Brainchip / Akida Image Classification

This is the finished Edge Impulse project for the tutorial 'Adding sight to your sensors'. From here you can acquire new training data, design impulses and train models.

Images

About this project

Lamps and Plants and Bears, oh my!

Plant classification OpenMV

Want to make the most of your device's camera and predict what is/is not present in your device's environment? Clone this project to build an embedded ML project to add intelligent sight to your device.

You can also follow our tutorial to guide you through building your image classification model, from data collection to deployment on embedded devices.

Sensor & Block Information

Image data
Image DSP block for normalizing image files
Transfer Learning block with prediction outputs: "lamp", "plant", "unknown"

Download block output

Title	Type	Size
Image training data	NPY file	221 windows
Image training labels	NPY file	221 windows
Image testing data	NPY file	58 windows
Image testing labels	NPY file	58 windows
Transfer learning with Akida model	TensorFlow Lite (float32)	3 MB
Transfer learning with Akida model	TensorFlow Lite (int8 quantized)	1 MB
Transfer learning with Akida model	MetaTF	1024 KB
Transfer learning with Akida model	TensorFlow SavedModel	3 MB
Transfer learning with Akida model	Keras h5 model	3 MB

Clone project

You are viewing a public Edge Impulse project. Clone this project to add data or make changes.

Summary

Data collected

289 items

Project info

Project ID	115634
Project version	6
License	Apache 2.0