Detection of disassembled items

A method, comprising:

• identifying, by processing circuitry, in an imaging representation, at least a first segment representing a first material;
• refining, by the processing circuitry, a boundary of the first segment such that the first segment with the refined boundary represents a first object;
• generating, by the processing circuitry, a feature profile for the first object based on the first material and the refined boundary of the first segment;
• generating, by the processing circuitry, using a detection network, in a feature space, an embedding vector for the first object based on the feature profile, wherein the feature space includes a plurality of embedding vectors representing a plurality of objects;
• and
• determining, by the processing circuitry, in the feature space, a plurality of distances between the embedding vector and the plurality of embedding vectors, wherein each of the plurality of distances is indicative of a degree of association of the first object with a corresponding object of the plurality of objects.