Property analysis using sensors

A system comprising:

• a sensor system comprising sensors, wherein the sensor systems is configured to generate sensor data representing one or more molecular properties of a physical object, the sensor system comprising:
• a plurality of sensors, each tuned to measure electromagnetic radiation in a particular frequency band;
• one or more radiation emitters, each configured to emit electromagnetic radiation in a different frequency band towards the physical object;
• one or more signal processors configured to generate sensor data indicating an intensity and a frequency of electromagnetic radiation detected by an associated detector;
• and
• a network interface configured to transmit the sensor data to computing systems;
• a computer system comprising one or more processors, wherein the computer system is configured to:
• control the sensor system to emit electromagnetic radiation;
• receive the sensor data from the sensor system;
• determine, using the one or more processors, molecular properties of the physical object by applying the sensor data as input to a machine learning algorithm trained to correlate patterns of electromagnetic intensity and frequency measurements from the sensors with molecular compositions from a library of known radiation patters for elements and molecules;
• store, in a memory of the computer system, the molecular properties of the physical object and the sensor data as a signature of the physical object;
• obtain, from the memory of the computer system, test model data representing one or more molecular properties of a test model object;
• generate, using the one or more processors and based on the sensor data and the test model data, a comparison between the molecular properties of the physical object and the one or more molecular properties of the test model object, the comparison comprising a statistical representation of how closely the molecular properties of the physical object compare with molecular properties of the test model object contained in the test model data;
• and
• determine an authenticity of the physical object based whether the statistical representation indicates a molecular match greater than or equal to a predefined threshold value;
• automatically generate, based at least in part on the authenticity of the physical object, an insurance policy for physical object;
• send the insurance policy to a computing device associated with an insurance agent or owner of the physical object;
• and
• adjusting, by a policy generator, an insurance policy associated with the physical object.