add README.md

# Lane Detection from Video Clips Using  Binarization and Sliding Window

![](https://img.shields.io/badge/license-MIT-green)![](https://img.shields.io/badge/Anaconda-CPython3.8-blue)![](https://img.shields.io/badge/SUSTech-EE326-%23F2935C)

![lane_wall](assets/lane_wall.jpg)

This repo contains the code and the report document of the final project in SUSTech EE326 digital image processing course (2021 Spring).

## Run the program

### Initialize the conda environment

```shell

cd code/

```

```shell

conda env create -f environment.yml

```

### Replace images

You can find the image you want to test and put it into `input_img` folder as the current format. We recommend using [TuSimple dataset](https://github.com/TuSimple/tusimple-benchmark) as the validation dataset.

### Run the code

```shell

python3 main.py

```

The output with overlaid lane mark will be appeared in `output_img` folder.

## View the report

Checkout the `document` folder.