Perhaps the most requested feature. If a suspect’s face is only 20 pixels wide, standard zooming results in a blocky mess. Ikena’s Super Resolution analyzes multiple consecutive frames to synthesize a higher-resolution image. It reconstructs missing detail by measuring sub-pixel shifts between frames. The result? A pixelated blob can sometimes be resolved into a recognizable face or license plate number.
Many agencies fail to realize that a poorly trained analyst using Ikena can destroy the evidentiary value of video. Certification is not optional; it is mandatory for court credibility. Ikena forensic video enhancement software
As of 2025, the forensic community is torn on AI. has incorporated limited, conservative AI (machine learning for object detection, e.g., automatically finding all license plates in a video). However, the developers have deliberately avoided "Generative AI" (filling in missing pixels with a guess). Perhaps the most requested feature
Noise (the static or grain seen in low-light video) is a major obstacle for facial recognition. Simple noise reduction filters often smooth out the grain by blurring the image, which destroys facial features. Ikena’s noise reduction is spatial and temporal, meaning it looks at the noise patterns across space and time to distinguish between random noise and actual image detail. This results in a clean image that retains the edge details necessary for identification. It reconstructs missing detail by measuring sub-pixel shifts
