International Research Journal of Engineering and Technology (IRJET) Volume: 09 Issue: 05 | May 2022
www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
SMART WASTE MANAGEMENT SYSTEM Shraddha Golande1, Kaushal Purohit2, Aalok Marathe3, Anuja Ghotekar4, Ms. Sonam Borhade5 1,2,3,4 B.Tech
student, Computer Science, Sinhgad Institute of Technology and Science , Pune, India Professor, Computer Science, Sinhgad Institute of Technology and Science , Pune, India ---------------------------------------------------------------------***--------------------------------------------------------------------5B.Tech
Abstract - Current waste management systems are
data can then be used to study and assess the limitations of the existing waste management system and hence improve the system’s efficiency as a whole. The Internet of Things application in the waste bin is one step towards a smart city. With the help of Internet of Things, the workers can get to know the actual fill level of each compartment of the segregated waste, which will further help them in the management phase and also cut the management costs considerably. The waste can also be segregated in solid and wet waste. With the help of Internet of Things one can also record the number of times the bin has been emptied.
inefficient and costly. Electronic sensors can be used to efficiently manage the waste and significantly reduce the cost of man labor. In this project raspberry Pi is used along with servo motors and conveyor belt. TensorFlow is used to build the waste object detection model. Waste will be carried by the conveyor belt and servo motors will turn the flaps in order to segregate the waste into solid and wet waste.
Key Words: Object detection, waste management system, TensorFlow
2. Requirements
1. INTRODUCTION
Following are the requirements for the smart waste management system.
2.01 billion tons of municipal waste is being generated worldwide every year. It is projected to increase at least by 70% to reach 3.4 billion tons by 2050 due to rapid urbanization, population growth and economic development unless urgent measures are not taken. Segregation at the source level is critical to its recycling and disposal. Lack of segregation, collection and transportation of unsegregated mixed waste being dumped in the landfills has an impact on the environment. During such times reducing man labor and automating the tasks of segregation and collection to increase the efficiency can be helpful and this project helps in achieving the same. It helps in reducing the labor cost as well as helps in segregating the waste in the form of solid and wet waste.
Moisture sensor
●
Servo motors
●
Conveyor belt
●
Pi camera (5MP or more)
The camera module is connected to Raspberry Pi to capture the waste image for the purpose of object detection and identification. Waste is dumped on the conveyor belt and the raspberry pi detects it. After this the waste will come in front of the moisture detector and then the detector will check whether there is moisture present or not. After the waste is detected, we will carry out classification with the help of SVM (Support Vector Machine) algorithm. And then after applying CNN (Convolution neural network) model we can detect the waste and determine whether it is solid or wet waste. Waste detected is added to the database by calling the api. Then the flaps will turn accordingly to dump the waste in the specified bin. If there is moisture present in the waste ,then that particular waste will be dropped off at the wet waste compartment. Speed of the conveyor is to be considered with reference to the time required by the object detection model to identify the waste.
Advancements in the field of Internet of Things have made it possible to improve the existing waste management system. Implementation of the sensors in the waste bin along with Internet of Things connectivity allows for the real-time monitoring of the waste being collected, which is absent in the existing waste management system. Data such as the type of waste, and any necessary data can be collected from the sensors. This data can then be uploaded to the cloud for storage and processing. The processed
Impact Factor value: 7.529
●
3.1 IMPLEMENTATION
The Internet of Things helps in acknowledging the objects in the real world with the help of cameras, sensors and various softwares who have the ability to process it over the internet or other networks used in communication.
|
Raspberry Pi 4B (4GB ram or more)
3. METHODOLOGY
1.1 INTERNET OF THINGS
© 2022, IRJET
●
|
ISO 9001:2008 Certified Journal
|
Page 861