessary calculation algorithms could be made available in the form of a smartphone app.
  • a dedicated calculator special processor or special hardware
  • a smartphone for digital recording.
  • the farmer could also use a conventional digital camera and transmit the digital recording of the ear of corn in another form to the computer for analysis: for example, via wired communication technologies or relay stations having known communication channels such as WLAN, Bluetooth or other comparable means of communication.
  • the template matching method used to determine the number of spindle stages provides by the nature of the digital recordings of the ear - in the form of the spindle or spindle stage view and a potential second digital recording, the 90 ° to the Longitudinal axis is rotated (flower view) - a good basis for the further image processing and investigation steps.
  • the algorithm used for differentiating between image pixels of the ears, as opposed to other image pixels may be a local binary pattern algorithm.
  • a local binary pattern algorithm Basically, such algorithms are known.
  • An example is published in: DG. He and L. Wang, Texture Unit, Texture Spectrum, and Texture Analysis, IEEE Transactions on Geoscience and Remote Sensing, vol. 28, pp. 509-512, 1990; T. Müenpüü. M. Pietikäinen, and T. Ojala, "Texture Classification by multi-predicate local binary pattern operators", Proceedings 15th International Conference on Pattern Recognition, Barcelona, Spain, 3: 951-954, 2000.
  • This provides a powerful computational algorithm, which can be used directly and easily in the form of existing program libraries for the proposed method.
  • the algorithm for differentiating between image pixels of the ears, as opposed to other image pixels may be a texture image analysis method.
  • texture image analysis method Such methods are also known in principle, can be adapted according to the requirements of the presented method and are described, for example, in: F. Cointault, D. Guerin, J-P. Guillemin & B. Chopinet, "In-field Triticum aestivum ear counting using color-texture image analysis", New Zealand Journal of Crop and Horticultural Science, vol. 36, pp. 117-130, 2008. This algorithm is also easy to apply adjust the task shown here.
  • the algorithm for differentiating between picture pixels of the ears, as opposed to other picture pixels may comprise or consist of a brightness difference filter.
  • a brightness difference filter it must be taken into account that differences in recognition may well be present depending on the lighting, time of day, color components in the sky light, as a function of rain, fog and / or solar radiation. For this reason, it may be advantageous to always produce the ear or the field with the help of artificial lighting such as artificial lightning.
  • the defined distance between the digital camera and the surface of the cornfield by a spacer between the digital camera and a middle levels of the ears of the Kornfeldes be determined.
  • the spacer may consist of a flexible element - such as a twine - one end of which is attached to the digital camera and at the other end may be a color contrast ball, which is positioned in the middle plane of the ears of cornfield. Due to the color contrast of the color contrast sphere to the environment (ie essentially the ears), this is well visible or recognizable in the digital image by means of pattern recognition.
  • the color contrast sphere may have, for example, a blue or blue-green color value.
  • regular geometric shapes are considered, such as a pyramid, a barrel, a cube, a box-shaped element or even irregular objects that have a good color contrast to the color values of the ears.
  • the defined distance may be determined by a spacer between the digital camera and a middle level of the cornfield ears.
  • the digital camera may be mounted at one end of the spacer at a predetermined angle other than 90 degrees, with the other end of the spacer positioned on a central plane of the cornfield ears.
  • the digital image can be captured when the image plane of the digital camera is horizontally aligned.
  • the exposure can be triggered automatically by position or acceleration sensors that can be connected to the digital camera. It is assumed that the middle level of the ears of the cornfield is horizontal.
  • the size of the surface of the recorded cornfield section is determined by positioning the digital camera and taking a digital image of a cornfield section ("determining an area of the recorded cornfield section from the defined distance and a viewing angle of the digital camera").
  • the total area of the ears in the digital image is determined from the digital image acquisition ("determining a total area of the ears in the digital image compared to a total area of the digital image"). The result is, for example, that a certain percentage of the pixels of the image capture ears.
  • the number of ears in the digital image acquisition can be determined. This requires knowing which area (how many pixels) a single ear of corn occupies on average. This quantity can be represented by the first calibration factor, which is usually determined empirically. If one divides the total area of the picture, which is due to spikes, by the average size of the area, which occupies a single spike, one receives the number of spikes in the picture.
  • Dividing the number of ears in the picture by the size of the area of the cornfield in the image section gives the number of ears per unit area of the cornfield. If the number of ears per unit area of the cornfield is multiplied by the total area of the field, the number of ears in the entire field is calculated. Multiply the number of Ears throughout the field with the average grain weight of an ear, gives the grain weight of the entire field - and thus the yield ("Determining a yield of the field from the total area of the ears in the digital image compared to a total area of the digital image, the determined Area of the recorded cornfield section, an average grain weight of an ear, a field total area and a first calibration factor ").
  • the first calibration factor may have at least one pendency with respect to one of the factors variety, growth stage - in particular represented in the form of the BBCH code - weather, geographical location and / or fertilization status. Further dependencies are conceivable.
  • GPS Global Positioning System
  • the calibration factor itself can be a direct function of the input variables. Dedicated input values can be stored together with result values in a matrix and retrieved there by the method.
  • the determination of the total area of the ears in the digital image compared to a total area of the digital image may further include loading the areas of the ears with an area factor whose value is from a center of the digital image to the edge decreases. This is advantageous because the ears in the center of the image can be taken rather centrally from the top, while ears in marginal areas of the digital image - due to the other angle of view - can rather be taken from the side and thus occupy a larger portion of the image. Due to the area factor, this effect can be compensated.
  • the method may include providing a second digital image of a single ear of corn in a spindle-stepped view of the ear.
  • the ear in the digital image can be imaged in front of a reference card as a background.
  • the method according to this embodiment may further include determining a length of the ear along the longitudinal axis of the ear by separating image pixels of the digital image of the ear from the background and comparing pixel coordinates at one end of the ear with pixel coordinates of the ear at an opposite one Have the end of the ear in the longitudinal direction of the ear with picture marks on the reference card.
  • the method according to this embodiment may include determining a number of spindle stages of the ear by a template matching method, determining a grain number of the ear by multiplying the detected spindle levels by a factor, and determining the weight of all the grains of the ear by multiplying the determined number of grains by a second calibration factor.
  • the grain weight of an ear can be determined elegantly without the need for weighing.
  • the optical method allows elegant and direct determination of the grain weight either within a smartphone or in a separate data center to which the captured digital image has been transmitted in the spindle stage view.
  • the same technical device - namely the smartphone - can be used both for determining the grain weight and for determining the field yield.
  • the template matching method may comprise a pixel-by-pixel displacement of a selected image template consisting of a central portion of the ear of the entire ear in the longitudinal direction of the ear.
  • the method may each determine a respective similarity factor of the
  • the selected portion of the ear can occupy about 15-25% of the ear in a central region of the ear.
  • this embodiment may include determining the number of spindles from the x-y plot. Since the template matching method is a well-known method in the field of image processing, conventional program library functions and modules can be used. The use of this matching method provides good accuracy and robustness against variations in the illumination geometry in the determination of the spindle stages. This is advantageous because the number of spindle stages has a significant influence on the number of grains of the ear. An additional spindle level of the spike can be equivalent to 4 additional grains, which can increase the total number of grains of honor by up to 10%. Consequently, the most accurate possible detection of the number of spindle stages is synonymous with the accuracy of the proposed method.
  • determining the number of spindles from the xy representation may include determining the number of relative maxima of a similarity value by simple counting. This procedure requires little computing power, but is not the most accurate compared to other methods because the degree of similarity to the ends of the ear decreases and accordingly the maxima are no longer as pronounced as in the central ear area.
  • determining the number of spindles from the xy plot may include determining a mean period length from the distances of the relative maxima of a similarity value to each other and determining the number of spindles by dividing the ear length by the period length.
  • this embodiment now described may have a higher degree of accuracy in determining the number of spindle stages. This is because the relative maxima in the x-y representation may be more pronounced than in the previous embodiment. This results in a higher accuracy in determining the number of spindle stages of an ear.
  • the second calibration factor may have at least one dependence on one of the following factors: variety of the ear, growth stage of the ear, weather (long term and short term), geographical location, and fertilization status. Further influencing parameters can be taken into account at any time.
  • the method for yield prediction for a cornfield may include a sub-method, in particular a grain weight determination method for determining a weight of all grains of an ear of a cereal grain.
  • This sub-procedure would be an alternative to the sub-procedure, in which a spindle-level view of the ear of corn is used.
  • This grain weight determination method may include providing a digital image of the spike in a flower view of the ear in front of a reference map, and determining an area of the flower view of the spike by separating image pixels of the digital image of the spike from the background by a color histogram method.
  • this grain weight determination method may include comparing an area occupied by the ear as compared to image marks on the reference map.
  • the partial weighting method of the yield prediction method for determining grain weight includes determining the weight of all the grains of the ear of corn by multiplying the determined ones
  • This calibration factor can have various dependencies, such as a dependence on the type of grain, the growth stage, the weather, a fertilization status, a known pest infestation, etc.
  • This sub-method has the advantage that it can be easily applied.
  • the computational intensity may be lower than in the sub-method for determining the grain weight using the spindle stage view of the ear. This would make it easier to implement this sub-procedure directly in the mobile device in the field. Alternatively, it would also be possible to transmit the recorded digital images to a Ausenserechentechnik and the result to receive again with the mobile device on the field.
  • This sub-method also has the advantage that the digital recording of the flower view is easier to make than the spindle stage view, since the ear comes to rest in a natural position in a flower view. This would be a relief for the one who makes the digital recording. It has been found that relatively accurate estimates of the grain weight of an ear of corn can be made with this grain weight determination method presented here.
  • the system for yield analysis of the cornfield on a transmitting and receiving unit which is adapted to send the captured digital image of the cornfield section - or the second digital recording - to a data center, which the partial area determination unit, the Ear area determining unit and the yield determination module has. After calculating the determination units and the module, the result can then be sent back to the smartphone, the digital camera or the other mobile device and reused directly on the field.
  • embodiments may take the form of an associated compute rogramm areas that can be accessed by a computer-usable or computer-readable medium.
  • the instructions may cause a computer - such as a smartphone, a server, or a combination of both - to perform processing steps in accordance with the presented method.
  • the computer usable or computer readable medium may be any device having elements for storing, communicating, transporting or forwarding the program together with the instruction processing system.
  • Fig. 1 shows a block diagram of an embodiment of the inventive method for determining the yield prediction of a corn field.
  • Fig. 2 shows a positioning of a digital recording device over ears of cornfield.
  • Fig. 3 shows a tool for the desirable positioning of the camera over the middle plane of the cornfield ears.
  • FIG. 4 shows an alternative for a reproducible distance positioning of the camera from the middle level of the ears of the cornfield.
  • FIG. 5 shows an exemplary photograph of the cornfield in accordance with a method that has been illustrated in connection with FIGS. 4 and 5.
  • Fig. 6 shows a block diagram of the partial method for determining the grain weight of an ear.
  • Fig. 7 shows a first part of a block diagram of an implementation nearer embodiment of the proposed method.
  • FIG. 8 shows a second part of the block diagram of the implementation of the proposed method of FIG. 7.
  • FIG. 9 shows an ear of wheat shown in abstract and an example of a reference card together with an ear of corn lying thereon.
  • Fig. 9a shows an image of an ear and a view of the spindle stages of an ear.
  • Fig. 10 shows an exemplary diagram for determining the ear length.
  • FIG. 11 shows an exemplary diagram of a cross-correlation function for determining the number of spindle stages.
  • Fig. 12 is a block diagram of a subsystem for determining the total grain number of an ear of a cereal grain.
  • Fig. 13 shows a block diagram of a crop field yield prediction system.
  • FIG. 14 shows a block diagram of an example of a computer system together with the system corresponding to FIG. 13 and / or FIG. 12.
  • cereal haulm or "ear of a cereal haulm” need not be interpreted further. It can be an ordinary crop that grows on an agricultural field. Typically, the grain may be wheat, rye or barley.
  • digital image describes a digital image of a real scene that typically can be captured with a digital camera.
  • the digital image or image may be composed of pixels of different color values to create a visual overall impression
  • a digital recording of the surface of the cornfield from a bird's-eye view and, if necessary, another digital recording of a single ear to determine the grain weight of a typical ear of corn are recorded.
  • flower view of the ear describes a view of the ear in which the grains are clearly visible.
  • the flower view can also be referred to as a grain view of the ear, as in this view the grains of the ear are most visible.
  • the view of the awns extends most to the left and to the right of the spike, in contrast to the flower view, the term “Spindle Step View” describes a view of the spike rotated 90 ° along the longitudinal axis of the spike. So a view of the narrow view of the ear. Here you look on the narrower side of the ear or on the awns of the ear when the longitudinal axis of the ear is vertical.
  • a "reference map” in the context of this description is a flat object - for example a monochrome map - whose color value is well different from that of the ear.
  • a complementary hue - e.g., blue - to a typical hue color value has been found to be advantageous.
  • template matching method is known to the person skilled in the art as a method for detecting a structure of a digitally displayed object, for example, a detailed description can be found in S. Kim, J. McNames, Automatic spike detection based on adaptive template matching for extracellular neuronal recordings ", Journal of Neuroscience Methods 165 pp. 165-174, 2007.
  • the term "developmental stage” or “developmental stage” describes a stage in the natural life cycle of a plant - here a cereal ear - from sowing to harvesting It has been found that the use of the "BBCH code” to describe the developmental stage a plant is helpful.
  • the abbreviation "BBCH” officially stands for "Biologische Bundesweg, Bundessortenamt and CHemische Industrie”.
  • the BBCH code describes a phenomenological developmental stage of plants. The code starts at 00 and ends at 89. For example, a BBCH code between 10 and 19 describes an early stage of development of a leaf. From a BBCH code of 60, the flower of the plant appears (up to 69).
  • the next ten steps each describe the development of the fruit (70-79), the seed maturity (80-89) and the death (90-99 - of annual plants) of the plant.
  • the term "digital camera” describes a camera which uses a digital storage medium as the recording medium instead of a photographic film. The digital image is previously digitized by means of an electronic image converter (image sensor).
  • color contrast sphere describes an article which has a spatial extent which is of the same order of magnitude as the average length of the ears (eg a few centimeters in size) and a weight of the order of about 10 to 100 g Its color is ideally complementary to a dominating color of a grain field surface, advantageously a complementary color of a cornfield in a mature state, for example the color contrast sphere may have a blue color value on its surface, it is not necessary that it actually be Other geometrical shapes are also possible It is important that the color contrast sphere be easily distinguishable by means of optical recognition methods of cornfield pixels
  • texture image analysis describes a process in which the texture of an object of a digital image image is analyzed.
  • FIG. 1 shows a block diagram of one embodiment of the inventive method 100 for yield forecasting for a cornfield.
  • the method includes positioning, 102, a digital camera at a defined distance above a central plane of ears of cornfield - ie, a bird's-eye view - and taking, 104, a digital image of a section of the cornfield with the positioned digital camera.
  • the middle level of the ears and one image plane of the digital camera should be parallel to each other. This can be done automatically by using camera acceleration or position sensors. Exactly when the image plane is oriented horizontally, the automated triggering of the digital camera can take place.
  • the method comprises determining, 106, an area of the picked cornfield section from the defined distance and an angle of view of the digital camera and determining, 108, a total area of the ears in the digital image compared to a total area of the digital image by a Algorithm for differentiating between picture pixels of the ears in contrast to other picture pixels that are not associated ears.
  • the method comprises determining, 110, an output of the field from the total area of the ears in the digital image as compared to a total area of the digital image, the particular area of the picked cornfield section, an average grain weight of an ear, a field total area, and a first one Calibration factor on.
  • This first calibration factor may be variety, growth, weather, location and / or fertilizer dependent.
  • the area of the recorded grain field cutout could be 4 m 2 .
  • the area of the recorded grain field cutout could be 4 m 2 .
  • 20% of the pixels are due to ears.
  • 41472 pixels would be due to spikes.
  • an ear under the present conditions typically has an average size of 208 pixels. Then about 200 ears would be displayed on the digital image.
  • a corn field cut of 4 m 2 that would correspond to about 50 ears per m 2 field surface. If the total area of the field were 1 km 2 , 50 million ears would be present in the entire field. If the average grain weight were 3 grams per ear, a total weight of 150 tons would be present in the field.
  • Fig. 2 shows a positioning of a digital recording device 202 via ears 210 of a cornfield.
  • the digital recording device may, for example, be a single digital camera or a digital camera in a mobile telephone-for example a smartphone.
  • the camera 202 picks up a defined section of the surface of the cornfield. The section is essentially determined by the distance of the image plane 204 of the camera 202 and the angle of view ⁇ 206 of the camera 202.
  • the digital image is the Image plane 204 of the camera 202 advantageously parallel to a central horizontal plane 208 through the ears 210 of the cornfield.
  • Figure 3 shows an aid to the desirable positioning of the camera 202 over the median plane 208 of the ears 210 of the cornfield.
  • the tool may be a spacer 302 between the camera 202 and a weight 304.
  • the spacer 302 is attached to both the camera 202 and the weight 304.
  • the weight 304 may for example consist of a ball.
  • the color of the sphere 304 should be such that it differs well from the color of the ears 210 or the cornfield.
  • As a complementary color to the color of mature or almost mature grain is a blue tone for the ball 304 into consideration (color contrast ball).
  • the spacer 302 may be configured as a thread or a thread-like structure.
  • FIG. 4 shows an alternative for a reproducible distance positioning of the camera 202 from the central plane 208 of the ears 210 of the cornfield.
  • the digital recording can be triggered in various ways. On the one hand, it is possible to integrate a trigger in the handle 404 of the rod 402. In this variant, however, it might be difficult to align the image plane 204 of the camera 202 so that it is parallel to the central plane 208 of the ears 210. More elegant would be a solution in which the image is automatically triggered as soon as the image plane 204 is horizontal after a signal 204 for triggering the camera 202 was triggered. In this way, a parallelism of the image plane 204 and the central plane 208 of the ears 210 can be ensured. A Detection of the horizontal alignment of the image plane 204 can be detected via acceleration sensors (or other sensors) of the camera.
  • FIG. 5 shows an exemplary receptacle 500 of a cornfield cutout.
  • the size of the area of the cornfield corresponding to the digital image can be determined by the methods mentioned above in connection with FIGS. 2 to 4 or by comparable methods. It can be seen clearly that the ears 502 in the exemplary recording 500. In addition, it can be seen that those ears 502, which lie in the center of the image, are recorded at a different angle than those ears 502, which are located in outer areas of the receptacle 500. This results from simple optical considerations. This effect can be compensated by an area factor decreasing towards the edges of the digital recording 500. Furthermore, it can be seen in FIG. 5 that individual ears of corn overlap. Accordingly, it is not possible, for example, to determine the number of ears present in the image recording.
  • the total area of the ears in the digital image is determined in comparison to the total area of the digital image, and in a further step the number of ears present is determined using a typical size of an ear in a digital image.
  • the information about the typical size of an ear in a digital image is provided by the first calibration factor, which is usually determined empirically.
  • the field yield of the entire field can be extrapolated.
  • the grain weight of an ear of wheat 502 is taken into account for extrapolation to the field yield.
  • a grain weight can also be determined by means of an additional digital recording of a single ear of corn.
  • the method initially comprises providing 602 a digital image of the ear in a spindle-end view of the ear.
  • the ear should be in the digital image in front of a reference card as a background when recording.
  • the reference card will, for convenience, have the complementary color (e.g., blue) to a typical color of an ear of corn (yellowish).
  • the method comprises determining 604 a length of the ear along the longitudinal axis of the ear of corn by separating image pixels of the digital image of the ear from the background. This separation can advantageously by means of a Color histogram procedure done. In this way, a contiguous area of the ear can be distinguished from the background of the reference map.
  • the method comprises comparing, 606, pixel coordinates at one end of the ear with pixel coordinates of the ear at an opposite end of the ear in the lengthwise direction of the ear with image marks on the reference card. By using a scale on the reference card it is easy to determine the length of the ear. For this purpose, only corresponding y-coordinates have to be subtracted from each other.
  • determining 608 a number of spindle stages of the ear can be done by a template matching method, which includes determining a grain number of the ear (step 610) by multiplying the determined spindle stages by a factor , which indicates the number of grains per spindle step, and has, for example, a value of 4 follows.
  • determining 612 the weight of all grains of the ear is achieved by multiplying the determined number of grains by a calibration factor.
  • the calibration factor can take into account a large number of variables.
  • Fig. 7 shows a first part of a block diagram of an implementation nearer embodiment of the proposed method.
  • a digital image of an ear of corn 708 is received along with a reference card.
  • Geometry correction 702 also includes a corner end detection 704 of the corners of a colored area on the reference map. This is followed by a transformation 706 of perspective and image section 710 so that areas outside the colored background with the ear of corn lying on it are ignored.
  • the image section 710 thus obtained is forwarded to an ear detection function 712.
  • the actual ear detection is done by means of an analysis 714 by means of a color histogram method to distinguish pixels of the ear and the colored background (716 segmentation foreground / background).
  • the recognized Ear object masked, 718.
  • detected image pixels of the background may be represented as a logical "0".
  • a spike preprocessing 722 is performed. This may include a step of lighting and contrast optimization 724.
  • a transformer straightening 726 of the ear and a further reduction of the image detail to be processed can be performed. By optically removing the awning, it is possible to detect the appearance of the ear (step 726). Ideally, the view of the ear is a spindle stage view. The further processing of the received digital image is based on FIG. 8.
  • Fig. 8 shows a second part of a block diagram of an implementation nearer embodiment of the proposed method.
  • the actual ears analysis 802 takes place.
  • a geometry analysis 804 is required, the result of which is a determination of the ear length 808.
  • corner marks - allows in conjunction with the distance to the Halmansatz at the lower end of the ear, a determination of the length of the ear in the longitudinal direction, as shown in Fig. 9.
  • a middle selected area 810 of the ear 728 in the shape shown at this time is moved pixel-wise in a vertical direction along the vertical longitudinal axis of the ear 728.
  • a similarity factor is determined, which is determined mathematically by the cross-correlation function 810 between template and image function.
  • a correlation analysis 808 is shown in Fig. 10, which provides the period length and thus the distance of the spindle stages with each other.
  • the ratio of ear length 808 to period length leads to a very accurate measure for half the number of spindle stages, because the periodicity of the spindle stages is very pronounced and with a constant distance pronounced.
  • the grain analysis 812 is carried out with the determination of the number of grains 816 and the determination of the 1000 grain weight 814.
  • a yield calculation 818 of the entire field, or a subarea thereof, can follow via a yield formula 820 .
  • the 1000-grain weight - also known as Tauskorntec (TKG) - is a common calculation value for the estimation of yields in agricultural environment and indicates the weight of 1000 grains of a grain lot. It can be calculated from the grain weight of an ear and the determined number of grains of the ear.
  • FIG. 9 shows an ear of corn 708 shown in abstract and an example of a colored area 902 (not recognizable in black and white representation) of a reference map (which may be larger than the area 902) together with an ear of corn 708 lying thereon
  • Surface 902 has image marks such as a scale 904 and, for example, bookmark marks 914.
  • the Jardineckmarken 914 may have different forms.
  • awns 906 are shown symbolically, which can be different lengths depending on the type of cereal.
  • a piece of the straw 910 is still shown, which has only for the detection of the lower ear field its meaning for the presented method.
  • the ear 708 should be aligned on the colored surface 902 of the reference card so that the longitudinal axis 912 of the ear 708 is aligned as parallel as possible to a side line of the colored surface 902.
  • a typically curved shape of the ear of corn 708 may be adapted by transforming the representation of the ear of corn 708 such that the longitudinal axis of the ear of the hair is actually aligned parallel to a side line of the colored area 902 of the reference card.
  • the reference map is typically a little larger than the colored area 902 contained on it, whose color is "blue", for example.
  • an actual image of an ear of corn 708 represents a contiguous region (shown, for example, in FIGS. 7, 720, 728).
  • the form of representation of the ear of corn 708 as used herein is intended to be illustrative only of the orientation of the ear of corn 708 relative to the reference map become.
  • FIG. 9a shows an illustration of an ear of corn 708 and a view 926 of the spindle stages of an ear of corn.
  • the illustration of the ear 708 clearly shows the different grains 916, 918, 920, 922 in the lower part of the ear and the stem 910. Accordingly, the more abstract shape of the ear on the right side of FIG. 9a shows the different spindle stages 424 of the ear 708th
  • FIG. 10 shows an exemplary diagram 1000 for determining the length of the ear. It can be seen here that the width of the ear (y-axis) is plotted per pixel row (x-axis) that relates to the ear.
  • the individual relative maxima - or a closely related group of relative maxima - refer to one spindle stage each. By simply counting the relative maxima or the groups of the relative maxima, the number of spindle stages can be identified. From the beginning of the pixels of the ear at about line 60 and the end of the pixels of the ear At approximately line 1710, the grain length 1002 results with the aid of the scale of the reference card or by a knowledge of the width of an individual pixel or a pixel row.
  • FIG. 11 shows an exemplary diagram 1100 of a cross-correlation function for determining the number of spindle stages on the basis of the template matching method.
  • the respective position of the template pattern from the middle of the ear
  • a correlation value similarity value
  • Match - a correlation score of practically 1.
  • the template is exactly in its original place.
  • a period length 1104 can be determined which corresponds to the distance of the individual spindle stages from one another. From the determined length of the ear and the average determined period length 1104, the number of spindle stages can also be calculated by division and rounding.
  • the process begins with providing a digital image of the ear of corn.
  • a photograph of the spike in the flower view - in the view in which the grains of the spike are clearly visible - is made in front of a reference card.
  • determining an area of the flower view of the ear of corn by separating image pixels of the digital image of the ear from the background, e.g. by means of a color histogram method, and comparing the area occupied by an ear of the ear with image marks on the reference map.
  • the image marks may be the scale of the reference map or the known distances of other image marks on the reference map.
  • the weight of all grains of the ear is determined by multiplying the determined area of the ear with a calibration factor. It has been shown that there is a pronounced direct correlation between the projection area of the spike in flower view and the number of grains of the spike. This phenomenon is used here to simply and elegantly determine the grain weight of the ear. This alternative method can also be used particularly well from a growth stage that is greater than 60 BBHC. However, it works even at lower BBHC values.
  • Fig. 12 shows a block diagram of a system for determining the total grain number of an ear of a cereal grain.
  • the system includes a receiving unit 1202 for receiving a digital image of the ear of wheat in a side view of the ear.
  • the receiving unit is a digital camera.
  • a digital Recording of the ear taken by a digital camera and to the receiving unit 1202 - optionally wirelessly - transmitted.
  • the ear of the digital image is recorded in front of a reference card as a background.
  • the system may include a display unit 1204.
  • the system has a surveying unit 1206.
  • the surveying unit 806 is adapted for comparing pixel coordinates at one end of the ear with pixel coordinates of the ear at an opposite end of the ear in the lengthwise direction of the ear with picture marks on the reference card.
  • the system includes a spindle step calculating unit 1208 adapted to determine a number of spindle stages of the ear by a template matching method, and a grain number determining unit 1210 adapted to determine a grain number of the ear by multiplying the ear determined spindle stages with a factor.
  • a weight determining unit 1212 adapted to determine the weight of all grains of the ear by multiplying the determined number of grains by a calibration factor.
  • the system may be part of a server system which receives the digital image (s) from a digital camera, such as a smartphone.
  • a digital camera such as a smartphone
  • This system of FIG. 12 may be integrated with a grain field yield analysis system 1300 shown in FIG.
  • This system includes a digital camera 1302 positioned at a defined distance above a central plane of ears of cornfield, the digital camera being adapted for taking a digital image of a section of the cornfield with the positioned digital camera, which may be the same as in FIG Fig. 12 is shown.
  • the middle level of the ears and one image plane of the digital camera should be parallel to each other.
  • the display / screen 1304 may be identical to the display unit 1204 of FIG. 12.
  • the yield prediction system 1300 further includes a patch determining unit 1306 for determining an area of the picked corn patch from the defined distance and an angle of view of the digital camera as well an ear area determining unit 1308 for determining a total area of the ears in the digital image as compared to a total area of the digital image by an algorithm for differentiating between picture pixels of the ears, as opposed to other picture pixels that are not associated with ears.
  • a patch determining unit 1306 for determining an area of the picked corn patch from the defined distance and an angle of view of the digital camera as well an ear area determining unit 1308 for determining a total area of the ears in the digital image as compared to a total area of the digital image by an algorithm for differentiating between picture pixels of the ears, as opposed to other picture pixels that are not associated with ears.
  • the system 1300 includes a yield determination module 1310 for determining an output of the field from the total area of the ears in the digital image as compared to a total area of the digital image, the particular area of the picked cornfield section, an average grain weight of an ear, a field total area and the first calibration factor.
  • Embodiments of the invention may be implemented in conjunction with virtually any type of computer, including, but not limited to, a smartphone, regardless of the platform used to store and execute program code.
  • Fig. 14 exemplifies a computer system 1400 suitable for executing program code related to the proposed method.
  • the computer system 1400 is only one example of a suitable computer system and is not intended to be a limitation on the scope of any use or functionality of the invention described herein. On the contrary, the computer system 1400 is adapted to implement each feature of each of the embodiments described herein.
  • the computer system 1400 includes components that may cooperate with a variety of other general or dedicated computer system environments and / or configurations.
  • Examples of known computer systems, environments, and / or configurations that may be suitable for cooperating with computer system 1400 include, but are not limited to, tablet computers, notebook computers, and / or other mobile computing systems, and / or Smartphones as well as multiprocessor systems, microprocessor-based systems, programmable consumer electronics or even digital cameras or PDAs (Personal Digital Assistant).
  • the computer system 1400 is described herein in a general context of computer system executable instructions. These may also be program modules that are executed by the computer system 1400. Generally, program modules include program routines, subprograms, objects, components, processing and / or decision logic, data structures, etc. that perform a particular task or represent a particular abstract data type. As already mentioned, the computer system 1400 may be implemented in the form of a "general purpose" computing system, wherein the components of the computer system 1400 include, but are not limited to, one or more processing units 1402 (CPU), a storage system 1404, and a System bus 1418, which connects various system components including main memory 1404 to processor 1402.
  • CPU processing units
  • storage system 1404 storage system 1404
  • System bus 1418 System bus 1418
  • the computer system 1400 also includes various computer-readable media. Such media includes all media accessible by the computer system 1400. This includes both volatile and non-volatile media, which can be removable as well as permanently installed.
  • Main memory 1404 may also include computer readable media in the form of a volatile memory. This can be, for example, a Random Access Memory (RAM) or a cache memory.
  • the computer system 1400 may further comprise removable and non-removable storage media.
  • storage system 1412 may include the ability to store data on a non-removable memory chip.
  • the storage media may be connected to the system bus 1406 through one or more data interfaces.
  • memory 1404 may include at least one program product having a plurality of program modules (at least one) configured and configured to configure the computer system to perform the functions of embodiments of the invention.
  • a program comprising a plurality of program modules may, for example, be stored in the memory 1404, as well as an operating system, one or more application programs, program modules and / or program data.
  • the computer system 1400 may further communicate with a plurality of external devices, such as a keyboard 1408, a pointing device (“mouse") 1410, a display (not shown), etc. These devices may be embodied, for example, in a touch-sensitive screen 1412 (touch-screen ) may be combined to facilitate interaction with the computer system 1400.
  • the computer system 1400 may also include acoustic input / output devices 1416.
  • other connections may be present to communicate with one or more other data processing devices (modem, network ports, etc.).
  • such communication may be via I / O interfaces 10.
  • the computer system 1400 may be over one or more networks, such as a local area network (LAN), a wide area network (WAN), and / or via a public (mobile) network eg Internet) via the adapter 1414 communicate.
  • the network adapter 1414 may communicate with other components of the computer system 1400 via the system bus 1418.
  • other hardware and / or software components may be used in conjunction with the computer system 1400. This concerns, for example, micro-code, device drivers, redundant processing units, etc.
  • system 1200 for determining a weight of all grains of an ear of a cereal health or an individual or integrated yield prediction system 1300 for a corn field may be connected to the bus system 1418.
  • computer system 1300 for determining field yield may receive the digital image, determine the weight of an ear of corn, and thus perform field yield prediction, and transmit the result back to the mobile device that acquires the digital image (s). was made (n).
  • systems 1200 and / or 1300 may also be integrated into a mobile computer system (e.g., high performance smartphone).
  • the present invention can be realized as a system, a method and / or a computer program product or a combination thereof.
  • the computer program product may comprise a computer readable storage medium (or simply "medium") containing computer readable program instructions for causing a processor to implement aspects of the present invention.
  • the medium can be based on electronic, magnetic, electromagnetic waves, infrared light or semiconductor systems, which are also suitable for forwarding. These include solid state memory, random access memory (RAM) as well as read only memory (ROM).
  • the computer readable program instructions described herein may be downloaded to the appropriate computer system from a potential service provider over a mobile network connection or a stationary network.
  • the computer readable program instructions for performing operations of the present invention may be any type of machine dependent or machine independent instructions, microcode, firmware, status code or object code, in any combination one or more programming languages is written, exhibit.
  • the programming languages may be C ++, Java or similar modern programming languages or conventional procedural programming languages such as the "C" programming language or similar programming languages.
  • the computer readable program instructions may be executed entirely on the computer system Circuits such as programmable logic devices, field programmable gate array (PGA) or programmable logic arrays (PLA) execute the instructions using status information in the computer readable program instructions to customize the electronic circuit (s) to accommodate aspects of the present invention to carry out the present invention.
  • These computer readable program instructions may be provided to a processor of a "general purpose computer” or more specifically to computer hardware or other programmable data processing devices to generate a machine such that the instructions executed by the respective processor will generate means for implementing the functions / actions illustrated in the corresponding flowchart and / or block diagram or blocks thereof.
  • These computer readable program instructions may also be stored on a computer-readable storage medium so as to cause a computer or programmable data processing device to execute the instructions stored in the medium by the respective processor, so that aspects or actions described in that document Procedure be performed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Tourism & Hospitality (AREA)
  • Animal Husbandry (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Mining & Mineral Resources (AREA)
  • General Business, Economics & Management (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Human Resources & Organizations (AREA)
  • Agronomy & Crop Science (AREA)
  • Geometry (AREA)
  • Botany (AREA)
  • Environmental Sciences (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

L'invention concerne un procédé pour établir des prévisions de rendement pour un champ de blé. Le procédé présente le positionnement d'un appareil photo numérique à une distance définie sur un plan intermédiaire des épis d'un champ de blé, l'acquisition d'une image numérique d'un secteur du champ de blé, la détermination d'une surface du secteur du champ de blé pris en photo, la détermination de la surface totale des épis sur l'image numérique comparé à la surface totale de l'image numérique, et la détermination d'un rendement du champ à partir de la surface totale des épis sur l'image numérique comparé à la surface totale de l'image numérique, à la surface déterminée du secteur du champ de blé pris en photo, au poids moyen des grains d'un épi, à la surface totale du champ et à un facteur d'étalonnage.
EP17790724.3A 2016-10-19 2017-10-16 Prévisions de rendement pour un champ de blé Withdrawn EP3528609A1 (fr)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
EP16194687 2016-10-19
EP16194686 2016-10-19
PCT/EP2017/076320 WO2018073163A1 (fr) 2016-10-19 2017-10-16 Prévisions de rendement pour un champ de blé

Publications (1)

Publication Number Publication Date
EP3528609A1 true EP3528609A1 (fr) 2019-08-28

Family

ID=60037630

Family Applications (2)

Application Number Title Priority Date Filing Date
EP17780770.8A Withdrawn EP3529771A1 (fr) 2016-10-19 2017-10-12 Détermination du point des grains d'un épi
EP17790724.3A Withdrawn EP3528609A1 (fr) 2016-10-19 2017-10-16 Prévisions de rendement pour un champ de blé

Family Applications Before (1)

Application Number Title Priority Date Filing Date
EP17780770.8A Withdrawn EP3529771A1 (fr) 2016-10-19 2017-10-12 Détermination du point des grains d'un épi

Country Status (6)

Country Link
US (3) US11010913B2 (fr)
EP (2) EP3529771A1 (fr)
CN (2) CN109863530B (fr)
BR (2) BR112019007937A8 (fr)
RU (1) RU2769288C2 (fr)
WO (2) WO2018073093A1 (fr)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10537071B2 (en) * 2013-07-26 2020-01-21 Blue River Technology Inc. System and method for individual plant treatment based on neighboring effects
CN113011220A (zh) * 2019-12-19 2021-06-22 广州极飞科技股份有限公司 穗数识别方法、装置、存储介质及处理器
US11823408B2 (en) * 2020-03-13 2023-11-21 Oregon State University Apparatus and method to quantify maize seed phenotypes
CN112101681B (zh) * 2020-09-25 2023-04-28 北京师范大学 基于遥感ndpi时间序列监测冬小麦干热风灾害的方法
CN113139934B (zh) * 2021-03-26 2024-04-30 上海师范大学 一种水稻谷粒计数方法
TWI818334B (zh) * 2021-10-19 2023-10-11 國立中興大學 以手機影像推估穀粒含水量及收穫時期的方法
DE102022110185A1 (de) 2022-04-27 2023-11-02 Deere & Company Verfahren und Anordnung zur Messung einer kornspezifischen Größe an einer Erntemaschine
DE102022129876A1 (de) 2022-11-11 2024-05-16 Deere & Company Verfahren und Anordnung zur Ermittlung einer massen- und/oder größenspezifischen Größe von Körnerfrüchten

Family Cites Families (39)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4076633A (en) * 1974-10-18 1978-02-28 The Procter & Gamble Company Fabric treating articles with improved conditioning properties
JP2617014B2 (ja) * 1990-04-26 1997-06-04 日本リライアンス 株式会社 シート測長システム
US8031910B2 (en) * 2003-09-17 2011-10-04 Syngenta Participations Ag Method and apparatus for analyzing quality traits of grain or seed
RU2308679C2 (ru) * 2005-09-08 2007-10-20 ИПБОЮЛ Ващенко Юрий Ефимович Способ картографирования земель
WO2007068056A1 (fr) * 2005-12-14 2007-06-21 Grains Research And Development Corporation Évaluation de coloration pour des grains de céréales
US9508007B2 (en) * 2006-11-07 2016-11-29 The Curators Of The University Of Missouri Method of predicting crop yield loss due to N-deficiency
US9195891B2 (en) * 2006-11-07 2015-11-24 The Curators Of The University Of Missouri Method of predicting crop yield loss due to n-deficiency
US8208680B2 (en) * 2006-11-07 2012-06-26 The Curators Of The University Of Missouri Method of predicting crop yield loss due to N-deficiency
JP2011167163A (ja) * 2010-02-22 2011-09-01 Pasuko:Kk 水稲収量予測モデル生成方法、及び水稲収量予測方法
CN101944231B (zh) * 2010-08-19 2012-01-04 北京农业智能装备技术研究中心 小麦穗部形态参数提取方法
EP2752808A4 (fr) 2011-08-30 2015-02-18 Fujitsu Ltd Dispositif d'imagerie, programme de support de travail, procédé de fourniture d'informations et programme de fourniture d'informations
EP2752807A4 (fr) 2011-08-30 2015-02-18 Fujitsu Ltd Dispositif d'imagerie, programme de support d'imagerie, procédé de fourniture d'informations et programme de fourniture d'informations
BR112014015990B1 (pt) 2011-12-30 2021-09-14 Pioneer Hi-Bred International, Inc Métodos para a valiar um vegetal de milho, método para a análise de produção elevada e método para avaliar uma espiga imatura de um vegetal de milho
CN102612892B (zh) * 2012-03-02 2014-05-07 北京农业信息技术研究中心 小麦麦穗发芽情况识别方法
CN103632157B (zh) 2012-08-24 2018-01-05 南京农业大学 单株小麦穗部籽粒数计数方法
US10234439B2 (en) * 2012-11-07 2019-03-19 Airscout Inc. Methods and systems for analyzing a field
JP5950166B2 (ja) 2013-03-25 2016-07-13 ソニー株式会社 情報処理システム、および情報処理システムの情報処理方法、撮像装置および撮像方法、並びにプログラム
CN103190224B (zh) * 2013-03-26 2015-02-18 中国农业大学 基于计算机视觉技术的玉米果穗考种方法、系统和装置
KR20140125229A (ko) * 2013-04-18 2014-10-28 한국전자통신연구원 과수의 생산량 예측 시스템 및 그것의 구동 방법
FR3013874A1 (fr) * 2013-11-26 2015-05-29 Invivo Agrosolutions Procede de traitement de donnees agronomiques pour l'estimation du rendement d'une culture de la famille des poacees
CN103810522B (zh) * 2014-01-08 2017-02-08 中国农业大学 一种玉米果穗籽粒计数方法和装置
CN103793686B (zh) * 2014-01-16 2017-12-15 中国农业大学 一种果树产量早期预测的方法
WO2015160837A2 (fr) * 2014-04-15 2015-10-22 Raven Industries, Inc. Système de surveillance de rendement basé sur un moissonnage et procédé associé
CN104021369A (zh) * 2014-04-30 2014-09-03 南京农业大学 基于数字图像处理技术的单株水稻穗部籽粒数计数方法
US9903979B2 (en) * 2014-09-23 2018-02-27 Deere & Company Yield estimation
US10126282B2 (en) * 2014-09-23 2018-11-13 Deere & Company Yield estimation
US10186029B2 (en) * 2014-09-26 2019-01-22 Wisconsin Alumni Research Foundation Object characterization
US9734400B2 (en) * 2015-01-30 2017-08-15 AgriSight, Inc. System and method for field variance determination
WO2016123525A1 (fr) * 2015-01-30 2016-08-04 Raytheon Company Appareil et procédés pour la classification et le comptage de grains de maïs
US10028426B2 (en) * 2015-04-17 2018-07-24 360 Yield Center, Llc Agronomic systems, methods and apparatuses
CA2990438A1 (fr) * 2015-06-30 2017-01-05 The Climate Corporation Systemes et procedes pour la capture d'images et l'analyse de champs agricoles
US10019791B2 (en) * 2015-07-24 2018-07-10 Raytheon Company Apparatus and methods for estimating corn yields
CN105427275B (zh) * 2015-10-29 2017-11-24 中国农业大学 大田环境麦穗计数方法及装置
CN105574853B (zh) * 2015-12-07 2018-05-15 中国科学院合肥物质科学研究院 一种基于图像识别的麦穗粒数计算的方法及系统
CN205607838U (zh) * 2016-05-06 2016-09-28 北京农业信息技术研究中心 一种小麦穗数测量装置
WO2018049289A1 (fr) * 2016-09-09 2018-03-15 Cibo Technologies, Inc. Systèmes d'ajustement d'entrées agronomiques à l'aide d'une détection à distance, appareil et procédés associés
WO2018049290A1 (fr) * 2016-09-09 2018-03-15 Cibo Technologies, Inc. Systèmes destinés à déterminer des sorties agronomiques d'une région cultivable, et appareil et procédés associés
WO2018049288A1 (fr) * 2016-09-09 2018-03-15 Cibo Technologies, Inc. Systèmes d'apprentissage de zones exploitables, et procédés et appareil associés
US10909367B2 (en) * 2017-03-02 2021-02-02 Basecamp Networks, LLC Automated diagnosis and treatment of crop infestations

Also Published As

Publication number Publication date
BR112019007962A2 (pt) 2019-07-02
US20210241482A1 (en) 2021-08-05
BR112019007937A2 (pt) 2019-07-02
RU2769288C2 (ru) 2022-03-30
US10984548B2 (en) 2021-04-20
CN109843034B (zh) 2022-08-23
WO2018073093A1 (fr) 2018-04-26
CN109843034A (zh) 2019-06-04
US11010913B2 (en) 2021-05-18
RU2019114132A3 (fr) 2021-01-27
RU2019114132A (ru) 2020-11-20
US20190286904A1 (en) 2019-09-19
US20190236800A1 (en) 2019-08-01
CN109863530A (zh) 2019-06-07
BR112019007937A8 (pt) 2023-04-25
WO2018073163A1 (fr) 2018-04-26
EP3529771A1 (fr) 2019-08-28
CN109863530B (zh) 2022-11-15

Similar Documents

Publication Publication Date Title
EP3528609A1 (fr) Prévisions de rendement pour un champ de blé
EP3560185B1 (fr) Système et procédé de saisie d'images de mesure d'un objet de mesure
DE102015207676B4 (de) Verfahren und Vorrichtung zur Hinderniserkennung auf Basis einer monokularen Kamera
DE202017007481U1 (de) System zur Überwachung von Nutzpflanzen
EP3701449A1 (fr) Estimation du rendement de production de plantes cultivées
EP2353146B1 (fr) Mesurer la croissance de disques de feuille de plantes
CN111582055A (zh) 一种无人机的航空施药航线生成方法及系统
DE112021000243B4 (de) Verfahren, system und computerprogrammprodukt zur bestimmung von kulturpflanzentyp und/oder aussaattermin
DE102015221085A1 (de) Verfahren und Informationssystem zum Erkennen zumindest einer auf einem Feld angepflanzten Pflanze
DE112015002189T5 (de) Bildfilter auf der Grundlage von Reihenidentifizierung
DE102012010912B3 (de) Verfahren zum berührungslosen Bestimmen des aktuellen Ernährungszustandes eines Pflanzenbestandes und zum Verarbeiten dieser Information
EP3782467A1 (fr) Procédé d'identification des mauvaises herbes dans un rang défini de plantes d'une surface agricole
DE102021107320A1 (de) Informationenbereitstellungsvorrichtung, mähfahrzeug und mähverwaltungssystem
EP3673425A1 (fr) Estimation du rendement de production de plantes cultivées
DE102015221092A1 (de) Verfahren und Informationssystem zum Erfassen zumindest eines Pflanzenparameterdatensatzes einer auf einem Feld wachsenden Pflanze
AT520253A2 (de) Verfahren zur selektiven Holzernte
DE102017217258A1 (de) Verfahren zum Klassifizieren von Pflanzen
Qu et al. Deep Learning-Based Weed–Crop Recognition for Smart Agricultural Equipment: A Review
DE102023126750A1 (de) Kalibriereinstellung für landwirtschaftliche sprühvorrichtung mit echtzeitzielsensor an der maschine
Tripathy et al. Image processing techniques aiding smart agriculture
Jiang et al. MIoP-NMS: Perfecting crops target detection and counting in dense occlusion from high-resolution UAV imagery
de Ocampo et al. Integrated Weed Estimation and Pest Damage Detection in Solanum melongena Plantation via Aerial Vision-based Proximal Sensing.
DE202022102591U1 (de) System zur Überwachung des Gesundheitszustands von Pflanzen in der Präzisionslandwirtschaft mittels Bildverarbeitung und Faltungsneuronalem Netz
EP3468339B1 (fr) Procede de determination de caracteristiques d'une plante utile
Tripathy Deepak Kumar Sharma

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20190520

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20201202

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: GRANT OF PATENT IS INTENDED

INTG Intention to grant announced

Effective date: 20230320

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20230801