1. Pothole Visual Detection using Machine Learning Method integrated with Internet of Thing Video Streaming Platform
- Author
-
Alfandino Rasyid, Rizqi Putri Nourma Budiarti, Mochammad Rifki Ulil Albaab, Alviansyah Arman Yusuf, Yohanes Yohanie Fridelin Panduman, Hendro Wicaksono, Dwi Kurnia Basuki, Muhammad Fajrul Falah, Anang Tjahjono, Sritrusta Sukaridhoto, and Firman Yudianto
- Subjects
030506 rehabilitation ,Computer science ,business.industry ,media_common.quotation_subject ,Real-time computing ,Process (computing) ,030229 sport sciences ,Hazard (computer architecture) ,Object detection ,Portable computer ,03 medical and health sciences ,0302 clinical medicine ,ComputerSystemsOrganization_MISCELLANEOUS ,Global Positioning System ,Pothole ,Wireless ,Quality (business) ,0305 other medical science ,business ,media_common - Abstract
A good condition of road was very needed by the people, because the movement of goods and services, and also a lot of people activities is indirectly depending on the road condition. So for keeps the road on good quality, early detection of road damage, especially pothole must be held. This thing is very important because a small damage that is not immediately handled can be large, so that the danger will decreased. It also make the cost that needed for repair it become greater. In the other hand, detecting road damage cannot be done with a manual checking, because it will take a lot of time and money. So supporting technology is needed to detect this kind of roads hazard. One of the technology easily used to detect road damage is computer vision. In this research, we build a system that can detect a pothole on the road which is captured by the camera. The camera used here is a wireless portable camera. Also for the location tagging, the GPS sensor is used here. The vision object detection system using imageZMQ library for stream the frames and process it in the processor PC. The capturing and streaming activity performed very well from the mini portable computer camera which is attached in the vehicle. the next step of this research is making the detection performed well in the more robust computational device.
- Published
- 2019
- Full Text
- View/download PDF