Information graphics, such as graphs and plots, are used in technical documents to convey information to humans and to facilitate greater understanding. Usually, graphics are a key component in a technical document, a...
详细信息
ISBN:
(纸本)9781479918058
Information graphics, such as graphs and plots, are used in technical documents to convey information to humans and to facilitate greater understanding. Usually, graphics are a key component in a technical document, as they enable the author to convey complex ideas in a simplified visual format. However, in an automatic text recognition system, which are typically used to digitize documents, the ideas conveyed in a graphical format are lost. We contend that the message or extracted information can be used to help better understand the ideas conveyed in the document. In scientific papers, line plots are the most commonly used graphic to represent experimental results in the form of correlation present between values represented on the axes. The contribution of our work is in the series of image processing algorithms that are used to automatically extract relevant information, including text and plot from graphics found in technical documents. We validate the approach by performing the experiments on a dataset of line plots obtained from scientific documents from computer science conference papers and evaluate the variation of a reconstructed curve from the original curve. Our algorithm achieves a classification accuracy of 91 % across the dataset and successfully extracts the axes from 92 % of line plots. Axes label extraction and line curve tracing are performed successfully in about half the line plots as well.
Dealing with visual data is the key for environmental monitoring tasks in Wireless Multimedia Sensor Networks (WMSNs). Tasks such as object detection, recognition, and/or tracking do require extracting and using the r...
详细信息
Dealing with visual data is the key for environmental monitoring tasks in Wireless Multimedia Sensor Networks (WMSNs). Tasks such as object detection, recognition, and/or tracking do require extracting and using the right information from the inherently large amount of visual data. The widely accepted solution of legacy WSNs, transmitting the acquired data to a central base station for further processing, would render a WMSN totally useless because of the unacceptable use of bandwidth and energy. Therefore, we consider the in situ processing as a viable solution for WMSNs. However, processing power and memory capacity restrictions of existing multimedia sensor nodes along with their power consumption are the limiting factors for wide-spread use of in situ processing. Nevertheless, recent technological improvements and introduction of the new ARM cores encourage us to evaluate the imageprocessing capabilities of ARM7/ARM9/ARM11 based micro-controllers for in situ processing in WMSNs. In this work, we first discussed the architectural design differences among the various ARM cores. Then we classified image processing algorithms into three categories. Then, we evaluated the performance of each microcontroller by running a set of basic image processing algorithms necessary for object detection, recognition, and/or tracking. The test results show that ARM11 runs up to 6-30 times faster than ARM9 and ARM7, respectively. Besides, ARM11 consumes up to 5-7 times less energy than ARM9 and ARM7 for the same type of operations.
Isotropic image edge enhancement with high contrast can be achieved using a spiral phase filter (SPF) in a 4f optical system. However, real-time application of edge enhancement with SPF has generally been limited due ...
详细信息
Isotropic image edge enhancement with high contrast can be achieved using a spiral phase filter (SPF) in a 4f optical system. However, real-time application of edge enhancement with SPF has generally been limited due to the requirement of coherent light or complex phase-shifting operation. In this paper, we demonstrate a real-time image edge enhancement method using a SPF and a graphic processing unit (GPU). By implementing the process of virtual spiral phase filtering on GPU, we are able to speed up the whole procedure by more than 8.3x with respect to CPU processing, and ultimately achieve video rate for megapixel images. In particular, our implementation can achieve higher speedup for more multiple images. These developments are increasing the potential for image edge enhancement of moving objects. (C) 2014 Optical Society of America
In this paper, the night sky star imageprocessing algorithm, consisting of image preprocessing, star pattern recognition, and centroiding steps, is improved. It is shown that the proposed noise reduction approach can...
详细信息
In this paper, the night sky star imageprocessing algorithm, consisting of image preprocessing, star pattern recognition, and centroiding steps, is improved. It is shown that the proposed noise reduction approach can preserve more necessary information than other frequently used approaches. It is also shown that the proposed thresholding method unlike commonly used techniques can properly perform image binarization, especially in images with uneven illumination. Moreover, the higher performance rate and lower average centroiding estimation error of near 0.045 for 400 simulated images compared to other algorithms show the high capability of the proposed night sky star imageprocessing algorithm. (c) 2014 Optical Society of America
A three-dimensional (3D) digital reconstruction method for integral imaging with high random-error tolerance based on statistics is proposed. By statistically analyzing the points reconstructed by triangulation from a...
详细信息
A three-dimensional (3D) digital reconstruction method for integral imaging with high random-error tolerance based on statistics is proposed. By statistically analyzing the points reconstructed by triangulation from all corresponding image points in an elemental images array, 3D reconstruction with high random-error tolerance could be realized. To simulate the impacts of random errors, random offsets with different error levels are added to a different number of elemental images in simulation and optical experiments. The results of simulation and optical experiments showed that the proposed statistic-based reconstruction method has relatively stable and better reconstruction accuracy than the conventional reconstruction method. It can be verified that the proposed method can effectively reduce the impacts of random errors on 3D reconstruction of integral imaging. This method is simple and very helpful to the development of integral imaging technology. (C) 2015 Optical Society of America
Non-local means (NLM) filter is one of the state-of-the-art denoising filters. It exploits the presence of similar features in an image and averages those similar features to remove noise. However, a conventional NLM ...
详细信息
Non-local means (NLM) filter is one of the state-of-the-art denoising filters. It exploits the presence of similar features in an image and averages those similar features to remove noise. However, a conventional NLM filter shows somewhat inferior performance of noise reduction around edges, suffering from low efficiency of collecting similar features to be averaged. In order to overcome this phenomenon, we propose a NLM filter with double Gaussian anisotropic kernels as a substitute for the conventional homogeneous kernel to effectively remove noise from OCT images corrupted by speckle noise. The proposed filter was evaluated by comparing with various denoising filters such as conventional NLM filter, median filter, bilateral filter, and Wiener filter. The fingertip OCT images, which were processed with the different denoising filters, indicated that the proposed NLM filter provides superior denoising performance, among the filters in terms of the contrast-to-noise ratio (CNR), the equivalent number of looks (ENL), and the speckle suppression index (SSI). A human retina OCT image was also used to compare and show the performances of noise reduction among different filters. In addition, the denoising performance with the proposed NLM filter was also investigated in the synthetic images for fair comparison among the filters by calculating the peak signal-to-noise ratio (PSNR). The proposed NLM filter outperformed the conventional NLM filter as well as the other filters. (C) 2015 Optical Society of America
Determining the authenticity of integrated circuits is paramount to preventing counterfeit and malicious hardware from being used in critical military, healthcare, aerospace, consumer, and industry applications. Exist...
详细信息
Determining the authenticity of integrated circuits is paramount to preventing counterfeit and malicious hardware from being used in critical military, healthcare, aerospace, consumer, and industry applications. Existing techniques to distinguish between authentic and counterfeit integrated circuits (ICs) often include destructive testing requiring subject matter experts. We present a nondestructive technique to detect ICs using x- ray microscopy and advanced imaging analysis with different pattern recognition approaches. Our proposed method is completely automated, and runs in real time. In our approach, images of an integrated circuit are obtained from an x- ray microscope. Local binary pattern features are then extracted from the x- ray image, followed by dimensionality reduction through principal component analysis, and alternatively through a nonlinear principal component methodology using a stacked autoencoder embedded in a deep neural network. From the reduced dimension features, we train two types of learning machines, a support vector machine with a nonlinear kernel and a deep neural network. We present experiments using authentic and ICs to demonstrate that the proposed approach achieves an accuracy of 100% in distinguishing between the counterfeit and authentic samples. (C) 2015 Optical Society of America
When a marine target is detected by a mid-infrared detector on a sunny day, the target's information could be lost if it is located in sun glint. Therefore, we developed a new mid-infrared imaging system capable o...
详细信息
When a marine target is detected by a mid-infrared detector on a sunny day, the target's information could be lost if it is located in sun glint. Therefore, we developed a new mid-infrared imaging system capable of effectively detecting marine targets in regions of strong sun glint, which is presented in this report. Firstly, the theory of the analysis methods employed in different detection scenarios is briefly described to establish whether one or two polarizers should be utilized to suppress further the p-polarized component of sun glint. Secondly, for the case in which a second polarizer is employed, the formula for the optimum angle between the two polarizers is given. Then, the results of our field experiment are presented, demonstrating that the developed system can significantly reduce sun glint and can enhance the contrast of target images. A commonly used imageprocessing algorithm proved capable of identifying a target in sun glint, confirming the effectiveness of our proposed mid-infrared polarization imaging system.
Accurate measurements of oceanic whitecap coverage from whitecap images are required for better understanding air-gas transfer and aerosol production processes. However, this is a challenging task because whitecap pat...
详细信息
Accurate measurements of oceanic whitecap coverage from whitecap images are required for better understanding air-gas transfer and aerosol production processes. However, this is a challenging task because whitecap patches are formed immediately after a wave breaks and are spread over a wide area. The main challenges in designing whitecaps imaging instrument are the small field of view of the camera lens, processing huge numbers of images, recording data over long time periods, and deployment difficulties in stormy conditions. This paper describes the hardware design of a novel high-resolution optical instrument for imaging oceanic whitecaps and the automated algorithm processing the collected images. The instrument was successfully deployed in 2013 as part of the HiWINGS campaign in the North Atlantic Ocean. The instrument uses a fish-eye camera lens to image the whitecaps in a wide angle of view (180°).
In the field of non-destructive testing, Terahertz (THz) tomography is a modern imaging technique permitting 3D inspection of opaque objects. A tomographic procedure reconstructs a 3D volume of the acquired object by ...
详细信息
In the field of non-destructive testing, Terahertz (THz) tomography is a modern imaging technique permitting 3D inspection of opaque objects. A tomographic procedure reconstructs a 3D volume of the acquired object by intersecting the data contained on the projection set. This depth information is extracted from several projections acquired around the object at different viewing angles. This property has made THz tomography a complementary technique to X-Rays. Several reconstruction methods have been investigated for THz tomography [1] but they must be adapted to THz wave properties. For monochromatic source, Beer-Lambert law describes the attenuation encountered by THz waves through the sample and measured on the detector, when reflexion and refraction effects are neglected. Then, as their complements in X-Ray tomography, we investigate in this paper on a Maximum-Likelihood expectation-maximization for TRansmission tomography (ML-TR) compatible with THz radiations. This method, based on Poisson distribution model of measured radiations, takes into account the Gaussian propagation of THz beam [2] and allows introducing some a priori knowledge about the imaged object. Especially, we focus on the implementation denoted Ordered Subsets Convex (OSC) algorithm since it has an efficient convergence rate despite of the noise level and the sparsity of acquired data [3]. We discuss how this new reconstruction is able to estimate physical properties of samples acquired with a 100GHz/300GHz scanner in addition to the 3D reconstruction. Moreover, we present an innovative data and imageprocessing sequence to perform non-destructive inspection from 3D terahertz (THz) images. After a 3D tomographic reconstruction of a sample, a preset segmentation affords the different regions of interest (ROI) composing the inert part of sample. Then a 3D visualization and dimensional measurements could be performed on these ROI, separately, in order to provide new informations of the studied sam
暂无评论