Search

Automation of 3D LiDAR point cloud processing is expected to increase the production rate of many applications including automatic map generation. Fast development on high-end hardware has boosted the expansion of deep learning research for 3D classification and segmentation. However, deep learning requires large amount of high quality training samples. The generation of training samples for accurate classification results, especially for airborne point cloud data, is still problematic. Moreover, which customized features should be used best for segmenting airborne point cloud data is still unclear. This paper proposes semi-automatic point cloud labelling and examines the potential of combining different tailor-made features for pointwise semantic segmentation of an airborne point cloud. We implement a Dynamic Graph CNN (DGCNN) approach to classify airborne point cloud data into four land cover classes: bare-land, trees, buildings and roads. The DGCNN architecture is chosen as this network relates two approaches, PointNet and graph CNNs, to exploit the geometric relationships between points. For experiments, we train an airborne point cloud and co-aligned orthophoto of the Surabaya city area of Indonesia to DGCNN using three different tailor-made feature combinations: points with RGB (Red, Green, Blue) color, points with original LiDAR features (Intensity, Return number, Number of returns) so-called IRN, and points with two spectral colors and Intensity (Red, Green, Intensity) so-called RGI. The overall accuracy of the testing area indicates that using RGB information gives the best segmentation results of 81.05% while IRN and RGI gives accuracy values of 76.13%, and 79.81%, respectively.

The effect of particulate matter is increasingly gaining significance due to its harmful effects on human and urban ecosystems. In view of it, many communities worldwide are collecting air quality data privately to influence their policy makers to make stricter provisions for reducing harmful emissions and thereby improving their quality of life. Likewise, in many German cities, a community of air quality monitors which rely on low-cost PM Sensors is gathering momentum. Such communities possess privately-owned & low-cost air quality monitoring devices that claim to accurately measure PM concentrations and are openly accessible via internet. One such initiative is an air quality monitoring network viz. “luftdaten.info”, which contains of more than 300 low-cost sensors that consistently obtains PM data, colloquially referred as fine dust, in the city of Stuttgart as well as its surrounding districts. Besides, eight stations are continuously monitoring PM concentration in Stuttgart; these are operated by the State Environmental Agency (LuBW- Landesanstalt für Umwelt Baden-Württemberg). Stuttgart University of Applied Sciences (HFT) has currently installed 7 low-cost PM sensors to monitor and study PM concentration in one of its projects. This study endeavors to relate PM 2.5 and PM 10.0 using low-cost sensors. It intends to investigate the reliability of the measured PM concentration using such low-costs sensors once these are placed horizontally and vertically apart and comparing the measures of the 7 sensors. Another objective is to compare the PM concentration measurements with a meteorological station operated by the State of Baden-Wuerttemberg in the vicinity. A correlation analysis is performed to develop understanding of relationships of PM concentration with meteorological parameters, viz. with respect to ambient temperature, air pressure, humidity, wind speed and wind direction. Furthermore, it attempts to develop a regression model using above listed meteorological parameters. Finally, deficiencies in the measurement of low-costs and its placement effects are commented. Further suggestions are made for improving the data capturing and analytical procedures while using low-cost sensors.

Digital elevation model serves to illustrate the appearance of the earth's surface. DEM can be produced from a wide variety of data sources including from radar data, LiDAR data, and stereo satellite imagery. Making the LiDAR DEM conducted using point cloud data from LiDAR sensor. Making a DEM from stereo satellite imagery can be done using same temporal or multitemporal stereo satellite imagery. How much the accuracy of DEM generated from multitemporal stereo stellite imagery and LiDAR data is not known with certainty. The study was conducted using LiDAR DEM data and multitemporal stereo satellite imagery DEM. Multitemporal stereo satellite imagery generated semi-automatically by using 3 scene stereo satellite imagery with acquisition 2013–2014. The high value given each of DEM serve as the basis for calculating high accuracy DEM respectively. The results showed the high value differences in the fraction of the meter between LiDAR DEM and multitemporal stereo satellite imagery DEM.