Miguel Ángel Sotelo

Combination of Feature Extraction Methods for SVM Pedestrian Detection

This paper describes a comprehensive combination of feature extraction methods for vision-based pedestrian detection in Intelligent Transportation Systems. The basic components of pedestrians are first located in the image and then combined with a support-vector-machine-based classifier. This poses the problem of pedestrian detection in real cluttered road images. Candidate pedestrians are located using a subtractive clustering attention mechanism based on stereo vision. A components-based learning approach is proposed in order to better deal with pedestrian variability, illumination conditions, partial occlusions, and rotations. Extensive comparisons have been carried out using different feature extraction methods as a key to image understanding in real traffic conditions. A database containing thousands of pedestrian samples extracted from real traffic images has been created for learning purposes at either daytime or nighttime. The results achieved to date show interesting conclusions that suggest a combination of feature extraction methods as an essential clue for enhanced detection performance

Fast traffic sign detection and recognition under changing lighting conditions

In this work a system for traffic-sign detection and classification is shown. It is intended for both prohibition and obligation circular signs and for advertising triangular ones. The system is divided into three stages: first, detection, using the Hough transform from the information of the edges of the image; second, classification, using a neural network, and third, tracking, making use of a Kalman filter, which provides the system with memory. Some results are presented, obtained by real images recorded by only one camera placed on board a conventional vehicle, in sunny days, and also cloudy, rainy ones or at night, in order to show the reliability and robustness of the system. The average processing time is 30 ms per frame, what makes the system a good approach to work in real time conditions

A Color Vision-Based Lane Tracking System for Autonomous Driving on Unmarked Roads

This work describes a color Vision-based System intended to perform stable autonomous driving on unmarked roads. Accordingly, this implies the development of an accurate road surface detection system that ensures vehicle stability. Although this topic has already been documented in the technical literature by different research groups, the vast majority of the already existing Intelligent Transportation Systems are devoted to assisted driving of vehicles on marked extra urban roads and highways. The complete system was tested on the BABIECA prototype vehicle, which was autonomously driven for hundred of kilometers accomplishing different navigation missions on a private circuit that emulates an urban quarter. During the tests, the navigation system demonstrated its robustness with regard to shadows, road texture, and weather and changing illumination conditions.

Adaptive road crack detection system by pavement classification.

This paper presents a road distress detection system involving the phases needed to properly deal with fully automatic road distress assessment. A vehicle equipped with line scan cameras, laser illumination and acquisition HW-SW is used to storage the digital images that will be further processed to identify road cracks. Pre-processing is firstly carried out to both smooth the texture and enhance the linear features. Non-crack features detection is then applied to mask areas of the images with joints, sealed cracks and white painting, that usually generate false positive cracking. A seed-based approach is proposed to deal with road crack detection, combining Multiple Directional Non-Minimum Suppression (MDNMS) with a symmetry check. Seeds are linked by computing the paths with the lowest cost that meet the symmetry restrictions. The whole detection process involves the use of several parameters. A correct setting becomes essential to get optimal results without manual intervention. A fully automatic approach by means of a linear SVM-based classifier ensemble able to distinguish between up to 10 different types of pavement that appear in the Spanish roads is proposed. The optimal feature vector includes different texture-based features. The parameters are then tuned depending on the output provided by the classifier. Regarding non-crack features detection, results show that the introduction of such module reduces the impact of false positives due to non-crack features up to a factor of 2. In addition, the observed performance of the crack detection system is significantly boosted by adapting the parameters to the type of pavement.

VIRTUOUS: vision-based road transportation for unmanned operation on urban-like scenarios

This work presents an intelligent transportation system (ITS) that was implemented on an autonomous vehicle designed to perform global navigation missions on a network of unmarked roads. This is the first step toward the complete implementation of ITS in urban environments, which is the long-term goal of this work. Using a global positioning system, global navigation is achieved by means of a global planner and a task manager that recurrently coordinate the execution of vision-based perception tasks for the road tracking of nonstructured roads and the navigation of intersections. In addition, a vision-based vehicle-detection task has been developed, which endows the global navigation system with a reactive capacity. The complete system has been tested on the BABIECA prototype vehicle, which was autonomously driven for hundreds of kilometers around a private circuit, designed to emulate an urban quarter, at speeds of up to 50 km/h, successfully carrying out different navigation missions. During the tests, the vehicle drove itself across crossroads and performed the appropriate turning maneuvers at intersections. It also demonstrated its robustness with regard to shadows, road texture, weather conditions, and changing illumination.

Using Fuzzy Logic in Automated Vehicle Control

Automated versions of a mass-produced vehicle use fuzzy logic techniques to both address common challenges and incorporate human procedural knowledge into the vehicle control algorithms. In-vehicle computing has been largely relegated to auxiliary tasks such as regulating cabin temperature, opening doors, and monitoring fuel, oil, and battery-charge levels. However, computers are increasingly assuming driving-related tasks in some commercial models. Among those tasks are: maintaining a reference velocity or keeping a safe distance from other vehicles; improving night vision with infrared cameras; and building maps and providing alternative routes

Unsupervised and adaptive Gaussian skin-color model

Abstract In this article a segmentation method is described for the face skin of people of any race in real time, in an adaptive and unsupervised way, based on a Gaussian model of the skin color (that will be referred to as Unsupervised and Adaptive Gaussian Skin-Color Model, UAGM). It is initialized by clustering and it is not required that the user introduces any initial parameters. It works with complex color images, with random backgrounds and it is robust to lighting and background changes. The clustering method used, based on the Vector Quantization (VQ) algorithm, is compared to other optimum model selection methods, based on the EM algorithm, using synthetic data. Finally, real results of the proposed method and conclusions are shown.

Night time vehicle detection for driving assistance lightbeam controller

In this paper we present an effective system for detecting vehicles in front of a camera-assisted vehicle (preceding vehicles traveling in the same direction and oncoming vehicles traveling in the opposite direction) during night time driving conditions in order to automatically change vehicle head lights between low beams and high beams avoiding glares for the drivers. Accordingly, high beams output will be selected when no other traffic is present and will be turned on low beams when other vehicles are detected. Our systemuses a B&W micro-camera mounted in the windshield area and looking at forward of the vehicle. Digital image processing techniques are applied to analyze light sources and to detect vehicles in the images. The algorithm is efficient and able to run in real-time. Some experimental results and conclusions are presented.

Lateral control strategy for autonomous steering of Ackerman-like vehicles

Abstract This paper presents the results of a lateral control strategy that has been applied to the problem of steering an autonomous vehicle using vision. The lateral control law has been designed for any kind of vehicle presenting the Ackerman kinematic model, accounting for the vehicle velocity as a crucial parameter for adapting the steering control response. This makes the control strategy suitable for either low or high speed vehicles. The stability of the control law has been analytically proved, and experimentally tested by autonomously steering Babieca, a Citroen Berlingo prototype vehicle.

Adaptive Fuzzy Sliding Mode Controller for the Kinematic Variables of an Underwater Vehicle

This paper address the kinematic variables control problem for the low-speed manoeuvring of a low cost and underactuated underwater vehicle. Control of underwater vehicles is not simple, mainly due to the non-linear and coupled character of system equations, the lack of a precise model of vehicle dynamics and parameters, as well as the appearance of internal and external perturbations. The proposed methodology is an approach included in the control areas of non-linear feedback linearization, model-based and uncertainties consideration, making use of a pioneering algorithm in underwater vehicles. It is based on the fusion of a sliding mode controller and an adaptive fuzzy system, including the advantages of both systems. The main advantage of this methodology is that it relaxes the required knowledge of vehicle model, reducing the cost of its design. The described controller is part of a modular and simple 2D guidance and control architecture. The controller makes use of a semi-decoupled non-linear plant model of the Snorkel vehicle and it is compounded by three independent controllers, each one for the three controllable DOFs of the vehicle. The experimental results demonstrate the good performance of the proposed controller, within the constraints of the sensorial system and the uncertainty of vehicle theoretical models.