# US 11,304,355 B2 · Methods and systems for reducing fitness of weed > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,304,355 B2 - Document ID: US11304355B2 - Status: Granted - Kind: B2 - Application number: 17/053,089 - Filed: 2019-05-06 - Published: 2021-03-11 - Granted: 2022-04-19 - Earliest claimed: 2018-05-06 - Assignees: Weedout Ltd. - Applicants: Weedout Ltd. - Inventors: Orly Noivirt-Brik; Efrat Lidor-Nili - Web page: https://patentgrep.com/patent/US11304355B2 ## CPC - B64D 1/16: Dropping or releasing powdered, liquid, or gaseous matter, e.g. for fire-fighting (jettisoning fuel B64D37/26) - A01B 79/005: {Precision agriculture} - A01H 1/027: {Apparatus for pollination} - A01M 7/00: Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass - A01M 21/00: Apparatus for the destruction of unwanted vegetation, e.g. weeds (control of undesirable vegetation on roads or permanent ways of railways E01H11/00) - A01M 21/043: {by chemicals} - B64D 1/18: by spraying, e.g. insecticides - G06F 18/241: relating to the classification model, e.g. parametric or non-parametric approaches - G06N 20/00: Machine learning - G06V 10/764: using classification, e.g. of video objects - G06V 20/188: {Vegetation} - B64U 2101/30: for imaging, photography or videography - B64U 2101/45: for releasing liquids or powders in-flight, e.g. crop-dusting - B64U 2201/104: using satellite radio beacon positioning systems, e.g. GPS - G06V 20/68: Food, e.g. fruit or vegetables - Y02P 60/21: Dinitrogen oxide [N2O], e.g. using aquaponics, hydroponics or efficiency measures - B64D: EQUIPMENT FOR FITTING IN OR TO AIRCRAFT; FLIGHT SUITS; PARACHUTES; ARRANGEMENT OR MOUNTING OF POWER PLANTS OR PROPULSION TRANSMISSIONS IN AIRCRAFT - A01B: SOIL WORKING IN AGRICULTURE OR FORESTRY; PARTS, DETAILS, OR ACCESSORIES OF AGRICULTURAL MACHINES OR IMPLEMENTS, IN GENERAL (making or covering furrows or holes for sowing, planting or manuring A01C5/00; mowers convertible to soil working apparatus or capable of soil working A01D42/04; mowers combined with soil working implements A01D43/12; soil working for engineering purposes E01, E02, E21) - A01H: NEW PLANTS OR {NON-TRANSGENIC} PROCESSES FOR OBTAINING THEM; PLANT REPRODUCTION BY TISSUE CULTURE TECHNIQUES - A01M: CATCHING, TRAPPING OR SCARING OF ANIMALS (appliances for catching swarms or drone-catching A01K57/00; fishing A01K69/00 - {A01K99/00}; biocides, pest repellants or attractants A01N); APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - B64U: UNMANNED AERIAL VEHICLES [UAV]; EQUIPMENT THEREFOR - Y02P: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS ## Abstract A method of weed control is provided. The method comprising: (a) determining development of flowers of at least one weed species of interest; and (b) artificially pollinating at flowering the flowers of the at least one weed species of interest with pollen that reduces fitness of the at least one weed species of interest. Also provided are systems for executing the method. ## Claims ### Claim 1 A system for reducing fitness of at least one weed species of interest, comprising: a non-transitory memory having stored thereon a code for execution by at least one hardware processor, the code comprising: code for inputting into at least one classifier, at least one image captured by at least one imaging sensor, the at least one image depicting at least one weed species of interest; code for computing by the at least one classifier, an indication of likelihood of the at least one weed being at a flowering state suitable for being artificially pollinated by artificial pollination of flowers of the at least one weed species of interest at a certain time interval, wherein the at least one classifier is trained on a training dataset comprising a plurality of images of respective at least one weed species of interest captured by the at least one imaging sensor and associated indication of a time interval of when the respective at least one weed species of interest is at the target flowering state suitable for being artificially pollinated by artificial pollination of flowers thereof relative to the time of capture of the respective image; and code for generating according to the indication of likelihood and the certain time interval, instructions for execution by pollination controller of a pollinator device for artificially pollinating flowers of the at least one weed species of interest with pollen that reduces fitness of the at least one weed species of interest. ### Claim 2 (depends on 1) The system of claim 1, wherein the certain time interval is a future time interval relative to the time at which the at least one image is captured. ### Claim 3 (depends on 1) The system of claim 1, wherein the certain time interval denotes a current time. ### Claim 4 (depends on 1) The system of claim 1, wherein the at least one image captured by at least one imaging sensor depicts the at least one weed species of interest at a flowering stage prior the start of flowering. ### Claim 5 (depends on 1) The system of claim 1, further comprising code for execution of the generated instructions for artificially pollinating flowers of the at least one weed species of interest with pollen that reduces fitness of the at least one weed species of interest. ### Claim 6 (depends on 1) The system of claim 1, wherein the instructions for execution by pollination controller are computed by at least one classifier according to the training dataset that further includes, in association with each of the plurality of images, respective instructions for execution by pollination controller of a pollinator device for artificially pollinating flowers of the at least one weed species of interest with pollen that reduces fitness of the at least one weed species of interest. ### Claim 7 (depends on 1) The system of claim 1, wherein the at least one image sensor is coupled to a first machinery navigating over the field, and the instructions for execution comprise code instruction for automated execution by the pollination controller of the pollinator device coupled to a second machinery navigating over the field, wherein the at least one image is iteratively acquired by the at least one sensor and the pollination controller of the pollinator device iteratively executes the code instructions as the first machinery and the second machinery navigate over the field for real-time pollination of identified at least one weed species of interest. ### Claim 8 (depends on 1) The system of claim 1, wherein the at least one image sensor is captured during a first time interval, and the instructions for execution comprise code instruction for automated execution by the pollination controller of the pollinator are executed at a second time interval spaced apart in time after the first time interval, wherein the second time interval is according to the certain time interval computed by the classifier. ### Claim 9 (depends on 1) The system of claim 1, wherein the at least one classifier further computes for the at least one image, a classification indication of at least one of a female of the at least one weed species of interest, or a male of the at least one weed species of interest, wherein the instructions for artificially pollinating flowers are generating according to the classification indication, and wherein images of the training set respectively include a classification indication of a female of the at least one weed species of interest. ### Claim 10 (depends on 1) The system of claim 1, wherein the at least one image is further associated with a tag storing a geographical location indication of the at least one weed species of interest captured within the at least one image, the geographic location outputted by at least one geographic positioning sensor, and wherein the instructions for pollination of the at least one weed species of interest are generated according to the geographic location of the at least one weed, wherein the instructions for pollination are created according to a plurality of images, each image associated with a different geographical location, the instructions for pollination comprise sub-sets of instructions for administering pollen to each at least one weed species of interest located at respective geographical locations of a field, wherein each of the sub-sets of instructions is selected for execution by the pollination controller of the pollinator device according to a real-time geographic location of a machinery navigating over the field. ### Claim 11 (depends on 1) The system of claim 1, wherein the instructions are generated in real time according to a real time location of the pollination device outputted by a geographical location sensor that senses the location of the pollination device. ### Claim 12 (depends on 1) The system of claim 1, wherein the instructions are generated in real time according to at least one environmental condition parameter outputted by at least one environmental condition sensor that senses the environmental condition in proximity to the at least one weed species of interest, wherein the instructions are generated according to a predicted dispersion pattern of the applied pollen according to the at least one environmental condition parameter. ### Claim 13 (depends on 1) The system of claim 1, wherein the at least one imaging sensor is selected from the group consisting of: a reflectance sensor that captures light applied by a light source reflected from the at least one weed species of interest, a spectral image sensor, multispectral image sensor, hyperspectral image sensor, and a visible light sensor. ### Claim 14 (depends on 1) The system of claim 1, wherein the pollinator device comprises a sprayer located at a predefined height above the ground, and above crops growing in the field where the at least one weed species of interest is growing. ### Claim 15 (depends on 1) The system of claim 1, wherein the pollinator device is coupled to an aerial based machinery selected from the group consisting of: an unmanned aerial vehicle (UAV), a remote-piloted vehicle (RPV), a drone, a specialized robot, a robot bee, a robobee, and a robo-bee. ### Claim 16 (depends on 1) The system of claim 1, wherein the pollinator device comprises a mechanism to attract bees, and couple the pollen to the bees for administration to the flower of the at least one weed species of interest. ### Claim 17 (depends on 1) The system of claim 1, wherein the instructions are generated according to a supplemental data profile including multiple supplemental data parameters indicative of the environment of the weed and/or details of the pollination to be applied and/or hardware of the pollinator device. ### Claim 18 A system for training at least one classifier for reducing fitness of at least one weed species of interest, comprising: a non-transitory memory having stored thereon a code for execution by at least one hardware processor, the code comprising: code for accessing a training dataset comprising a plurality of images of respective at least one weed species of interest captured by at least one imaging sensor and associated indication of a time interval of when the respective at least one weed species of interest is at a target flowering state suitable for being artificially pollinated by artificial pollination of flowers thereof relative to the time of capture of the respective image; and code for training at least one classifier according to the training dataset, for computing an indicative of likelihood of a target at least one weed being at a target flowering state suitable for being artificially pollinated by artificial pollination of flowers of the at least one weed species of interest at a certain time interval according to at least one target image captured by at least one imaging sensor. ### Claim 19 A method of reducing fitness of at least one weed species of interest, comprising: inputting into at least one classifier, at least one image captured by at least one imaging sensor, the at least one image depicting at least one weed species of interest; computing by the at least one classifier, an indication of likelihood of the at least one weed being at a target flowering state suitable for being artificially pollinated by artificial pollination of flowers of the at least one weed species of interest at a certain time interval, wherein the at least one classifier is trained on a training dataset comprising a plurality of images of respective at least one weed species of interest captured by the at least one imaging sensor and associated indication of a time interval of when the respective at least one weed species of interest is at the target flowering state suitable for being artificially pollinated by artificial pollination of flowers thereof relative to the time of capture of the respective image; and generating according to the indication of likelihood and the certain time interval, instructions for execution by pollination controller of a pollinator device for artificially pollinating flowers of the at least one weed species of interest with pollen that reduces fitness of the at least one weed species of interest.