A sewer disease grading method based on entropy power theory

Abstract

Nowadays, urban sewers are becoming more and more dense, and a large number of pipes are inevitably suffering from various kinds of diseases due to their old age and long exposure to a wet environment, which creates various hidden dangers for people’s property safety. The identification of the type and degree of damage to sewers is still mainly manual, which requires a lot of accumulated experience to determine the level of damage and is time-consuming and laborious. To address these problems, we collected and constructed a sewer disease dataset named sewerdata, which contains over 300K images labeled by professionals and contains three common disease categories: crack, corrosion, and misalignment. Meanwhile, a sewer disease grading method based on entropy power theory is proposed. The method designs three disease grading indicators for the characteristics of sewer diseases, and its experimental results show that the detection accuracy is significantly improved compared with the traditional detection method. A disease grading method based on entropy power theory proposed in this paper aims to provide a new scientific research idea in the field of sewer disease grading and detection.