Image Object Detection

Total Dataset : 800 Images
Type of Data : extruder, spaghetti

Precision:

This release contains:

Name	Type	Size
Image Training Model	NPY File	600 Windows
Image Training Labels	NPY File	600 Windows
Object Detection Model	TensorFlow Lite [float32]	11 MB
Object Detection Model	TensorFlow Lite [int8 quantized]	4 MB
Object Detection Model	TensorFlow SavedModel	10 MB
Compiled C++ Library	C++ SDK	8 MB
Dataset	RAW Images with labels	73 MB

Image Classification

Total Dataset : 710 images
Type of data : none, normal

Accuracy:

This dataset contains:

• RAW Images
• Sony Spresense Firmware

Audio predictive maintenance

Total dataset : 22m 30s

Type of data : background, fanNoise, fanDisturbance

Type	Time [m, s]
background	6m 40s
fanNoise	9m 10s
fanDisturbance	6m 40s
Total Dataset	22m 30s

Accuracy:

This release contains:

• Audio Dataset in WAV
• C++ Library
• Sony Spresense firmware