Method and system of detecting foreign materials within an agricultural product stream
US-2018100810-A1 · Apr 12, 2018 · US
US2021170452A1 · US · A1
| Field | Value |
|---|---|
| Publication number | US-2021170452-A1 |
| Application number | US-201916973928-A |
| Country | US |
| Kind code | A1 |
| Filing date | Jun 10, 2019 |
| Priority date | Jun 11, 2018 |
| Publication date | Jun 10, 2021 |
| Grant date | — |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A process for detecting foreign objects in a food-containing product stream comprises: forwarding the product stream, illuminating the product stream, generating raw data based on electromagnetic energy reflected from the product stream using a camera, and processing the raw data to generate classified image data corresponding with food product, foreign object(s), and background. A system for detecting foreign objects in the product stream comprises a forwarding device, an illuminator, a camera, and instructions in memory that form image data and classify the data as corresponding with food product, foreign objects, and background.
Opening claim text (preview).
1 - 26 . (canceled) 27 . A system for detecting a foreign object present in a product stream comprising a food product, the system comprising: A) a forwarding device configured to forward the product stream; B) an illuminator configured to generate incident electromagnetic energy and to direct the electromagnetic energy against the product stream; C) a camera arranged to generate raw data based on reflected electromagnetic energy from the product stream; D) instructions stored in memory to form image data from the raw data using a machine learning classification algorithm with regression analysis and including unsupervised learning comprising grouping pixels into similar categories by their spectra, in order to generate classified data, wherein the instructions, in response to execution by a processor, cause the processor to: (i) classify a first portion of the raw data as corresponding with the food product; (ii) classify a second portion of the raw data as corresponding with the foreign object; (iii) classify a third portion of the raw data as corresponding with a background which is behind the product stream. 28 . The system according to claim 27 , wherein the forwarding device comprises a conveyor belt. 29 . The system according to claim 28 , wherein the conveyor belt is a first conveyor belt having a first color, and the system further comprises a second conveyor belt upstream of the first conveyor belt, the second conveyor belt having a second color, the second color being different from the first color. 30 . The system according to claim 27 , wherein the camera comprises a hyperspectral camera selected from the group consisting of line scan camera, whisk broom camera, and snapshot camera. 31 . The system according to claim 27 , wherein the illuminator comprises a first illuminator upstream of a field of view of the camera, and a second illuminator downstream of the field of view of the camera, with the first and second illuminators each generating electromagnetic energy and directing it onto the product stream in the field of view of the camera. 32 . The system according to claim 27 , wherein the system comprises covers for the camera and the illuminator so that the system is washable with pressurized water without having liquid contact the camera or the illuminator. 33 . The system according to claim 27 , wherein the camera comprises a plurality of cameras configured to generate the image data at one or more different wavelength regions. 34 . The system according to claim 33 , wherein the plurality of cameras includes a hyperspectral camera and a visible light camera. 35 . The system according to claim 34 , wherein the hyperspectral camera and the visible light camera are proximate to each other in the direction of motion of the forwarding device. 36 . The system according to claim 35 , wherein the hyperspectral camera and the visible light camera are mounted at angles so that the hyperspectral camera and the visible light camera are directed at approximately the same spot on the forwarding device. 37 . The system according to claim 34 , wherein the instructions, in response to execution by the processor, further cause the processor to compensate for an offset in the image data taken by the hyperspectral camera and the visible light camera so that the image data from the hyperspectral camera and the visible light camera can be overlaid into one seamless image.
using video scanning devices, e.g. TV-cameras · CPC title
Validation; Performance evaluation · CPC title
Validation; Performance evaluation; Active pattern learning techniques · CPC title
Illumination specially adapted for pattern recognition, e.g. using gratings · CPC title
Recognition of objects for industrial automation · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.