Abstract: Embodiments of the disclosed technologies are capable of inputting, to a machine-learned model that has been trained to recognize a horticultural product in digital imagery, digital video data comprising frames that represent a view of the horticultural product in belt-assisted transit from a picking area of a field to a harvester bin; outputting, by the machine-learned model, annotated video data; using the annotated video data, computing quantitative data comprising particular counts of the individual instances of the horticultural product associated with particular timestamp data; using the timestamp data, mapping the quantitative data to geographic location data to produce a digital yield map; causing display of the digital yield map on a field manager computing device.

Abstract: Systems, methods and apparatus for detecting stalks processed by a combine harvester, for measuring stalk diameters, and for displaying harvest metrics and yield data to a user based on stalk locations and stalk diameters.

Abstract: Described herein are systems and methods for capturing images of a field and performing agricultural data analysis of the images. In one embodiment, a computer system for monitoring field operations includes a database for storing agricultural image data including images of at least one stage of crop development that are captured with at least one of an apparatus and a remote sensor moving through a field. The computer includes at least one processing unit that is coupled to the database. The at least one processing unit is configured to execute instructions to analyze the captured images, to determine relevant images that indicate a change in at least one condition of the crop development, and to generate a localized view map layer for viewing the field at the at least one stage of crop development based on at least the relevant captured images.

Abstract: A method and system for decontaminating raw yield maps by combining filters with spatial outlier detectors is provided. In an embodiment, the method comprises receiving over a computer network electronic digital data comprising first yield data representing crop yields harvested from an agricultural field; applying one or more filters to the first yield data to identify, from the first yield data, first outlier data; generating first filtered data from the first yield data by removing the first outlier data from the first yield data; identifying, in the first filtered data, second outlier data representing outlier values based on one or more outlier characteristics; generating second outlier data from the first filtered data by removing the second outlier data from the first filtered data; generating and causing displaying on a mobile computing device a graphical representation of the crop yields harvested from the agricultural field using only the second outlier data.

Abstract: A computer-implemented method for managing crop harvesting activities is implemented by a harvest advisor computing device in communication with a memory. The method includes receiving an initial date of a crop within a field, receiving an initial moisture value associated with the crop and the initial date, and receiving a target harvest moisture value associated with the crop. The method also includes receiving field condition data associated with the field. The method further includes computing, by the harvest advisor, a target harvest date for the crop based at least in part on the initial date, the initial moisture value, the field condition data, and the target harvest moisture value, and displaying the target harvest date for the crop to the grower for harvest planning. The target harvest date indicates a date at which the crop will have a present moisture value approximately equal to the target harvest moisture value.

Abstract: Methods are provided for improving performance of a computing system used to model potential crop yield. In one example embodiment, a computer-implemented method includes generating a model of potential crop yield, as a function of planting date and relative maturity based, at least in part, on one or more relative maturity maps, one or more planting date maps, and one or more actual production history maps, and storing the model in a memory of the server computer system. The method also includes receiving, via an interface at a field manager computing device, a selection of a particular field and computing, from the model of potential crop yield, a potential yield for the particular field based, at least in part, on a planting date for the particular field, a relative maturity value, and values representing actual production history for the particular field.

Abstract: A method comprising while traversing, by a current agricultural implement during a current agricultural operation, a plurality of current geo-referenced locations: receiving, from one or more sensors, mounted in the current agricultural implement, current data at the plurality of current geo-referenced locations, wherein the plurality of current geo-referenced locations includes a plurality of swath locations; dynamically displaying a visual correlation between previous data collected from a previous agricultural operation and the current data, comprising: rendering a bitmap from bitmap values corresponding to the current data, wherein the bitmap includes only those bitmap values that are associated with a threshold number of swath locations; generating a plurality of values of the previous data corresponding to a current geo-referenced location of the current agricultural implement; and displaying, on a display device, the plurality of values of the previous data as visually correlated with the current data

Abstract: Described herein are methods and systems for generating shared collaborative maps for planting or harvesting operations. A method of generating a collaborative shared map between machines includes generating a first map for a first machine based on a first set of data and generating a second map for a second machine based on a second set of data. The method further includes generating at least one shared collaborative map for at least one of the first and second machines based on the first and second maps.

Abstract: Systems and methods for improving the use of precipitation sensors, such as radar and rain gauges, are described herein. In an embodiment, an agricultural intelligence computer system receives one or more digital precipitation records comprising a plurality of digital data values representing precipitation amount at a plurality of locations. The system additionally receives one or more digital forecast records comprising a plurality of digital data values representing precipitation forecasts, each of which comprising predictions of precipitation at a plurality of lead times. The system identifies a plurality of forecast values for a plurality of locations at a particular time, each of the plurality of forecast values corresponding to a different lead time. The system uses the plurality of forecast values to generate a probability of precipitation at each of the plurality of locations.

Abstract: A method for controlling application of agrichemical products, comprises acquiring remotely sensed digital image data; developing a prescription to apply at least one agrichemical product in a variable manner based on at least the digital image data, wherein the prescription describes a plurality of passes of a particular autonomous vehicle over a field to apply the at least one agrichemical product; applying the at least one agrichemical product to a crop in the variable manner by the particular autonomous vehicle according to the prescription.

Abstract: Systems and methods for identifying operational abnormalities based on data received from an agricultural implement performing a task in an agricultural field are described herein. In an embodiment, a system receives time-series data captured from an agricultural implement performing an agronomic activity on an agricultural field, the time-series data including, for each of a plurality of timestamps, a location of the agricultural implement. The system identifies a plurality of passes in the time-series data and using the identified plurality of passes, identifies a plurality of location on the agricultural field in which the activity performed by the agricultural implement included a particular operational abnormality. The system generates a map of operational abnormalities for the agricultural field, the map of operational abnormalities including the plurality of locations on the agricultural field in which the activity performed by the agricultural implement included the particular operational abnormality.

Abstract: Embodiments of the disclosed technologies are capable of inputting, to a machine-learned classifier that has been created using a set of neural network-based models, multi-band digital image data that represents an aerial view of an agricultural field location containing an horticultural product; outputting, by the classifier, annotated image data, the annotated image data comprising annotation data indicative of individual instances of the horticultural product in the agricultural field location; using the annotated image data, computing a first predicted yield for the agricultural field location; adjusting the first predicted yield by a scaling factor to produce a second predicted yield for the agricultural field location.

Abstract: A computer-implemented method for recommending agricultural activities is implemented by an agricultural intelligence computer system in communication with a memory. The method includes receiving, over a network, field definition data and environmental data; determining soil data from at least the environmental data; determining, based on at least the soil data, carbon load data indicating carbon load present in the field during a planting date; receiving first input indicating a plurality of starter application criteria; based on the first input and the carbon load present in the field, generating application data that include at least an amount of the starter fertilizer that provides nutrients to seeds throughout a nutrient immobilization period between the planting date and a beginning of a microorganism mineralization period; generating control instructions for controlling an agricultural implement based on the application data.

Abstract: A computer system and computer-implemented techniques for determining and presenting improved seeding rate recommendations for planting seeds in a field are provided. In an embodiment, a computer-implemented method includes receiving digital data representing planting parameters including seed type information and planting row width, and retrieving a set of seeding models based upon the planting parameters, where each of the seeding models includes a regression model defining a relationship between plant yield and seeding rate on a specific field. The method also includes generating an empirical mixture model as a composite distribution of the set of seeding models, generating a seeding rate distribution for the planting parameters based upon the empirical mixture model, and calculating a seeding rate recommendation based on the seed rate distribution. The method then also includes planting plant seeds in the specific field consistent with the seeding rate recommendation.

Abstract: Systems and methods are provided for generating a plurality of data files, by: displaying a field map corresponding to an agricultural field through a graphical user interface, selecting multiple locations on the field map, identifying, for each selected location, a corresponding data file, generating and displaying a geographic region comprising locations in the identified corresponding data files, and updating the graphical user interface to include a data panel corresponding to the geographic region.

Abstract: Described herein are implements that include a vehicle having a system for sensing or testing soil and/or vegetation as the vehicle traverses a field.

Abstract: Techniques for recommending side-by-side plantings of pairs of seeds include a server computer receiving agricultural data records that represent crop seed data describing seed and yield properties of seeds and first data for agricultural fields where the seeds were planted. The server receives second data for available seeds and automatically calculates a dataset of success probability scores that describe the probability of a successful yield on the target fields. Data is organized as pairs to facilitate comparison of actual plantings to optimized plantings that have a probability of success (POS), in terms of yield lift or increased yield season-over-season, for different yield values. Confidence values are generated and stored in association with the POS values and can be used as a basis of visual output to support planting and/or field management decisions as part of an automated intelligent agricultural decision support system.

Abstract: A server computer system receives a dataset of candidate hybrid seeds for planting on one or more target fields, which comprises data regarding one or more hybrid seeds, probability of success values associated with each of the one or more hybrid seeds that describe a probability of a successful yield, and historical agricultural data associated with each of the one or more hybrid seeds further receives one or more properties for one or more target fields, and selects a subset of the one or more hybrid seeds based on the probabilities of success values.

Abstract: A computer system is provided comprising a classification model management server computer configured, by instructions, to: receive a new image from a user device; apply a first digital model to first regions within the new image for classifying each of the first regions into a particular class; apply a second digital model to second regions within the new image for classifying each of the second regions into a particular class; and transmit classification data related to the class of the first regions and the class of the second regions to the user device. In connection therewith, the second regions each generally correspond to a combination of multiple first regions.

Abstract: A system and processing methods for configuring and utilizing a convolutional neural network (CNN) for plant disease recognition are disclosed. In some embodiments, the system is programmed to collect photos of infected plants or leaves where regions showing symptoms of infecting diseases are marked. Each photo may have multiple marked regions. Depending on how the symptoms are sized or clustered, one marked region may include only one lesion caused by one disease, while another may include multiple, closely-spaced lesions caused by one disease. The system is programmed to determine anchor boxes having distinct aspect ratios from these marked regions for each convolutional layer of a single shot multibox detector (SSD). For certain types of plants, common diseases lead to relatively many aspect ratios, some having relatively extreme values. The system is programmed to then train the SSD using the marked regions and the anchor boxes and apply the SSD to new photos to identify diseased plants.