IDS391 capstone seminar

Project Explanation

1. The Problem

In many public places like mosques, parks, streets, and stations, trash or spilled drinks can appear on the ground.

The problem is that cleaning staff cannot see every area all the time. Because of this:

  • Trash may stay on the ground for a long time
  • Spilled liquids can create bad smells or safety problems
  • People may complain before cleaners notice the problem

So the main issue is slow detection of messes in public spaces.


2. The Idea of the Project

My project is to create an AI camera system that automatically detects trash or spills on the ground.

When the system detects something that needs cleaning, it will send an alert to the cleaning staff so they can clean it quickly.

The goal is to detect problems earlier and improve cleanliness in public spaces.


3. How the System Works (Step by Step)

The system works in several steps.

Step 1 Camera observes the area

A camera is placed in a public space such as:

  • mosque entrance
  • hallway
  • park area
  • transport station

The camera captures images of the ground area.


Step 2 AI analyzes the image

The images are analyzed using computer vision, which is a type of artificial intelligence that allows computers to understand images.

The AI looks for patterns that match trash objects such as:

  • plastic bottles
  • paper
  • food wrappers
  • cups
  • spilled liquids

The AI model learns these patterns by training on many images of trash.


Step 3 Detection

If the AI recognizes trash or a spill in the image, the system marks that area as unclean.

This process is called object detection.

The model identifies:

  • what the object is
  • where it is located in the image

Step 4 Alert system

After detection, the system sends a notification to the cleaning team.

The alert could include:

  • location of the camera
  • image of the detected trash
  • time the problem was detected

This allows cleaners to respond quickly instead of waiting to discover the mess manually.


4. Technology Used in the Project

The project combines several technologies.

Computer Vision

Allows the system to understand images.

Machine Learning

Trains the AI model to recognize trash patterns.

Python Programming

Used to build the AI system.

OpenCV

Used for image processing.

Deep Learning Models

Object detection models such as YOLO may be used to detect trash.


5. Privacy Protection

Because cameras are used, privacy is very important.

The system will follow these rules:

  • focus mainly on ground areas
  • avoid recording faces
  • blur faces if they appear
  • no audio recording
  • follow UAE privacy and CCTV regulations

The goal is to detect trash without collecting personal information.


6. How the Project Will Be Tested

To know if the system works well, it will be evaluated using several measures.

Detection Accuracy

How often the system correctly detects trash.

False Detection Rate

How often the system detects trash when there is none.

Response Time

How fast the system sends alerts.

These results will be compared with traditional manual inspection by cleaning staff.


7. Why the Project Is Important

This project can improve several things:

Cleaner public spaces

Trash is detected faster.

Faster cleaning response

Cleaners receive alerts immediately.

Better hygiene and safety

Spills and litter are removed quickly.

Smart city development

The system supports the UAE vision of using AI and smart technology to improve services.


8. Where It Could Be Used

The system could be used in places such as:

  • mosques
  • parks
  • airports
  • train stations
  • shopping malls
  • university campuses

These places often have many people, which increases the chance of litter appearing.


9. Final Goal of the Project

The final goal is to build a prototype AI system that can:

  1. Detect trash in images
  2. Send alerts automatically
  3. Help cleaners respond faster

If the system works well, it could be expanded in the future to support smart city cleaning systems.

WRITE MY PAPER


Leave a Reply