Back to results
US 2020/0367422 A1Published application
Systems and methods for sorting of seeds
AssigneeSeedX Technologies Inc.
InventorsMordekhay SHNIBERG, Elad CARMON, Sarel ASHKENAZY, David GEDALYAHO VAISBERGER, Sharon AYAL
Life of this patent
Estimating the term…
Abstract
A system for sorting seeds based on their resistance to a stress is disclosed. Batches of purified seeds sorted using the system are also disclosed.
Claims31
A system for sorting of seeds, comprising:
- at least one hardware processor for executing a code, the code comprising:
- code for inputting into at least one neural network, at least one image including a plurality of seeds which have statistically similar extractable at least one visual feature, the at least one image captured by at least one imaging sensor,
- wherein the at least one visual feature extracted from an image of one of the plurality of seeds are statistically similar to corresponding at least one visual feature extracted from another image of another seed of the plurality of seeds,
- code for computing by the at least one neural network, an indication of one classification category for which visual features are not explicitly defined, for each of the plurality of seeds selected from the group consisting of:
- stress resistant, and non-stress resistant,
- wherein the indication of at least one classification category is computed at least according to weights of the at least one neural network,
- wherein the at least one neural network classifies the plurality of seeds which have similar extractable at least one visual feature into one classification category selected from the group consisting of:
- stress resistant, and non-stress resistant, for which visual features are not explicitly defined,
- wherein the at least one neural network is trained according to a training dataset comprising a plurality of training images of a plurality of seeds which have statistically similar extractable at least one visual feature captured by the at least one imaging sensor, wherein each respective training image of the plurality of training images is associated with an indication of at least one classification category for which visual features are not explicitly defined of at least one seed depicted in the respective training image selected from the group consisting of:
- stress resistant, and non-stress resistant;
- and
- code for generating according to the indication of at least one classification category selected from the group consisting of:
- stress resistant, and non-stress resistant, instructions for execution by a sorting controller of an automated sorting device for automated sorting of seeds.
10 more elements
2-3. (canceled)
7-10. (canceled)
13-15. (canceled)
17-21. (canceled)
23-26. (canceled)
(canceled)
32-33. (canceled)
(canceled)
(canceled)
A system for training at least one neural network for sorting of seeds, comprising:
- at least one hardware processor executing a code, the code comprising:
- code for accessing a training dataset comprising a plurality of training images of a plurality of seeds which have statistically similar extractable at least one visual feature captured by at least one imaging sensor, wherein each respective training image of the plurality of training images is associated with an indication of at least one classification category for which visual features are not explicitly defined of at least one seed depicted in the respective training image selected from the group consisting of:
- stress resistant and stress non-resistant,
- wherein the at least one visual feature extracted from an image of one of the plurality of seeds are statistically similar to corresponding at least one visual feature extracted from another image of another seed of the plurality of seeds;
- and
- code for training at least one neural network according to the training dataset, the at least one neural network trained for computing an indication of one classification category for which visual features are not explicitly defined, selected from the group consisting of:
- stress resistant and stress non-resistant according to at least one target image comprising at least one seed captured by at least one imaging sensor,
- wherein the indication of at least one classification category of the at least one target image is computed at least according to weights of the at least one trained neural network,
- wherein the at least one neural network classifies the plurality of seeds which have similar extractable at least one visual feature into one classification category selected from the group consisting of:
- stress resistant and stress non-resistant, for which visual features are not explicitly defined.
7 more elements
A container comprising a plurality of seeds, and at least one member selected from the group consisting of:
- (i) wherein at least 90% of the seeds are stress-resistant seeds, (ii) wherein said plurality of seeds comprises more than 1000 seeds, and (iii) wherein said plurality of seeds weights more than 100 grams.
42-43. (canceled)
45-46. (canceled)
Dependencies are inferred from the claim text.