Skip to content

cadrev/corrupt-check

Repository files navigation

Image Segregator

This is a sample script for determining the image quality sent by DOST-ASTI's UrbanFlood Sensor Networks. DOST-ASTI's project on urban flood monitoring involves a tandem of water-level sensor and serial camera to monitor street level flooding. Data from these sensors are sent via GPRS every 5 minutes.

The challenge in archiving the camera images is that sometimes the images are corrupted upon receiving. To segregate between the good and bad images, a simple machine learning model is made to clean the data set worth of 30000 images.

About

Classifying if images are corrupt or not

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages