Dissertations / Theses on the topic 'Detection of road surface conditions'

This dissertation presents an approach to reconstruct degenerate near-planar road surface in three-dimensional (3D) while automatically detect road defects. Three techniques are developed in this dissertation to establish the proposed approach. The first technique is proposed to reconstruct the degenerate near-planar road surface into 3D from one camera. Unlike the traditional Structure from Motion (SfM) technique which has the degeneracy issue for near-planar object 3D reconstruction, the uniqueness of the proposed technique lies in the use of near-planar characteristics of surfaces in the 3D reconstruction process, which solves the degenerate road surface reconstruction problem using only two images. Following the accuracy-enhanced 3D reconstructed road surface, the second technique automatically detects and estimates road surface defects. As the 3D surface is inversely solved from 2D road images, the detection is achieved by jointly identifying irregularities from the 3D road surfaces and the corresponding image information, while clustering road defects and obstacles using a mean-shift algorithm with flat kernel to estimate the depth, size, and location of the defects. To enhance the physics-driven automatic detection reliability, the third technique proposes and incorporates a self-supervised learning structure with data-driven Convolutional Neural Networks (CNN). Different from supervised learning approaches which need labeled training images, the road anomaly detection network is trained by road surface images that are automatically labeled based on the reconstructed 3D surface information. In order to collect clear road surface images on the public road, a road surface monitoring system is designed and integrated for the road surface image capturing and visualization. The proposed approach is evaluated in both simulated environment and through real-world experiments. The parametric study of the proposed approach shows the small error of the 3D road surface reconstruction influenced by different variables such as the image noise, camera orientation, and the vertical movement of the camera in a controlled simulation environment. The comparison with traditional SfM technique and the numerical results of the proposed reconstruction using real-world road surface images then indicate that the proposed approach effectively reconstructs high quality near-planar road surface while automatically detects road defects with high precision, accuracy, and recall rates without the degenerate issue.
Doctor of Philosophy
Road is one of the key infrastructures for ground transportation. A good road surface condition can benefit mainly on three aspects: 1. Avoiding the potential traffic accident caused by road surface defects, such as potholes. 2. Reducing the damage to the vehicle initiated by the bad road surface condition. 3. Improving the driving and riding comfort on a healthy road surface. With all the benefits mentioned above, it is important to examine and check the road surface quality frequently and efficiently to make sure that the road surface is in a healthy condition. In order to detect any road surface defects on public road in time, this dissertation proposes three techniques to tackle the road surface defects detection problem: First, a near-planar road surface three-dimensional (3D) reconstruction technique is proposed. Unlike traditional 3D reconstruction technique, the proposed technique solves the degenerate issue for road surface 3D reconstruction from two images. The degenerate issue appears when the object reconstructed has near-planar surfaces. Second, after getting the accuracy-enhanced 3D road surface reconstruction, this dissertation proposes an automatic defects detection technique using both the 3D reconstructed road surface and the road surface image information. Although physics-based detection using 3D reconstruction and 2D images are reliable and explainable, it needs more time to process these data. To speed up the road surface defects detection task, the third contribution is a technique that proposes a self-supervised learning structure with data-driven Convolutional Neural Networks (CNN). Different from traditional neural network-based detection techniques, the proposed combines the 3D road information with the CNN output to jointly determine the road surface defects region. All the proposed techniques are evaluated using both the simulation and real-world experiments. Results show the efficacy and efficiency of the proposed techniques in this dissertation.

Modern day cars are often equipped with a variety of sensors that collect information about the car and its surroundings. The stereo camera is an example of a sensor that in addition to regular images also provides distances to points in its environment. This information can, for example, be used for detecting approaching obstacles and warn the driver if a collision is imminent or even automatically brake the vehicle. Objects that constitute a potential danger are usually located on the road in front of the vehicle which makes the road surface a suitable reference level from which to measure the object's heights. This Master's thesis describes how an estimate of the road surface can be found to in order to make these height measurements. The thesis describes how the large amount of data generated by the stereo camera can be scaled down to a more effective representation in the form of an elevation map. The report discusses a method for relating data from different instances in time using information from the vehicle's motion sensors and shows how this method can be used for temporal filtering of the elevation map. For estimating the road surface two different methods are compared, one that uses a RANSAC-approach to iterate for a good surface model fit and one that uses conditional random fields for modeling the probability of different parts of the elevation map to be part of the road. A way to detect curb lines and how to use them to improve the road surface estimate is shown. Both methods for road classification show good results with a few differences that are discussed towards the end of the report. An example of how the road surface estimate can be used to detect obstacles is also included.

Le domaine des véhicules autonomes a suscité un grand intérêt ces dernières années. Afin de garantir au passager une expérience sûre et confortable sur les véhicules autonomes, des systèmes d'obstacles avancés doivent être mis en œuvre. Bien que les solutions actuelles de détection d'obstacles aient montré de bonnes performances, elles doivent être encore améliorées pour une sécurité accrue des véhicules autonomes sur route, de jour comme de nuit. En particulier, les véhicules autonomes dans la vie réelle peuvent rencontrer de la glace, de la neige ou des flaques d'eau, qui peuvent être la cause de collisions graves et d'accidents de la circulation. Les systèmes de détection doivent donc permettre de détecter les changements d'état de la route pour anticiper la réaction du véhicule et/ou désactiver les fonctions automatisées. L'objectif de cette thèse est de proposer un système pour les véhicules autonomes afin de détecter les conditions de chaussée induites par la météo. Après une étude approfondie de l'état de l'art, un système proche infrarouge (NIR) basé sur des LED et un système d'apprentissage automatique sont proposés pour la détection diurne et nocturne. Le système NIR a été conçu puis validé expérimentalement et, les spécifications techniques du système ont été définies. Le système d'apprentissage automatique est de plus proposé comme solution complémentaire au système NIR. Différents modèles d'apprentissage ont été testés et comparés en termes de performance. Enfin, les résultats sont discutés et une combinaison des deux systèmes est proposée afin de garantir une performance accrue pour la reconnaissance des conditions de route
The field of autonomous vehicles has aroused great interest in recent years. In order to ensure the passenger to get a safe and comfortable experience on autonomous vehicles, advanced obstacle systems have to be implemented. Although current solutions for detecting obstacles have shown quite good performances, they have to be improved for an increased safety of autonomous vehicles on road, both in day-time and night-time conditions. In particular, autonomous vehicles in real life may encounter ice, snow or water puddles, which may be the cause of severe crashes and traffic accidents. The detection systems must hence allow detecting changes in road conditions to anticipate the vehicle reaction and/or deactivate the automated functions. The aim of this thesis is to propose a system implemented on the autonomous vehicles in order to detect the road surface conditions induced by the weather. After deep investigation of the state of art, a near infrared (NIR) system based on LEDs and a machine learning system were proposed for daytime and night-time detection. The NIR systems with three LEDs were investigated with experimental validations. In addition, the specifications of the NIR systems are carefully discussed. Furthermore, the machine learning system is proposed as a supplementary system. The performance of different models is compared in terms of classification accuracy and model complexity. Finally, the results are discussed and a combination of the two systems is proposed

The remote real time identification of road surfaces is an increasingly important task in the automotive world. The development of automotive active safety system requires a remote sensing technology that alerts drivers to potential hazards such as slippery surfaces caused by water, mud, ice, snow etc. This will improve the safety of driving and reduce the road accidents all over the world. This thesis is dedicated to the experimental study of the feasibility of an affordable short-range ultrasonic and radar system for road surface recognition ahead of a vehicle. It introduces a developed novel system which can recognize the surfaces for all terrains (both on-road and off-road) based on the analysis of backscattered signals. Fundamental theoretical analysis, extensive modelling and practical experiments demonstrated that the use of pattern recognition techniques allows for reliable discrimination of the surfaces of interest. The overall classification system is described, including features extraction and their number reduction, as well as optimization of the algorithms. The performance of 4 classification algorithms was assessed and evaluated to confirm the effectiveness of the system. Several aspects like the complexity of the classification algorithms and the priori knowledge of the environment were investigated to explore the potential of this research and the possibility of introducing the surface classification system into the automotive market in the nearest future.

The need for improved cloud parameterisations in a road surface temperature model is demonstrated. Case studies from early 1994 are used to investigate methods of tracking cloud cover using satellite imagery and upper level geostrophic flow. Two of these studies are included in this thesis. Errors encountered in cloud tracking methods were investigated as well as relationships between cloud height and pixel brightness in satellite imagery. For the first time, a one dimensional energy balance model is developed to investigate the effects of erroneous cloud forecasts on surface temperature. The model is used to determine detailed dependency of surface freezing onset time and minimum temperature on cloud cover. Case studies from the 1995/96 winter in Devon are undertaken to determine effects of differing scenarios of cloud cover change. From each study, an algorithm for predicting road surface temperature is constructed which could be used in future occurrences of the corresponding scenario of the case study. Emphasis is strongly placed on accuracy of predictions of surface freezing onset time and minimum surface temperature. The role o f surface and upper level geostrophic flow, humidity and surface wetness in temperature prediction is also investigated. In selected case studies, mesoscale data are also analysed and compared with observations to determine feasibility of using mesoscale models to predict air temperature. Finally, the algorithms constructed from the 1995/96 studies are tested using case studies from the 1996/97 winter. This winter was significantly different from its preceding one which consequently meant that the algorithm from only one scenario of the 1995/96 winter could be tested. An algorithm is also constructed from a 1996/97 winter case study involving a completely different scenario Recommendations for future research suggest testing of existing algorithms with guidance on additional scenarios.

Several past studies have attempted to quantify the impact of winter weather conditions on highway mobility in terms of traffic volume, speed, and capacity. While consistent in their general findings, these studies have shown considerably different results in terms of effect size and contributing factors. More importantly, most of these studies have not attempted to model the effects of winter maintenance operations on mobility or isolate these effects from those due to snowstorm characteristics, rendering their results and the proposed methods of limited use for estimating the benefits of maintenance activities. This research attempts to address this gap through a statistical analysis of a data set that is unique in terms of spatial and temporal coverage and data completeness. The data set includes both event based and hourly observations of road weather and surface conditions, maintenance operations, traffic volume and speed, as well as several other measures, from 21 highway sections across the province of Ontario. Event based information is available for six winter seasons (2000 to 2006) at 19 of the sites. For this event based data a matched pair technique was employed to determine the changes in traffic volumes and speeds under matched conditions with and without snow events. A regression analysis was subsequently performed to relate the changes in traffic volume and speed over an event to changes in various contributing factors such as highway type, snow event characteristics and road surface conditions. A case study was conducted to illustrate the application of the developed models for quantifying the mobility impact of road surface condition and the mobility benefit of winter maintenance operations. Complete hourly records were available for all 21 sites for three winter seasons. This was used to perform the evaluation on an hourly basis. A matching technique is employed to assign hour-by-hour median speeds observed under typical weather and road surface conditions to each hour of a snowstorm event. A regression analysis is subsequently performed to relate changes from average hourly speed to various contributing factors such as highway type, weather conditions and maintenance operations. Effects of maintenance operations are represented by an intermediate variable called road surface condition index (RSI). A case study is conducted to illustrate the application of the developed models for quantifying the mobility impact of winter snowstorms and the mobility benefit of maintenance operations. The models developed in these analyses confirmed the relationships between weather variables and traffic volume and speed described in the literature. In addition a strong association between road surface condition and traffic volumes and speed was identified.

碩士
國立中興大學
電機工程學系所
105
In this thesis, the high efficiency sensing technology for road surface detection based on the deformation of light grating captured by the mobile camera mounted on motor or bike is proposed. The proposed algorithm is composed of two steps: cross points generation and road surface detection. In cross points generation, projected checkerboard is extracted with adaptive threshold to get a clear checkerboard without noise. Moreover, we record cross point position by vector projection to reduce the storage of golden pattern. For road surface detection, light grid image is extracted with adaptive threshold to get a clear light grid without noise. According to the step of cross points generation, a search window is extended based on the relative position of the recorded cross point to search the cross point features. If the feature is matched, light grid is not deformed and judge there are no obstruction in the search range. If the feature is not matched, light grid is deformed and judge there are obstruction in the search range. Experiment results show that for FULL HD resolution, the accuracy of road surface detection in nighttime is achieve to 99.2%, and the average processing time is 36.5 ms per frame.

碩士
國立中山大學
機械與機電工程學系研究所
103
Nowadays, the drivers cannot dodge bumpy road because of unfamiliar with traffic and poor visibility and to cause the traffic accident, therefore the situation of the roadway and driving safety are the most interested topic. In this thesis, the main experimental equipment is golf car which equip the webcam, laser range finder, IMU and RTK-DGPS to construct an intelligent roadway detection system. In this system, we divide it into three functions, terrain classification, pothole detection and roadway quality analysis. In terms of terrain classification, the experimental equipment captures the front of image through the webcam, and this information as the inputs of Back Propagation Neural Network (BPNN) is the training of the terrain classification and the final classification mechanism. In terms of pothole detection and roadway quality analysis, the experimental equipment gauges the pothole and analyze the roadway quality through laser range finder, webcam and IMU. At the end, the system will gather the outcome of functions and then mark on the latitude and longitude of Google Map through RTK-DGPS on user interface to notify the drivers of nearby traffic. This thesis, intelligent roadway detection system, is the first system which integrates the terrain classification, pothole detection and roadway quality analysis. This system provides the information of the front of roadway for drivers to avoid the rough roadway and to decrease the chance of the traffic accident, and the more safety and comfortable driving environment.

In this modern era, land transports are increasing dramatically. Moreover, self-driven car or the Advanced Driving Assistance System (ADAS) is now the public demand. For these types of cars, road conditions detection is mandatory. On the other hand, compared to the number of vehicles, to increase the number of roads is not possible. Software is the only alternative solution. Road Conditions Detection system will help to solve the issues. For solving this problem, Image processing, and machine learning have been applied to develop a project namely, Detection of Road Conditions Using Image Processing and Machine Learning Techniques for Situation Awareness. Many issues could be considered for road conditions but the main focus will be on the detection of potholes, Maintenance sings and lane. Image processing and machine learning have been combined for our system for detecting in real-time. Machine learning has been applied to maintains signs detection. Image processing has been applied for detecting lanes and potholes. The detection system will provide a lane mark with colored lines, the pothole will be a marker with a red rectangular box and for a road Maintenance sign, the system will also provide information of aintenance sign as maintenance sing is detected. By observing all these scenarios, the driver will realize the road condition. On the other hand situation awareness is the ability to perceive information from it’s surrounding, takes decisions based on perceived information and it makes decision based on prediction.

碩士
國立中正大學
電機工程研究所
105
In recent years, even though the average using a navigation system for destination user has gradually increased, we continue to witness on news a slight increase in getting lost with a GPS. Most of the getting lost events occur due to the navigation system usually using the shortest paths as their recommended route. Supposedly, we use the shortest path may encounter the following situation:(1)GPS device might lead you down the narrow alley or down a closed road. If we are driving down a narrow alley, we have to pay more attention to our surrounding environment, as well as the inconvenience caused by the vehicle passing cross form the opposite lane. (2)If we are driving down a brick road or driving into some areas of damaged roads, we must lower the speed. The traffic is often dense and crowds because most of brick road area is laid near to the tourist attractions. In addition, the vehicle driving on the brick road areas might more easily skid than driving on the asphalt road areas when it is raining. Furthermore, the driver may feel bumpy when driving pass through some areas of damaged roads or driving on a brick road. To address the above problems, in this thesis, we present a framework of system for vision-based path recommend from street view images. The entire system is a combination of the following three steps: (1)Data collection, the path is selected by graphical user interface. Also, we download the street view images of the whole path. (2)Surface extraction, the vanishing point is calculated from the collection images. The area of the pavement is extracted by the Grow-Cut in super-pixel level, and the road width of the area is calculated finally. (3)Surface classification, training step is carried out on the types of brick pavement and asphalt pavement collected in advance. Finally, the pavement categories for each image were predicted by using our previously trained modules. In the experimental results, in this thesis, we use the surface extraction and classification method, the results of the extraction and classification are displayed, then we explain the rules of its distribution. After that, we will discuss the accuracy of surface classification. Since we calculate the recommend scores for each path through our own scoring rules, we will examine whether the results of the path proposed by our system can effectively solve the above problems.

碩士
國立臺北科技大學
電腦與通訊研究所
97
Similar to the function of human eyes, autonomous land vehicle (ALV) uses camera to acquire road information. In this paper, we adopt disparity map (DM) to detect ALV''s march path and various obstacles if may face to. Then we develop several road surface voters to recognize what kind of road surface the ALV drives on. Finally, according to the information we collected, the best navigation can be achieved. After experimenting with our ALV in an outdoor road without pavement markings, the proposed algorithms really work well.

碩士
國立虎尾科技大學
機械與機電工程研究所在職專班
98
Because the automobile under the long time travel, and the chassis itself causes regarding the road surface the vibration, will take to drives the personnel and the passenger felt that is not comfortable and is weary. Takes advantage of this provides drives the personnel and the passenger under the different state of roads, can maintain good while the place comfortableness and the security, has been the goal which the people pursue. Regarding automobile''s vibration, may divide into on the path undulation and chassis''s origin generally, but affects the biggest vibration comes from the vibration which the path causes, its transfer mode, by way of the tire, the suspension system, the seat cushion elastic damping part arrives at the human body again, the human body appraises comfortableness again based on the vibration response. The vehicles travel undulating quantity makes the comparative analysis affiliation, also provides for is engaged in the automobile to be related the personnel and the social populace makes the reference. This research take the SAAB 9000 of vehicle types as the experiment vehicle, the use infrared range finder erects vehicle of side in the SAAB 9000 vehicle type, records the vehicles in the travel to vibrate bending down, discusses the vehicles from the practice stratification plane, after the different pressure, the state of roads, the travel time fills the nitrogen and the common air, performance of data the change and compares the vibration analysis in the travel, this research analyzes difference of reason from the nitrogen tire and the general compressed air tire. Fills the nitrogen and the common air obviously after this research, the backfill nitrogen vibration obviously small which comes compared to the backfill air, thus it may be known fills the nitrogen the tire to improve the tire absorption vibration ability, reduces vehicles'' vibration therefore to lengthen the vehicles to shock proof effect and the life the system.

Τα αυτοκινητιστικά δυστυχήματα αποτελούν μια από τις κυριότερες αιτίες θανάτου παγκοσμίως. Ο αυξανόμενος αριθμός τους οδήγησε στην συνειδητοποίηση ότι η χρήση προηγμένης τεχνολογίας για την κατασκευή ασφαλέστερων οχημάτων είναι απαραίτητη για την μείωση των ατυχημάτων και κατά συνέπεια των θανάτων που οφείλονται σε αυτά. Από τη στιγμή που οι τεχνολογικές εξελίξεις επέτρεψαν την ενσωμάτωση φθηνών, χαμηλής κατανάλωσης συστημάτων με μεγάλη επεξεργαστική ταχύτητα σε οχήματα, κατέστη προφανές ότι περίπλοκες τεχνικές υπολογιστικής όρασης μπορούσαν πλέον να χρησιμοποιηθούν για την υποβοήθηση της οδήγησης. Σε αυτή την κατεύθυνση, η παρούσα διατριβή εστιάζει στην ανάπτυξη καινοτόμων λύσεων για διαφορετικά κομμάτια που εμπλέκονται στα προηγμένα συστήματα υποβοήθησης του οδηγού. Πιο συγκεκριμένα, σε αυτή την διατριβή προτείνονται καινοτόμα υποσυστήματα για την αναγνώριση σημάτων οδικής κυκλοφορίας, την αναγνώριση φωτεινών σηματοδοτών, τον εντοπισμό προπορευόμενου οχήματος και τον εντοπισμό δρόμου. Οι τεχνικές που χρησιμοποιήθηκαν για την ανάπτυξη των προτεινόμενων λύσεων βασίζονται στην χρωματική επεξεργασία εικόνας με έμφαση στην ανεξαρτησία από την φωτεινότητα της σκηνής, στην χρήση πληροφορίας συμμετρίας για τον εντοπισμό χαρακτηριστικών αντικειμένων (όπως σήματα οδικής κυκλοφορίας, φωτεινοί σηματοδότες και οχήματα), στην χώρο-χρονική παρακολούθηση των εντοπισμένων αντικειμένων και στην αυτόματη κατάτμηση εικόνας για τον εντοπισμό δρόμου. Τα προτεινόμενα συστήματα αναπτύχθηκαν με στόχο την ανθεκτικότητα σε αλλαγές της φωτεινότητας ή τις καιρικές συνθήκες, καθώς και στην οδήγηση σε απαιτητικά περιβάλλοντα. Επίσης, έχει δοθεί ιδιαίτερη έμφαση στην προοπτική υλοποίησης συστημάτων πραγματικού χρόνου. Τα αποτελέσματα που παρουσιάζονται σε αυτή την διατριβή αποδεικνύουν την ανωτερότητα των προτεινόμενων μεθόδων έναντι αντίστοιχων της σχετικής βιβλιογραφίας, ειδικά στις περιπτώσεις του εντοπισμού προπορευόμενου οχήματος και του εντοπισμού δρόμου. Ελπίζουμε ότι μέρη της έρευνας αυτής θα εμπνεύσουν νέες προσεγγίσεις για τις μελλοντικές υλοποιήσεις αντίστοιχων συστημάτων.
Traffic accidents are one of the main reasons for the loss of human lives worldwide. Their increasing number has led to the realization that the use of advanced technology for manufacturing safer vehicles is imperative for limiting casualties. Since technological breakthroughs allowed the incorporation of cheap, low consumption systems with high processing speeds in vehicles, it became apparent that complex computer vision techniques could be used to assist drivers in navigating their vehicles. In this direction, this thesis focuses on providing novel solutions for different tasks involved in advanced driver assistance systems. More specifically, this thesis proposes novel sub-systems for traffic sign recognition, traffic light recognition, preceding vehicle detection and road detection. The techniques used for developing the proposed solutions are based on color image processing with a focus on illumination invariance, using symmetry information for man-made objects (like traffic signs, traffic lights and vehicles) detection, spatiotemporal tracking of detected results and automated image segmentation for road detection. The proposed systems were implemented with a goal of robustness to changes of illumination and weather conditions, as well as to diverse driving environments. A special focus on the prospect for real-time implementation has also been given. The results presented in this thesis indicate the superiority of the proposed methods to their counterparts found in relevant literature in both normal and challenging conditions, especially in the cases of preceding vehicle detection and road detection. Hopefully, parts of this research will provide new insights for future developments in the field of intelligent transportation.