This paper represents super resolution technique for higher level of scale. It describes categorization like if lower the resolution, the lesser details are captured and for higher resolution the image details are higher and with greater image quality. Image sensors (e.g. CCD) and optics, camera, lens are the parameter which limits the imagine system. In real applications to capture high resolution images using constructing imaging tools like optical chips, lens, and optical component is not practical due to its high cost. Where, it is all time acceptable to improve quality of images or video captured already as the legacy from the development of digital imaging. Key way to overcome the resolution problem is to use machine learning techniques to post-process the captured images i.e. Super-resolution (SR).