No abstract

The use of power cepstrum analysis in image registration is explored. Rotational shifts and translational shifts are corrected separately. The technique involves two main ideas. First, after preprocessing to remove extraneous information and information which could result in false registration parameters, a rotational shift is changed into a translational shift by using the shift-invariant property of the power spectrum. Second, power cepstrum analysis is used to correct the translational shift. Because of the introduction of these ideas, this new algorithm can work very fast and accurately compared to conventional correlation techniques. This registration technique is applied to sequential fundus images with potential application in detecting changes in fundus anomalies.

No abstract

A feature tracker is only as good as the features found by the feature detector. Common feature detectors such as Harris, Sobel, Canny, and Difference of Gaussians convolve an image with a specific kernel in order to identify "corners" or "edges". This convolution requires, however, that the source image contain only one value (or color channel) per pixel. This requirement has reduced the scope of feature detectors, trackers, and descriptors to the set of gray scale (and other single-channel) images. Due to the standard 3-channel RGB representation for color images, highly useful color information is typically discarded or averaged to create a gray scale image that current detectors can operate on. This removes a large amount of useful information from the image. We present in this paper the color Difference of Gaussians algorithm which outperforms the gray scale DoG in number and quality of features found. The color DoG utilizes the YCbCr color space to allow for separated processing of intesity and chrominance values. An embedded vision sensor based on a low power field programmable gate array (FPGA) platform is being developed to process color images using the color DoG with no reduction in processing speed. This low power vision sensor will be well suited for autonomous vehicle applications where size and power consumption are paramount.

Precise registration techniques are essential for quantitative evaluation of sequential fundus images to make early detection of fundus anomalies feasible.The familiar correlation techniques for achieving such image registration are computationally intensive and suffer from non -uniqueness of solution. We have developed an accurate, yet fast, algorithm for image registration by using a combination of power spectrum and power cepstrum analyses.In this new algorithm rotational shifts and translational shifts are corrected separately.The technique involves two main ideas. First, a rotational shift is corrected and changed into a translational shift by computing Fourier power spectrums. After the rotational shift has been corrected, i.e., images are parallel, the remaining translational shifts are handled.Because of the accuracy characteristics of the power cepstrum and the speed of theFFT, this new algorithm can work very fast and accurately compared to conventional techniques.Also, the cepstrum technique has better tolerance of image noise than the traditional correlation measures.The accuracy obtained and computational time required for the cepstrum -based registration techniques will be illustrated by operating on sequential fundus images used in early detection of glaucoma.