Nowadays it becomes more and more critical to process the increasingly large amounts of data in timely manner. In order to meet this requirement and ensure the reliable processing of streaming data, a variety of distr...
详细信息
ISBN:
(纸本)9781450364607
Nowadays it becomes more and more critical to process the increasingly large amounts of data in timely manner. In order to meet this requirement and ensure the reliable processing of streaming data, a variety of distributed stream processing architectures and platforms have been developed, which handles the fundamental task of allocating processing tasks to the currently available physical resources and routing streaming data between these resources. However, many stream processingsystems lack an intelligent scheduling mechanism, in which their default schedulers allocate tasks without taking resource demands and availability, or the transfer latency between resources into consideration. Besides stream processing has a strict request for latency. Thus it's important to give latency guarantee for distributed stream processing. In this paper, we propose a new algorithm for stream processing with latency guarantee, the algorithm both consider transfer latency and resource demand in the process of task allocation. Extensive experiments verify the correctness and effectiveness of our approach. Under the condition of satisfying the latency constraints, the heuristic algorithm AHA on average, reduce more than 21.3% and 58.9% resources compared with the greedy and the round-robin algorithms.
The intent of the digital image fusion is a process to obtain important information from acquired images and then form as a distinct fused image. image fusion algorithms are popular in transform domain than spatial do...
详细信息
ISBN:
(纸本)9789811055089;9789811055072
The intent of the digital image fusion is a process to obtain important information from acquired images and then form as a distinct fused image. image fusion algorithms are popular in transform domain than spatial domain methods. The usual methods in transform domain are block-based and multi-resolution transforms. Commonly used orthogonal transforms for imageprocessing are SVD, DCT, KLT, CT, and DWT, but hardware implementation of these transforms is difficult because of the floating-point arithmetic operations. Hadamard transform (HT) is preferred, where the computational speed is the criterion for real-time implementation. In general, block-based methods suffer from blocking artifacts. It influences the features of the fused image. To reduce these problems, statistical measures like mean, contrast, and variance are applied. In the current proposal, statistical measures like entropy and uniformity are explored in HT domain. Further, all statistical measures in HT domain are compared and analyzed. Application of statistical measures in HT domain gives better image fusion results than conventional HT domain fused techniques. Dominance of the uniformity measure in HT domain is observed based on the experimental results.
The present paper proposes image edge detection algorithm utilizing a reaction-diffusion network. The network consists of two-dimensionally coupled FitzHugh-Nagumo type neurons, whose behavior is described by a set of...
详细信息
aSummary: This note presents the design of a scalable software package named imagePy for analysing biological images. Our contribution is concentrated on facilitating extensibility and interoperability of the software...
详细信息
aSummary: This note presents the design of a scalable software package named imagePy for analysing biological images. Our contribution is concentrated on facilitating extensibility and interoperability of the software through decoupling the data model from the user interface. Especially with assistance from the Python ecosystem, this software framework makes modern computer algorithms easier to be applied in bioimage analysis. Availability and implementation: imagePy is free and open source software, with documentation and code available at https://***/image-Py/imagepy under the BSD license. It has been tested on the Windows, Mac and Linux operating systems. Contact: yxdragon@*** or wzjdlut@***
Since digital images require a large space on the storage devices and the network bandwidth, many compression methods have been used to solve this problem. Actually, these methods have, more or less, good results in t...
详细信息
ISBN:
(纸本)9783319625218
Since digital images require a large space on the storage devices and the network bandwidth, many compression methods have been used to solve this problem. Actually, these methods have, more or less, good results in terms of compression ratio and the quality of the reconstructed images. There are two main types of compression: the lossless compression which is based on the scalar quantization and the lossy compression which rests on the vector quantization. Among the vector quantization algorithms, we can cite the Kohonen's network. To improve the compression result, we add a pre-processing phase. This phase is performed on the image before applying the Kohonen's network of compression. Such a phase is the wavelet transform. Indeed, this paper is meant to study and model an approach to image compression by using the wavelet transform and Kohonen's network. The compression settings for the approach to the model are based on the quality metrics rwPSNR and MSSIM.
An automatic recognizer system based in Artificial Intelligence for thermographic images of the electric power distribution network is proposed in this article. The infrared thermography is usually used to conduct ins...
详细信息
An automatic recognizer system based in Artificial Intelligence for thermographic images of the electric power distribution network is proposed in this article. The infrared thermography is usually used to conduct inspections in electrical power distribution lines, assisted by a human operator, which is usually responsible for operating all the equipment, selecting the hottest spots in the image (corresponding to the places needing maintenance), making reports and calling the technical team, which will do the repairs. The proposed automatic diagnosis system aims to replace the manual inspection operation using imageprocessingalgorithms. An old method of segmentation for thermal images known as JSEG is implemented and tested and a Deep Learning Neural Network is responsible to recognize the segmented elements. A comparison between the exclusive Deep Learning based image recognition with the same method preceded by the JSEG segmentation algorithm is done in this article, showing better performance with this previous segmentation of the thermographic images. (C) 2018 The Authors. Published by Elsevier Ltd.
This study introduces an innovative non-contact sensing technique for vision-based displacement measurement. Existing vision-based displacement measurement techniques utilizes physical target panels or physical featur...
详细信息
ISBN:
(数字)9781510616936
ISBN:
(纸本)9781510616936
This study introduces an innovative non-contact sensing technique for vision-based displacement measurement. Existing vision-based displacement measurement techniques utilizes physical target panels or physical features to compute relative displacement between the target and the observation point. Instead, the proposed method exploits the optical reference of a speckle pattern. A coherent light that is diffusely reflected on the surface of the target structure creates the speckle pattern. In this study, a camera records the changes in the speckle pattern in real time. Because the speckle pattern is sensitive to small changes of surface, the ambient vibration is enough to affect it. To estimate the displacement of the target from the raw speckle images, speckle contrast imaging (SCI), speckle flow imaging (SFI), and k-means clustering algorithm were used. After SCI and SFI quantifies the blurring effect in each image, the k-means clustering algorithm creates virtual sensing node from each image. The connection of virtual nodes from frame to frame highlights the displacements of the surface in time domain. Because the algorithms are time-consuming and computationally intensive, a GPU executes the entire post-processing operation in parallel and identifies the natural frequencies of the structure.
The Screen content images (SCIs) are images containing textual and pictorial regions, which have become more and more connected with our daily life with the widespread adoption of multimedia applications. In particula...
详细信息
ISBN:
(数字)9781510622005
ISBN:
(纸本)9781510622005
The Screen content images (SCIs) are images containing textual and pictorial regions, which have become more and more connected with our daily life with the widespread adoption of multimedia applications. In particular, the image quality assessment (IQA) of SCIs is important because of its good property to guide and optimize lots of imageprocessingsystems. However, the no-reference (NR) IQA algorithms receive little attention and achieve unsatisfactory performance. Hence, this paper proposes a novel no-reference IQA method based on patch-wise multi-order derivatives for SCIs. This method includes two stages: patch-wise image quality evaluation and quality pooling. The first stage focuses on learning visual quality of local regions. Two features of image patches are extracted: multi-order derivative statistics, multi-order derivative histograms, which respectively describe the global and local information of the multi-order derivatives. Then the support vector regression (SVR) is applied to measure visual quality of image patches given a set of extracted features. The second stage aims at pooling patch-wise quality to an overall quality score with weights derived from entropy of gradient information of SCIs. Experimental results show that our method obtains superior performance against state-of-the-art NR-IQA approaches on the SIQAD database of SCIs, and also achieves competitive performance against state-of-the-art FR-IQA methods for SCIs.
Background: The male-based infertility test known as a spermiogram involves a manual count using a Makler counting chamber. There is a need to develop an automated sperm-counting system to provide more precise diagnos...
详细信息
Background: The male-based infertility test known as a spermiogram involves a manual count using a Makler counting chamber. There is a need to develop an automated sperm-counting system to provide more precise diagnoses. To that end, the automatic detection of Regions of Interest (ROI) in Makler images constitutes the first phase to using the advantages of the Makler chamber in a computerized counting system. Methods: ROI are defined between grids, hence, another challenging issue, that of exact grid detection, is examined. In this study, initially we reviewed several line detection algorithms with their applications and possible usage on the grid-detection problem of Makler images. Next, a combinational grid-detection technique, particularly for Makler images, was improved upon. Results: In summary, the Hough transform method has been enhanced by a combined approach of using Line Segment Detector, the clustering of slope angles, and post processing. The K-means method is deployed to refine the grids and to find the direction of grid lines to use in Hough transform. In the grid-detection step, the presented technique is evaluated with a template-matching technique following the Sorensen-Dice index. It gives 95.3% accuracy and 88.5% F-measure scores. Discussion: ROI extraction is performed based on grid detection output by multiple logical queries. Each extracted region, clarified from the grid lines, was identically examined for sperm count. Fuzzy c-means clustering was first performed to segment the objects in ROI, then blob analysis was utilized to eliminate non-sperm objects. Coclusion: The proposed sperm analysis approach was then compared to the visual assessment technique. Results indicate that the proposed system might be useful in laboratories, but still needs to be improved in the feature extraction process.
This book constitutes the refereed proceedings of the 12th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2016, and three parallel workshops, held in Thessaloniki, ...
ISBN:
(纸本)9783319831688
This book constitutes the refereed proceedings of the 12th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2016, and three parallel workshops, held in Thessaloniki, Greece, in September 2016. The workshops are the Third Workshop on New Methods and Tools for Big Data, MT4BD 2016, the 5th Mining Humanistic Data Workshop, MHDW 2016, and the First Workshop on 5G - Putting Intelligence to the Network Edge, 5G-PINE 2016. The 30 revised full papers and 8 short papers presented at the main conference were carefully reviewed and selected from 65 submissions. The 17 revised full papers and 7 short papers presented at the 3 parallel workshops were selected from 33 submissions. The papers cover a broad range of topics such as artificial neural networks, classification, clustering, control systems - robotics, data mining, engineering application of AI, environmental applications of AI, feature reduction, filtering, financial-economics modeling, fuzzy logic, genetic algorithms, hybrid systems, image and video processing, medical AI applications, multi-agent systems, ontology, optimization, pattern recognition, support vector machines, text mining, and Web-social media data AI modeling.
暂无评论