Iván García Daza

Gaze Fixation System for the Evaluation of Driver Distractions Induced by IVIS

We present a method to monitor driver distraction based on a stereo camera to estimate the face pose and gaze of a driver in real time. A coarse eye direction is composed of face pose estimation to obtain the gaze and drivers fixation area in the scene, which is a parameter that gives much information about the distraction pattern of the driver. The system does not require any subject-specific calibration; it is robust to fast and wide head rotations and works under low-lighting conditions. The system provides some consistent statistics, which help psychologists to assess the driver distraction patterns under influence of different in-vehicle information systems (IVISs). These statistics are objective, as the drivers are not required to report their own distraction states. The proposed gaze fixation system has been tested on a set of challenging driving experiments directed by a team of psychologists in a naturalistic driving simulator. This simulator mimics conditions present in real driving, including weather changes, maneuvering, and distractions due to IVISs. Professional drivers participated in the tests.

Expert video-surveillance system for real-time detection of suspicious behaviors in shopping malls

Tracking-by-detection based on segmentation, Kalman predictions and LSAP association.Occlusion management: SVM kernel metric for GCH+LBP+HOG image features.Overall performance near to 85% while tracking under occlusions in CAVIAR dataset.Human behavior analysis (exits, loitering, etc.) in naturalistic scenes in shops.Real-time multi-camera performance with a processing capacity near to 50fps/camera. Expert video-surveillance systems are a powerful tool applied in varied scenarios with the aim of automatizing the detection of different risk situations and helping human security officers to take appropriate decisions in order to enhance the protection of assets. In this paper, we propose a complete expert system focused on the real-time detection of potentially suspicious behaviors in shopping malls. Our video-surveillance methodology contributes several innovative proposals that compose a robust application which is able to efficiently track the trajectories of people and to discover questionable actions in a shop context. As a first step, our system applies an image segmentation to locate the foreground objects in scene. In this case, the most effective background subtraction algorithms of the state of the art are compared to find the most suitable for our expert video-surveillance application. After the segmentation stage, the detected blobs may represent full or partial people bodies, thus, we have implemented a novel blob fusion technique to group the partial blobs into the final human targets. Then, we contribute an innovative tracking algorithm which is not only based on people trajectories as the most part of state-of-the-art methods, but also on people appearance in occlusion situations. This tracking is carried out employing a new two-step method: (1) the detections-to-tracks association is solved by using Kalman filtering combined with an own-designed cost optimization for the Linear Sum Assignment Problem (LSAP); and (2) the occlusion management is based on SVM kernels to compute distances between appearance features such as GCH, LBP and HOG. The application of these three features for recognizing human appearance provides a great performance compared to other description techniques, because color, texture and gradient information are effectively combined to obtain a robust visual description of people. Finally, the resultant trajectories of people obtained in the tracking stage are processed by our expert video-surveillance system for analyzing human behaviors and identifying potential shopping mall alarm situations, as are shop entry or exit of people, suspicious behaviors such as loitering and unattended cash desk situations. With the aim of evaluating the performance of some of the main contributions of our proposal, we use the publicly available CAVIAR dataset for testing the proposed tracking method with a success near to 85% in occlusion situations. According to this performance, we corroborate in the presented results that the precision and efficiency of our tracking method is comparable and slightly superior to the most recent state-of-the-art works. Furthermore, the alarms given off by our application are evaluated on a naturalistic private dataset, where it is evidenced that our expert video-surveillance system can effectively detect suspicious behaviors with a low computational cost in a shopping mall context.

Fusion of Optimized Indicators from Advanced Driver Assistance Systems (ADAS) for Driver Drowsiness Detection

This paper presents a non-intrusive approach for monitoring driver drowsiness using the fusion of several optimized indicators based on driver physical and driving performance measures, obtained from ADAS (Advanced Driver Assistant Systems) in simulated conditions. The paper is focused on real-time drowsiness detection technology rather than on long-term sleep/awake regulation prediction technology. We have developed our own vision system in order to obtain robust and optimized driver indicators able to be used in simulators and future real environments. These indicators are principally based on driver physical and driving performance skills. The fusion of several indicators, proposed in the literature, is evaluated using a neural network and a stochastic optimization method to obtain the best combination. We propose a new method for ground-truth generation based on a supervised Karolinska Sleepiness Scale (KSS). An extensive evaluation of indicators, derived from trials over a third generation simulator with several test subjects during different driving sessions, was performed. The main conclusions about the performance of single indicators and the best combinations of them are included, as well as the future works derived from this study.

Vehicle Model Recognition Using Geometry and Appearance of Car Emblems from Rear View Images

In this paper a novel vehicle model recognition approach is presented modelling the geometry and appearance of car emblems (model, trim level, etc.) from rear view images. The proposed system is assisted by LPR and VMR modules. Thus, a generic methodology is defined to build a hierarchical structure of car-make-dependent vehicle model classifiers. The emblems location, size and variations are firstly learnt. Then, the appearance of each badge is modelled using a linear SVM binary classifier with HOG features and the outputs of each individual classifier are converted to an estimate of posterior probabilities. A specific probability is computed for each hypothesis (model) integrating the posterior probabilities of all the emblems using the geometric mean. Inference about the most probable car model is finally carried out selecting the model with the maximum probability. We evaluate this approach on a dataset composed of 1.342 images (910/432 for training/test) corresponding to 8 different car makes and 28 different car models (52 considering generations) achieving an overall accuracy of 93.75%.

Assistive Intelligent Transportation Systems: The Need for User Localization and Anonymous Disability Identification

The main goal of Assistive Technology (AT) is to ensure the functional independence of disabled individuals. This paper proposes the definition of a new concept of AT within the context of the ITS, Assistive Intelligent Transportation System (AITS), analyzing its intrinsic requirements and providing a set of examples. We demonstrate that AITS must localize users with disabilities and identify their specific type of impairment in order to provide an efficient response, and we propose a specific procedure to guarantee anonymity while identifying the type of disability. Moreover, this new type of AT is illustrated by means of a new assistive intelligent pedestrian crossing application that is capable of localizing pedestrians with disabilities, identifying the specific type of impairment and providing an adaptive response to enhance functional capabilities of impaired pedestrians while crossing. By combining stereo-based object detection with radio-frequency identification technology (RFID and Bluetooth Low Energy), a specific solution to the problem of user localization and anonymous disability identification is proposed. Our approach has been validated in a real crosswalk scenario and it may be extended to other types of AITS, depending on the localization accuracy requirements and the range of operation of the specific application.

Extended Floating Car Data system - experimental study

This paper presents the results of a set of extensive experiments carried out in daytime and nighttime conditions in real traffic using an enhanced or extended Floating Car Data system (xFCD) that includes a stereo vision sensor for detecting the local traffic ahead. The detection component implies the use of previously monocular approaches developed by our group in combination with new stereo vision algorithms that add robustness to the detection and increase the accuracy of the measurements corresponding to relative distance and speed. Besides the stereo pair of cameras, the vehicle is equipped with a low-cost GPS and an electronic device for CAN Bus interfacing. The xFCD system has been tested in a 198-minutes sequence recorded in real traffic scenarios with different weather and illumination conditions, which represents the main contribution of this paper. The results are promising and demonstrate that the system is ready for being used as a source of traffic state information.