Information-client server built on a rapid material identification platform
Items are identified in a waste stream for purposes of recycling, using deterministic and/or probabilistic techniques. Imagery of the waste stream from multiple viewpoints permit creation of a 3D depth draped image representation, from which one or more 2D planes can be synthesized. Phase-coherent patches of recoverable encoded data can be identified from among soiled and crumpled object surfaces, and used in combination to recover object identification information. Recognition of certain items can trigger further image processing that is specific to such items. (Detection of a catsup bottle, for example, can trigger image analysis to discern the presence of catsup residue.) Information about recognized objects can be provided to external data customers, e.g., to track grey market diversion of particular products into unlicensed territories. These and other features and advantages, which can be used alone or in combination, are detailed herein.
1. A method comprising the acts:
(a) identifying an object in a waste stream conveyed on a conveyor, by processing imagery captured under illumination provided by a light source, said light source producing illumination in one or more different spectral bands;
(b) identifying a region within the imagery that depicts an area associated with the object, and associating an identifier with said region;
(c) identifying an analysis algorithm associated with said identifier; and
(d) applying said identified analysis algorithm to said identified region.
2. The method of claim 1 wherein the analysis algorithm is a contamination detection algorithm.
3. The method of claim 1 wherein the analysis algorithm extracts serialization data encoded in the imagery.
4. A system comprising:
a camera;
a conveyor that conveys containers past the camera;
a processing system configured to identify a container on the conveyor from information collected by the camera and to launch a contaminant detection algorithm for a contaminant product or closure that is particularly associated with the identified container.
5. The system of claim 4 in which said processing system is further configured to execute said contaminant detection algorithm.
6. A method comprising the acts:
capturing imagery depicting items in a waste stream;
analyzing the imagery to produce an item identifier for an item, and to identify an excerpt of image data that depicts said item;
choosing which of plural contamination detection algorithms should be applied to said item, said determining including accessing a data structure with said item identifier;
applying the chosen contamination detection algorithm to said excerpt of image data; and
sorting said item from the waste stream based, in part, on data produced by the chosen contamination detection algorithm.
7. The method of claim 6 that includes choosing two contamination detection algorithms that should be applied to said item, a first of said algorithms detecting a contaminant that is water-soluble, and a second of said algorithms detecting a contaminant that is not water-soluble, and sorting said item from the waste stream based, in part, on data produced by said first and second algorithms.
8. The method of claim 6 in which the chosen contamination detection algorithm is tuned to a class of contamination, and is also tuned to a class of plastic used in said item.
9. The method of claim 6 in which the chosen contamination detection algorithm produces a contamination score for the item, said score being based at least in part on a spatial extent of contamination detected in said excerpt, an apparent transparency of the item depicted in said excerpt, or a saturation of a color depicted in said excerpt.