Pedestrian detection in infrared images

Bertozzi, Massimo; Broggi, Alberto; Grisleri, Paolo; Graf, T.; Meinecke, Marc-Michael

doi:10.1109/ivs.2003.1212991

Cited by 86 publications

(59 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The output of those cameras is a projection on the sensor plane of the emissions of heat of the objects, that is proportional to the temperature. Most systems take advantage of this characteristic and select the regions of interest based on the distribution of warm areas on the image [2] [3] [4].…”

Section: State Of the Artmentioning

confidence: 99%

Contrast invariant features for human detection in far infrared images

Olmeda

Escalera

Armingol

2012

2012 IEEE Intelligent Vehicles Symposium

View full text Add to dashboard Cite

Abstract-In this paper a new contrast invariant descriptor for human detection in long-wave infrared images is proposed. It exploits local information histogram of orientations of phase coherence. Contrast in infrared images depends on the temperature of the object and the background, which makes gradient based descriptors less robust, especially in daylight conditions. The objective is to obtain a scale, brightness and contrast invariant descriptor that can successfully detect pedestrians in images taken with a cheap, temperature-sensitive, uncooled microbolometer. The descriptor, packed into grids is feed to a Support Vector Machine classifier. The algorithm has been tested in night and day sequences and its performance is compared with a day only descriptor: the histogram of oriented features (HOG).

show abstract

Section: State Of the Artmentioning

confidence: 99%

Contrast invariant features for human detection in far infrared images

Olmeda

Escalera

Armingol

2012

2012 IEEE Intelligent Vehicles Symposium

View full text Add to dashboard Cite

show abstract

“…1.b shows the size of the BBs that are examined (green), and highlights which pedestrian cannot be detected; the distance range in which a 170 cm tall pedestrian can be detected is shown as well (red). The detection distance range can be computed as a function of pedestrian height specification and the BBs height range [12]. With the setup and design choices used for this work, the distance range is 7 m ÷ 20 m.…”

Section: Problem Analysis and Design Choicesmentioning

confidence: 99%

IR Pedestrian Detection for Advanced Driver Assistance Systems

Bertozzi

Broggi

Carletti

et al. 2003

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. This paper describes a system for pedestrian detection in infrared images implemented and tested on an experimental vehicle. A specific stabilization procedure is applied after image acquisition and before processing to cope with vehicle movements affecting the camera calibration. The localization of pedestrians is based on the search for warm symmetrical objects with specific size and aspect ratio. A set of filters is used to reduce false detections. The final validation process relies on the human shape's morphological characteristics.

show abstract

“…As a result, the lowest possible accuracy level of object state observations, the maximum object state evolution and the maximum system latency that can occur in an advanced driver assistance system have to be taken into account when determining which feature specific accuracy demand can be satisfied [34]. Because the accuracy level of an object state observation from a single-sensor may be subject to fluctuations [49], [7], [28], single-sensor advanced driver assistance systems are often limited to low feature service specific accuracy demands. One approach to deal with this problem comprises updating the real-time images of the object states with redundant object state observations derived from heterogeneous sensors [16], [14].…”

Section: Introductionmentioning

confidence: 99%

State-of-the-art Versus Time-triggered Object Tracking in Advanced Driver Assistance Systems

Koplin

Elmenreich

2013

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

Most state-of-the-art driver assistance systems cannot guarantee that real-time images of object states are updated within a given time interval, because the object state observations are typically sampled by uncontrolled sensors and transmitted via an indeterministic bus system such as CAN. To overcome this shortcoming, a paradigm shift toward time-triggered advanced driver assistance systems based on a deterministic bus system, such as FlexRay, is under discussion.In order to prove the feasibility of this paradigm shift, this paper develops different models of a state-of-the-art and a time-triggered advanced driver assistance system based on multi-sensor object tracking and compares them with regard to their mean performance. The results show that while the state-of-the-art model is advantageous in scenarios with low process noise, it is outmatched by the time-triggered model in the case of high process noise, i.e., in complex situations with high dynamic.

show abstract

Pedestrian detection in infrared images

Cited by 86 publications

References 10 publications

Contrast invariant features for human detection in far infrared images

Contrast invariant features for human detection in far infrared images

IR Pedestrian Detection for Advanced Driver Assistance Systems

State-of-the-art Versus Time-triggered Object Tracking in Advanced Driver Assistance Systems

Contact Info

Product

Resources

About