Realistic Image Degradation with Measured PSF

Wittpahl, Christian; Zakour, Hatem Ben; Lehmann, Matthias; Braun, Alexander

doi:10.2352/issn.2470-1173.2018.17.avm-149

Cited by 10 publications

(12 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, high-fidelity camera models often use rendered images as the input and perform postprocessing steps in order to transform the ideal image into more realistic camera raw data. Examples for such high-fidelity camera models were given by Carlson et al [29,30], Schneider and Saad [31], Wittpahl et al [32]. Schneider and Saad [31] applied optical distortion, blur, and vignetting to modify the ideal image from the environment simulation.…”

Section: Previous Work On Automotive Camera Modelingmentioning

confidence: 99%

“…Schneider and Saad [31] applied optical distortion, blur, and vignetting to modify the ideal image from the environment simulation. Wittpahl et al [32] used point spread functions and neural networks to reduce the gap between synthetic and real images. Carlson et al [29,30] presented an augmentation pipeline including chromatic aberration, blur, exposure, noise, and color temperature to simulate the image formation process and artifacts of a real camera.…”

Section: Previous Work On Automotive Camera Modelingmentioning

confidence: 99%

“…Sensor datasets help to understand the capabilities and limitations of perception sensors and, therefore, play an important role in assessing sensor performance and sensor modeling. In particular, sensor models based on probabilistic functions [28] or neural networks [30,32] require a representative labeled dataset to build realistic relationships between the ground-truth and sensor output.…”

Section: Datasets For Automotive Camera Sensorsmentioning

confidence: 99%

See 2 more Smart Citations

Development and Experimental Validation of an Intelligent Camera Model for Automated Driving

Genser

Muckenhuber

Solmaz

et al. 2021

Sensors

View full text Add to dashboard Cite

The virtual testing and validation of advanced driver assistance system and automated driving (ADAS/AD) functions require efficient and realistic perception sensor models. In particular, the limitations and measurement errors of real perception sensors need to be simulated realistically in order to generate useful sensor data for the ADAS/AD function under test. In this paper, a novel sensor modeling approach for automotive perception sensors is introduced. The novel approach combines kernel density estimation with regression modeling and puts the main focus on the position measurement errors. The modeling approach is designed for any automotive perception sensor that provides position estimations at the object level. To demonstrate and evaluate the new approach, a common state-of-the-art automotive camera (Mobileye 630) was considered. Both sensor measurements (Mobileye position estimations) and ground-truth data (DGPS positions of all attending vehicles) were collected during a large measurement campaign on a Hungarian highway to support the development and experimental validation of the new approach. The quality of the model was tested and compared to reference measurements, leading to a pointwise position error of 9.60% in the lateral and 1.57% in the longitudinal direction. Additionally, the modeling of the natural scattering of the sensor model output was satisfying. In particular, the deviations of the position measurements were well modeled with this approach.

show abstract

Section: Previous Work On Automotive Camera Modelingmentioning

confidence: 99%

Section: Previous Work On Automotive Camera Modelingmentioning

confidence: 99%

Section: Datasets For Automotive Camera Sensorsmentioning

confidence: 99%

See 1 more Smart Citation

Development and Experimental Validation of an Intelligent Camera Model for Automated Driving

Genser

Muckenhuber

Solmaz

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…In order to enhance the fidelity of camera simulation, a complex camera model is proposed that mimics the physics of imaging processes in [29,30] optical situations (e.g., optical distortion, blur, and vignetting) and additionally the image processing modules (e.g., signal amplification, objects or features identification, and detection) are modelled. In [31], an optical model was presented to validate the functional and safety limits of camera-based ADAS, which is based on the real, measured lens used in the product. In addition, Carlson et al [32] proposed an efficient, automatic, and physically based augmentation pipeline to vary sensor effects to augment camera simulation performance.…”

Section: Introductionmentioning

confidence: 99%

Phenomenological Modelling of Camera Performance for Road Marking Detection

Tarik

Arefnezhad

et al. 2021

Energies

View full text Add to dashboard Cite

With the development of autonomous driving technology, the requirements for machine perception have increased significantly. In particular, camera-based lane detection plays an essential role in autonomous vehicle trajectory planning. However, lane detection is subject to high complexity, and it is sensitive to illumination variation, appearance, and age of lane marking. In addition, the sheer infinite number of test cases for highly automated vehicles requires an increasing portion of test and validation to be performed in simulation and X-in-the-loop testing. To model the complexity of camera-based lane detection, physical models are often used, which consider the optical properties of the imager as well as image processing itself. This complexity results in high efforts for the simulation in terms of modelling as well as computational costs. This paper presents a Phenomenological Lane Detection Model (PLDM) to simulate camera performance. The innovation of the approach is the modelling technique using Multi-Layer Perceptron (MLP), which is a class of Neural Network (NN). In order to prepare input data for our neural network model, massive driving tests have been performed on the M86 highway road in Hungary. The model’s inputs include vehicle dynamics signals (such as speed and acceleration, etc.). In addition, the difference between the reference output from the digital-twin map of the highway and camera lane detection results is considered as the target of the NN. The network consists of four hidden layers, and scaled conjugate gradient backpropagation is used for training the network. The results demonstrate that PLDM can sufficiently replicate camera detection performance in the simulation. The modelling approach improves the realism of camera sensor simulation as well as computational effort for X-in-the-loop applications and thereby supports safety validation of camera-based functionality in automated driving, which decreases the energy consumption of vehicles.

show abstract

“…Another alternative is to have a fixed-focus, passively athermalized system in which the optical system is designed such that sharp focus is obtained at the same image plane location (for a given object distance) over a range of temperatures, eliminating the need for an external actuator. Typically, ADAS and autonomous driving cameras are fixed-focus [16], therefore passive athermalization is essential for operation over a wide range of temperatures.…”

Section: Introductionmentioning

confidence: 99%

Long-Range, High-Resolution Camera Optical Design for Assisted and Autonomous Driving

Sahin

2019

Photonics

View full text Add to dashboard Cite

High-quality cameras are fundamental sensors in assisted and autonomous driving. In particular, long-range forward-facing cameras can provide vital information about the road ahead, including detection and recognition of objects and early hazard warning. These automotive cameras should provide high-resolution images consistently under extreme operating conditions of the car for robust operation. This paper aims to introduce the design of fixed-focus, passively athermalized lenses for next-generation automotive cameras. After introducing an overview of essential and desirable features of automotive cameras and state-of-the-art, based on these features, two different camera designs that can achieve traffic sign recognition at 200 m distance are presented. These lenses are designed from scratch, with a unique design approach that starts with a graphical lens material selection tool and arrives at an optimized design with optical design software. Optical system analyses are performed to evaluate the lens designs. The lenses are shown to accomplish high contrast from − 40 °C to 100 °C and allow for a 4 × increase in resolution of automotive cameras.

show abstract

Realistic Image Degradation with Measured PSF

Cited by 10 publications

References 6 publications

Development and Experimental Validation of an Intelligent Camera Model for Automated Driving

Development and Experimental Validation of an Intelligent Camera Model for Automated Driving

Phenomenological Modelling of Camera Performance for Road Marking Detection

Long-Range, High-Resolution Camera Optical Design for Assisted and Autonomous Driving

Contact Info

Product

Resources

About