Method and device for determining lane of traveling vehicle by using artificial neural network, and navigation device including same

A lane determination method using a pre-trained lane prediction artificial neural network with driving image information as input information and with lane prediction information of a vehicle and confidence information for the lane prediction information as output information, the lane determination method for a driven vehicle using an artificial neural network, comprising:

• acquiring driving image information of the vehicle from at least one camera module installed in the vehicle;
• calculating a data distribution map of the output information to thereby generate a first data distribution map, wherein the first data distribution map indicates a ratio of predictions corresponding to each of a plurality of lanes among total predictions outputted by the artificial neural network based on the driving image information;
• collecting reference information for actual traveling lane prediction information of the vehicles;
• calculating a data distribution map of the reference information, and thereby generating a second data distribution map, wherein the second data distribution map indicates a ratio of labels corresponding to each of the plurality of lanes among labels of the reference information corresponding to the driving image information;
• and
• updating parameters of the artificial neural network so as to reduce a difference between the first data distribution map and the second data distribution map based on the second data distribution map.

A navigation device comprising:

• a display,
• at least one processor including processing circuitry,
• memory storing instructions that, when executed by the at least one processor individually or collectively, cause the navigation device to:
• acquire driving image information of a vehicle from at least one camera module installed in the vehicle,
• process the driving image information using a pre-trained lane prediction artificial neural network to output lane prediction information of the vehicle and confidence information for the lane prediction information,
• calculate a data distribution map of the output information to thereby generate a first data distribution map, wherein the first data distribution map indicates a ratio of predictions corresponding to each of a plurality of lanes among total predictions outputted by the artificial neural network based on the driving image information,
• collect reference information for actual traveling lane prediction information of the vehicle,
• calculate a data distribution map of the reference information, and thereby generate a second data distribution map, wherein the second data distribution map indicates a ratio of labels corresponding to each of the plurality of lanes among labels of the reference information corresponding to the driving image information,
• update parameters of the artificial neural network so as to reduce a difference between the first data distribution map and the second data distribution map based on the second data distribution map, and
• control the display to display externally the output information outputted from the artificial neural network and driving images of the vehicle together.