In portable, 3-D, or ultra-fast ultrasound (US) imaging systems, there is an increasing demand to reconstruct high quality images from limited number of data. However, the existing solutions require either hardware ch...
详细信息
ISBN:
(纸本)9781538646595
In portable, 3-D, or ultra-fast ultrasound (US) imaging systems, there is an increasing demand to reconstruct high quality images from limited number of data. However, the existing solutions require either hardware changes or computationally expansive algorithms. To overcome these limitations, here we propose a novel deep learning approach that interpolates the missing RF data by utilizing the sparsity of the RF data in the Fourier domain. Extensive experimental results from sub-sampled RF data from a real US system confirmed that the proposed method can effectively reduce the data rate without sacrificing the image quality.
This article presents two scheduling algorithms applied to the processing of astronomical images to detect cosmic rays on distributed memory high performance computing systems. We extend our previous article that prop...
详细信息
ISBN:
(纸本)9783319579726;9783319579719
This article presents two scheduling algorithms applied to the processing of astronomical images to detect cosmic rays on distributed memory high performance computing systems. We extend our previous article that proposed a parallel approach to improve processing times on image analysis using the image Reduction and Analysis Facility IRAF software and the Docker project over Apache Mesos. By default, Mesos introduces a simple list scheduling algorithm where the first available task is assigned to the first available processor. On this paper we propose two alternatives for reordering the tasks allocation in order to improve the computational efficiency. The main results show that it is possible to reduce the makespan getting a speedup=4.31 by adjusting how jobs are assigned and using Uniform processors.
Accurate license plate localization is the most important prerequisite in ANPR (Automatic Number Plate Recognition) systems. Majority of the existing algorithms use a single feature to obtain the license plate locatio...
详细信息
ISBN:
(纸本)9781728128436
Accurate license plate localization is the most important prerequisite in ANPR (Automatic Number Plate Recognition) systems. Majority of the existing algorithms use a single feature to obtain the license plate location which causes to potential false detections. In this article we propose a robust methodology using 16 statistical features while we still preserve real-time processing of the system which is a requirement for such applications. The proposed method uses a Vertical Projection technique and Discrete Fourier Transform (DFT) in order to extract multiple statistical features, as well as K-means clustering and multilayer perceptron neural network technique to identify the location of a license plate in an image. The method is compared with the state-of-the-art research in the field and the effectiveness of the method is evaluated for various types of license plates with different scripts.
The paper employs theoretic and methodological generalization in order to develop new algorithms of orthogonal transformation in the field of aerospace photo processing. Employing systems of Vilenkin-Chrestenson funct...
详细信息
ISBN:
(纸本)9781509065301
The paper employs theoretic and methodological generalization in order to develop new algorithms of orthogonal transformation in the field of aerospace photo processing. Employing systems of Vilenkin-Chrestenson functions (VCF) in a non-trigonometrical space, we obtain a minimally possible shape of their construction. The paper describes algorithms and filtration samples (quasi-two-dimensional filtering and correlation analysis of aerospace photos when applied to distortions and image noise.
Diabetic Retinopathy (DR) is an eye ailment affecting millions of diabetes patients globally. DR is a systemic pathology, one that occurs due to problems in the insulin levels of the body. Glucose intolerance badly af...
详细信息
ISBN:
(纸本)9781467387651
Diabetic Retinopathy (DR) is an eye ailment affecting millions of diabetes patients globally. DR is a systemic pathology, one that occurs due to problems in the insulin levels of the body. Glucose intolerance badly affects the blood circulatory network including the micro-vascular structure of the retina where damaged vessels dump unwanted liquids into it. Early detection is the key here and if not detected and treated in time, DR results in partial to complete blindness in a high percentage of all diabetics worldwide [1]. This review is an attempt at a rundown of various imageprocessingalgorithms and systems employed for the recognition of these retinal pathologies. Although the grading efficacy is examined, most of the algorithms are honed in for a target set of patients so abstraction among these is not easy. It is also exhibited that the feature extraction of these pathology detection algorithms has seen stable improvement so much so that their efficiency now competes human expert levels diagnosis.
In practical media distribution systems, visual content often undergoes multiple stages of quality degradations along the delivery chain between the source and destination. By contrast, current image quality assessmen...
详细信息
ISBN:
(纸本)9781509021758
In practical media distribution systems, visual content often undergoes multiple stages of quality degradations along the delivery chain between the source and destination. By contrast, current image quality assessment (IQA) models are typically validated on image databases with a single distortion stage. In this work, we construct two large-scale image databases that are composed of more than 2 million images undergoing multiple stages of distortions and examine how state-of-the-art IQA algorithms behave over distortion stages. Our results suggest that the performance of existing IQA models degrades rapidly with distortion stages, especially when the distortion types of different stages vary. We also find that full-reference and no-reference frameworks, though both readily applicable, have major drawbacks at predicting the quality of images at middle distortion stages. However, when the quality level of the previous stage is accessible, significantly improved quality prediction performance may be achieved. This study points out a new avenue of degraded-reference IQA research that is both practically desirable and technically challenging.
The issues of enhanced vision multispectral systems application for robotic complexes control in the Arctic are considered. The existing contrast enhancement methods are observed. Probability characteristics of images...
The issues of enhanced vision multispectral systems application for robotic complexes control in the Arctic are considered. The existing contrast enhancement methods are observed. Probability characteristics of images being subject to contrast enhancement parameters are estimated. Based on these characteristics, the authors concluded that image areas requiring the greatest contrast enhancement are the areas with low saturation and magnitude gradients, at certain brightness values. image quality improvement method is proposed. It performs processing only in the areas where it is necessary to enhance the contrast, practically without affecting the most homogeneous or structured image parts. The processed image saturation remains due to the processing of both luminance channel and saturation channel. The algorithm proposed also provides contrast enhancement of shaded image areas. The calculated values of various objective image quality indices indicate that the contrast enhancement algorithm proposed provides better results than known approaches. In addition, different spectral range image fusion algorithm ensuring visibility in the presence of interfering factors is proposed. It differs from known methods by adaptive weight adjustment in different areas of image. The example confirming the effectiveness of the fusion method proposed is shown. For its comparison with known methods, the values of fusion objective quality indices are calculated. The fusion algorithm proposed is shown to surpass known methods by various quality assessments. The conclusion about the expediency of using the algorithms developed in technical vision systems of robotic complexes in the Arctic is made.
Advances in modern medical imaging technologies such as X-Ray, Computed Tomography (CT), Ultra Sound (US) imaging, Magnetic Resonance Imaging (MRI), Positron emission tomography (PET) and Single Photon Emission Comput...
详细信息
ISBN:
(纸本)9781538658741
Advances in modern medical imaging technologies such as X-Ray, Computed Tomography (CT), Ultra Sound (US) imaging, Magnetic Resonance Imaging (MRI), Positron emission tomography (PET) and Single Photon Emission Computed tomography (SPECT) enable better disease diagnoses and treatment assessment. This paper explains a layered architecture suitable for design and development of application software for medical imaging devices. The generic nature of medical imaging devices in the process of data acquisition, signals processing and image reconstruction is the major inspiration behind conceptualization of this architecture. Also the currently available medical imaging software follow a multi-stage interlinked processing workflow which can be directly mapped to the layered software approach. Layered software approach facilitates quick and easy customization, configuration and feature enhancements of the medical imaging software. The architecture facilitates the academic community or researchers to build end user solutions based on research outputs, which can be directly integrated and used in the main software workflow. This indeed provides opportunity for utilizing the technical expertise available for implementation of algorithms which can directly be used to interface with medical imaging devices. The architecture discussed in this paper, has been employed in the design of MRI imaging software as a case study to further illustrate its applicability in signal generation, data acquisition and image reconstruction.
This paper presents an effective and robust algorithm to detect the lanes in highway. It uses Hough Transform to fit the lane line of top view of the road and extracts the most representative lane line in each categor...
详细信息
ISBN:
(纸本)9781538626191
This paper presents an effective and robust algorithm to detect the lanes in highway. It uses Hough Transform to fit the lane line of top view of the road and extracts the most representative lane line in each category after clustering all the lines, which is then followed by a post-processing step. The results show that this algorithm can effectively reduce the disturbance of vehicles and guardrails to achieve 90% correct rate.
Latent fingerprints are fingerprint impressions unintentionally left on surfaces at a crime scene. Such fingerprints are usually incomplete or partial, making it challenging to match them to full fingerprints register...
详细信息
Latent fingerprints are fingerprint impressions unintentionally left on surfaces at a crime scene. Such fingerprints are usually incomplete or partial, making it challenging to match them to full fingerprints registered in fingerprint databases. Latent fingerprints may contain few minutiae and no singular structures. Matching algorithms that entirely rely on minutiae or alignment of singular structures fail when those structures are missing. This paper presents an approach for matching latent to rolled fingerprints using the (a) similarity of learned representations of patches and (b) the minutiae on the correlated patches. A deep learning network is used to learn optimized representations of image patches. Similarity scores between patches from the latent and reference fingerprints are determined using a distance metric learned with a convolutional neural network. The matching score is obtained by fusing the patch and minutiae similarity scores. The proposed system was tested by matching fingerprints segmented from the 258 latent fingerprints in the NIST SD27 database against a database of 2,257 rolled fingerprints from NIST SD27 and SD4 databases. Experimental results show a rank-1 identification rate of 81.35% and highlights the promise of our proposed approach.
暂无评论