Skip to content

ariesiitr/Lane_detection_model

Repository files navigation

Lane Detection Using Deep Learing

The aim is to build a machine learning model to distinguish the drivable lane road from the obstacles and background in the active road image.

Dataset

BDD100K dataset is used as reference (BDD100K -https://bdd-data.berkeley.edu/). The label images of BDD100K dataset were modified in such a way that, there are only three colors (pink,blue and black) in the label representing lane area,obstacles and background respectively.

Library Used

TensorFlow
PyTorch
Matplotlib
Numpy
Pandas

Unet Model Architecture

The architecture consists of two main parts: the contracting path and the expanding path.
The contracting path is a series of convolutional and pooling layers that gradually reduce the spatial resolution of the input image while increasing the number of feature channels.
Once the contracting path has extracted the features, the expanding path uses a series of upsampling and convolutional layers to restore the spatial resolution of the image while decreasing the number of feature channels.
In addition to the contracting and expanding paths, UNet also includes skip connections that directly connect the corresponding feature maps from the contracting and expanding paths. These skip connections allow the network to preserve spatial information from the contracting path, which can be important for accurate segmentation. image

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors