Most long range imaging systems fail to perform on field because of large variation in environmental conditions like fog, heat, rain and low light conditions. Also, capturing the scene from a long distance further deteriorates the imaging performance. These features are among must have features for any military mission. Our state-of-the-art enhancement modules reveals hard to find information in an image while visually enhancing the image. Some of the realtime video enhancement modules we offer include
- Digital detail Enhancement
The detail enhancement module increases the details in the videos in real time. It applies a combination of enhancement filters to the input video and enhance the original content with details.
- Contrast Enhancement
Contrast stretching module attempts to improve the contrast of the target by intelligently stretching the range of “valuable” intensity values to a desired range of intensity values.
- Defog / Dehaze
Haze, fog and smoke cause degradation of images of outdoor scenes. The light incident on the camera is attenuated and blended with ‘airlight’ (ambient light). Single image haze removal relies on strong assumptions/priors. The algorithm uses the ‘dark channel ‘prior. ‘Dark pixels’, often found in hazy images, have low intensity in at least one of R-G-B channels. This intensity is almost entirely contributed by airlight. These dark pixels can be used to recover airlight and estimate haze transmission for the entire image.
- Atmospheric Turbulence Correction
Our turbulence removal algorithm removes the atmospheric turbulent deformations and spatially varying blur in frames captured through a turbulent atmospheric medium. Traditional algorithms fail to perform in degraded environment while not performing when the quality of image goes down. Our algorithm employs AI based learning network and performs the correction in real time even when the other image enhancement modules and tracker algorithms are in the pipeline.
- Digital Noise Reduction
Image noise is caused by disturbance in video signal or low light situations. It shows up in the image as grainy specks. Our real-time filtering algorithms can reduce the noise revealing the underlying details with more clarity.
- Low Light Enhancement
Many real-world applications like video surveillance, automated driving, long range monitoring, image tracking etc. are to perform even in low visibility conditions. We estimate the illumination level in the image and refine the images iteratively to obtain better visibility. Our algorithm also tries to minimize the level of over-saturation while enhancing the images.
- Digital Sensor Fusion
Fusion enables real time pixel level fusion between different imaging channels. Pixel level fusion is a technique of intelligently integrating coherent spatial and temporal information, from sensors operating in different spectrum of wavelength, into a compact form. Fusion can increase informational content to the viewer when images from different sensors working in different wavelength spectrum are combined seamlessly together.
- Electronic Stabilisation
Video stabilisation is the removal of unwanted, parasitic vibrations in the video sequence induced by camera motion. For the purpose of image alignment, we employ hierarchical image processing algorithms for sub-pixel accurate finer alignment. We apply advanced motion filters to segregate intended motion from jitter and vibrations in the video.
