Resolution Of Input Image Research Articles

A fully pipelined and parallel hardware architecture for real-time BRISK salient point extraction

Scale and rotation invariant salient point detection and matching algorithms are variously used in computer vision applications such as image matching, 3D localization and pose estimation. Recently, hardware implementation of image and video processing algorithms has emerged as a viable solution to handle the high computational complexity of applications like 3D pose estimation with several processing stages. The hardware implementation of various stages of theses algorithms can be executed in a pipelined manner to ensure the reality of time. In this paper, a new and fully pipelined hardware architecture is proposed for salient point detection using Binary Robust Invariant Scalable Keypoints (BRISK) algorithm. BRISK algorithm is a binary keypoint extractor that detects salient points by constructing a scale-space pyramid; therefore, its fixed-point hardware implementation in a pipelined manner is challenging because of the required synchronization for various layers in scale domain. The proposed hardware architecture was implemented using Verilog Hardware Description Language, and the functionality of the design was validated through several experiments. The proposed design was synthesized by using an ASIC digital design flow utilizing 180 nm CMOS technology as well as a Virtex-4 FPGA. The design is clocked at 90.91 MHz in ASIC implementation and achieves processing rate of 169.29 frames/s while running on input images with 800 × 600 resolution. The throughput of FPGA implementation is 180.44 frames/s with 96.89 MHz clock frequency for the same input image resolution. Experimental results confirm the efficiency of the proposed hardware architecture in comparison with software implementation.

PatchMatch belief propagation meets depth upsampling for high‐resolution depth maps

For stereo matching, PatchMatch belief propagation (PMBP) gives an efficient way of inferencing continuous labels on the Markov random field. Nevertheless, it still requires considerable time when the resolution of input images is high. To handle high-resolution images, a two-step stereo method is proposed that efficiently exploits PMBP by depth upsampling. In the first step, PMBP is conducted on the random field whose nodes correspond to the downsampled pixels from an input image. As a result, accurate low-resolution disparity maps are efficiently obtained by taking advantage of PMBP. In the second step, the low-resolution disparity map is upsampled while considering depth boundaries and sub-pixel accuracy. Experimental results show that the proposed method provides more accurate disparity maps than the original PMBP while reducing computation time remarkably.

Single image super-resolution via blind blurring estimation and anchored space mapping

It has been widely acknowledged that learning-based super-resolution (SR) methods are effective to recover a high resolution (HR) image from a single low resolution (LR) input image. However, there exist two main challenges in learning-based SR methods currently: the quality of training samples and the demand for computation. We proposed a novel framework for single image SR tasks aiming at these issues, which consists of blind blurring kernel estimation (BKE) and SR recovery with anchored space mapping (ASM). BKE is realized via minimizing the cross-scale dissimilarity of the image iteratively, and SR recovery with ASM is performed based on iterative least square dictionary learning algorithm (ILS-DLA). BKE is capable of improving the compatibility of training samples and testing samples effectively and ASM can reduce consumed time during SR recovery radically. Moreover, a selective patch processing (SPP) strategy measured by average gradient amplitude |grad| of a patch is adopted to accelerate the BKE process. The experimental results show that our method outruns several typical blind and non-blind algorithms on equal conditions.

ANTHROPOMETRIC MEASUREMENT BASED ON STRUCTURE FROM MOTION IMAGING TECHNIQUE

The aim of this work was to estimate limitations related of the actual resolution of images taken into the object reconstruction procedure (using Structure from Motion method) applied for anthropometric measurements. This is necessary to define technical parameters of equipment used for this application. A series of 42 images of the sample static object were taken with use of high resolution camera. The object was in a stable position, properly illuminated with stable, diffused light sources. With use of VisualSFM software a virtual cloud of points representing real points of examined object was calculated. During tests a resolution of input images were changed then the calculation time and influence of the number of points in the cloud onto deviation of selected anthropometric parameters were analyzed. In conditions of the performed experiment decreasing of the resolution of input images to the level of 80 % of initial value (or lower) has significant influence onto results of the measurements of selected anthropometric parameters. Structure from Motion (SfM) technique can be adopted for anthropometric measurement systems when the input images are taken with high resolution imaging device (over a dozen of million pixels). Presented measurement method for anthropometry is characterized by ease of use, do not need any calibration before measurements or specialized equipment. Thank to this can be adopted in practice in any conditions and operated by unskilled operator.

All Common Subsequences for Face Recognition

In recent years, face recognition has become one of the hottest research topics aimed at biometric applications. Comparing with other biometrics recognition, face recognition provides more natural means for perceptual interface. However, face recognition algorithms weakly perform under some common conditions, which include the variation of facial expressions or lightening conditions, the occlusion of faces like wearing glasses or mask, the low resolution or noises of input images, and the like. The other problem is the recognition efficiency, especially when the facial database is tremendous. This paper presents all common subsequences (ACS) as the kernel function (similarity method) to solve the time series problem. Experiments on 4 public face databases: Caltech, Jaffe, Orl and Yale databases, demonstrate that ACS can achieve higher recognition accuracy than some classic face recognition methods, e.g. 2DPCA and 2DLDA.These instructions give you basic guidelines for preparing camera-ready papers for conference proceedings.

Design and Implementation of Real-time Fabric Defect Detection System

Fabric defect detection is a key part to ensure quality of product in textile industry. The paper addressed on structure design and implementation of real-time fabric defect detection system based on Gabor filtering. The system mainly consists of three parts of image acquisition, defect detection and defect classification. Defect detection is core of whole system, the basic structure of which is PCIbased image data transmission as well as process system based on FPGA and external RAM. The Gabor filtering algorithm is used for defect detection, which is implemented with high performance FPGA assisted by external RAM to buffer image data. The CCD camera based rapid high-resolution image acquisition, high-speed PCI transmission as well as defect detection algorithm implemented with FPGA meet real-time needs of system. Practices show that the system can accurately detect more than 20 kinds of fabric defects in real time. When the resolution of input image is 2048×512 and fabric speed is 60 m/s, defect detection accuracy can be up to 0.5mm and accuracy about 90%.

Application of Bayesian image superresolution to spectral reflectance estimation

Hyperspectral imaging provides us with high-dimensional, scene-independent color information but faces problems such as long image acquisition time and severe lighting and focusing conditions. To achieve efficient spectral imaging, this study presents an extended method of Bayesian image superresolution. The proposed method increases both the spatial and the wavelength resolution of input images and enables the processing of hyperspectral images to a higher resolution from easily acquired low-spatial-resolution multispectral images. In an experiment using acquired multispectral images of a Japanese traditional painting, visible and near-infrared hyperspectral images were produced, and the obvious effect of superresolution was validated.

Super-Resolution Without Dense Flow

Super-resolution is a widely applied technique that improves the resolution of input images by software methods. Most conventional reconstruction-based super-resolution algorithms assume accurate dense optical flow fields between the input frames, and their performance degrades rapidly when the motion estimation result is not accurate enough. However, optical flow estimation is usually difficult, particularly when complicated motion is presented in real-world videos. In this paper, we explore a new way to solve this problem by using sparse feature point correspondences between the input images. The feature point correspondences, which are obtained by matching a set of feature points, are usually precise and much more robust than dense optical flow fields. This is because the feature points represent well-selected significant locations in the image, and performing matching on the feature point set is usually very accurate. In order to utilize the sparse correspondences in conventional super-resolution, we extract an adaptive support region with a reliable local flow field from each corresponding feature point pair. The normalized prior is also proposed to increase the visual consistency of the reconstructed result. Extensive experiments on real data were carried out, and results show that the proposed algorithm produces high-resolution images with better quality, particularly in the presence of large-scale or complicated motion fields.

MEDICAL IMAGE SEGMENTATION TECHNIQUES FOR VIRTUAL ENDOSCOPY

Virtual endoscopes give internal views of the human body without penetrating it, based on a set of parallel cross-sections produced with any computer tomography method. This paper presents some ideas concerning the design and implementation of a software system, which acts like a virtual endoscope. It takes into account the general requirements of the system, gives a solution that uses a multi-step algorithm, and finally shows the resulting 3-D images. Most of the algorithmic steps have several possible solutions. Our virtual endoscope establishes 3-D internal views based on sets of 2-D slices, which originate from magnetic resonance imaging devices. The chain of the applied image processing methods consists of the followings: (1) Adequate pre-filtering to eliminate the low-frequency intensity non-uniformity (INU) artifact, and high-frequency “salt-and-pepper” disturbances; (2) Segmentation of the stack of MRI slices using an enhanced fuzzy C-means algorithm; (3) 3-D surface recovery algorithm based on level set methods and fast marching methods; (4) Interactive visualization using modern computer graphics technologies, providing the possibility to measure distances, areas, volumes as well. The quality of service provided by the chosen method mainly depends on the resolution of input images.

Deformable model with a complexity independent from image resolution

We present a parametric deformable model which recovers image components with a complexity independent from the resolution of input images. The proposed model also automatically changes its topology and remains fully compatible with the general framework of deformable models. More precisely, the image space is equipped with a metric that expands salient image details according to their strength and their curvature. During the whole evolution of the model, the sampling of the contour is kept regular with respect to this metric. By this way, the vertex density is reduced along most parts of the curve while a high quality of shape representation is preserved. The complexity of the deformable model is thus improved and is no longer influenced by feature-preserving changes in the resolution of input images. Building the metric requires a prior estimation of contour curvature. It is obtained using a robust estimator which investigates the local variations in the orientation of image gradient. Experimental results on both computer generated and biomedical images are presented to illustrate the advantages of our approach.

Multiprocessor system for high-resolution image correlation in real time

In this work, the correlation-based image localization problem is targeted toward implementation for real-time operation on high-resolution images. The main contribution resides in a global approach to the two-dimensional correlation problem which concurrently considers the algorithm, the system architecture, and its implementation. Analyzing and studying the alternatives concerning algorithm specification, VLSI architecture, and system architecture at different levels of the design phase, a scalable multiprocessor system, based on multiple instances of identical modules, is derived which achieves a real-time image correlation performance exceeding that of currently available solutions by an order of magnitude concerning frame-rate and input image resolution. As a result, the relevance of image correlation traditionally limited to the image processing domain is now expanded into applied machine vision applications. Addressing packaging and reusability aspects, an MCM implementation of the relevant system components is presented.

Radiometric normalization, compositing, and quality control for satellite high resolution image mosaics over large areas

An objective normalization procedure has been developed to create image mosaics of radiometric equalization radiometric normalization for image mosaics (RNIM). The procedure employs a band-specific principal component analysis for overlap areas to achieve accurate and consistent radiometric transforms in each spectral band. It is demonstrated that the result of radiometric equalization is independent of the order of images to be mosaicked after the radiometric normalization adjustment is made. The selection of corresponding pixel pairs in the overlap area is controlled by using band-specific linear correlation coefficients, and the criteria for rejecting the cloudy and land-cover changed pixels. The final result is controlled quantitatively by employing the first and second principal components for the input data, which in turn depend on the selection of corresponding pixel pairs in the overlap area. In general, the radiometric resolution of input images can be conserved as long as gain /spl ges/1 and offset /spl ges/0 because of the stored format of the unsigned integer. The RNIM procedure accommodates these conditions. To take the best advantage of the data in the overlap areas, a pixel-based composite technique is employed in the production of the final mosaic. The selection of corresponding pixel pairs and the final result can be controlled and assessed with quantitative criteria. Therefore, this approach produces an objective, analyst-independent result and can be automated. The method has been successfully applied to six Landsat TM images of the BOREAS transect in Saskatchewan and Manitoba, Canada.

Sensitivity of transvenous defibrillation models to adaptive mesh density and resolution: the potential for interactive solution times

The voltage gradients induced in ventricular myocardium by an electric shock have been shown to correlate to the probability of the shock producing a successful defibrillation. Finite element modeling is one method for computing these voltage gradients, although the meshing of complex biomedical domains is difficult on a patient-specific basis. We recently described an adaptive algorithm that automates the generation of finite element meshes for complex 3-D domains from bitmapped images. This article examines the sensitivity of the computed distribution of ventricular voltage gradients to the resolution of the images and to the adapted density of the mesh. The results allow us to establish an adaptation stopping criterion and a minimum input image resolution for modeling transvenous defibrillation. The sensitivity to adapted mesh density was analyzed by comparing voltage gradient histograms from successively finer meshes to histograms from a uniform mesh at the maximum possible density. Comparisons were made using the Kolmogorov-Smirnov test with the number of samples required to detect a 5% difference in the histograms at the 0.05 significance level. Adaptation to a global current density error estimate of 5% or less was required in order to achieve acceptance of the null hypothesis that the distributions were the same in all cases. Defibrillation efficacy, however, is predicted from the voltage gradient in the first quartile, and the results suggest that this region of the cumulative histogram converges faster during mesh adaptation than the histogram as a whole. We also compared histograms from models generated from successively finer input images. The histogram of each model was compared with the histogram obtained from the finest possible resolution. In all cases, the null hypothesis of no difference was accepted at resolutions of 2.3 × 2.3 × 3.0 mm. The average time required to build and adapt models to a 5% accuracy at the first quartile at this resolution was 1.8 min. on a common workstation. We believe that this demonstrates a potential for the eventual synthesis of finite element computations into interactive electrode placement tools on a subject-specific basis.