Artificial vision parameter learning and automating method for improving visual prosthetic systems

An artificial vision parameter learning and automating system, comprising:

• a sensor configured to take images;
• a communicator configured to be communicated with the sensor;
• a database configured to store the images received from the communicator;
• an image processor configured to convert an image selected from the database into a low-resolution phosphene image, to generate test images with various artificial vision parameters;
• a learning module configured to extract important features from a high-resolution training images, and to train numerous machine learning models to identify the visual stimuli from low-resolution phosphene images with a classifier based on the extracted features;
• an automatic storer configured to automatically select and store primary test data, based on the learning performed by the learning module, and
• a tester configured to perform a human psychophysical experiment,
• wherein the learning module is configured to additionally learning of the machine learning model based on the small-scale human psychophysical experiment in the tester,
• wherein the automatic storer is configured to automatically select and store primary test data, based on the performance of machine learning models from the learning module.