Identification of Asbestos Slates in Buildings Based on Faster Region-Based Convolutional Neural Network (Faster R-CNN) and Drone-Based Aerial Imagery

Abstract

Asbestos is a class 1 carcinogen, and it has become clear that it harms the human body. Its use has been banned in many countries, and now the investigation and removal of installed asbestos has become a very important social issue. Accordingly, many social costs are expected to occur, and an efficient asbestos investigation method is required. So far, the examination of asbestos slates was performed through visual inspection. With recent advances in deep learning technology, it is possible to distinguish objects by discovering patterns in numerous training data. In this study, we propose the use of drone images and a faster region-based convolutional neural network (Faster R-CNN) to identify asbestos slates in target sites. Furthermore, the locations of detected asbestos slates were estimated using orthoimages and compiled cadastral maps. A total of 91 asbestos slates were detected in the target sites, and 91 locations were estimated from a total of 45 addresses. To verify the estimated locations, an on-site survey was conducted, and the location estimation method obtained an accuracy of 98.9%. The study findings indicate that the proposed method could be a useful research method for identifying asbestos slate roofs.

Enhancing Drones for Law Enforcement and Capacity Monitoring at Open Large Events

“…High-definition cameras are also the most commonly used payload sensor, together with infrared cameras or LIDAR. The processing of images from the payload cameras can be used in missions such as building inspection (i.e., the identification of harmful asbestos slates [8] or the identification of damage to multiple steel surfaces from panorama images [9]), road traffic surveillance [10], and search and rescue [11]. Automatically labelling the images with significant information helps law enforcement agents to detect situations that need to be corrected or to find people that are lost.…”

Selection of Take-Off and Landing Sites for Firefighter Drones in Urban Areas Using a GIS-Based Multi-Criteria Model

“…Hence, the data provided by a Building Ledger include the lot number address and road address, the main building and annex, the area of land and of building, the building-to-land ratio and floor area ratio, the building structure and purpose, the roof structure, as well as the number of parked cars. For complementation of the Building Registry, the building data from the open API of the Building Ledger were incorporated based on the parcel number (PNU) code, a code used for the management of buildings in Republic of Korea alongside the unique feature identifier (UFID) code assigned to each district nationwide in preparation for computerization according to the Cadastral Act [52].…”

“…Drones 2022, 6, x FOR PEER REVIEW 9 of 20 parcel number (PNU) code, a code used for the management of buildings in South Korea alongside the unique feature identifier (UFID) code assigned to each district nationwide in preparation for computerization according to the Cadastral Act [52].…”

Construction of Asbestos Slate Deep-Learning Training-Data Model Based on Drone Images

“…To supplement these weaknesses, object detection techniques using deep learning, which are accurate and reliable, have been widely applied according to the recent development of object recognition techniques. Seo et al [15] targeted a residential area where asbestos slate was widely used, and utilized deep learning, which can automatically detect asbestos. Comparative verification with the traditional visual inspection method enabled the rapid detection of asbestos slate roofs with high accuracy.…”