Theoretical Accuracy Prediction and Validation of Low-End and High-End Mobile Mapping System in Urban, Residential and Rural Areas

Mattheuwsen, L.; Bassier, Maarten; Vergauwen, Maarten

doi:10.5194/isprs-archives-xlii-2-w18-121-2019

Cited by 4 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…clouds at 500 kHz. A more detailed summary and analysis of this mobile mapping systems is presented in (Mattheuwsen et al, 2019). The testing dataset is a 2.5 km residential area located north of the city Ghent in Belgium.…”

Section: Datasetmentioning

confidence: 99%

Storm Drain Detection and Localisation on Mobile Lidar Data Using a Pre-Trained Randla-Net Semantic Segmentation Network

Mattheuwsen

Bassier

Vergauwen

2022

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. As the expansion of cities and urban areas results in the construction of more impermeable road surfaces, a well designed urban drainage system becomes of greater importance. However, the accurate and up-to-date mapping of storm drains necessary to create accurate drainage models is often lacking. In recent years, mapping of the road infrastructure is increasingly carried out by highly efficient mobile mapping systems but which lack automatic interpretation of the massive amount data. In this paper we present a fully automatic storm drain detection method to extract and locate storm drain inlets in mobile mapping lidar data. The point cloud is first segmented by a pre-trained RandLa-Net model, which although untrained to segment storm drains, is able the segment storm drain clusters in the hardscape class. The results from this class are further processed by enforcing different requirements to only extract and locate storm drain clusters. Our approach is evaluated on a large testing dataset with 171 storm drains and achieves 81.9%, 95.2% and 88.1% for recall, precision and F1-score respectively. The majority of the false positive and false negative detections are due to incorrect point cloud segmentation of the RandLa-Net. In terms of localisation, our approach achieves an RMSE of 5.5 cm on the centre location while the dimensions of the bounding box are on average 23% off compared to the ground truth.

show abstract

Section: Datasetmentioning

confidence: 99%

Storm Drain Detection and Localisation on Mobile Lidar Data Using a Pre-Trained Randla-Net Semantic Segmentation Network

Mattheuwsen

Bassier

Vergauwen

2022

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…The Lynx system was mounted on the roof of a Mitsubishi SUV driving at a speed of approximately 25-40 km/h during data acquisition. A more detailed summary and analysis of this system is presented in our previous work [30]. As no dedicated training or testing dataset for manhole cover detection is currently publicly available, we captured our own.…”

Section: Training Validation and Testing Datasetsmentioning

confidence: 99%

“…As no dedicated training or testing dataset for manhole cover detection is currently publicly available, we captured our own. The three mobile mapping datasets (rural, residential and urban datasets) from [30] were used to create the training and validation dataset to transfer learn the networks. Using the manhole covers positions in the GRB, the ground truth data were automatically extracted in the ground image.…”

Section: Training Validation and Testing Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

Manhole Cover Detection on Rasterized Mobile Mapping Point Cloud Data Using Transfer Learned Fully Convolutional Neural Networks

Mattheuwsen

Vergauwen

2020

Remote Sensing

View full text Add to dashboard Cite

Large-scale spatial databases contain information of different objects in the public domain and are of great importance for many stakeholders. These data are not only used to inventory the different assets of the public domain but also for project planning, construction design, and to create prediction models for disaster management or transportation. The use of mobile mapping systems instead of traditional surveying techniques for the data acquisition of these datasets is growing. However, while some objects can be (semi)automatically extracted, the mapping of manhole covers is still primarily done manually. In this work, we present a fully automatic manhole cover detection method to extract and accurately determine the position of manhole covers from mobile mapping point cloud data. Our method rasterizes the point cloud data into ground images with three channels: intensity value, minimum height and height variance. These images are processed by a transfer learned fully convolutional neural network to generate the spatial classification map. This map is then fed to a simplified class activation mapping (CAM) location algorithm to predict the center position of each manhole cover. The work assesses the influence of different backbone architectures (AlexNet, VGG-16, Inception-v3 and ResNet-101) and that of the geometric information channels in the ground image when commonly only the intensity channel is used. Our experiments show that the most consistent architecture is VGG-16, achieving a recall, precision and F2-score of 0.973, 0.973 and 0.973, respectively, in terms of detection performance. In terms of location performance, our approach achieves a horizontal 95% confidence interval of 16.5 cm using the VGG-16 architecture.

show abstract

Accuracy Evaluation and Comparison of Mobile Laser Scanning and Mobile Photogrammetry Data

Kalvoda

Nosek

Kuruc

et al. 2020

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

Mobile mapping systems (MMS) are becoming used in standard geodetic tasks more common in the last years. This paper deals with the accuracy evaluation of two types of data acquired by MMS RIEGL VMX-450, and their comparison. The first type is data from mobile laser scanning (MLS). The second type is mobile photogrammetry data. The new high accurate test point field was built in area of Advanced Materials, Structures and Technologies (AdMaS) research centre that is part of Brno University of Technology. Geodetic network and test point field were measured by Trimble R8s GNSS system and Trimble S8 HP total station. The estimate of the 3D standard deviation determined by an adjustment is 2 mm. The accuracy of MLS and mobile photogrammetry data was tested based on the differences between the coordinates of the points determined from the MMS data and determined by before mentioned high precise measurement. The resulting coordinates from photogrammetric data were determined by manual detection of targets in the images. The estimate of the 3D standard deviation is 0.017 m from the MLS data, and 0.061 m from the mobile photogrammetry data. As we supposed, the mobile laser scanning data are significantly more accurate than mobile photogrammetry data. Achieved accuracy of MLS exceeds the original expectations with respect to the GNSS/IMU positioning accuracy, which is according to the manufacturer RIEGL between 0.02-0.05 m. The same scene is often scanned with multiple scanning passes to ensure high quality of the scanned point cloud, therefore we tested the relative accuracy of mobile laser scanning data from two MMS vehicle passes in the same locality of interest. Two different data sets were evaluated, first data set contains points on roads, second data set on buildings. The standard deviation estimate does not exceed 0.008 m and the maximum absolute deviation does not exceed 0.030 m for both data sets. The difference between the two passes is not significant in comparison with the accuracy criteria required for standard mapping purposes. We also compared automatic point cloud production from photogrammetry data processed in Bentley ContextCapture to the point cloud from laser scanning. The MLS data has been used as a reference because it is significantly more accurate as mentioned before. This comparison was done only on the second data set (buildings). The standard deviation estimate is 0.16 m and the maximum absolute deviation is 0.25 m. Our evaluation contains also statistical testing of outliers and stragglers. In contrast to many authors, we don’t use the simplified approach 3σ rule, and in 1D. We use more exact approach using critical values of the statistics for significance levels α = 5 % and α = 1 % to stragglers and outliers test in 3D.

show abstract

Theoretical Accuracy Prediction and Validation of Low-End and High-End Mobile Mapping System in Urban, Residential and Rural Areas

Cited by 4 publications

References 16 publications

Storm Drain Detection and Localisation on Mobile Lidar Data Using a Pre-Trained Randla-Net Semantic Segmentation Network

Storm Drain Detection and Localisation on Mobile Lidar Data Using a Pre-Trained Randla-Net Semantic Segmentation Network

Manhole Cover Detection on Rasterized Mobile Mapping Point Cloud Data Using Transfer Learned Fully Convolutional Neural Networks

Accuracy Evaluation and Comparison of Mobile Laser Scanning and Mobile Photogrammetry Data

Contact Info

Product

Resources

About