Vidinger
How does it work?
How does it work?
The nominal value of the ENF (which can be 50/60 Hz depending on the geographical location), does not remain constant in practice and fluctuates across time. These variations in the ENF are caused by the load control mechanism used to stabilize the power grid, yielding a unique and unpredictable evolution of the ENF throughout time, which is common to all the interconnected geographical points. As a particular characteristic, these fluctuations of the ENF are “filtered” in recorded video sequences when the recording is conducted in indoor lighting environments.
VIDINGER first needs to record and store a reference ENF signal to locate in time the ENF extracted from a given video under analysis. The reference ENF signal can be acquired either through our Tensiunator device (which provides a very fine estimation of the ENF fluctuations), or from any public database, such as Swissgrid or RTE. Then, VIDINGER processes the input video sequence to extract the filtered ENF fluctuations in the lighting. Finally, VIDINGER compares the extracted signal from the video with the reference ENF signal in order to elucidate the actual video time-of-recording.
VIDINGER compensates the detrimental effects of motion and scene-dependent characteristics for a better extraction of the ENF signal from a given video sequence. In addition, VIDINGER is able to estimate the actual frame rate used to record the video, which is a critical parameter for the ENF extraction process.
Finally, the design of VIDINGER meets the standard practice required in forensic science, ensuring that the derived algorithm is explainable, which is essential to guarantee the credibility of VIDINGER in any legal context and also to establish its limitations.
Request a folder, a demo, or a quote