Digital image filters and related methods for image contrast enhancement
Digital image filters and related methods for image contrast enhancement are disclosed. According to one aspect of the method, an invariant brightness level is initially determined. For each pixel of an input image, the invariant brightness level is subtracted from the input brightness of the pixel. The resulting value is multiplied with a contrast adjustment constant. After that, the invariant brightness level is added. Further aspects of the method can involve histogram equalization.
1. A visual prosthesis comprising:
a video input device for providing video data;
a video processing unit, including a digital signal processor, receiving the video data from the video input device and converting the video data into a video output image, including an array of pixels, the video processing unit including:
an image contrast enhancement filter based on an invariant brightness level, and the formula Iout=(Iin−L) C+L, where Iout is output intensity, Iin is input intensity, L is the invariant brightness level, and C is a predetermined multiplier, and each pixel's difference from the invariant brightness level, where L is determined by segmentating the image to find objects, calculating an average brightness for each object, and setting L to be the darkest average brightness;
a video filter processor producing the video output image;
a telemetry engine transmitting the video output image; and
a neural stimulation system suitable to receive the video output image and to stimulate visual percepts according to the video output image.
2. The visual prosthesis of claim 1 , wherein the image contrast enhancement filter is based on image histogram equalization, so that an output image output by the image contrast enhancement filter has an output image histogram which is an equalized version of the input image histogram.
3. The visual prosthesis of claim 2 , wherein image histogram equalization operates through a nonlinear monotonic scaling function.
4. The visual prosthesis of claim 2 , wherein the image contrast enhancement filter operates in the time domain.
5. The visual prosthesis of claim 1 , wherein filtering through the contrast enhancement filter is automatically set.
6. The visual prosthesis of claim 2 , wherein filtering through the contrast enhancement filter is applied automatically on every video frame of the video data.
7. The visual prosthesis of claim 2 , wherein filtering through the contrast enhancement filter is an adaptive filtering.
8. The visual prosthesis of claim 7 , wherein the adaptive filtering depends on a luminance distribution of input images at any instant of time.
9. The visual prosthesis of claim 2 , wherein the image histogram equalization is based on a temporal average.