Note recognition and management using multi-color channel non-marker detection

A computer system configured to:

• receive an input image of an environment having a plurality of physical notes;
• separate the input image into a plurality of channelized input images, each of the channelized input images associated with a different color;
• generate a plurality of transformed channelized input images by applying edge detection to each of the plurality of channelized input images;
• process the plurality of transformed channelized input images to identify, based on edges detected within the transformed channelized input images, a plurality of polygons corresponding to the physical notes within the input image;
• group the plurality of polygons from the plurality of transformed channelized input images into a plurality of groups, each of the groups comprising one or more of the polygons and corresponding to different one of the physical notes;
• and
• select a representative one of the polygons from each of the plurality of groups for each respective physical note in the input image.

A non-transitory computer readable medium comprising instructions that cause a programmable processor to:

• receive an input image of an environment having a plurality of physical notes;
• separate the input image into a plurality of channelized input images, each of the channelized input images associated with a different color;
• generate a plurality of transformed channelized input images by applying edge detection to each of the plurality of channelized input images;
• process the plurality of transformed channelized input images to identify, based on edges detected within the transformed channelized input images, a plurality of polygons corresponding to the physical notes within the input image;
• group the plurality of polygons from the plurality of transformed channelized input images into a plurality of groups, each of the groups comprising one or more of the polygons and corresponding to different one of the physical notes;
• and
• select a representative one of the polygons from each of the plurality of groups for each respective physical note in the input image.

A computing device comprising:

• a processor;
• and
• a note identification module executable on the processor and configured to separate an input image into a plurality of channelized input images, each of the channelized input images associated with a different color,
• wherein the note identification module is configured to apply edge detection and feature extraction to identify polygons within each of the channelized input images, and
• wherein the note identification module is configured to select, from the polygons from the channelized input images, a representative polygon for each of the physical notes in the input image.