Image-Based Dry and Wet Waste Recognition with Location-Aware Disposal Guidance: A Deep Learning Web Platform |
Author(s): |
| Vijay Kumar S , ATME College of Engineering; Harshitha H B, ATME College of Engineering; Yeshashwini Bhandari K R, ATME College of Engineering |
Keywords: |
| Waste Classification, Convolutional Neural Network, Tensorflow, Geospatial Analysis, Dustbin Recommendation, Flask, Sustainable Waste Management |
Abstract |
|
Rapid urbanisation has increased the volume of municipal waste, while manual segregation at source remains slow, inconsistent and error prone. Citizens also lack immediate guidance on how to dispose of an item and where the nearest suitable bin is. This paper presents Binify ECO Route, a web platform that combines a convolutional neural network (CNN) for image-based waste recognition with a GPS-based dustbin recommendation module. A user uploads or captures a photograph of waste; the image is pre-processed and classified by a TensorFlow model, which returns the waste category, confidence scores, the recommended bin colour and disposal instructions. The platform then reads the user’s location, filters a JSON dustbin database by accepted waste type, computes geodesic distances with the Geopy library and lists the nearest suitable bins on an interactive map. A knowledge centre and sign-in module complete the application. The paper describes the architecture, implementation and test plan, shows the working interfaces, and discusses limitations and future work such as sensor-based bin monitoring and wider waste categories. |
Other Details |
|
Paper ID: IJSRDV14I90062 Published in: Volume : 14, Issue : 9 Publication Date: 01/12/2026 Page(s): 136-139 |
Article Preview |
|
|
|
|
