The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Increasing application of Gabor feature space in various computer vision tasks and its high computational demand, encourages using parallel computing technologies. In this work we have designed a high throughput GPU based Gabor kernel that mimics the function of initial biological visual cortex layers namely `Simple' and `Complex' cells. The kernel is basically a Gabor filter bank with adjustable...
Clinical research suggests that changes in the retinal blood vessels (e.g., vessel caliber) are important indicators for earlier diagnosis of diabetes and cardiovascular diseases. Reliable vessel detection or segmentation is a prerequisite for quantifiable retinal blood vessel analysis for predicting these diseases. However, the segmentation of blood vessels is complicated by its huge variations such...
Kernel regression has been previously proposed as a robust estimator for a wide range of image processing tasks, including image denoising, interpolation and super resolution. In this article we propose a kernel formulation that relaxes the usual symmetric and unimodal properties to effectively exploit the smoothness characteristics of natural images. The proposed method extends the kernel support...
Medical diagnosis is the major challenge faced by the medical experts. Highly specialized tools are necessary to assist the experts in diagnosing the diseases. Gestational Diabetes Mellitus is a condition in pregnant women which increases the blood sugar levels. It complicates the pregnancy by affecting the placental growth. The ultrasound screening of placenta in the initial stages of gestation helps...
This paper presents a novel technique of obtaining the Region of Interest (ROI) which is square-shaped. After palm print segmentation, we perform texture feature extraction using Local Binary Pattern (LBP) based on the directional gradient response. With the feature vectors so obtained we perform matching using Support vector machines (SVM) and also using the Chi-square distance measurement. The experimental...
This paper proposes an embedded vision system for real-time moving object tracking using modified mean-shift algorithm for mobile robot application. This design of modified mean-shift algorithm fully utilizing the advanced parallelism of Field Programmable Gate Arrays (FPGA) is capable of processing real-time PAL video of 720*576 at 25 fps. This hardware implementation realizes time-consumed color...
Classification is an important task in Hyperspectral data analysis. Hyperspectral images show strong correlations across spatial and spectral neighbors. Theoretically, classifier designed with a joint spectral and spatial correlations can improve classification performance than classifier which only utilize one of the correlations. Gaussian Processes(GPs) have been used for Hyperspectral imagery classification...
While the computational power of Field Programmable Gate Arrays (FPGA) makes them attractive as code accelerators, the lack of high-level language programming tools is a major obstacle to their wider use. Graphics Processing Units (GPUs), on the other hand, have benefitted from advanced and widely used high-level programming tools. This paper evaluates the performance, throughput and energy of both...
The most annoying artifacts in image deconvolution are ringing and amplified noise. These artifacts can be reduced significantly by regularization using the Maximum a Posteriori (MAP) method that exploits not only the likelihood but also the image prior in image deconvolution. Although ringing and noise can be reduced significantly with strong regularization, image details are also reduced, so the...
This paper introduces certain innovative algorithms to mask pixel defects in image sensors. Pixel defectivity rates scale with pixel architecture and process nodes. Smaller pixel and process nodes introduce more defects in manufacturing. Brief introduction to causes for pixel defectivity at lower pixel nodes is explained. Later in the paper, popular defect correction schemes used in image processing...
Scan-converting lines for raster-based displays is a basic building block of many computer graphics tasks. While many fast line drawing algorithms are known, most produce thin and 'jagged' lines due to aliasing. Wu's algorithm includes a crude approximation of antialiasing, which still includes noticeable step edges. Even hardware multisampling cannot entirely eliminate aliasing. Instead, the proper...
In this paper a novel Gabor filter hardware scheme for the fingerprint image enhancement is presented. For each pixel of the image, we use accurate local frequency and orientation to generate the corresponding convolution kernel and thus achieve a better enhancement effect. And compared to the previous works, our design yields a higher throughput which is due to the pipeline techniques. Moreover the...
The underwater robots called Unmanned Underwater Vehicles (UUVs) take over complex and dangerous underwater missions that were previously performed by humans. These vehicles operate in the unknown environments and make their own decisions within the mission based on the readings of the sensors, without any link with a human operator. Independent of the mission, it is critical for the AUVs to be able...
Although trivial background subtraction (BGS) algorithms (e.g. frame differencing, running average...) can perform quite fast, they are not robust enough to be used in various computer vision problems. Some complex algorithms usually give better results, but are too slow to be applied to real-time systems. We propose an improved version of the Extended Gaussian mixture model that utilizes the computational...
Background subtraction is a traditional method for detecting objects in stationary background. However, this traditional method is difficult to detect objects accurately in the real world, because the background is usually cluttered and not completely static. In this paper, we propose an object detection approach using Ant Colony System (ACS) in a MAP-MRF framework. For object segmentation, a MAP-MRF...
Handwritten character recognition has received extensive attention in academic and production fields. The recognition system can be either on-line or off-line. Off-line handwriting recognition is the subfield of Optical Character Recognition . In this paper, We introduce the fundamental principles of Chinese handwritten numerals, including digital image preprocessing, segmentation, features extraction...
An improve method of the block wise non-local means (BNL - means) is proposed for removing noise in the digital image. This method consists of the spectral decomposition of the Gaussian weighted matrix, the pseudo filter constructions, computations of the pseudo weighted coefficients and image denoising using the weighted sum of Gaussian. Experimental results show that this method is simpler, more...
Paper often includes pre-printed ruling lines to help people write more neatly. This particular example of real- world noise can have a serious impact on applications such as handwriting recognition and writer identification, however. In this work, we investigate the effects of ruling lines on writer ID. We study a method for detecting and removing ruling lines and test its utility for Arabic writer...
We present a statistical framework to merge the information from silhouettes segmented in multiple view images to infer the 3D shape of an object. The approach is generalising the robust but discrete modelling of the visual hull by using the concept of averaged likelihoods. One resulting advantage of our framework is that the objective function is continuous and therefore an iterative gradient ascent...
Optical flow methods, such as Lucas-Kanade and Horn-Schunck algorithms, are popular in motion estimation. However, they fall short on accuracy when they are applied to blurred videos. Some people utilize hybrid camera system to get a low resolution image to suppress the blurring effect so that more accurate optical flow for blurred high resolution image can be further derived, though in most of the...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.