patrick-perkins-3wylDrjxH-E-unsplash.jpg

WECARE

Plants Self-Care System

By Xinjia Pang

We believe that

maintaining plant life in a household is challenging. Different kinds of plants require different amounts of attention, care, water, fertilizer and sunlight to be healthy. But the current products still have a lot of areas that need to be improved and we consider this to be a problem worth solving. If people have better relationships with nature, it could have a positive effect on the environment because plant life is more existent in people's lives.

 

 

HOW MIGHT WE HELP THE AVERAGE HOMEOWNER MAINTAIN THEIR PLANTS TO IMPROVE THE PLANT’S SURVIVABILITY?

 
81LpOUYfPOL._SL1500_.jpg

MARKET RESEARCH

At present, even the high-end central control irrigation system still controls the opening and closing of the valve by a manual input program. There are major problems such as an unreasonable irrigation system and difficult control system management. At the same time, water and fertilizer are wasted, causing soil compaction and groundwater pollution.

There is some research that shows a technology that using the moisture sensor tip, water reservoir, and bottom irrigation opening work together to create a vacuum technology that delivers the precise amount of water necessary for the plant.

 

CONCEPT IDEA

Idea 1
Task Flow

  • Coral dev board / or ML5.JS to quickly prototype on-device ML products.

  • Object Classification Algorithms in ML

  • Soil Moisture Sensor to detect the water level of the soil

  • Photodetector to detect the darkness and brightness environment for the plants

  • Grow LED optimize lights for plant growth

 
CULTURE DESIGN TOOLS.png
 

CULTURAL AI DESIGN TOOL

How it works

  • The products will be automated indoor green plant pots. When people want to add plants for their living space, they can choose to buy plants online or go to a nearby store. We also provide door-to-door delivery or pick-up service.

  • Receive the plant and put it in forint the AI sensor to identify which type the plant is. Then the best pots for the plants to grow will light up, and the user just needs to put the plants into the pot.

  • The system will automatically water plants.

  • Moving the pots toward the sun when it's daytime, and providing growLED at night time.

  • The system helps the user to reuse compost (waste food) and grow the plants. When people finish eating, they just need to put their waste food into a container in the system. The system will process the waste as compost for plants. When plants grow taller, people can use the plans as food materials.

 

ML PROTOTYPE TOOL EXPLORATIONS

YOLACT in Runway ML

Real-time Instance Segmentation

Uses ResNet-101 with FPN (Feature Pyramid Networks) helps in creating pyramids of feature maps of high-resolution images rather than the Conventional Pyramid of Images approach, therefore faster than the other competitive frameworks

Insufficient:

The program is trained to identify several types of plants. The database is limited and cannot support the identification of a large number of varieties.

p5.js

ML5.JS/P5.JS

ML Online Edit

Using MobileNet architecture (proposed by Google) to analyze the images classier.

The database is used to train ML is from Imagenet. The list in the link shows the items can be identified by the program:

https://github.com/ml5js/ml5-library/blob/master/src/utils/IMAGENET_CLASSES.js

Code: https://github.com/CodingTrain/website/blob/master/Courses/beginner_ml5/03_video_classification/sketch.js

Insufficient:

Not accurate enough to easily identify other objects around and cause errors solution in the results.

The program is trained to identify several types of plants. The database is limited and cannot support the identification of a large number of varieties.

 

IMAGE PROJECT IN TEACHABLE MACHINE

Web tool to create machine learning models (no coding required)

Train a computer to recognize your images, then export your model for sites, apps, and more.

Proposed by Google Creative Lab

The models make with Teachable Machine are real Tensorflow.js models that work anywhere javascript runs, so they play nice with tools like Glitch, P5.js, Node.js & more.

Insufficient:

Training by yourself(but easy)

https://teachablemachine.withgoogle.com/models/IYhwSB6P/

 

ML PROTOTYPE

Combination of Teachable Machine with P5.JS. Training my own models on Teachable Machine and export the coding for the model to P5.JS.

Overview

Video

 
Poster.png

POSTER

This project shared resources with IXDSN-300- Object & Space.

Group Member: Xinjia Pang, Ron Alunan, Dayeon Kim