Many applications like forensic, surveillance, medical, and satellite imaging require zooming the area of interest in an image. However, digital zooming is limited by the spatial resolution of an imaging system. A plenoptic camera captures the radiance and provides both spatial and angular information. A Plenoptic camera is different from a conventional camera. It has an array of micro lens placed between objective lens and image sensor. Each Micro Lens Array (MLA) forms a whole image. We develop a plenoptic camera that has 28 \(\times\) 28 (MLA) and an objective lens of 25 mm. The designed plenoptic camera results in 784 low resolution sub-aperture images in single frame. Images were captured under lab-generated artificial turbulence at the receiver plane. We developed super-resolution algorithms to obtain turbulence-free high-resolution image. Our Super-resolution technique extracts sup-pixel information from multiple low-resolution images to get high-resolution images. We implemented noise adaptive parameters with the steepest gradient optimization to avoid over fitting and noise amplification. It was observed that the resolution of an image has increased without artifacts. The proposed algorithm is fast, easy to implement, and gives better results compared to other interpolation methods. We successfully obtained super resolved image up to 8 times (8 \(\times\) ) the original image with increased Signal-to-Noise Ratio and better Structured Similarity Index than other super-resolution methods.
