To produce images with a high dynamic range (HDR), various techniques can be used according to the application in question.The problem: loss of information in scenes with large differences in brightness. Bright areas appear overexposed, while dark areas are underexposed, leaving details in these areas unrecognizable. Merely adjusting the exposure time cannot improve the entire image in both areas.
HDRadapt masters all lighting situations in real time: dawn, high noon, dusk, scenes in parking garages and tunnels, nighttime as well as highlights and backlighting. No matter how extreme the light is or how extremely it changes, the adaptive control algorithm keeps up and delivers what it promises. With intuition and awareness of end users’ needs, the Kappa development team created a formulation for all extreme light situations that gets the best from gain control, exposure time control, luminance and aperture control, color saturation, and edge enhancement, and even the interactions between them. The gain control works across a wide range, the electronic shutter operates the shortest possible exposure times, and the combined external luminance control is precisely positionable.
The result: the best image with the most information within a maximum dynamic range of 1:1,000,000 / 120 dB in all conditions is produced. The adaptive algorithm is a module available for all Kappa camera profiles, consisting of an FPGA and a microcontroller.
Other methods are based on the use of sensors with nonlinear characteristics, extending the achievement of saturation capacity, whereby overexposure occurs only in extreme brightness. The choice of suitable techniques, appropriate sensors and their parameterization is entirely dependent upon the application in question.
Alongside the actual generation of HDR images, tone mapping is particularly important. The goal of tone mapping is to reduce the dynamic range of an HDR image to the point where it can be displayed on a monitor. The challenge in so doing is to retain details in the image that are of interest to the user and to create a realistic looking image at the same time.
For low-contrast scenes – such as those in haze or fog, or images recorded during long-range observation – Kappa developed a special algorithm for automatic contrast control (ACC). In so doing, it is possible even in recording low-contrast scenes to represent a great deal of image information and to visibly increase image quality.
In this process, which works in real time, an image’s luminance and chrominance are viewed separately. The actual contrast adjustment is performed on the luminance signal. It is histogrambased and scales the luminance such that the minimum value lies at 0 percent (black) and the maximum at 100 percent (white). The complete range of available luminance values is fully utilized and the sharpness within a low-contrast image is noticeably greater. In processing the color components however, the focus remains on the most realistic recreation possible of the colors in the recorded scene.
In reproducing scenes containing dark as well as light areas with little contrast, the ACC algorithm has only a limited effect. Some camera models offer a dynamic histogram balance as an alternative. In this process, a nonlinear contrast change takes place dependent on the brightness distribution so that a concurrent improvement in the contrast of both the dark as well as light areas is achieved. Contrast control can take place then either across the entire image or within a freely selectable window.