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

Abstract

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.

Towards rain detection through use of in-vehicle multipurpose cameras

“…Nashashibi et al [6] present an approach to detect unfocused raindrops on a transparent screen. They used a spatiotemporal approach to achieve detection in real-time.…”

A complete system to determine the speed limit by fusing a GIS and a camera

“…8). We therefore plan to couple our vision module with a precipitation detector under development [22] to estimate the accuracy loss due to bad visibility conditions. Moreover, we work on a system able to detect and recognize any type of subsign in order to fully manage the traffic sign indications.…”

“…If information from the CAN bus of the vehicle is available, like wipers or fog lamp state, this step is trivial. An alternative solution would be to implement image-based algorithms to estimate the weather conditions ( [20], [21], [22]). The advantages are twofold: detecting precipitations and traffic signs with a single piece of equipment and quantifying the expected reliability of the camera.…”

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

“…Furthermore, another approach comes from [24] who presented an unprecedented approach to detect unfocused raindrops in spherical deformations which are the cause for a blurred vision through the windshield. This approach does not require any additional devices.…”

“…However, that approach does not support classification templates. In this paper, we were inspired by the idea of [24] and we did improve it by providing a novel approach which is based on Cellular Neural Networks (CNN) to accelerate the Support Vector Machine (SVM) classification.…”