MODISSA: a multipurpose platform for the prototypical realization of vehicle-related applications using optical sensors

Borgmann, Björn; Schatz, Volker; Hammer, Marcus; Hebel, Marcus; Arens, Michael; Stilla, Uwe

doi:10.1364/ao.423599

Cited by 9 publications

(5 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data used in this paper was recorded with one of the four thermal infrared cameras of the experimental MODISSA platform. 26 From the recorded images, one image per second is chosen and only, if our target class "car" is present. This results in a total of 1324 images.…”

Section: Datasetmentioning

confidence: 99%

APMD: Adversarial Pixel Masking Derivative for multispectral object detectors

Bayer

Münch²,

Arens³

2022

Counterterrorism, Crime Fighting, Forensics, and Surveillance Technologies VI

View full text Add to dashboard Cite

Recently, adversarial patches have been used successfully, to fool object detectors by hiding a specific or suppress almost all relevant detections in an image. Although there are various ways to harden against or identify those attacks in the visual spectrum, there is only a small fraction that actually evaluates these mechanisms on thermal infrared input data. Thermal infrared object detectors and classifiers cannot be fooled with pixel optimized adversarial patches, but they are still prone to Gaussian function patches. This paper (I) investigates two methods for hardening real-time infrared object detectors against adversarial patches. One of these methods is our novel (II) APMD, an extension of an already existing adversarial robustness mechanism, that relies on (unsupervised) adversarial training, to clear adversarial patches for deep learning object detectors in the infrared spectrum. We therefore (III) generate adversarial patches, that fool object detectors in the infrared spectrum in three different ways, and evaluate them with real-world data, recorded with the experimental platform MODISSA. Our results show, that the hardened system is fast enough to be used in a real-time environment and successfully detects and inhibits adversarial attacks.

show abstract

Section: Datasetmentioning

confidence: 99%

APMD: Adversarial Pixel Masking Derivative for multispectral object detectors

Bayer

Münch²,

Arens³

2022

Counterterrorism, Crime Fighting, Forensics, and Surveillance Technologies VI

View full text Add to dashboard Cite

show abstract

“…For the generation of the global pole map, we used all LiDAR sensors and all direct georeferencing capabilities of Fraunhofer IOSB's sensor platform MODISSA 19 (see Figure 1, left side). MODISSA is equipped with two 360°scanning LiDAR sensors Velodyne HDL-64E and an inertial navigation system (INS) Applanix POS LV 5.…”

Section: Generation Of the Pole Mapmentioning

confidence: 99%

“…For our experiments we used three different datasets, all recorded with the MODISSA sensor platform 19 (see Section 3.2 and Figure 1). Dataset one (Et I ) and two (Et II ) cover the same, mostly residential area located on the outskirts of the city of Ettlingen, Germany.…”

Section: Datasets Used For the Experimentsmentioning

confidence: 99%

Extraction and matching of 3D features for LiDAR-based self-localization in an urban environment

Hammer

Hebel²,

Arens³

2022

Electro-Optical Remote Sensing XVI

View full text Add to dashboard Cite

Geolocation of vehicles, objects or people is commonly done using global navigation satellite system (GNSS) receivers. Such a receiver for GNSS-based positioning is either built into the vehicle, or separate handheld devices like a smartphone or similar are used. Self-localization in this way is simple and accurate up to a few meters.Environments where no GNSS service is available require other strategies for self-localization. Especially in the military domain, it is necessary to be prepared for such GNSS-denied scenarios. Awareness of the own position in relation to other units is crucial in military operations, especially where joint operations have to be coordinated geographically and temporally. However, even if a common map-like representation of the terrain is available, precise self-localization relative to this map is not necessarily easy.In this paper, we propose an approach for LiDAR-based localization of a vehicle-based sensor platform in an urban environment. Our approach is to use 360°scanning LiDAR sensors to generate short-duration point clouds of the local environment. In these point clouds, we detect pole-like 3D features such as traffic sign poles, lampposts or tree trunks. The relative distance and orientation of these features to each other is rather unique, and the matrix of these individual distances and orientations can be used to determine the position of the sensor relative to a current map. This map can either be created in advance for the entire area, or a cooperative preceding vehicle with an equivalent sensor setup can generate it. By matching the found LiDARbased 3D features with those of the map, not only the position of the sensor platform but also its orientation can be determined. We provide first experimental results of the proposed method, which were achieved with measurements by Fraunhofer IOSB's sensor-equipped vehicle MODISSA.

show abstract

“…A fair share of work on that topic includes data from thermal infrared cameras, e.g. [1][2][3][4]. Combining data from multiple sensors to a common data set requires referencing them in a common spatial coordinate system as well as on a common time scale.…”

Section: Introductionmentioning

confidence: 99%

Measuring timing properties of thermal infrared cameras

Schatz

2023

Meas. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

A measurement setup and method is presented that is serves to determine constants characterising the timing behaviour of thermal infrared cameras. A continuously rotating chopper wheel is used to periodically obscure and reveal the cutout of a mask plate. The exposed fraction of the cutout is the main measured quantity. Irregularities in the chopper wheel are determined from timestamps of the light beam detectors in the chopper head. The method is suitable for both kinds of thermal cameras, cooled cameras and microbolometer cameras, and for both snapshot and rolling-frame cameras. Example results are presented for a cooled snapshot camera and a rolling-frame microbolometer camera.

show abstract

MODISSA: a multipurpose platform for the prototypical realization of vehicle-related applications using optical sensors

Abstract: Please cite as "MODISSA: a multipurpose platform for the prototypical realization of vehicle-related applications using optical sensors", Applied Optics 60 ( 22), F50-F65, 2021.

Cited by 9 publications

References 33 publications

APMD: Adversarial Pixel Masking Derivative for multispectral object detectors

APMD: Adversarial Pixel Masking Derivative for multispectral object detectors

Extraction and matching of 3D features for LiDAR-based self-localization in an urban environment

Measuring timing properties of thermal infrared cameras

Contact Info

Product

Resources

About