Search Results (11)

The rapid evolution of unmanned aerial vehicle (UAV) technology and low-altitude economic development have propelled drone applications in critical infrastructure monitoring, particularly in intelligent transportation systems where real-time aerial image processing has emerged as a pressing requirement. We address the pivotal challenge of highway lane extraction from low-altitude UAV perspectives by applying a novel three-stage framework. This framework consists of (1) a deep learning-based semantic segmentation module that employs an enhanced STDC network with boundary-aware loss for precise detection of roads and lane markings; (2) an optimized polynomial fitting algorithm incorporating iterative classification and adaptive error thresholds to achieve robust lane marking consolidation; and (3) a global optimization module designed for context-aware lane generation. Our methodology demonstrates superior performance with 94.11% F1-score and 93.84% IoU, effectively bridging the technical gap in UAV-based lane extraction while establishing a reliable foundation for advanced traffic monitoring applications. Full article

Pedestrians, as the most vulnerable road users in traffic crashes, prompt transportation researchers and urban planners to prioritize pedestrian safety due to the elevated risk and growing incidence of injuries and fatalities. Thorough pedestrian crash data are indispensable for safety research, as the most detailed descriptions of crash scenes and pedestrian actions are typically found in crash narratives and diagrams. However, extracting and analyzing this information from police crash reports poses significant challenges. This study tackles these issues by introducing innovative image-processing techniques to analyze crash diagrams. By employing cutting-edge technological methods, the research aims to uncover and extract hidden features from pedestrian crash data in Michigan, thereby enhancing the understanding and prevention of such incidents. This study evaluates the effectiveness of three Convolutional Neural Network (CNN) architectures—VGG-19, AlexNet, and ResNet-50—in classifying multiple hidden features in pedestrian crash diagrams. These features include intersection type (three-leg or four-leg), road type (divided or undivided), the presence of marked crosswalk (yes or no), intersection angle (skewed or unskewed), the presence of Michigan left turn (yes or no), and the presence of nearby residentials (yes or no). The research utilizes the 2020–2023 Michigan UD-10 pedestrian crash reports, comprising 5437 pedestrian crash diagrams for large urbanized areas and 609 for rural areas. The CNNs underwent comprehensive evaluation using various metrics, including accuracy and F1-score, to assess their capacity for reliably classifying multiple pedestrian crash features. The results reveal that AlexNet consistently surpasses other models, attaining the highest accuracy and F1-score. This highlights the critical importance of choosing the appropriate architecture for crash diagram analysis, particularly in the context of pedestrian safety. These outcomes are critical for minimizing errors in image classification, especially in transportation safety studies. In addition to evaluating model performance, computational efficiency was also considered. In this regard, AlexNet emerged as the most efficient model. This understanding is precious in situations where there are limitations on computing resources. This study contributes novel insights to pedestrian safety research by leveraging image processing technology, and highlights CNNs’ potential use in detecting concealed pedestrian crash patterns. The results lay the groundwork for future research, and offer promise in supporting safety initiatives and facilitating countermeasures’ development for researchers, planners, engineers, and agencies. Full article

High-definition (HD) maps for autonomous driving rely on data from mobile mapping systems (MMS), but the high cost of MMS sensors has led researchers to explore cheaper alternatives like low-cost LiDAR sensors. While cost effective, these sensors produce sparser point clouds, leading to poor feature representation and degraded performance in deep learning techniques, such as convolutional neural networks (CNN), for tasks like road marking extraction and classification, which are essential for HD map generation. Examining common image segmentation workflows and the structure of U-Net, a CNN, reveals a source of performance loss in the succession of resizing operations, which further diminishes the already poorly represented features. Addressing this, we propose improving U-Net’s ability to extract and classify road markings from sparse-point-cloud-derived images by introducing a learnable resizer (LR) at the input stage and learnable resizer blocks (LRBs) throughout the network, thereby mitigating feature and localization degradation from resizing operations in the deep learning framework. Additionally, we incorporate Laplacian filters (LFs) to better manage activations along feature boundaries. Our analysis demonstrates significant improvements, with F1-scores increasing from below 20% to above 75%, showing the effectiveness of our approach in improving road marking extraction and classification from sparse-point-cloud-derived imagery. Full article

The documentation of roadway factors (such as roadway geometry, lane marking retroreflectivity/classification, and lane width) through the inventory of lane markings can reduce accidents and facilitate road safety analyses. Typically, lane marking inventory is established using either imagery or Light Detection and Ranging (LiDAR) data collected by mobile mapping systems (MMS). However, it is important to consider the strengths and weaknesses of both camera and LiDAR units when establishing lane marking inventory. Images may be susceptible to weather and lighting conditions, and lane marking might be obstructed by neighboring traffic. They also lack 3D and intensity information, although color information is available. On the other hand, LiDAR data are not affected by adverse weather and lighting conditions, and they have minimal occlusions. Moreover, LiDAR data provide 3D and intensity information. Considering the complementary characteristics of camera and LiDAR units, an image-aided LiDAR framework would be highly advantageous for lane marking inventory. In this context, an image-aided LiDAR framework means that the lane markings generated from one modality (i.e., either an image or LiDAR) are enhanced by those derived from the other one (i.e., either imagery or LiDAR). In addition, a reporting mechanism that can handle multi-modal datasets from different MMS sensors is necessary for the visualization of inventory results. This study proposes an image-aided LiDAR lane marking inventory framework that can handle up to five lanes per driving direction, as well as multiple imaging and LiDAR sensors onboard an MMS. The framework utilizes lane markings extracted from images to improve LiDAR-based extraction. Thereafter, intensity profiles and lane width estimates can be derived using the image-aided LiDAR lane markings. Finally, imagery/LiDAR data, intensity profiles, and lane width estimates can be visualized through a web portal that has been developed in this study. For the performance evaluation of the proposed framework, lane markings obtained through LiDAR-based, image-based, and image-aided LiDAR approaches are compared against manually established ones. The evaluation demonstrates that the proposed framework effectively compensates for the omission errors in the LiDAR-based extraction, as evidenced by an increase in the recall from 87.6% to 91.6%. Full article

Lane detection plays a pivotal role in the successful implementation of Advanced Driver Assistance Systems (ADASs), which are essential for detecting the road’s lane markings and determining the vehicle’s position, thereby influencing subsequent decision making. However, current deep learning-based lane detection methods encounter challenges. Firstly, the on-board hardware limitations necessitate an exceptionally fast prediction speed for the lane detection method. Secondly, improvements are required for effective lane detection in complex scenarios. This paper addresses these issues by enhancing the row-anchor-based lane detection method. The Transformer encoder–decoder structure is leveraged as the row classification enhances the model’s capability to extract global features and detect lane lines in intricate environments. The Feature-aligned Pyramid Network (FaPN) structure serves as an auxiliary branch, complemented by a novel structural loss with expectation loss, further refining the method’s accuracy. The experimental results demonstrate our method’s commendable accuracy and real-time performance, achieving a rapid prediction speed of 129 FPS (the single prediction time of the model on RTX3080 is 15.72 ms) and a 96.16% accuracy on the Tusimple dataset—a 3.32% improvement compared to the baseline method. Full article

This research aimed to develop a system for classifying horizontal road signs as correct or with poor visibility. In Poland, road markings are applied by using a specialized white, reflective paint and require periodic repainting. Our developed system is designed to assist in the decision-making process regarding the need for repainting. It operates by analyzing images captured by a standard car camera or driving recorder. The image data undergo initial segmentation and classification processes, facilitated by the utilization of the YOLOv4-Tiny neural network model. The input data to the network consist of frames extracted from the video stream. To train the model, we established our proprietary database, which comprises 6250 annotated images and video frames captured during driving. The annotations provide detailed information about object types, their locations within the image, and their sizes. The trained neural network model effectively identifies and classifies objects within our dataset. Subsequently, based on the classification results, the identified image fragments are subjected to further analysis. The analysis relies on assessing pixel-level contrasts within the images. Notably, the road surface is intentionally designed to be dark, while road signs exhibit relatively lighter colors. In conclusion, the developed system serves the purpose of determining the correctness or visibility quality of horizontal road signs. It achieves this by leveraging computer vision techniques, deep learning with YOLOv4-Tiny, and a meticulously curated database. Ultimately, the system provides valuable information regarding the condition of specific horizontal road signs, aiding in the decision-making process regarding potential repainting needs. Full article

This study aims to assess the potential of Sentinel-2 NDVI time series and Google Earth Engine to detect small land-use/land-cover changes (at the pixel level) in fire-disturbed environs. To capture both slow and fast changes, the investigations focused on the analysis of trends in NDVI time series, selected because they are extensively used for the assessment of post-fire dynamics mainly linked to the monitoring of vegetation recovery and fire resilience. The area considered for this study is the central–southern part of the Italian peninsula, in particular the regions of (i) Campania, (ii) Basilicata, (iii) Calabria, (iv) Toscana, (v) Umbria, and (vi) Lazio. For each fire considered, the study covered the period from the year after the event to the present. The multi-temporal analysis was performed using two main data processing steps (i) linear regression to extract NDVI trends and enhance changes over time and (ii) random forest classification to capture and categorize the various changes. The analysis allowed us to identify changes occurred in the selected case study areas and to understand and evaluate the trend indicators that mark a change in land use/land cover. In particular, different types of changes were identified: (i) woodland felling, (ii) remaking of paths and roads, and (ii) transition from wooded area to cultivated field. The reliability of the changes identified was assessed and confirmed by the high multi-temporal resolution offered by Google Earth. Results of this comparison highlighted that the overall accuracy of the classification was higher than 0.86. Full article

Light detection and range (LiDAR) intensity is an important feature describing the characteristics of a target. The direct use of original intensity values has limitations for users, because the same objects may have different spectra, while different objects may have similar spectra in the overlapping regions of airborne LiDAR intensity data. The incidence angle and range constitute the geometric configuration of the airborne measurement system, which has an important influence on the LiDAR intensity. Considering positional shift and rotation angle deviation of the laser scanner and the inertial measurement unit (IMU), a new method for calculating the incident angle is presented based on the rigorous geometric measurement model for airborne LiDAR. The improved approach was applied to experimental intensity data of two forms from a RIEGL laser scanner system mounted on a manned aerial platform. The results showed that the variation coefficient of the intensity values after correction in homogeneous regions is lower than that obtained before correction. The overall classification accuracy of the corrected intensity data of the first form (amplitude) is significantly improved by 30.01%, and the overall classification accuracy of the corrected intensity data of second form (reflectance) increased by 18.21%. The results suggest that the correction method is applicable to other airborne LiDAR systems. Corrected intensity values can be better used for classification, especially in more refined target recognition scenarios, such as road mark extraction and forest monitoring. This study provides useful guidance for the development of future LiDAR data processing systems. Full article

Land use and land cover (LULC) are diverse and complex in urban areas. Remotely sensed images are commonly used for land cover classification but hardly identifies urban land use and functional areas because of the semantic gap (i.e., different definitions of similar or identical buildings). Social media data, “marks” left by people using mobile phones, have great potential to overcome this semantic gap. Multisource remote sensing data are also expected to be useful in distinguishing different LULC types. This study examined the capability of combined multisource remote sensing images and social media data in urban LULC classification. Multisource remote sensing images included a Chinese ZiYuan-3 (ZY-3) high-resolution image, a Landsat 8 Operational Land Imager (OLI) multispectral image, and a Sentinel-1A synthetic aperture radar (SAR) image. Social media data consisted of the hourly spatial distribution of WeChat users, which is a ubiquitous messaging and payment platform in China. LULC was classified into 10 types, namely, vegetation, bare land, road, water, urban village, greenhouses, residential, commercial, industrial, and educational buildings. A method that integrates object-based image analysis, decision trees, and random forests was used for LULC classification. The overall accuracy and kappa value attained by the combination of multisource remote sensing images and WeChat data were 87.55% and 0.84, respectively. They further improved to 91.55% and 0.89, respectively, by integrating the textural and spatial features extracted from the ZY-3 image. The ZY-3 high-resolution image was essential for urban LULC classification because it is necessary for the accurate delineation of land parcels. The addition of Landsat 8 OLI, Sentinel-1A SAR, or WeChat data also made an irreplaceable contribution to the classification of different LULC types. The Landsat 8 OLI image helped distinguish between the urban village, residential buildings, commercial buildings, and roads, while the Sentinel-1A SAR data reduced the confusion between commercial buildings, greenhouses, and water. Rendering the spatial and temporal dynamics of population density, the WeChat data improved the classification accuracies of an urban village, greenhouses, and commercial buildings. Full article

The objective to fast-track the mapping and registration of large numbers of unrecorded land rights globally has led to the experimental application of Artificial Intelligence in the domain of land administration, and specifically the application of automated visual cognition techniques for cadastral mapping tasks. In this research, we applied and compared the ability of rule-based systems within Object-Based Image Analysis (OBIA), as opposed to human analysis, to extract visible cadastral boundaries from very high-resolution World View-2 images, in both rural and urban settings. From our experiments, machine-based techniques were able to automatically delineate a good proportion of rural parcels with explicit polygons where the correctness of the automatically extracted boundaries was 47.4% against 74.24% for humans and the completeness of 45% for the machine compared to 70.4% for humans. On the contrary, in the urban area, automatic results were counterintuitive: even though urban plots and buildings are clearly marked with visible features such as fences, roads and tacitly perceptible to eyes, automation resulted in geometrically and topologically poorly structured data. Thus, these could neither be geometrically compared with human digitisation, nor actual cadastral data from the field. The results of this study provide an updated snapshot with regards to the performance of contemporary machine-driven feature extraction techniques compared to conventional manual digitising. In our methodology, using an iterative approach of segmentation and classification, we demonstrated how to overcome the weaknesses of having undesirable segments due to intra-parcel and inter-parcel variability, when using segmentation approaches for cadastral feature delineation. We also demonstrated how we can easily implement a geometric comparison framework within the Esri’s ArcGIS software environment and firmly believe the developed methodology can be reproduced. Full article

Automated processing of the data provided by a laser-based mobile mapping system will be a necessity due to the huge amount of data produced. In the future, vehiclebased laser scanning, here called mobile mapping, should see considerable use for road environment modelling. Since the geometry of the scanning and point density is different from airborne laser scanning, new algorithms are needed for information extraction. In this paper, we propose automatic methods for classifying the road marking and kerbstone points and modelling the road surface as a triangulated irregular network. On the basis of experimental tests, the mean classification accuracies obtained using automatic method for lines, zebra crossings and kerbstones were 80.6%, 92.3% and 79.7%, respectively. Full article