A New Image Model for Predicting Cracks in Sewer Pipes based on Time

Abstract

Sewer overflows may cause communities to be vulnerable to various health problems and other monetary losses. This puts a lot of burden on responsible to minimize end user complaints. Therefore, crack prediction would be helpful to facilitate decision makers to control sewer overflow problems and prioritize inspection and rehabilitation needs .The accurate prediction of current underground sewer pipe cracks must be done before any pipe crashing with enough period of time to enable rehabilitation and replacement intervals, appropriate remedial methods and preventing sewer pipes crashing. Unfortunately, traditional technologies and models approaches have been limited to predict the development of sewer pipe cracks. In this paper, we address the problem of crack prediction of such cracks. This paper provides a proposed model which predict crack and cracks developments in any period of time that may occur in weak areas of a network of pipes.. We evaluate our results by comparing them with those obtained by many other models. The accuracy percentage of this model exceeds 90% and outperforms other approaches.

Robust Sewer Defect Detection With Text Analysis Based on Deep Learning

“…However, this approach highly emphasized traditional computer vision techniques and took a long time to perform an operation on an image. Iraky et al [15] developed a Markov-based model to detect defect and used the canny edge algorithm to recognize faults. However, this study was limited to several classes, and the procedure to detect defects was time-consuming.…”