Kutluhan Aktar / AI-assisted Pipeline Diagnostics Public

AI-assisted Pipeline Diagnostics

About this project

Project Description

This model detects pipeline diagnostic classes based on the mmWave data parameters extracted from a 60GHz mmWave radar module:

  • Clogged
  • Cracked
  • Leakage

After building my neural network model, I deployed my model as an Arduino library and uploaded it to Nicla Vision. Also, I employed Nicla Vision to capture images of deformed pipes for further examination and communicate with the web application I developed to generate a pre-formatted CSV file from the stored data records and display the model detection results.

home_3.jpg

data_records.s123
data_records.s49
data_records.s23
data_records.s12
data_records.s17
data_records.s70
data_records.s139
data_records.s53

Run this model

Scan QR code or launch in browser

Dataset summary

Data collected
3m 30s
Sensors
p_1, p_2, p_3, p_4, p_5, p_6, p_7 @ 1Hz
Labels
Clogged, Cracked, Leakage

Project info

Project ID 214371
Project version 1
License Apache 2.0
No. of views 24,576
No. of clones 2