Skip to content

Latest commit

 

History

History
34 lines (22 loc) · 964 Bytes

File metadata and controls

34 lines (22 loc) · 964 Bytes

Trash Type Detecion Software

Classify trash by computer vision. A ssd-mobilenet based deep learning model and a kivy based GUI is included in this repository. Currently, the software can detect bottles in video frame.

demo

demo

demo2

Installation

pip install -r requirements.txt

Run App

python app/main.py

Capture image without detection

  1. Run App
  2. Click Capture button in window

Training model

Follow this instruction

The notebook for training is here, you can open the notebook in Google Colab: https://github.com/deepdiy/trash-type-detection-software/blob/master/tf_ssd_mobilenet/tensorflow_object_detection_training_colab_trash_detection.ipynb

Use new model

put frozen_inference_graph.pb in /app/model