Detecting illuminant flicker
A method of determining when an image capture device with a rolling shutter is in an environment having a flickering illuminant by using autocorrelation or frequency analysis of difference vectors produced from first and second captured images.
1. A method of determining when an image capture device with a rolling shutter is in an environment having a flickering illuminant, comprising:
(a) storing rows of pixels from a first captured digital image in a memory;
(b) using the rows of pixels of the first capture digital image to provide a first column vector;
(c) providing a first-order difference vector from the first column vector; and clipping the elements in the first-order difference vector of the first column vector and reconstructing those clipped elements into a first modified column vector;
(d) storing rows of pixels from a second captured digital image in the memory;
(e) using the rows of pixels of the second captured digital image to provide a second column vector;
(f) providing a first-order difference vector from the second column vector; and clipping the elements in the first-order difference vector of the second column vector and reconstructing those clipped elements into a second modified column vector;
(g) subtracting the first modified column vector from the second modified column vector to provide a refined difference vector; and
(h) computing autocorrelations of the refined difference vector at specified lag(s) and using the autocorrelations to determine when the image capture device with the rolling shutter is in an environment having a flickering illuminant.
2. The method of claim 1 further including adjusting the image capture device to change the exposure time when flicker has been determined.
3. The method of claim 1 wherein lag of step (f) is for 100 Hz or 120 Hz.
4. A method of determining when an image capture device with a rolling shutter is in an environment having a flickering illuminant, comprising:
(a) storing rows of pixels from a first captured digital image in a memory;
(b) converting each pixel of each row of the first capture digital image into a logarithmic digital representation providing a first converted digital image and using the rows of the first converted digital image to provide a first column vector;
(c) providing a first-order difference vector from the first column vector; and clipping the elements in the first-order difference vector of the first column vector and reconstructing those clipped elements into a first modified column vector;
(d) storing rows of pixels from a second captured digital image in the memory;
(e) converting each pixel of each row of the second captured digital image into a logarithmic digital representation providing a second converted digital image and using the rows of the second converted digital image to provide a second column vector;
(f) providing a first-order difference vector from the second column vector; and clipping the elements in the first-order difference vector of the second column vector and reconstructing those clipped elements into a second modified column vector;
(g) subtracting the first modified column vector from the second modified column vector to provide a refined difference vector; and
(h)computing autocorrelations of the refined difference vector at specified lag(s) and using the autocorrelations to determine when the image capture device with the rolling shutter is in an environment having a flickering illuminant.
5. The method of claim 4 further including adjusting the image capture device to change the exposure time when flicker has been determined.
6. The method of claim 4 wherein lag of step (f) is for 100 Hz or 120 Hz.
7. A method of determining when an image capture device with a rolling shutter is in an environment having a flickering illuminant, comprising:
(a) storing rows of pixels from a first captured digital image in a memory;
(b) converting each pixel of each row of the first capture digital image into a logarithmic digital representation providing a first converted digital image and using the rows of the first converted digital image to provide a first column vector;
(c) providing a first-order difference vector from the first column vector; and clipping the elements in the first-order difference vector of the first column vector and reconstructing those clipped elements into a first modified column vector;
(d) storing rows of pixels from a second captured digital image in the memory;
(e) converting each pixel of each row of the second captured digital image into a logarithmic digital representation providing a second converted digital image and using the rows of the second converted digital image to provide a second column vector;
(f) providing a first-order difference vector from the second column vector; and
clipping the elements in the first-order difference vector of the second column vector and reconstructing those clipped elements into a second modified column vector;
(g) subtracting the first modified column vector from the second modified column vector to provide a refined difference vector; and
(h) computing using frequency analysis on the refined difference vector to determine when the image capture device with the rolling shutter is in an environment having a flickering illuminant.
8. The method of claim 7 further including adjusting the image capture device to change the exposure time when flicker has been determined.
9. The method of claim 7 wherein the frequency analysis of step (f) is at a frequency for 100 Hz or 120 Hz.