Design Framework of Image Sensor System Based on Dynamic Range Extension by Adding Noise for Saturated Conditions

It is a well-known fact that the light input to an image sensor is often out of its dynamic range under the strong light conditions of halation. In this case, the sensor cannot output any information. Therefore, various countermeasures have been proposed for sensor devices and signal processing. These countermeasures increase the dynamic range using additional components and/or computations, which increase the system cost. In this paper, we propose a simple alternative method of just adding noise. Inspired by the concept of supra-threshold stochastic resonance, which is often discussed in the context of nonlinear physics, a fluctuation in sensor response is introduced in the proposed method. We analytically describe the mechanism by which the dynamic range is increased, which is intuitively understandable and can be used in practical image sensors.