Sökning: "Outlier removal"
Visar resultat 1 - 5 av 7 uppsatser innehållade orden Outlier removal.
1. Comparing the effect of random and contextual removal of images on object detection performance
Kandidat-uppsats, Högskolan i Skövde/Institutionen för informationsteknologiSammanfattning : As datasets grow, the need for automated methods to ensure dataset quality arises. This report presents an experiment conducted on the MSCOCO train2017 dataset to identify image outliers using a force-directed graph built from a co-occurrence context, focusing on the mean average precision and average precision. LÄS MER
2. Modelling the full-scale N2O emissions from wastewater treatment plants for identifying mitigation strategies
Master-uppsats, Lunds universitet/Kemiteknik (CI)Sammanfattning : Nitrous oxide (N2O) emissions, which primarily originate from the biological nitrogen removal process, dominate the release of greenhouse gases (GHGs) in wastewater treatment plants (WWTPs). N2O production occurs through dynamic microbial pathways that have a significant impact on the environment compared with other GHG emissions. LÄS MER
3. LiDAR-Equipped Wireless Sensor Network for Speed Detection on Classification Yards
Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Varje dag kopplas tusentals godsvagnar om på de olika rangerbangårdarna i Sverige. För att kunna automatiskt bromsa vagnarna tillräckligt mycket är det nödvändigt att veta deras hastigheter. En teknik som har blivit populär på sistone är Light Detection and Ranging (LiDAR) som använder ljus för att mäta avstånd till objekt. LÄS MER
4. Maps element prediction from raw time-series position data using machine learning techniques
Master-uppsats, Uppsala universitet/Institutionen för informationsteknologiSammanfattning : Having geopositional tracking data of users moving in the work space, we would like to detect map elements such as paths, zones of interest and borders and to create abroad map of the working area from this time series raw data. In order to do so, an experiment to generate data was conducted to emulate workers moving around thework-space. LÄS MER
5. Automated error matching system using machine learning and data clustering : Evaluating unsupervised learning methods for categorizing error types, capturing bugs, and detecting outliers.
Master-uppsats, Linköpings universitet/Programvara och systemSammanfattning : For large and complex software systems, it is a time-consuming process to manually inspect error logs produced from the test suites of such systems. Whether it is for identifyingabnormal faults, or finding bugs; it is a process that limits development progress, and requires experience. LÄS MER