Selective darkening of the windshield using high speed video analysis for driving assistance

Abstract

The paper presents the concept of a new visual based driving assistance system for increasing the safety and comfort in traffic. Using a high speed camera observing the vehicle's driver, an embedded image analysis module and a selective darkening panel attached to the windshield, the system is able to detect the appropriate area to be darkened in order to shade the driver's eyes and avoid the dazzling effect. The driver has a better view over the scene, since the rest of the windshield stays in high transparent mode. In this contribution, the main structure of the algorithm is presented. A patent is pending.

An algorithm for glare detection via photometric, colorimetric, and global positioning features

“…where X r = 0.950456, Y r = 1, Z r = 1.088754 are the CIE XY Z tristimulus values of the D65 reference white point; and the function g is given by: Note that equations (1) to (8) and all the coefficients used in them are standard conversion formulas between RGB and CIELAB color spaces.…”

“…Although the detection of glare is an important problem, there are relatively few publications on glare detection using digital cameras. [3][4][5][6][7][8][9][10] The existing approaches to glare detection generally rely on one or two simplistic image properties, followed by ad-hoc thresholding. The obvious and most commonly used feature is light intensity, [3,4] which can certainly detect the light source(s), but which cannot differen-tiate these sources from other bright regions (e.g., clouds), as demonstrated in Figure 1.…”

“…Instead of analyzing the scene, another class of approach has focused on analyzing an image of the driver's face. [7,8] Unfortunately, this approach cannot be used in autonomous systems nor in scenarios where the driver's environment is different from the actual to-be-driven location (e.g., in remote navigation). Previous approaches have used intensity and hue; part of our approach uses intensity and saturation, the latter being more reliable at detecting the spatial extent of each glare region Deep learning has been utilized in the work by Jatzkowski et al [13] to determine whether overexposure (which is not always as detrimental to the driver's vision as glare) exists in an image.…”

Automatic Glare Detection via Photometric, Geometric, and Global Positioning Information

“…Although the detection of glare is an important problem, there are relatively few publications on glare detection using digital cameras [2,3,4,5,1,6]. The existing approaches to glare detection generally rely on one or two simplistic image properties, followed by ad-hoc thresholding.…”

“…However, hue often fails to delineate the boundaries of the glare region(s), particularly when the image is subject to compression, as demonstrated in Figure 2. Instead of analyzing the scene, another class of approach has focused on analyzing an image of the drivers face [1,6]. Unfortunately, this approach cannot be used in autonomous systems nor in scenarios where the drivers environment is different from the actual to-be-driven location (e.g., in remote navigation).…”