Contents

This week, we worked on completing the diagram involved in the re-identification system for stray cats post-TNR. Defined System Overview : Sorted the basic purpose of the application in order to minimize repetitive medical procedures for stray cats and assist better tracking post-TNR. Planned out interaction between administrators , volunteers , animal hospitals and central database. Use Case Diagram : Spotted 19 functional use cases including sign up , upload image , view cat profile , manage ID results. The use case diagram demonstrated actor-function relationships. Group actors are : Volunteers Animal Hospital Staff Administrators ...

Summary of Activities On May 13, our team focused on two major tasks: Finalizing the interview questions for animal shelters and organizations. Mapping responsibilities and categorizing non-functional requirements (NFRs) for each team member, aligned with key stakeholders. Interview Question Design – Animal Shelter Staff We brainstormed and refined a comprehensive list of questions to ask shelter staff in our upcoming interviews. These questions aim to deepen our understanding of real-world workflows, adoption procedures, and the shelter’s technical readiness to collaborate with our system. ...

Summary of Activities On May 6, our group focused on clarifying key system behaviors and preparing in-depth technical questions for our upcoming consultation with the professor. This work session was dedicated to translating our conceptual design into more concrete technical concerns, especially related to cat image handling, recognition, and re-identification processes. Technical Planning and Discussion We discussed the expected photo upload process from general users, particularly from smartphones. While the feature itself was already listed as a requirement, we focused on exploring the technical details and potential user behaviors. We identified the need to accept .jpg images from mobile users and started researching what image formats would be optimal for internal processing and model input. We agreed to seek expert input on whether conversion to other formats (e.g., .webp, .tiff) is recommended for performance or quality reasons. We talked about challenges users might introduce, such as blurry images or inconsistent angles, and decided to ask about data pre-processing techniques to handle these cases effectively. ...

Smart Re-Identification Application for Stray Cats Post-SNR Program Problem Elicitation / Background Urban animal shelters, particularly in regions like Taiwan, are heavily burdened and face legal or ethical restrictions against euthanizing animals. Following the Shelter-Neuter-Return (SNR) program, many cats are released back into the urban environment without systematic tracking, making it difficult to monitor their wellbeing or promote their adoption. There is a growing societal need for a solution that maintains a connection with these cats after their return, fostering animal welfare and enabling post-release care and adoption opportunities. ...

Recognition Application for Cats after SNR Program Problem Statement Urban animal shelters in many regions—such as Taiwan—are overburdened and face legal or ethical restrictions against euthanizing animals. As a result, many cats undergo SNR (Shelter, Neuter, Return) programs and are returned to city streets. Post-release, shelters lose track of these cats, making it difficult to monitor their health or facilitate potential adoptions. There’s a need for a data-driven system that allows society to stay connected to these animals, even after release. ...

Citizen-Based Participatory Budget System (general idea for now, we have a few ideas for data processing, but don’t know which one to choose) Problem Statement Participatory budgeting (PB) empowers citizens to influence how public funds are allocated. However, challenges persist in ensuring fairness, transparency, and effective data utilization within these processes. Traditional methods may not adequately capture diverse community needs or leverage data-driven insights for optimal decision-making.​ Data Acquisition Sources of Data: ...

Ideas: Toilet Finder App It helps users find nearby public restrooms. real-time crowd data, wheelchair accessibility, user reviews, location maps, and elevator info. Parking Finder App It shows available parking spaces near sightseeing spots. real-time availability, price comparison, distance to destination. Community feedback participatory budgetary system Crowd sourcing Use geolocation Monitor progress, suggest, discuss, vote etc Image processing Machine learning based rating system -> Agree or Disagree -> Motivates their transparency Citizen Based Participatory Budgeting System Data Sets from Government website Gather feedbacks from Citizens Analyze feedback to improve the quality and motivate their transparency by being monitored Smart Food Scanner: A Personalised Ingredient Detection App Many international students and residents in Japan, including those who follow halal, vegetarian, vegan diets, or have specific food allergies, often face difficulties identifying suitable food options in convenience stores due to language barriers and unclear ingredient labelling. ...

Ideas: Highway Route Optimization Would optimize the route across an expressway for vehicles to avoid traffic. Low-Light Area Detectors Use location of areas that are not lit up during the night to advise the government where to install street-lighting systems to improve nighttime visibility and increase safety of such low-light areas. Urban Shadow-Tracker Detects public areas with too much and too less shade to predict and provide recommendations for the government to plant trees and install solar panels to reduce public over-exposure to solar radiation. ...

Ideas: Toilet Finder Application This is a smart app feature designed to help users easily and effectively find public toilets in their environment. It’s especially helpful for those with special needs, such as the disabled or even clean and accessible toilet hunters. Here’s how it works: Rush Hour Detection: The application cross-checks the time and location information to find out how busy certain restrooms are. At rush hour, it will suggest less busy ones around so the users don’t have to wait. ...