Automated detection of chemical component of moving object

A system comprising:

• one or more processors;
• and
• one or more computer-readable media storing instructions that are operable, when executed by the one or more processors, to cause the system to perform operations comprising:
• obtaining, by the one or more processors, image data that indicates an extent to which one or more objects reflect, scatter, or absorb light at each of multiple wavelength bands, wherein the image data was collected while a conveyor belt was moving the one or more objects;
• preprocessing the image data to generate preprocessed image data, wherein the preprocessing includes performing a first analysis across frequencies and/or performing a second analysis across a representation of a spatial dimension, wherein the preprocessing further includes selecting a subset of the multiple wavelength bands and generating one or more synthetic bands by applying one or more operations to combine the image data from two or more selected bands, wherein each synthetic band is different from the subset of the multiple wavelength bands selected during preprocessing;
• performing an automatic segmentation based on the preprocessed image data, the segmentation determines different types of regions of the one or more objects;
• generating, by the one or more computers, a set of feature values derived from segmented regions of the preprocessed image data;
• generating, by the one or more processors and based on an output produced by a machine learning model in response to the set of feature values being provided as input to the machine learning model, a prediction of an identity of a chemical in the one or more objects or a level of one or more chemicals in the one or more objects, wherein the prediction further includes an identity of contaminant or of a particular type of plastic or a level of contaminant or of the particular type of plastic in the one or more objects, wherein the machine learning model is trained using the set of feature values derived from the segmented regions of the preprocessed image data to classify the one or more objects as recyclable based on the particular type of plastic and the level of contaminant in the one or more objects;
• and
• providing, by the one or more processors, data indicating the prediction of the identity of the chemical in the one or more objects or the level of the one or more chemicals in the one or more objects.

One or more non-transitory computer-readable media storing instructions that are operable, when executed by the one or more computers, to cause the system to perform operations comprising:

• obtaining, by the one or more computers, image data that indicates an extent to which one or more objects reflect, scatter, or absorb light at each of multiple wavelength bands, wherein the image data was collected while a conveyor belt was moving the one or more objects;
• preprocessing the image data to generate preprocessed image data, wherein the preprocessing includes performing a first analysis across frequencies and/or performing a second analysis across a representation of a spatial dimension, wherein the preprocessing further includes selecting a subset of the multiple wavelength bands and generating one or more synthetic bands by applying one or more operations to combine the image data from two or more selected bands, wherein each synthetic band is different from the subset of the multiple wavelength bands selected during preprocessing;
• performing an automatic segmentation based on the preprocessed image data, the segmentation determines different types of regions of the one or more objects;
• generating, by the one or more computers, a set of feature values derived from segmented regions of the preprocessed image data;
• generating, by the one or more computers and based on an output produced by a machine learning model in response to the set of feature values being provided as input to the machine learning model, a prediction of an identity of a chemical in the one or more objects or a level of one or more chemicals in the one or more objects, wherein the prediction further includes an identity of contaminant or of a particular type of plastic or a level of contaminant or of the particular type of plastic in the one or more objects, wherein the machine learning model is trained using the set of feature values derived from the segmented regions of the preprocessed image data to classify the one or more objects as recyclable based on the particular type of plastic and the level of contaminant in the one or more objects;
• and
• providing, by the one or more computers, data indicating the prediction of the identity of the chemical in the one or more objects or the level of the one or more chemicals in the one or more objects.