Soil is a significant natural resource composed of organic and inorganic material. Nitrogen, one of the essential elements, is traditionally measured using laboratory methods. The development of hyperspectral imaging enables the cost-effective acquisition of both spectral and spatial information for detecting physical, chemical, and biological attributes of the soil samples. The presented work evaluates the suitability of airborne hyperspectral imaging for determining soil nitrogen content and producing a soil nitrogen map on a pixel-wise basis. The measurement of spatial variability of the soil nitrogen content was taken at two fields located at Rudice, in northeast Brno, Czech Republic, using laboratory methods and a handheld spectrometer. The soil reflectance was also recorded using airborne-mounted imaging spectroscopy sensors. A partial least squares regression was used to develop a model for the calibration of the data collected with a portable spectrometer and to predict the total nitrogen in the soils based on hyperspectral images from airborne sensors. The determination factor for the PLSR model presented in this paper reached an R² of 0.44. The model’s performance could be improved by using a handheld spectrometer with a wider spectral range, using the same acquisition period for field data collection and hyperspectral imaging, and enlarging the sample size.

Faulting measurements have traditionally been conducted manually by means of fault meters. However, operating any manual device such as a fault meter close to vehicular traffic is hazardous to both the operator and the traveling public. Automated methods, such as those associated with high-speed profilers, offer a safer, more efficient, and cost-effective alternative. Therefore, there is a need to develop an automated method for measuring joint faulting with longitudinal profiles from high-speed profilers. A study was initiated with the primary objective of determining an appropriate profiler sampling interval to accurately locate transverse joints. A secondary objective was to determine how well faulting estimated from profile elevation compares with faulting measured with a Georgia fault meter. An algorithm was developed: it can accurately detect on average 95% of transverse joints from profile data collected at highway speed with a 0.68-in. (17.3-mm) sampling interval. This algorithm was also adapted to estimate faulting measured with a Georgia fault meter (AASHTO R36-04). Although the algorithm results are repeatable, the algorithm overestimated the faulting at joints by 0.05 in. (1.3 mm) to 0.06 in. (1.5 mm) compared with faulting measured with the Georgia fault meter.

Traditionally, the Florida Department of Transportation (DOT) has measured faulting with a manual fault meter. However, this method is slow and labor intensive, disrupts traffic, and presents safety hazards. Extracting the fault magnitude from a pavement profile collected with an automated high-speed inertial profiler is a more efficient and cost-effective alternative. Thus, the Florida DOT developed the Florida automated faulting method (FAFM), which detects joints and calculates faulting from longitudinal profile data. A study was conducted to establish the accuracy and precision of the FAFM. In addition, an improved manual fault meter was developed by the Florida DOT and used as a reference device for the FAFM in the field. Because accuracy and precision measures for the new fault meter were not readily available, the study assessed the accuracy of the fault meter under controlled laboratory conditions, as well as the precision in both laboratory and field conditions. The results of the study indicated that the manual fault meter demonstrated no bias and a repeatability of 0.06 mm (0.002 in.) in laboratory conditions. Under field conditions, the fault meter showed a repeatability of 0.42 mm (0.02 in.). Also under field conditions, the FAFM achieved accuracy in terms of bias ranging between 0.2 mm (0.01 in.) and 0.7 mm (0.03 in.). The repeatability and reproducibility of the FAFM were determined to be 0.6 mm and 0.9 mm (0.04 in.), respectively.

Manual surveying methods have long been used in the field of highway engineering to determine the cross slope and the longitudinal grade of an existing roadway. However, these methods are slow, tedious, and labor intensive. Moreover, manual survey methods almost always require partial or full lane closure, which results in traffic delays, increases in costs, and inconvenience to the traveling public. In 2003, the Florida Department of Transportation acquired a state-of-the-art multipurpose survey vehicle (MPSV) that combines the advantage of an inertial profiler with the additional ability of simultaneously collecting images, cross slope, longitudinal grade, curvature, rutting, and roughness at highway speed. To complement the MPSV system, an automated analysis tool has been developed to process pavement geometry data for identifying areas with deficiencies in cross slope and longitudinal grade in a fast, efficient manner. In addition, this analytical tool computes a drainage path along the pavement surface. The program also displays the analysis results in two- and three-dimensional graphical formats, allowing for an easier interpretation of the data. The MPSV technology coupled with the developed cross-slope analysis tool provides an effective, practical, and cost-effective method to identify quickly potential problem-prone areas along a highway. This paper presents an innovative technique for the analysis of data on cross slope and longitudinal profile that helps identify areas with deficiencies in cross slope and surface drainage. Highway agencies can successfully implement similar systems and analysis methodology to complement and enhance their existing safety and pavement management programs.

Many municipalities and local road authorities seek to implement automated evaluation of road damage. However, they often lack technology, know-how, and funds to afford state-of-the-art data collection equipment for collection and analysis of road deficiencies. The paper describes the development of a localized road damage detection model using the transfer learning method and assessment of its usability for training the detection model from a local road image dataset of a limited size. Localized road damage dataset is created by capturing 3,923 Czech and Slovak road images containing 5,072 instances of detected road damage using a smartphone installed on the vehicle's windshield. Then, a supervised neural network was trained using the road damage dataset labeled by experts. A pre-trained MobileNet model developed by the University of Tokyo and transfer learning method were employed to accelerate the training process and to improve the model's performance when a relatively small, localized dataset is used. Finally, the performance of the developed road damage detection model was analyzed. The results show that it is possible to capture road damage into preset classes with accuracy based on the F1-score ranging between 45% and 98%. Further improvement in the detection rate can be achieved by increasing the training dataset size. The developed road damage detection model is publicly available on https://github.com/amraz39/RoadDamage DetectorCZ and it shows the high potential of employing deep neural networks in the detection of road damage by local road agencies.

Several states have adopted profiler-based systems for automatic measurement of faulting in jointed concrete pavements. However, little published work exists that documents the validation process for these automated faulting systems. An alternative practice for making an initial assessment of a newly developed automated faulting method was documented. Findings from this experiment showed that a high-speed inertial profiler used in conjunction with a faulting reference device provides a practical validation method under controlled conditions. Furthermore, the algorithm that controls the automated faulting measurement method provides reliable and highly repeatable faulting results. The test equipment used in the experiment, as well as the data collection process, the analysis, the subsequent findings and recommendations, are documented.