Low-Light Image Enhancement Using Adaptive Digital Pixel Binning

Implement a low light camera by a smart pixel binning method.

The mission is to implement an algorithm for High Dynamic Range (HDR) enhancement of low light images, by method of smart pixel binning as described in an article.

[Abstract of the article: This paper presents an image enhancement algorithm for low-light scenes in an environment with insufficient illumination. Simple amplification of intensity exhibits various  undesired artifacts: noise amplification, intensity saturation, and loss of resolution. In order to enhance low-light images without undesired artifacts, a novel digital binning algorithm is proposed  that considers brightness, context, noise level, and anti-saturation of a local region in the image. The proposed algorithm does not require any modification of the image sensor or additional frame-memory; it needs only two line-memories in the image signal processor (ISP). Since the proposed algorithm does not use an iterative computation, it can be easily embedded in an existing digital camera ISP pipeline containing a high-resolution image sensor. ]

This project is supervised by a researcher in the new established Computational Photography Lab (Prof. Anat Levin).

Click here for more proposals