Object detection

Summery of EdjeElectronics https://github.com/EdjeElectronics/TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi/blob/master/Raspberry_Pi_Guide.md

Prepare Raspberry PI

• Download Raspbian Buster with desktop (without recommanded software)
• enable ssh on image
• Boot with new image
• login over ssh
• change password
• sudo apt-get update -y && sudo apt-get upgrade -y
• enable vnc
• reboot

Install Tensorflow

• git clone https://github.com/EdjeElectronics/TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi.git
• mv TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi tflite1
• cd tflite1
• Create an virtual enviroment
• sudo pip3 install virtualenv
• python3 -m venv tflite1-env
• source tflite1-env/bin/activate
• bash get_pi_requirements.sh

Setting up models

• Using Google
• https://www.tensorflow.org/lite/models/object_detection/overview
  OR
  wget https://storage.googleapis.com/download.tensorflow.org/models/tflite/coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip

https://towardsdatascience.com/simple-surveillance-system-with-the-tensorflow-object-detection-api-125e04d36446