Apeel Technology, Inc.

Described herein are systems and methods for determining quality levels for food items using image data, such as time lapse RGB, hyperspectral, thermal, and/or multispectral images. The method can include receiving, from imaging devices, image data of food items, performing object detection on the image data to identify a bounding box around each food item, and identifying a quality level of each food item by applying trained models to the bounding boxes. The models were trained using image training data of other food items that was annotated based on previous identifications of a first portion of the other food items as having poor quality features and a second portion as having good quality features. The other food items and the food items are a same type. The method also includes determining, for each food item, a quality level score based on the identified quality level of the food item.

Presented are compositions that can be used as antimicrobial coatings or protective coatings for agricultural (e.g., food) products. The compositions can comprise one or more saturated glycerides selected from monoglycerides and diglycerides; one or more surfactants; and one or more food-safe antimicrobials. The antimicrobial coatings or protective coatings can be used to prevent or delay food spoilage due to, for instance, infection by a foreign pathogen, moisture loss, or oxidation.

The disclosed technology provides for determining infection in food items using image data of the food items under ultra-violet (UV) light. A method includes performing object detection on the image data to identify a bounding box around each of the food items in the image data, determining, for each food item, an infection presence metric by applying a model to the bounding box, the model being trained using image training data of other food items under UV light, the image training data being annotated based on previous identifications of a first portion of the other food items having infection features and a second portion having healthy quality features, and determining, based on a determination that the infection presence metric for each of the food items indicates presence of an infection, an infection coverage metric for the food item.

A treatment system, including a conveyor bed configured to transport a plurality of products; a brushing device located along the conveyor bed and having one or more brushes that include a first antimicrobial compound; a surface treatment device including a lamp configured to emit light of a peak wavelength directed toward a portion of the conveyor bed; and a coating device configured to deliver a coating mixture onto the plurality of products on the conveyor bed.

Presented are compositions that can be used as protective coatings for agricultural (e.g., food) substrates. The compositions can comprise a compound of Formula I: and an additive, wherein the variables m, n, q, r, R a , R b , R 1 , R 2 , R 3 , R 4 , R 5 , R 6 , R 7 , R 8 , R 9 , R 10 , R 11 , R 12 and R 13 are defined herein. The protective coatings formed from the compositions can be used to prevent food spoilage due to, for instance, moisture loss, oxidation, or infection by a foreign pathogen.

The present disclosure relates to protective coatings e.g., on agricultural products, that can include a glycerophospholipid bilayer structure formed on the surface of the agricultural product.

A method for predicting a likelihood of infection in a set of similarly sourced plant products is disclosed. A subset of plant products is selected from the set of plant products. For each plant product in the subset, a level of expression of one or more infection biomarkers, and optionally a level of expression of one more housekeeping biomarkers, are determined. A set of biomarker expression statistics for the subset of plant products is determined based on the determined levels of expression of the one or more infection biomarkers and optionally the levels of expression of the one or more housekeeping biomarkers for each plant product in the subset. A likelihood of infection in the set of plant products is then predicted based at least in part on the determined set of biomarker expression statistics for the subset of plant products.

A produce treatment method to mitigate latent infections, such as fungal infections, and to delay ripening in produce is described. The produce treatment method includes providing an antimicrobial agent to an enclosure comprising a plurality of produce items, and contacting the plurality of produce items with the antimicrobial agent, wherein the antimicrobial agent is in gaseous form and is selected to deactivate latent microbes in the plurality of produce items.

Systems and methods described herein can non-destructively determine current produce quality. An evaluation system can, at a first time, receive image data of produce and identify a type of the produce using machine learning techniques. The produce can be sent to an end-consumer retail environment. A quality assessment system can, at a second time, receive, from a user device in the retail environment, a request for a current quality of a particular produce that includes scanned data of that produce. The quality assessment system can identify that produce, retrieve, from a data store, machine learning trained models for identifying quality features of that produce, determine, based on applying the models to the scanned data, a current quality of that produce, and transmit, to the user device, the current quality. The produce's price can also be dynamically adjusted based on the current quality of that produce.

A computer implemented method for reducing an amount of memory required to store hyperspectral images of an object include: obtaining tensor data representing a hyperspectral image including a first portion depicting an object and a second portion depicting at least a portion of a surrounding environment where the object is located; identifying a portion of the tensor data representing the hyperspectral image that corresponds to the first portion; providing the identified portion of the tensor data representing the hyperspectral image as an input to a feature extraction model; obtaining one or more matrix structures as output by the feature extraction model based on the feature extraction model processing the identified portion of the tensor data, the one or more matrix structures representing a subset of features extracted from the identified portion of the tensor data; and storing the one or more matrix structures in a memory device.