Rainy weather recognition from in-vehicle camera images for driver assistance

Kurihata, H.; Takahashi, Tomokazu; Ide, Ichiro; Mekada, Yoshito; Murase, Hiroshi; Tamatsu, Yukimasa; Miyahara, Takayuki

doi:10.1109/ivs.2005.1505103

Cited by 144 publications

(77 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regarding rain drop detection on a windshield [21], used raindrop templates known as eigendrop to detect raindrops on windshields. Related results show that the detection within the sky area is quite promising, but the named method produces a large number of false positive within non-sky regions of the image where raindrop appearance and background texture become less uniform.…”

Section: Related Workmentioning

confidence: 99%

Real-time raindrop detection based on cellular neural networks for ADAS

Machot

Ali

Mosa

et al. 2016

J Real-Time Image Proc

View full text Add to dashboard Cite

A core aspect of advanced driver assistance systems (ADAS) is to support the driver with information about the current environmental situation of the vehicle. Bad weather conditions such as rain might occlude regions of the windshield or a camera lens and therefore affect the visual perception. Hence, the automated detection of raindrops has a significant importance for video-based ADAS. The detection of raindrops is highly time critical since video pre-processing stages are required to improve the image quality and to provide their results in real-time. This paper presents an approach for real-time raindrops detection which is based on cellular neural networks (CNN) and support vector machines (SVM). The major idea is to prove the possibility of transforming the support vectors into CNN templates. The advantage of CNN is its ultra fast precessing on embedded platforms such as FPGAs and GPUs. The proposed approach is capable to detect raindrops that might negatively affect the vision of the driver. Different classification features were extracted to evaluate and compare the performance between the proposed approach and other approaches.

show abstract

Section: Related Workmentioning

confidence: 99%

Real-time raindrop detection based on cellular neural networks for ADAS

Machot

Ali

Mosa

et al. 2016

J Real-Time Image Proc

View full text Add to dashboard Cite

show abstract

“…Stronger limitations are imposed by adherent water droplets on the glass surface covering the camera lens. Kurihata et al [9] used a machine learning approach with raindrop templates to detect raindrops on windshields from inside a moving vehicle. Results within the sky area were promising, whereas the proposed method produced a large number of false positives within the non-sky regions of the image where raindrop appearance modeling becomes more challenging.…”

Section: Related Workmentioning

confidence: 99%

“…Especially in rainy weather it is often the case that adherent waterdrops on the lens-protecting glass disturb the view of a camera. Although a lot of research has been pursued in robotics [16], computer vision [5,7,11] and for driver assistance [9,12,19], raindrop detection still remains a challenging task. This might be for several reasons: Water droplets on a glass surface exhibit a large variety in shape and size.…”

Section: Introductionmentioning

confidence: 99%

Realistic Modeling of Water Droplets for Monocular Adherent Raindrop Recognition Using Bézier Curves

Röser

Kurz

Geiger

2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper, we propose a novel raindrop shape model for the detection of view-disturbing, adherent raindrops on inclined surfaces. Whereas stateof-the-art techniques do not consider inclined surfaces because they assume the droplets as sphere sections with equal contact angles, our model incorporates cubic Bézier curves that provide a low dimensional and physically interpretable representation of a raindrop surface. The parameters are empirically deduced from numerous observations of different raindrop sizes and surface inclination angles. It can be easily integrated into a probabilistic framework for raindrop recognition, using geometrical optics to simulate the visual raindrop appearance. In comparison to a sphere section model, the proposed model yields an improved droplet surface accuracy up to three orders of magnitude.

show abstract

“…Several approaches attempted the use of either learning processes or image processing. For instance in [7], Kurihata et al proposed to use PCA (Principal Component Analysis) using eigenspace features extracted from raindrop images in the learning stage. At run time template matching is used to determine whether a region is or not a drop on the windscreen.…”

Section: State Of Artmentioning

confidence: 99%

“…So far, methods which have been developed for ITS in the past years [7][8][9] seem to work quite well but they do require either a special optical device [9] or the focus to be set on the windscreen [7], [8]. However, ITS already have many applications using camera sensor (such as road lanes detection or obstacles detection) and both for the sake of their performance as well as for industrial constraints, the camera is usually installed behind the interior rear-view mirror.…”

Section: State Of Artmentioning

confidence: 99%

Detection of unfocused raindrops on a windscreen using low level image processing

Nashashibi

Charrette

Alexandre

2010

2010 11th International Conference on Control Automation Robotics &Amp; Vision

View full text Add to dashboard Cite

Abstract-In a scene, rain produces a complex set of visual effects. Obviously, such effects may infer failures in outdoor vision-based systems which could have important side-effects in terms of security applications. For the sake of these applications, rain detection would be useful to adjust their reliability. In this paper, we introduce the problem (almost unprecedented) of unfocused raindrops. Then, we present a first approach to detect these unfocused raindrops on a transparent screen using a spatio-temporal approach to achieve detection in real-time.We successfully tested our algorithm for Intelligent Transport System (ITS) using an on-board camera and thus, detected the raindrops on the windscreen. Our algorithm differs from the others in that we do not need the focus to be set on the windscreen. Therefore, it means that our algorithm may run on the same camera sensor as the other vision-based algorithms.

show abstract

Rainy weather recognition from in-vehicle camera images for driver assistance

Cited by 144 publications

References 5 publications

Real-time raindrop detection based on cellular neural networks for ADAS

Real-time raindrop detection based on cellular neural networks for ADAS

Realistic Modeling of Water Droplets for Monocular Adherent Raindrop Recognition Using Bézier Curves

Detection of unfocused raindrops on a windscreen using low level image processing

Contact Info

Product

Resources

About