Edge Impulse Inc. / Sound Weld Anomaly Detection Public

Target: Cortex-M4F 80MHz Clone this project

Welcome to Edge Impulse, the largest community of edge AI developers!
This is a public Edge Impulse project, use the navigation bar to see all data and models in this project; or clone to retrain or deploy to any edge device.

Sound Weld Anomaly Detection

Using audio data collected by the Belgian Welding Institute, this project uses sound signatures to detect defects during the welding process.

About this project

This public Edge Impulse project does not have a README yet. Clone this project to add new data or retrain this project, or to deploy this project to a device.

Data (130 samples)

View all

normal.FWA4-46-3.json.2f230nru

ID: 125971238

Label: normal

Sample length: 3s

error.FWA4-36-13.json.2f22the2

ID: 125971285

Label: error

Sample length: 6s

error.FWA4-HA-9.json.2f0s3r8g

ID: 125971381

Label: error

Sample length: 18s

error.FWA4-46-11.json.2f22tq3k

ID: 125971275

Label: error

Sample length: 4s

normal.FWA4-36-2.json.2f2301he

ID: 125971247

Label: normal

Sample length: 4s

normal.FWA4-36-3.json.2f0tligr

ID: 125971325

Label: normal

Sample length: 17s

error.FWA4-36-10.json.2f22sn8j

ID: 125971288

Label: error

Sample length: 4s

error.FWA4-STA-5.json.2f22utar

ID: 125971259

Label: error

Sample length: 4s

Model accuracy (Unoptimized (float32))

Validation set

95.2%

Test set

96.8%

On-device performance (Cortex-M4F 80MHz, Quantized (int8))

Latency

75 ms.

Peak RAM usage

42.1K

Flash usage

37.4K

Run this model

On any device

Clone project

Dataset summary

Data collected

24m 34s

Sensors

mic, ac ch1, ac ch2 @ 50KHz

Labels

error, normal

Project info

Project ID	47648
Project version	4
License	BSD 3-Clause Clear
No. of views	15,712
No. of clones	4