Parametric Extraction of the Road Conditions Spatial Data and Detection of Defeats using Pragmatic Clustering Method
Dara Anitha Kumari1, A. Govardhan2
1Anitha Kumar Dara, Department of Computer Science and Engineering, JNTUH CEH, Hyderabad, Telangana, India .
2Dr.A.Govardhan, Professor of Computer Science & Engineering, Registrar and Executive Council Member, Jawaharlal Nehru Technological University Hyderabad (JNTUH). India.
Manuscript received on May 06, 2020. | Revised Manuscript received on May 15, 2020. | Manuscript published on June 30, 2020. | PP: 1622-1630 | Volume-9 Issue-5, June 2020. | Retrieval Number: C5254029320/2020©BEIESP | DOI: 10.35940/ijeat.C5254.029320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Growth in the population and road transportation for any region can create a higher demand for better road conditions and the less safe road conditions can be a great bottleneck for the growth of the nation. Hence most of the progressive nations like India,builds better infrastructure for the road transportations.Nevertheless,the higher populations in countries like India,demand greater maintenance of the roads and for a gigantic demographic, the maintenance work can be very tedious. Also,the damage in the road surfaces can also lead to the increasing rate of the accidents. Henceforth,the demand of the road maintenance is always increasing.Nevertheless, the existing road maintenanceauthorities deploy a manual process for identification of the repair needs, which is naturally highly time consuming. Thus, this work identifies the demand for automation of the road condition monitoring system and identifies the defects based on three classes as cracks, patch works and potholes on the road surface. The work deploys a novel model for parametric extraction, in order to segregate the defect types. The segregation of the defect types can be highly challenging due to the nature of the data, which clearly hints to solve the problem using unsupervised methods. Thus, this work also deploys a pragmatic clustering method using a decisive factor, which is again generated from the extracted features of parameters. The work demonstrates nearly 96% accuracy on the benchmarked dataset with sophistication on the complexity of the model.
Keywords: Road condition identification, Pragmatic Clustering, Parametric Analysis,intensity Variation, Coordinate Mapping, Segmentation..