URBAN STREET CLEANLINESS ASSESSMENT USING MOBILE EDGE COMPUTING AND DEEP LEARNING |
Author(s): |
M. Jafar sathick |
Keywords: |
Convolutional Neural Networks (CNN) for Optical Character Recognition (OCR), Handwritten Text Recognition (HTR), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Natural Language Processing (NLP), Pytesseract, Deep Learning Models, BERT, and Automated Essay Scoring (AES). |
Abstract |
Object detection plays a vital role in computer vision, with numerous applications, including smart city development. City managers invest significant resources in cleaning street garbage due to its unpredictable appearance. As deep learning models grow more intricate, they are often constrained by the availability of training data. To address this, datasets like Open Images, released by OpenCV and Google AI, have been introduced to enhance image analysis at an unprecedented scale, following in the footsteps of PASCAL VOC, ImageNet, and COCO. In this project, we aim to implement the top-performing algorithm for automatic object detection, with a focus on assessing street cleanliness. Existing methods for evaluating street cleanliness face limitations, such as non- automated data collection and lack of real-time updates. By integrating detection results into a street cleanliness assessment framework, we enable city managers to better allocate cleaning resources based on real-time cleanliness data. |
Other Details |
Paper ID: IJSARTV Published in: Volume : 10, Issue : 11 Publication Date: 11/9/2024 |
Article Preview |
Download Article |