irenexychen / gpr-data-classifier

:radio: Proof-of-concept project for detecting and classifying hyperbolas in ground penetrating radar (GPR) data.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

gpr-data-classifier

Proof-of-concept side project for Sensors & Software Inc.

Background Information

What is a hyperbola?

Hyperbolas are artefacts and evidences of disturbances in the soil, collected by ground penetrating radars. Below is a classic hyperbola:

And here are some not so obvious hyperbolas:

Classifier implemented with MobileNet

Mobilenet was chosen for its potential for real-time detection of hyperbolas in ground penetrating data. To read more about the comparison process between the suitability of different pre-trained models, refer to the (WIP) report.

About

:radio: Proof-of-concept project for detecting and classifying hyperbolas in ground penetrating radar (GPR) data.


Languages

Language:Python 91.2%Language:TeX 5.4%Language:Jupyter Notebook 2.1%Language:Shell 1.1%Language:Makefile 0.2%