INTERNATIONAL JOURNAL OF NOVEL RESEARCH AND DEVELOPMENT International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2456-4184 | Impact factor: 8.76 | ESTD Year: 2016
Scholarly open access journals, Peer-reviewed, and Refereed Journals, Impact factor 8.76 (Calculate by google scholar and Semantic Scholar | AI-Powered Research Tool) , Multidisciplinary, Monthly, Indexing in all major database & Metadata, Citation Generator, Digital Object Identifier(DOI)
Pavement potholes pose a threat to both drivers and pedestrians. In many third world nations, it is a significant contributor to the tragedy that is traffic accidents, which often results in the loss of lives and property. So that vehicles may be alerted to alternative routes and the appropriate public authority can swiftly act to eliminate potholes for the advantage of travelers, it is necessary to regularly gather & refresh information on current road surfaces. Determine the proportion of road damage utilising the given factors: How deep the crater is. The highway has a lot of holes. To scare the government into action, rank the roads in order of importance, depending on which ones need fixing first. After signing up for an account on our hub, the user may submit a picture showing the location of a road along with some other metadata. When we have enough information, we'll create a database table that lists every street name and stores any user-submitted photos of that street. The ML model then makes a prediction of the percentage of damage. When calculating the proportion of damages to each indicated road, we will use an aggregate of the proportions shown in all linked pictures. A proportion of damage typical for that road type will be the end consequence. A straightforward and effective method for locating potholes on roadways is to use object identification techniques on photographs taken with a phone's cam. This project takes as its foundation a model of neural networks that has already been trained on a different set of data and then adapts it to perform the job of classifying road conditions. In order to improve performance and decrease the quantity of labelled data needed, models might transfer what they've learned from one job to another, a technique known as transfer learning. Image classification methods, such as the detection of potholes in road photographs, are ideally adapted for the convolutional network design.
Keywords:
CNN, Deep Learning, SVM, VGG, Pathole Detection
Cite Article:
"Classifying Road Conditions using Transfer Learning and Convolutional Neural Networks for Potholes and Smooth Surfaces ", International Journal of Novel Research and Development (www.ijnrd.org), ISSN:2456-4184, Vol.8, Issue 4, page no.a418-a426, April-2023, Available :http://www.ijnrd.org/papers/IJNRD2304051.pdf
Downloads:
000118746
ISSN:
2456-4184 | IMPACT FACTOR: 8.76 Calculated By Google Scholar| ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 8.76 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Facebook Twitter Instagram LinkedIn