The main goal for the project is to develop a comprehensive recycling scanner system powered by machine learning that accurately identifies and categorizes various types of plastics, including Multi-Layer Plastics (MLP) and other recyclable materials. The project aims to simplify and enhance recycling processes by providing instant, accurate recycling codes for different commercial products through image recognition, facilitating better sorting, recycling, and management of plastic waste, contributing to environmental sustainability efforts.

Tasks for learners to complete to achieve the project goal:

- Develop a user-friendly digital submission form for uploading photos of various commercial plastic products along with their recycling codes.

- Create a backend system to organize and store the submitted data efficiently for machine learning training purposes.

- Develop a robust material recognition service that utilizes machine learning to analyze images of plastic products and provide accurate recycling information.

- Test the recognition service rigorously to ensure high accuracy and reliability across a wide range of plastic products and packaging types.

Final deliverables:

1. A secure, accessible online submission form for uploading images of plastic products and their recycling codes.

2. A fully functional material recognition service accessible via web or mobile application, offering an intuitive interface for users to scan product images and receive instant feedback on the recycling category and code.

Access to Tools and Technology: We will grant students access to necessary development tools (AWS and etc.), software licenses, and platforms required for building the scanner and submission form. This includes cloud services for data storage and processing, machine learning frameworks, and any specific software needed for image processing and recognition.

Data Provision: Our team will supply the students with initial datasets to kickstart the machine learning model training process. This will include images of various plastics and commercial products along with their corresponding recycling codes. We will also guide them on how to augment this dataset through the digital submission form they will develop.

Collaborative Workspaces: We will facilitate collaborative workspaces, both physical (if geographically feasible) and virtual, to encourage teamwork, brainstorming sessions, and effective project management. This includes access to communication tools like Slack or project management tools like ShortCut.

Feedback and Iteration Support: There will be structured feedback sessions where students can present their progress and receive constructive feedback from both our internal team and external experts in the recycling and technology fields.