A method of characterizing and enhancing the pixels in an image, which is captured at all lighting conditions. The method enable dynamically enhancement of images captured at extremely low lighting conditions. Each pixel of the image, represented by I SUB (x, y) where x is the horizontal axis coordinate value and y is the vertical axis coordinate value, contains three color-components. Each pixel is characterized by defining a relationship between the three color-components of that pixel in its initial state. A nonlinear expansion function, which adaptively adjusts the expansion rate with respect to the statistical property of the original image, is used for enhancing the brightness of the image so that the poorly lighted pixels are enhanced in intensity more as compared to the amount of enhancement for the brighter pixels. The color property of each pixel in the enhanced image is then restored using the characteristic features previously defined by the relationship of the color components. This procedure provides an enhanced image, which is natural in its appearance and has sharp visibility to human observation. In addition, this new technique is fast enough to provide a real time enhancement of images from a video stream.