Edge enhancement

About: Edge enhancement is a research topic. Over the lifetime, 2324 publications have been published within this topic receiving 30962 citations.

Text Detection Using MSER and Stroke Width Transform

Abstract: Text extraction, detection and recognition in images aims at integrating advanced text-based searching technologies which is now recognized as a main component in the development of advanced image and video annotation and retrieval systems. It is also used for image in painting. In this paper, we present a novel algorithm for detecting text in natural and complex images. Basically, it is a two step approach combining maximally stable extremal regions (MSER) and stroke width transformation for the accurate detection of text. Firstly, the MSER image is obtained on which canny edge detection is performed for edge enhancement. To exclude non-text parts, the image is filtered using stroke width information as well as geometric filtering. Experimental results show the effective performance of the proposed method.

Deblurred Images Post-Processing by Poisson Warping

Abstract: In this work we develop a post-processing algorithm which enhances the results of the existing image deblurring methods. It performs additional edge sharpening using grid warping. The idea of the proposed algorithm is to transform the neighborhood of the edge so that the neighboring pixels move closer to the edge, and then resample the image from the warped grid to the original uniform grid. The proposed technique preserves image textures while making the edges sharper. The effectiveness of the method is shown for basic deblurring methods on LIVE database images with added blur and noise.

An Automatic White Balance Method Based on Edge Detection

Abstract: Automatic white balance is an important function of digital still and video cameras. The goal of white balance is to adjust the color of the pixels under different illumination automatically. In this paper, we proposed a new method of automatic white balance. In our algorithm, we detected the image edges on the value of C b and C r firstly, and then took the pixels near edges in the both side as references pixels to calculate the correction coefficients for R, G and B respectively. The experimental results show that our algorithm performs well even though the images are dominated by few chromatic regions and short of highlight near-white highlight pixels..

Fourier optical preprocessing in lieu of deep learning

Abstract: Deep learning convolutional neural networks generally involve multiple-layer, forward-backward propagation machine-learning algorithms that are computationally costly. In this work, we demonstrate an alternative scheme to convolutional neural nets that reconstructs an original image from its optically preprocessed, Fourier-encoded pattern. The scheme is much less computationally demanding and more noise robust, and thus suited for high-speed and low-light imaging. We introduce a vortex phase transform with a lenslet-array to accompany shallow, dense, “small-brain” neural networks. Our single-shot coded-aperture approach exploits the coherent diffraction, compact representation, and edge enhancement of Fourier-transformed spiral phase gradients. With vortex encoding, a small brain is trained to deconvolve images at rates 5–20 times faster than those achieved with random encoding schemes, where greater advantages are gained in the presence of noise. Once trained, the small brain reconstructs an object from intensity-only data, solving an inverse mapping without performing iterations on each image and without deep learning schemes. With vortex Fourier encoding, we reconstruct MNIST Fashion objects illuminated with low-light flux (5nJ/cm2) at a rate of several thousand frames per second on a 15 W central processing unit. We demonstrate that Fourier optical preprocessing with vortex encoders achieves similar accuracies and speeds 2 orders of magnitude faster than convolutional neural networks.

Use of Quadtrees for Edge Enhancement

Abstract: : A quadtree approximation of a 2 to the nth power by 2 to the nth power gray level image tends to enhance edges that are strong in the original image and suppress those that are weak. An edge detector applied to such an approximation produces edges that are in general stronger than those obtainable from the original, though slightly displaced. Methods of combining the edge maps obtained from the original image and its quadtree approximation are examined. It is shown that the choice of a suitable combination rule results in an edge map which is superior to that obtainable by conventional means.