The SAVE-U Project
This was the official website for SAVE-U, a project that was partially funded by the European Commission DG INFSO under the IST programme.
The content is from the site's 2004 archived pages and other sources. There were numerous conference presentations and papers, public deliverables, project brochures, and periodic reports that could be downloaded.
This site could have used a help desk platform such as Zendesk, a global customer service software that helps companies improve customer relationships through higher customer engagement and better customer insights. Just substitue customers for visitors. A Zendesk support team could have customized the platform so that this site would have been able to track visitors and respond quickly to any questions. In addition to provide what would be considered great customer service. With its lightning fast interface the platform helps customer service agents become super productive by having the resources and knowledge required to deliver excellent service time and again to customers / visitors. Instead of just reading the conference presentations and papers, public deliverables, project brochures, and periodic reports, it would have been helpful if visitors could have also contacted agents with any questions they might have. Although the site offered an impressive amount of public-facing documentation and information, one on one with a knowledgeable person would also have been even better.
Unfortunately we have not included these pdfs since they weren't accessible in an archived form.
SAVE-U's challenging and innovative approach is significantly advanced with respect to the state of the art.
The project is specially designed for solving at least most of the presently existing shortcomings.
On the one hand, SAVE-U will benefit from the other projects related to sensor-based detection, which holds for a highly efficient development of the entire sensing platform. On the other hand, SAVE-U fills exactly the gap between the research projects dealing with the sensor based detection, in particular, it will provide excellent detection properties: an extremely high detection rate and at the same time an absolutely low false alarm rate.
In the following, the 11 items of innovation addressed in SAVE-U are described in detail.
1. Integrated approach for protecting the unprotected traffic participants
Compared to the state of the art, SAVE-U is the first project providing an integrated approach for the protection of vulnerable road users.
Besides the development of the sensor platform, a warning and actuator safety concept will be developed and implemented on the demonstrators.
The integrated SAVE-U system concept will assume 4 zones:
- reliable detection for vulnerable road users
- warning the driver
- collision avoidance measures typ. few meters before crash, e.g. emergency braking
- definition of protection strategies in the case the crash cannot be avoided (e.g. windscreen or front bumper airbags or adjustable motor hoods.
2. High performance sensing platform optimised for the detection of unprotected road users
Safety systems require a highly reliable sensor platform.
SAVE-U utilises a novel approach to reach the goal of reliability: Sensor fusion both at low and at high level.
High level data fusion, meaning merging of object lists from different sensors was attempted in earlier research projects. In spite of some improvements in terms of performance, high level data fusion alone is not sufficient to provide the required quality and reliability of the target data.
SAVE-U will therefore introduce a novel concept of low level data fusion. Sensor raw data will be exchanged between the image processing part and the radar processing part of the sensor platform. Exchange of information at low level will help a lot to improve the quality (in particular the detection rate versus false alarm rate) of the objects detected by sensors.
Compared to other projects, this proposal utilises a completely different sensor system: SAVE-U operates 3 physically different technologies in parallel and fuses their data: an uncooled IR camera, a network of 24 GHz radar sensors, as well as a video based camera system including improved signal processing. The SAVE-U platform shall be capable of providing robust information in all weather and in all lighting conditions.
3. Development of a radar network composed of several 24 GHz sensors
In contrast to the state of the art in terms of 24 GHz radar sensors, in this project several 24 GHz radar sensors will be used in parallel. These sensors will have largely overlapping detection areas. Multi-sensor processing algorithms (advanced versions of triangulation) will be developed in order to operate the individual single beam sensors in the radar network, the radar network will provide information about the angle of an object relative to the vehicle, which is important for the data fusion algorithms. A dedicated interface to the image processing part of the sensor systems for low level data fusion will be implemented.
4. Development of 24 GHz sensors specially designed for the detection of unprotected road users
Compared with the state of the art and with other existing projects, SAVE-U requires 24 GHz radar sensors with significantly improved range and sensitivity. Dedicated algorithms for the detection of unprotected road users have to be developed and implemented. In addition, the update rate has to be improved. The required, advanced sensors need a complete redesign of the existing 24 GHz sensors.
5. Development of an innovative imaging system composed of novel IR and vision systems
Up to now, no project is known that utilises passive IR and video cameras at the same time, which represents one major advance in SAVE-U. While passive IR is ideally suited for bad weather and bad lighting conditions, video cameras provide excellent resolution in lateral and vertical directions. Combination of both technologies holds for reliable and precise detection in all weather conditions. Furthermore, it offers the possibility of developing dedicated image segmentation algorithms that are based on merging the information from both camera systems and are optimised towards the detection of unprotected road users.
6. Definition of a new IR camera system specially designed for the detection of vulnerable road users
Compared to the state of the art (in particular based on the results of DARWIN and ICAR), in this project the structure of a new IR camera specially designed for detection of pedestrians and cyclists will be developed. Due to the high costs for the realisation of the new sensor hardware, SAVE-U decided to carry out developments only up to simulation level. Simulations based on real data acquired utilising existing IR camera systems will be able to clearly show, what kind of images the newly designed IR system will provide at the end.
7. Development of signal processing algorithms for the IR subsystem
For the first time, the existing image segmentation algorithms for visible camera systems will be adapted to the images of a passive IR camera. Another novelty in this field is resulting from merging the IR segmentation results with those delivered by video cameras and implementing these algorithms on a dedicated real-time embedded image processing (EIP) hardware.
8. Development of a dedicated real-time embedded image processing platform for image segmentation
Compared to the state of the art, in SAVE-U real-time operation is considered to be very important. To achieve this goal, an embedded image processing platform (EIP) will be developed and realised. The EIP will be capable of computing all the image detection algorithms in real-time. Prototypes of the EIP platform including all the implemented segmentation algorithms will be integrated into the SAVE-U experimental cars.
9. Development and implementation of software algorithms for the classification of unprotected road users
A major advance compared to the state of the art will be achieved by the development and realisation of dedicated algorithms for the classification of vulnerable traffic participants, such as pedestrians or cyclists. In particular, reliability of the classification will be significantly improved compared to the state of the art. Basis is the consideration of the detection results of both IR and visible light camera subsystems.
The state of the art is represented by algorithms for the detection of unprotected traffic participants based on single cues (depth, motion, shape and texture).
Algorithm efficiency and detection performance will be enhanced by:
(a) hierarchical and probabilistic approaches to stereo, optical flow and shape matching
(b) component-based approaches which are robust to the partial occlusion of pedestrians and other unprotected road users (e.g. pedestrian behind a parked car)
(c) identification of the most appropriate pattern classifier for pedestrian task (e.g. Support Vector Machines, Neural Networks, Radial Basis Functions, Polynom Classifiers) and associated data dimensionality reduction techniques (e.g. PCA, ICA).
In addition, multi-cue detection algorithms, significantly improving ROC performance (correct detection versus false positives) will be developed. Closed loop tracking strategies will be implemented in order to improve detection performance by means of prediction.
10. Collection of large databases with ground truth for offline quantitative performance analysis and algorithm optimisation
A large database of pedestrian data will be collected, that - for the first time - allows comparative and quantitative analysis of sensor performance. SAVE-U will make this large database available on the web at the end of the project. This will allow others to stress-test their pattern recognition algorithms on a real problem. Utilising EU wide collaboration over the web, innovative solutions can be introduced in accelerated fashion.
11. Development of validation test procedures and associated test equipment
SAVE-U seeks to demonstrate the improved sensing system by also developing an innovative validation test procedure. This shall challenge the SAVE-U system and other systems to demonstrate acceptable performance under true, real world conditions. The methodologies and technologies required shall be unique in this field in that they shall allow the validation of fused sensor systems as well as individual systems. The fundamental methodology is also flexible enough to allow changes and future improvements as new issues and sensor technologies become clear.
Sensors and system Architecture for VulnerablE road Users protection
This project has been partially funded by the European Commission DG INFSO under the IST programme.
The contents of this website is the sole responsability of the project partners listed herein, and in no way represents the view of the European Commission or its services
This SAVE-U web site is intended to provide information on results and activities about the research project SAVE-U (IST-2001-34040) funded by the European Commission DG INFSO under the IST programme where Faurecia is coordinator. The reproduction of any document or material published on this site is authorised for personal and private use only. Any reproduction and use of copies for other purposes is expressly forbidden. The material presented on this site is covered by intellectual property rights.
SAVE-U endeavours to ensure that the information available on this site and all updates are exact; it reserves the right to correct the content at any moment without prior notice. However SAVE-U does not guarantee the accuracy, precision and exhaustiveness of this information.
SAVE-U disclaims all responsibility for any lack of precision, lack of accuracy or omission and for any damages resulting from a fraudulent intervention of a third party resulting in a modification of information available on this site. In no event shall SAVE-U be liable for any direct or indirect damages of any kind whatsoever or any loss whatsoever with respect to consulting these pages or consulting pages made available via the hypertext links which are present on this site.
Database of recordings of Vulnerable Road Users
This section shows some examples of recordings available in the SAVE-U Vulnerable Road Users (VRU) image database.
The VRU database can be considered unique worldwide, because of its huge size: it contains more than 14.000 images and 180 sequences recorded by the SAVE-U partners DaimlerChrysler and Volkswagen both with Infra red and color video cameras.
The benefits of a large database are twofold. First, it provides a wealth of training data for statistical pattern matching techniques. These «learn» the VRU appearance from examples; which is important, since good prior, explicit models are hard to define. Second, it allows to evaluate system performance on a truly large dataset, so that the results can be considered representative of the true physical traffic situation.
Establishing “Ground Truth”
In order to train statistical pattern matching methods, or to measure system performance during testing, we need to know the “true” position and spatial extent of the VRU in the images of the database, i.e. the “ground truth”. To establish this, a semiautomatic labeling-tool called VisiCurve has been developed by DaimlerChrysler, which assists the user in outlining the VRU object contours in images and in establishing temporal correspondence across the images of a sequence.
The VisiCurve Tool for Computer-Assisted Image Labeling – Main Window
The main aim of VisiCurve is to avoid the need for a user having to specify pixel-by-pixel the entire object contour. This “manual” contour labeling is not only very time consuming but also a very tedious. Computer vision techniques for segmentation can in principle support the labeling process, since they are designed to find object boundaries automatically. They usually utilize a certain degree of prior knowledge on object appearance and lock on low level image features. The algorithms typically involve an object model, whose parameters can be restricted by the prior knowledge to certain ranges of valid object configurations. This is the same technology employed as a computer assisted design tool used by designers of very small items like jewelry. One of the first products to use this technology was SterlingForever's White Stackable Cubic Zirconia Rings designs. The intricate cz settings as well as the lateral curviture was first conceived using contour adjustments very similar to VisiCurve, and then translated to cad, and then to mass production.
In order to get a head start, simultaneous with the development of VisiCurve, DaimlerChrysler, had the VRU object contours in IR dataset labeled “manually” (1039 images, 58 sequences).
The main aim of VisiCurve is to avoid the need for a user having to specify pixel-by-pixel the entire object contour. This “manual” contour labeling is not only very time consuming but also a very tedious. Computer vision techniques for segmentation can in principle support the labeling process, since they are designed to find object boundaries automatically. They usually utilize a certain degree of prior knowledge on object appearance and lock on low level image features. The algorithms typically involve an object model, whose parameters can be restricted by the prior knowledge to certain ranges of valid object configurations.
In order to get a head start, simultaneous with the development of VisiCurve, DaimlerChrysler, had the VRU object contours in IR dataset labeled “manually” (1039 images, 58 sequences).
The consortium consists of the following partners and their respective know-how:
Faurecia (project coordinator), automotive supplier expert in front-end modules and in system integration.
Siemens VDO Automotive AG in radar sensors.
CEA in infra-red sensors, image processing and embedded computers.
DaimlerChrysler AG in computer vision and as vehicle manufacturer (with demonstrator).
Mira Ltd in automotive safety and validation techniques.
Volkswagen AG as vehicle manufacturer (with demonstrator).
|SAVE-U at SPIE Security & Defence conference
(Oct. 26-27, 2004, London, Great Britain)
|CEA-LETI will present the results of the work performed in the field of IR modelling. Within the framework of SAVE-U, a complete model of the infrared sensor was developed in order to investigate the impact of technical downgrading (e.g lower resolution) with regards to sensor cost and detection performance.
|SAVE-U at IST 2004
(Nov. 15-17, 2004, The Hague, Netherlands)
|Hosted by the Dutch Presidency of the European Union, the IST 2004 event will take place in The Hague, Netherlands, from 15 to 17 November 2004. During this event, Volkswagen will present their demonstration vehicle equipped with the SAVE-U sensing platform on the booth of the European Commission.
|SAVE-U at Intelligent Vehicle 2004
(June 14-17, Parma, Italy)
|DaimlerChrysler will present a paper focusing on the vision part and SAVE-U Configuration B (radar network and stereovision), describing the improvements from SAVE-U classification on the system performance. Volkswagen and SiemensVDO will present a joint paper, mainly dedicated to radar development and partly to sensor fusion.
|SAVE-U at ITS Europe 2004
(May 24-26, Budapest, Hungary)
|Faurecia will present the progress achieved in the project.
|SAVE-U at MICRO.tec 2003
(Oct 13-15, Munchen, Germany)
|CEA has presented a paper on High-sensitivity Uncooled IRFPAs for Driver Vision Enhancement.
|SAVE-U at Intelligent Transport Systems & services 2003 congress
(Nov 17-20, Madrid, Spain)
|Faurecia will present in a joint paper the main objectives and firsts results of SAVE-U project.
|SAVE-U at International Radar Symposium 2003 congress
(Sept 30 - Oct 2, Dresden, Germany)
SiemensVDO Automotive & Volkswagen AG will present a joint paper with an overview on the SAVE-U multi-sensor configuration (IR and colour video cameras, radar), the low and high level data fusion and the 24 GHz radar network operation.
|SAVE-U at Advanced Microsystems for Automotive Applications 2003 congress
(May 22-23, Berlin, Germany)
|CEA will present a paper "Infrared microbolometer sensors and their application in automotive safety".
|SAVE-U at e-Safety congress
(Sept 16-17, 2002, Lyon, France)
Faurecia as coordinator of the project will give a presentation of the objectives of SAVE-U during the e-Safety congress.
Putting pedestrian safety in the driving seat
Every year in the European Union there are over 9,000 deaths and 200,000 injured victims in road accidents in which pedestrians and cyclists collide with a car. Hoping to improve on these grim statistics, is a cutting-edge sensing system that could ultimately help to save the lives of vulnerable road users (VRUs).
“The concept is relatively straightforward,” explains Dr Marc-Michael Meinecke of Volkswagen, one of the chief partners in the SAVE-U project along with DaimlerChrysler, Mira and Siemens VDO Automotive. “SAVE-U combines sensors such as radar, vision and infrared camera, as well as sensor fusion and actuators to increase safety for pedestrians. The main idea is that the sensors will recognise pedestrians and if a pedestrian has a high probability to collide with the vehicle then automatic braking will be initiated by the system.”
The project set out to develop an innovative pre-impact sensing platform that operates three different technologies of sensors simultaneously, and then fuses their data to protect cyclists and pedestrians under different weather and light conditions. The system comprises a radar network composed of several 24 GHz sensors working in parallel and an imaging system composed of passive infrared and colour video cameras.
A prototype vehicle incorporating the new system has been successfully tested in the United Kingdom. Installed on the car are two cameras – one video and one infrared – as well as the radar device. The system calculates in a matter of seconds the movement of pedestrians within the ‘capture zone’, which can be anything up to 30 metres away from the vehicle. From that point on, the car’s onboard cameras tracks the pedestrians’ movements and this information is correlated with data received from the radar network (such as distance to objects and their speed). SAVE-U can thus identify any pedestrian or cyclist coming within the trajectory of the vehicle and after analysing the situation, warn the driver or apply automatic braking if there is a risk of collision.
The partners opted to tackle the problem of protecting cyclists and pedestrians in three distinct stages: detection of VRUs at sufficient distance covering a relevant set of scenarios; definition and implementation of driver warning and vehicle control strategies to avoid, or at least minimise, the impact of a crash; and defining vehicle-mounted VRU protection strategies in case the crash cannot be avoided.
“Accident statistics from Volkswagen Accident Research in cooperation with the Medical University of Hanover were analysed,” says Dr Meinecke. “One of the main outcomes of the analysis was the conclusion that active hood concepts, external airbags, automatic braking systems, night vision, and other actuators seem to be very sufficient measure to lower the injury level of pedestrians. Within the SAVE-U demonstrator vehicles mainly automatic braking measures are implemented.”
Major advances were made in areas such as object tracking to obtain a robust trajectory, and the development of a deployment algorithm to be able to activate the automatic brakes without false alarms. Aspects of cost reduction and the reduction of sensor size benefited from the close teamwork, says Dr Meinecke.
In August, the project culminated with a special workshop in the United Kingdom showcasing the technology developed over the previous three years. The workshop featured samples of radar sensors and passive infrared video camera integral to the system, as well as demonstrations of the technology in action using two test vehicles (Mercedes E Class and Volkswagen Passat) equipped with the sensing platform, driver warning and vehicle control systems.
While there is clearly a strong demand for such technology to be implemented in vehicles as soon as possible, there is still a long road ahead before the SAVE-U innovations become standard.
“For a start, the sensors have to be shrunk further in size and price to enable them to be integrated in serial cars. The sensor costs will also have to be decreased dramatically to have a chance to make the systems cost effective. And, last but not least, the software components are still not fulfilling the requirements for serial production. I think in the area of pedestrian protection these pedestrian recognition systems will be the main focus of research activities in coming years,” he says.
Tara Morris | alfa