what if mosquitoes had attendance? We started with a completely unnecessary question and somehow turned it into a working computer-vision project. This is the story of how we built the Mosquito Attendance System. View GitHub β†— Start the story🦟 Mosquito Attendance TRAAASH Β· TINKERHUB PROJECT What if mosquitoes had attendance? We started with a completely unnecessary question and somehow turned it into a working computer-vision project. This is the story of how we built the Mosquito Attendance System. View GitHub β†— Start the story ↓ Mosquito Attendance System project preview CHAPTER 01 Β· THE SPARK The problem that didn't exist. People take attendance. Students take attendance. Even meetings take attendance. But mosquitoes? Nobody was keeping track of them. So naturally, we decided to fix that. 🦟 The question Who entered the monitored area, who stayed, and who flew away? πŸ‘€ The challenge Mosquitoes are tiny, fast and unpredictable β€” which makes tracking them surprisingly tricky. 🎯 The goal Turn movement in a video into a simple attendance report: Present or Absent. CHAPTER 02 Β· THE BUILD From β€œwhat if?” to β€œit actually works.” We broke the idea into a pipeline: capture the video, find moving objects, follow them, interpret their movement, then show the result in a dashboard. πŸŽ₯ Video β†’ πŸ” Detection β†’ 🧭 Centroid Tracking β†’ πŸšͺ Entry / Exit β†’ πŸ“‹ Attendance β†’ 🌐 Dashboard 01 Β· CAPTURE Give the system a video. The user uploads a mosquito video. The application accepts the video as the raw material for the computer-vision pipeline. 02 Β· DETECT Find movement in the scene. OpenCV background subtraction helps separate moving objects from the background, producing candidate mosquito-like regions. 03 Β· TRACK Give every object a trail. Centroid tracking follows detected objects across frames so the system can reason about movement rather than treating every frame as a brand-new mosquito. 04 Β· INTERPRET Translate movement into attendance. The tracked movement is used to determine whether a mosquito is considered present or absent in the monitored area. 05 Β· PRESENT Turn computer vision into a report. Flask connects the processing logic to a futuristic dashboard with attendance results, counts, search and sorting. CHAPTER 03 Β· THE TOOLBOX Meet the team behind the mosquito chaos. 🐍 Python The main programming language powering the application and detection logic. πŸ‘οΈ OpenCV Handles video processing, background subtraction and computer-vision operations. 🌐 Flask Connects the Python processing layer to the web interface. πŸ”’ NumPy Supports numerical processing used in the computer-vision pipeline. ✨ HTML Β· CSS Β· JS Creates the dark futuristic interface, dashboard interactions and Fun Mode. πŸ™ Git + GitHub Version control, collaboration and hosting for the project. $ git clone https://github.com/sarahbasil03-ctrl/Mosquito_attendence_system.git $ cd Mosquito_attendence_system $ python app.py βœ“ Flask server started βœ“ Open http://localhost:5000/ βœ“ Mosquito attendance dashboard ready 🦟 CHAPTER 04 Β· THE DASHBOARD Where mosquitoes meet spreadsheets. The final interface turns the detection output into something a human can actually understand. 🦟 Mosquito tracking Track detected objects across video frames. βœ“ Attendance report Present / Absent results in a readable table. ↕ Search & sort Quickly find and organize attendance records. Fun Mode Because a serious attendance system needs at least one unserious button. Make it fun 🦟 🦟 CHAPTER 05 Β· RUNNING IT Okay, but how do you actually use it? 1️⃣ Clone Download the project repository from GitHub. 2️⃣ Run Start the Flask application with python app.py. 3️⃣ Observe Open the local dashboard, provide a video and inspect the attendance report. # Clone git clone https://github.com/sarahbasil03-ctrl/Mosquito_attendence_system.git # Enter the project cd Mosquito_attendence_system # Start the application python app.py # Open in your browser http://localhost:5000/ CHAPTER 06 Β· WHAT WE LEARNED The bugs were part of the project. The interesting part wasn't only getting a final screen. It was figuring out how computer vision, tracking logic, a Flask backend and a web interface could work together as one system. 🎯 Detection β‰  tracking Finding an object in one frame is different from understanding where that object goes across many frames. 🧩 Integration matters A working detection algorithm still needs a usable interface and a clear way to communicate its results. πŸ› οΈ Build, test, repeat Project development is an iterative loop: try something, see what breaks, fix it, and try again. CHAPTER 07 Β· NEXT FLIGHT Where could this go next? The current project is a laptop-based prototype. Future versions could explore more robust detection, stronger tracking and real-time deployment. πŸ€– Better detection Explore trained object-detection models to improve identification in difficult scenes. πŸ“Ή Real-time camera Move from uploaded videos toward continuous webcam-based monitoring. ☁️ Deployment Package the application for easier sharing and demonstration beyond localhost. THE PEOPLE Team TRAAASH Sarah Basil Team Lead Β· College of Engineering Chengannur Theertha S B Team Member Β· College of Engineering Chengannur Built at TinkerHub Made with curiosity, code, and an unnecessarily ambitious question. Made with ❀️ by TRAAASH Β· Mosquito Attendance System Β· GitHub β†—