The proceedings contain 269 papers. The topics discussed include: high-speed multiview 3D structured light imaging technique;integration of advanced stereo obstacle detection with perspectively correct surround views;...
The proceedings contain 269 papers. The topics discussed include: high-speed multiview 3D structured light imaging technique;integration of advanced stereo obstacle detection with perspectively correct surround views;recreating Van Gogh's original colors on museum displays;modeling long range features from serial section imagery of continuous fiber reinforced composites;the quality of stereo disparity in the polar regions of a stereo panorama;do different radiologists perceive medical images the same way? some insights from representational similarity analysis;the intersection of artificial intelligence and augmented reality;self-calibrated surface acquisition for integrated positioning verification in medical applications;gradient management and algebraic reconstruction for single image super resolution;segmentation-based detection of local defects on printed pages;automated optical inspection for abnormal-shaped packages;towards combining domain knowledge and deep learning for computational imaging;a 360-degrees holographic true 3D display unit using a Fresnel phase plate;and visual analytic process to familiarize the average person with ways to apply machine learning.
Smart agriculture has involved evolution, judgment, and application of new methods of using modern technology. Technological advances in agriculture will enable farmers to enhance their skills in farming. We planned t...
详细信息
The proceedings contain 166 papers. The topics discussed include: data visualization and analysis of playing styles in tennis;crosstalk minimization method for eye-tracking-based 3D display;detecting deepfakes with Ha...
The proceedings contain 166 papers. The topics discussed include: data visualization and analysis of playing styles in tennis;crosstalk minimization method for eye-tracking-based 3D display;detecting deepfakes with Haralick’s texture properties;turkey behavior identification system with a GUI using deep learning and video analytics;projection methods for finding intersection of two convex sets and their use in signal processing problems;high-speed inline computational imaging for area scan cameras;interdisciplinary immersive experiences within artistic research, social and cognitive sciences;a single chip PPG sensor with enhanced IR sensitivity for low power and small size;single chip auto-valet parking system with TDA4VMID SoC;boosting computer vision performance by enhancing camera ISP;transfer learning with style transfer between the photorealistic and artistic domain;knowledge distillation from multispectral images for fish freshness estimation;next best light position: a self-configuring approach for the reflectance transformation imaging acquisition process;experimental study for revising visual noise measurement of ISO 15739;and computational identification of significant actors in paintings through symbols and attributes.
imageprocessing is usually associated with pattern recognition and is rather treated as a subject outside of the computer graphics interest. Basically computer graphics algorithms are used for the visualization of sc...
详细信息
imageprocessing is usually associated with pattern recognition and is rather treated as a subject outside of the computer graphics interest. Basically computer graphics algorithms are used for the visualization of scenes or models described using some abstract notation, while imageprocessing is used on the opposite way -i.e. when finding an abstract description of an analyzed pattern. This paper proposes to use the imageprocessing approach for rendering optical effects in computer graphics algorithms. Proposed algorithms can be used for the generation of realistic and artistic optical effects in real-time, in, for example, visual simulators, virtual reality or multi-media applications. (C) 1997 Elsevier Science B.V.
Many imageprocessing operations can be abstracted into matrix operations. With the help of matrix analysis, we can understand the inherent properties of the operations and thus design better algorithms. In this paper...
详细信息
Many imageprocessing operations can be abstracted into matrix operations. With the help of matrix analysis, we can understand the inherent properties of the operations and thus design better algorithms. In this paper, we propose a matrix decomposition method referred to as identity-plus-row decomposition, The decomposition is particularly useful in design of parallel projection algorithms on mesh-connected computers. Projection is a frequently used process in imageprocessing and visualization. In volume graphics, projection is used to render the essential content of a three-dimensional volume onto a two-dimensional image plane. For Radon transform, projection is used to transform the image space into a parameter space. By applying the identity-plus-row matrix decomposition method, we solve the data redistribution problem due to the irregular data access patterns present in those applications on single instruction stream, multiple data stream (SIMD) mesh-connected computers, developing fast algorithms for volume rendering and Radon transform on SIMD mesh-connected computers. (C) 2001 SPIE and IS&T.
This paper describes a set of image segmentation algorithms for mineral froth images, based on gray-value valley detection and a kind of image classification. The size, shape, texture and color of froth bubbles are ve...
详细信息
This paper describes a set of image segmentation algorithms for mineral froth images, based on gray-value valley detection and a kind of image classification. The size, shape, texture and color of froth bubbles are very important pieces of information for production optimization in mineral processing. In order to determine these parameters, bubbles in a froth image first have to be delineated. Froth images display a large variation of image patterns and quality, thus it is difficult to use only a single algorithm for segmenting all images. To achieve successful segmentation the images are first classified into image classes. Then sets of segmentation algorithms are used, based on the different image classes. The segmentation algorithms and classification algorithms have been tested in a laboratory and in industrial on-line systems for froth images, the test results show that they are robust for froth images. The processing speed for the segmentation algorithm is much faster than for a standard morphological segmentation algorithm. The processing accuracy is comparable to manual drawn result. This test shows that the algorithms work satisfactorily. (C) 2003 Elsevier Ltd. All rights reserved.
Signal, image and Synthetic Aperture Radar imagery algorithms in recent time are used in a daily routine. Due to huge data and complexity, their processing is almost impossible in a real time. Often imageprocessing a...
详细信息
ISBN:
(纸本)9781538669792
Signal, image and Synthetic Aperture Radar imagery algorithms in recent time are used in a daily routine. Due to huge data and complexity, their processing is almost impossible in a real time. Often imageprocessingalgorithms are inherently parallel in nature, so they fit nicely into parallel architectures multicore Central processing Unit (CPU) and Graphics processing Unit GPUs. In this paper imageprocessingalgorithms were evaluated, which are capable to execute in parallel manner on several platforms CPU and GPU. All algorithms were tested in TensorFlow, which is a novel framework for deep learning, but also for imageprocessing. Relative speedups compared to CPU were given for all algorithms. TensorFlow GPU implementation can outperform multi-core CPUs for tested algorithms, obtained speedups range from 3.6 to 15 times.
This paper presents an analytical performance prediction model and methodology that can be used to predict the execution time, speedup, scalability and similar performance metrics of a targe set of imageprocessing op...
详细信息
This paper presents an analytical performance prediction model and methodology that can be used to predict the execution time, speedup, scalability and similar performance metrics of a targe set of imageprocessing operations running on a p-processor parallel system. The model which requires only a few parameters obtainable on a minimal system can help in the systematic design, evaluation and performance tuning of parallel imageprocessingsystems. Using the model one can reason about the performance of a parallel imageprocessing system prior to implementation. The method can also support programmers in detecting critical parts of an implementation and system designers in predicting hardware performance and the effect of hardware parameter changes on performance. The execution of parallel imageprocessing operations was studied and operations were arranged in three main problem classes based on data locality and the communication patterns of the algorithms. The core of the method is the derivation of the overhead function, as it is the overhead that determines the achievable speedup. The overheads were examined and modelled for each class. The use of the method is illustrated by four class-representative imageprocessingalgorithms: image-scalar addition, convolution, histogram calculation and the Fast Fourier Transform. The developed performance model has been validated on a 16-node parallel machine and it has been shown that the model is able to predict the parallel run-time and other performance metrics of parallel imageprocessing operations accurately.
Application of a class of parallel processing architecture for implementation of imageprocessing and, more specifically, image pattern recognition is proposed. Noting the problem of optimally matching a wide variety ...
详细信息
Application of a class of parallel processing architecture for implementation of imageprocessing and, more specifically, image pattern recognition is proposed. Noting the problem of optimally matching a wide variety of possible processor configurations to a broad spectrum of different algorithms, a general framework for characterizing imageprocessingalgorithms is proposed, and the task of mapping the algorithm characteristics into processing system characteristics is begun. A simple image pattern recognition method is considered from this point of view. An algorithm is developed and alternative SIMD (single instruction stream - multiple data stream) parallel implementations are compared. Complexity analyses are presented to show the computational speedup made possible by the parallelism.
In this work we show the algorithms developed to extract the Cole parameters from multi-Frequency EIT. With these parameters it is possible to obtain information about various different tissues and their pathologies. ...
详细信息
In this work we show the algorithms developed to extract the Cole parameters from multi-Frequency EIT. With these parameters it is possible to obtain information about various different tissues and their pathologies. The algorithms developed obtain the Cole-model parameters from the real and imaginary parts of impedance, or using only the real part, without problems of convergence. A study of the influence of noise is performed with simulations. We find a correct solution in all cases with signal to noise ratio in the data higher than 40 dB. Finally, we show parametric images of the human abdomen obtained with these algorithms.
暂无评论