Demo Team / Training Models with Synthetic Data: End to End Conveyor Belt Counting Walkthrough with Edge Impulse Public

Target: Arduino Nicla Vision (Cortex-M7 480MHz) 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.

Training Models with Synthetic Data: End to End Conveyor Belt Counting Walkthrough with Edge Impulse

A simple walkthrough showing how a reliable model can be created with just a few images using data synthesis. This video looks at how you could count objects on a conveyor belt using an MCU with object detection and Edge Impulse.

Object detection

About this project

Training Models with Synthetic Data: End to End Conveyor Belt Counting Walkthrough with Edge Impulse

This project was created as part of this Youtube Demo showing an end-to-end example of object counting on a conveyor belt using synthetic data.

The files & BOM for this project can be found here: https://github.com/edgeimpulse/conveyor-counting-data-synthesis-demo

Data (400 items)

View all

new_image389

ID: 184269277

new_image323

ID: 184269343

new_image273

ID: 184269378

new_image354

ID: 184269312

new_image126

ID: 184269508

new_image295

ID: 184269362

new_image344

ID: 184269328

new_image18

ID: 184269469

Model accuracy (Unoptimized (float32))

Validation set

100.0%

Test set

-

On-device performance (Arduino Nicla Vision (Cortex-M7 480MHz), Quantized (int8))

Latency

79 ms.

Peak RAM usage

244.1K

Flash usage

73.5K

Run this model

Scan QR code or launch in browser

Launch in browser

Dataset summary

Data collected

400 items

Labels

m5_nut

Project info

Project ID	187740
Project version	2
License	BSD 3-Clause Clear
No. of views	17,622
No. of clones	35