Method and apparatus for detecting road lane

A device for detecting a road lane, comprising:

• a processor, and
• a memory, wherein
• the memory is configured to store a program code and transmit the program code to the processor;
• and
• the processor is configured to execute, in response to an instruction in the program code, the following steps:
• acquiring a current road image of a road in proximity of a vehicle;
• and
• inputting the current road image into a deep learning model and detecting a road lane region in the current road image based on a result outputted from the deep learning model,
• wherein the deep learning model is a model that has been trained based on correspondences between historical road images and known road lane regions in the historical road images,
• wherein the deep learning model comprises a first model device and a second model device, and an output of the first model device serves as an input of the second model device,
• wherein the first model device comprises at least one first model subdevice, the first model subdevice comprises a convolutional neural network and a first recurrent neural network, and an output of the convolutional neural network serves as an input of the first recurrent neural network, and
• wherein the second model device comprises at least one second model subdevice, the second model subdevice comprises a deconvolution neural network and a second recurrent neural network, and an output of the deconvolution neural network serves as an input of the second recurrent neural network.

A storage medium that stores a program code for executing the following steps:

• acquiring a current road image of a road around a vehicle;
• and
• inputting the current road image into a deep learning model and detecting a road lane region in the current road image based on a result outputted from the deep learning model,
• wherein the deep learning model is a model that has been trained based on correspondences between historical road images and known road lane regions in the historical road images,
• wherein the deep learning model comprises a first model device and a second model device, and an output of the first model device serves as an input of the second model device,
• wherein the first model device comprises at least one first model subdevice, the first model subdevice comprises a convolutional neural network and a first recurrent neural network, and an output of the convolutional neural network serves as an input of the first recurrent neural network, and
• wherein the second model device comprises at least one second model subdevice, the second model subdevice comprises a deconvolution neural network and a second recurrent neural network, and an output of the deconvolution neural network serves as an input of the second recurrent neural network.