A Real-Time Edge Detector: Algorithm and VLSI Architecture

A robust, real-time ellipse detector

“…We follow the same steps as that in Ref. [15]. A simpliÿed non-maximum suppression of edge strength is performed within a 3 × 3 mask to determine the possible edge points.…”

“…Our system ÿrst converts the color image in RGB to the gray-scale image. We then apply the Absolute Di erence Mask (ADM) algorithm [15] to the image to produce the edge image, together with the information about the edge strength and direction. After we have obtained the edge image, we post-process the edge image by removing spurious edge points to build an edge point map from which we estimate edge convexity based on neighborhood directions.…”

A Cellular Automaton Processor for Line and Corner Detection in Gray-Scale Images

“…Also the proposed architecture produces very satisfactory frame rate results when compared to other, similar architectures [9,10].…”

“…For the set of 8 masks and for a 2566256 image the frame rate of the proposed CA architecture with the chip operating at 100 MHz is 190 frames/s. This frame rate is quite satisfactory when compared to the 25 frames/s rate of the computer vision chip presented in [10] and the 30 frames/s rate of the edge detection architecture presented in [9].…”

CAST: An automated segmentation and tracking tool for the analysis of transcriptional kinetics from single-cell time-lapse recordings

“…As the first step towards segmentation, a background subtraction (as described in 2.1) can be applied, here affecting only this step of the analysis. Second, default de-noising is performed by applying a Gaussian filter with a radius of 0.6 px [59] and by subtracting the estimated mean of the uniform Gaussian white noise, replacing the homogeneity analyzer proposed in [60,61], which is utilized to identify the empty portions of the image, by the Absolute Difference Mask (ADM) edge detector [62]. Candidate object locations are then detected using the ''à trous'' wavelet transform [30], followed by a filtering step in which only single-pixel local maxima detections are kept.…”