Using Rasterization to Identify Traffic Signal Devices

A system for representing a traffic signal device in an image, the comprising:

• a processor;
• a memory storing a neural network;
• and
• a computer-readable medium storing programming instructions that are configured to instruct the processor to:
• receive map information that includes one or more traffic signal devices in an environment at a location of a vehicle;
• use the map information and an orientation of the vehicle to generate a raster image of the environment, wherein the raster image includes the one or more traffic signal devices, one of which labeled as a primary traffic signal device;
• and
• provide a real world image of the environment at the location and the raster image as inputs to the neural network to generate a classified real world image which includes a bounding box indicating a location of the primary traffic signal device.

A computer-readable medium storing programming instructions that are configured to instruct a processor to representing a traffic signal device in an image by:

• receiving map information that includes one or more traffic signal devices in an environment at a location of a vehicle;
• using the map information and an orientation of the vehicle to generate a raster image of the environment, wherein the raster image includes the one or more traffic signal devices, one of which labeled as a primary traffic signal device;
• and
• providing a real world image of the environment at the location and the raster image as inputs to a neural network to generate a classified real world image which includes a bounding box indicating a location of the primary traffic signal device.