Lane line recognition method, electronic device and storage medium

An electronic device comprising:

• a storage device;
• at least one processor;
• and
• the storage device storing one or more computer instructions, which when executed by the at least one processor, cause the at least one processor to:
• obtain a first foreground image while a vehicle is travelling, and converting the first foreground image into a bird's eye view (BEV) image;
• determine a first initial position of a left lane line and a second initial position of a right lane line of the BEV image;
• slide a first sliding window on the BEV image from the first initial position, and fit a first curve and a second curve after sliding the first sliding window each time, the first curve being fitted by a plurality of non-zero pixels covered by the first sliding window before arriving at a current position of the BEV image, the second curve being fitted by the first curve and a plurality of non-zero pixels covered by the first sliding window at the current position of the BEV image;
• slide a second sliding window on the BEV image from the second initial position, and fit a third curve and a fourth curve after sliding the second sliding window each time, the third curve being fitted by a plurality of non-zero pixels covered by the second sliding window before arriving at the current position of the BEV image, the fourth curve being fitted by the third curve and a plurality of non-zero pixels covered by the second sliding window at the current position of the BEV image;
• and
• recognize the left lane line according to the second curve and the right lane line according to the fourth curve.

A non-transitory storage medium having instructions stored thereon, when the instructions are executed by a processor of an electronic device, cause the processor to perform a lane line recognition method, the method comprising:

• obtaining a first foreground image while a vehicle is travelling, and converting the first foreground image into a bird's eye view (BEV) image;
• determining a first initial position of a left lane line and a second initial position of a right lane line of the BEV image;
• sliding a first sliding window on the BEV image from the first initial position, and fitting a first curve and a second curve after sliding the first sliding window each time, the first curve being fitted by a plurality of non-zero pixels covered by the first sliding window before arriving at a current position of the BEV image, the second curve being fitted by the first curve and a plurality of non-zero pixels covered by the first sliding window at the current position of the BEV image;
• sliding a second sliding window on the BEV image from the second initial position, and fitting a third curve and a fourth curve after sliding the second sliding window each time, the third curve being fitted by a plurality of non-zero pixels covered by the second sliding window before arriving at the current position of the BEV image, the fourth curve being fitted by the third curve and a plurality of non-zero pixels covered by the second sliding window at the current position of the BEV image;
• and
• recognizing the left lane line according to the second curve and the right lane line according to the fourth curve.