Commit 24538ac8 authored by Sandeep Fernando's avatar Sandeep Fernando

add README

parents
Pipeline #3009 failed with stages
# 2021-102
**GARBAGE DETECTION & CLASSIFICATION FOR MANAGEMENT & DISPOSAL**
**Main Objective**
*Our main goal is to make a mobile application to detect garbage with evidently,share garbage maps and schedule tasks to make the waste cleaning and disposing processes efficient.*
Our research project aims for a web-based solution for detection and classification of garbage/waste gathered unnecessarily in a region to make the garbage reporting, collecting, disposing and recycling tasks efficient. This benefits the personals of the direct users like waste collectors, recyclers of waste and the citizens living in the region. This encompasses the mechanisms for identifying garbage using CCTV cameras, client reporting, mapping the spots of interest and spots reported by clients using a regional map, alerting of new spots discovered in benefit of waste collectors to take actions and mechanisms related to notifying garbage reported has being collected successfully for the usage of customer. We hope to use image processing, machine learning technologies, and publisher/subscriber architectures in implementation of this solution.
**Main Research Problem**
Certainly, ignorance and disregard for the effects of waste underlie our wasteful practices. Over usage and mismanagement plays a big role to periodically discourage garbage disposal. In an urban environment, we can agree on the fact that people are not taking care of their surroundings. The authorities that are responsible for the cleaning process, is unable to finish their job to a satisfactory level when they follow the current established systems. In under developed nations, uncollected waste is typically heaviest near less affluent neighborhoods and slums. For scavengers, discarded food in the heaps of trash provides sustenance and a livelihood.
**Individual Research Questions**
*A.C.S. Fernando (IT18047066)*
- Machine Learning based garbage detection with the mobile and our camera feeds as inputs.
*B.R.M.N. Wijewardana (IT18017366)*
- Distribution of garbage map between clients and staff with peer-to-peer network.
*A.K.G.A.D. Jayasooriya (IT18397840)*
- Using Image processing Identify the human and vehicle number plates for Filtering and censoring the private data in results taken from visual inputs before sending them to the map.
*R.D.M.L.Y.I.B. Dissanayake (IT18534078)*
- Task scheduling via algorithm based on available, nearest staff based on the
nearest targets.
**Individual Objectives**
*A.C.S. Fernando (IT18047066)*
- Implementation of the function of garbage detection using python and ML. Import camera feeds to gather location details, times recorded and duration of occurrence details in which benefit the task scheduling functions also to train learning.
*B.R.M.N. Wijewardana (IT18017366)*
- Implementation of a peer-to-peer network to connect mobile users/citizens enabling communication with them. Distribution of the proof of garbage dump with data collected from citizens and make them available to review and testify with other peers.
*A.K.G.A.D. Jayasooriya (IT18397840)*
- Use image processing for human and vehicle number plate detection to develop a function that takes the inputs from client’s camera feeds then sensor and filter private data in real-time and by labeling obstacles making the ML resilient. Needs to implement a filtering system for exclude the privacy concerned data like vehicle number plates and faces of people that being captured.
*R.D.M.L.Y.I.B. Dissanayake (IT18534078)*
- Implementation of a task scheduling function based on priority scheduling algorithms and the factors like staff availability of citizens and monitor the processes.
**Technologies to be Used**
- Machine Learning
- Image Processing
- TensorFlow
- Flutter/Dart
- Python
- MongoDB
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment