Search

Non-rigid registration is crucial in imaging, in particular, to adjust deformities produced during image acquisition and improve the accuracy of datasets. However, conventional imaging systems lack the desired speed and computational bandwidth for additional non-rigid registration of the deformed images. Therefore, such functionality is usually unavailable in time-critical settings. Expensive computations and memory intensive characteristics of non-rigid image registration algorithms such as the Demons algorithm further limits the realization of such systems. In response, we propose an alternative and efficient custom hardware-based Demons registration algorithm which utilizes pipelined streaming models to minimize memory fetches for computation. Designed for highly customizable hardware, our design only requires single-pass of images to compute the Demons kernel. Implementation results on the Xilinx ML605 FPGA system is presented and quantitatively evaluated in clock cycle counts in contrast with a software-based implementation. [ABSTRACT FROM PUBLISHER]

The laser scanning confocal endomicroscope (LSCEM) is used to perform non-invasive, in vivo imaging at cell level. Online 3D LSCEM data visualization is thought of through a novel realtime computing system. We present such a system which performs three tasks in parallel: (i) provide automated controls for sequential acquisition of cross-sectional live images across depths on the tissue; (ii) manage and synchronize imaging and rendering tasks in real time; (iii) perform realtime rendering on incrementally accumulated LSCEM datasets. Experimental results are promising. [ABSTRACT FROM AUTHOR]

Our recent work on windowed Fourier transform for fringe patterns analysis is reviewed, including the theoretical analyses of the algorithms, windowed Fourier filtered and quality-guided phase unwrapping, denoising and phase extraction techniques for a single closed fringe pattern, and algorithm acceleration using multicore computers and graphics processing units. [ABSTRACT FROM AUTHOR]

Fringe patterns produced by various optical interferometric techniques encode the information of deformation, refractive index, vibration, etc. Noise as one of the key problems affects further processing of the fringe patterns and reduces the final measurement quality. Coherence enhancing diffusion (CED) is a partial differential equation based denoising model that suppresses the noise as well as preserves the flow-like structure. Windowed Fourier transform-based windowed Fourier filtering (WFF) is another useful fringe pattern denoising tool that removes noise by thresholding the windowed Fourier transform spectrum. An adaptive windowed Fourier filtering (AWFF) that denoises the fringe pattern based on pixels' local frequencies is proposed in this paper. The performance of AWFF is compared with WFF and CED by applying them to fringe patterns that contain speckle noise and different levels of frequencies. Quantitative results will be given on simulated fringe patterns. Experimental fringe pattern will also be tested to illustrate the performance of these methods. [ABSTRACT FROM AUTHOR]