Home

Awesome

Tin House Detection from Satellite/Aerial Images

TensorFlow 2.x

Detection of tin houses from satellite/aerial images, trained using the Tensorflow 2 Object Detection API.

Satellite images are rich with information about people and the Earth. The latest advancements in deep learning and object detection has allowed very accurate analysis of satellite images. This project is a proof-of-concept of how satellite images can be applied to understand rural Bangladesh better. The project have a lot of potential in revealing various insights about villages in Bangladesh.

<img src="https://github.com/yasserius/satellite_image_tinhouse_detector/blob/main/images/poster.png?raw=true" height=300px>

Demo

The Jupyter notebook for inference using the pretrained model: [Github] | [Colab]

Dataset

How to train

The Jupyter notebook for training: [Github] | [Colab]

The notebook contains detailed instructions and a link to the data. The training was done using the Tensorflow 2 Object Detection API.

Wrong Detections

See all 128 test output images: Google Drive

Missed detections

<img src="https://github.com/yasserius/satellite_image_tinhouse_detector/blob/main/images/bad_1.png?raw=true"> <img src="https://github.com/yasserius/satellite_image_tinhouse_detector/blob/main/images/bad_3.png?raw=true"> <img src="https://github.com/yasserius/satellite_image_tinhouse_detector/blob/main/images/bad_4.png?raw=true">

Double counting the same house

<img src="https://github.com/yasserius/satellite_image_tinhouse_detector/blob/main/images/bad_2.png?raw=true">

False positive

<img src="https://github.com/yasserius/satellite_image_tinhouse_detector/blob/main/images/bad_5.png?raw=true">

How to improve detection