Adaptive neural network selection to extract particular results

An electronic device comprising:

• an I/O interface configured to receive image data and corresponding metadata;
• and
• a processor operably coupled to the I/O interface, wherein the processor is configured to:
• manipulate the received image data based on a set of transform parameters;
• analyze the manipulated image data, without using ground truth data, to generate metadata predictive of content in the received image data;
• analyze the corresponding metadata and the generated metadata to identify one or more features of the received image data;
• select, based on the identified one or more features, a neural network from a plurality of pre-trained neural networks;
• and then
• analyze perform a second analysis of the received image data using the selected neural network to extract information from the received image data.

A non-transitory computer readable medium embodying a computer program, the computer program comprising computer readable program code that when executed by a processor of an electronic device causes the processor to:

• receive image data and corresponding metadata;
• manipulate the received image data based on a set of transform parameters;
• analyze the manipulated image data, without using ground truth data, to generate metadata predictive of content in the received image data;
• analyze the corresponding metadata and the generated metadata to identify one or more features of the received image data;
• select, based on the identified one or more features, a neural network from a plurality of pre-trained neural networks;
• and then
• analyze the received image data using the selected neural network to extract information from the received image data.