USING MACHINE LEARNING TECHNIQUES TO FILTER VEGETATION IN COLORIZED SFM POINT CLOUDS OF SOIL SURFACES

We present cost-effective photogrammetric systems for high-resolution rockfall monitoring. The paper outlines the components, assembly, and programming codes required. The systems utilize prime cameras to generate 3D models and offer comparable performance to lidar for change detection monitoring. Real-world applications highlight their potential in geohazard monitoring which enables accurate detection of pre-failure deformation and rockfalls with a high temporal resolution.

Low-cost sensors could fill gaps in existing observation networks. To ensure data quality, the quality of the factory calibration of a given sensor has to be evaluated, if the sensor is used out-of-the box. Here, the factory calibration of a widely used low-cost rain gauge type has been tested, both in the lab (66) and in the field (20). Results of the study suggest, that the calibration of this particular type should at least be checked for every sensor, before put into duty.

The rise of new technologies such as drones (unmanned aerial systems – UASs) has allowed widespread use of image velocimetry techniques in place of more traditional, usually slower, methods during hydrometric campaigns. In order to minimize the velocity estimation errors, one must stabilise the acquired videos. In this research, we compare the performance of different UAS video stabilisation tools and provide guidelines for their use in videos with different flight and ground conditions.

Intensified extreme weather events due to climate change can result in changes of soil erosion. These unclear developments make an improvement of soil erosion modelling all the more important. Assuming that soil erosion models cannot keep up with the current data, this work gives an overview of 44 models, their strengths and weaknesses and discusses their potential for further development with respect to new and improved soil and soil erosion assessment techniques.