Node-red Machine Learning module Tensorflow in Raspberry Pi4

Purpose:

這個專題想利用Node-Red結合USBCamera分析所得到的影像照片中物件所屬的類別, 而TensorFlow剛好可以符合所要的影片類別分析的部分. 再來就是在Node-Red的圖形介面中把所要的功能呈現出來.

This topic wants to use Node-Red combined with USBCamera to analyze the categories of objects in the images and photos obtained, and TensorFlow can just meet the required video category analysis. The next step is to present the required functions in the graphical interface of Node-Red.

Fundamental:

Node-RED

Node-RED is a programming tool for wiring together hardware devices, APIs and online services in new and interesting ways. It provides a browser-based editor that makes it easy to wire together flows using the wide range of nodes in the palette that can be deployed to its runtime in a single-click.
Node-RED 是 IBM 以 Node.js 為基礎, 開發出來的視覺化 IOT 開發工具, 透過「流程圖」方式的操作,Node-RED 完成許多後端才能做的事情.
Node-RED is a visual IOT development tool developed by IBM based on Node.js. Through the operation of "flow chart", Node-RED can accomplish many things that can only be done in the back end.

TensorFlow

TensorFlow is open source library for Machine Learning. It provides a platform that can be used to develop and train ML(Machine Learning)  models and integrate them into applications.

This is a custom Node-RED node that handles the prediction results of an Object Detection model.

Node-Red  [manage palette]:

1. node-red-dashboard

2. node-red-contrib-ui-webcam

3. node-red-contrib-tensorflow

4. node-red-contrib-image-output

5. node-red-contrib-image-tools

6. node-red-contrib-browser-utils

YouTube Demonstration: