dc.contributor.advisor | Eftestøl, Trygve Christian | |
dc.contributor.advisor | Vevle, Geir | |
dc.contributor.author | Holmesland, Christoffer René Haaland | |
dc.date.accessioned | 2021-09-29T16:26:43Z | |
dc.date.available | 2021-09-29T16:26:43Z | |
dc.date.issued | 2021 | |
dc.identifier | no.uis:inspera:73533758:14549297 | |
dc.identifier.uri | https://hdl.handle.net/11250/2786177 | |
dc.description.abstract | In this thesis, the goal has been to improve the data flow from the garbage trucks in Halden municipality. We begin by designing, building, and installing a data capture unit to collect images of the collected waste. Machine learning models were trained to analyze the images and detect whether customers use green bags to dispose of their organic waste. Models were also trained on data from a time of flight camera to measure the volume of the waste.
We discover that high accuracy object detection is possible on organic waste. Limitations on the use of time of flight technology are found when it is used in a garbage collection environment. The result is that volume measurement is not possible unless the environment changes. | |
dc.description.abstract | | |
dc.language | eng | |
dc.publisher | uis | |
dc.title | Digital Waste Management - detection technology | |
dc.type | Master thesis | |