System for automatic stop sign violation identification

A system, comprising:

• one or more processors configured to:
• obtain a plurality of frames of image data, location data, and sensor data associated with a vehicle traveling from a first location to a second location;
• identify a set of frames included in the plurality of frames that include an image of a traffic sign,
• the set of frames comprising a longest sequence of consecutive frames for which a probability that each frame of the set of frames satisfies a probability threshold;
• determine a maximum probability associated with a frame included in the set of frames:
• determine, based on the location data, a series of speeds of the vehicle;
• determine a time delay that, when applied to the location data, results in a maximum correlation between the series of speeds and the sensor data;
• time-align the image data, the location data, and the sensor data based on the time delay;
• determine a time window based on a time associated with the frame associated with the maximum probability;
• obtain a portion of the location data and a portion of the sensor data associated with the time window based on time-aligning the image data, the location data, and the sensor data;
• determine an occurrence of a traffic violation based on the portion of the location data and the portion of the sensor data associated with the time window;
• determine a type of the traffic violation based on a correlation of the portion of the location data and the portion of the sensor data;
• and
• perform one or more actions based on the type of the traffic violation.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors, cause the one or more processors to:
• obtain a plurality of frames of image data, location data, and sensor data associated with a vehicle traveling from a first location to a second location;
• utilize a first machine learning model to identify a set of frames, of the plurality of frames, that include an image of a stop sign,
• the set of frames comprising a longest sequence of consecutive frames for which a probability that each frame of the set of frames satisfies a probability threshold;
• determine a maximum probability associated with a frame of the image data included in the set of frames;
• determine, based on the location data, a series of speeds of the vehicle;
• determine a time delay that, when applied to the location data, results in a maximum correlation between the series of speeds and the sensor data;
• align the location data with the sensor data and the image data based on the time delay;
• determine a time window based on a time associated with the frame of the image data associated with the maximum probability;
• identify a portion of the location data associated with the time window and a portion of the sensor data associated with the time window based on aligning the location data with the sensor data and the image data;
• utilize a second machine learning model to determine an occurrence of a stop sign violation associated with the vehicle traveling from the first location to the second location based on the time window, the portion of the location data associated with the time window, and the portion of the sensor data associated with the time window;
• determine a type of the stop sign violation based on the portion of the location data associated with the time window and the portion of the sensor data associated with the time window;
• and
• perform one or more actions based on the type of the stop sign violation.