Neural network solution for a real-time no-reference video quality assessment of H.264/AVC video bitstreams

The ever-growing video streaming services require accurate quality assessment with often no reference to the original media. One primary challenge in developing no-reference (NR) video quality metrics is achieving real-timeliness while retaining the accuracy. A real-time no-reference video quality assessment (VQA) method is proposed for videos encoded by H.264/AVC codec. Temporal and spatial features are extracted from the encoded bit-stream and pixel values to train and validate a fully connected neural network. The hand-crafted features and network dynamics are designed in a manner to ensure a high correlation with human judgment of quality as well as minimizing the computational complexities. Proof-of-concept experiments are conducted via comparison with: 1) video sequences rated by a full-reference quality metric, and 2) H.264-encoded sequences from the LIVE video dataset which are subjectively evaluated through differential mean opinion scores (DMOS). The performance of the proposed method is verified by correlation measurements with the aforementioned objective and subjective scores. The framework achieves real-time execution while outperforming state-of-art full-reference and no-reference video quality assessment methods.

     

A quantitative comparison of the effects of modeling approaches on system verification using a controlled challenge problem

Requirements Engineering - To reduce program risks, engineering methods capitalizing on modeling and machine assistance have been extensively investigated within systems engineering (and more...     

A reappraisal of transition elements in linear elastic fracture mechanics

This article offers a detailed comparison of the transition elements described by P.P. Lynn and A.R. Ingraffea [International Journal for Numerical Methods in Engineering 12,1031–1036] and C. Manu[Engineering Fracture Mechanics 24,509–512]. The source of a numerical phenomenon in using Manu's transitionelement (TE) is explained. The effect of eight-noded TEs with differentquarter-point elements (QPE) on the calculated stress intensity factors (SIFs) isinvestigated. Strain at the crack tip is shown to be singular for any ray emanating from the crack tip within an eight-noded TE, but strain has bothr ^–1/2andr ^–1singularities, withr ^–1/2dominating for large TEs. Semi-transition elements (STEs) are defined and shown to have a marginal effect on the calculated SIFs. Nine-nodedtransition elements are formulated whose strain singularity is shown to be the same as that of eight-noded TEs. Then the effect of eight-noded and nine-noded TEs with collapsed triangular QPEs, and rectangular and nonrectangular quadrilateral eight-noded and nine-noded QPEs, is studied, and nine-noded TEs are shown to behave exactly like eight-noded TEs with rectangular eight-noded and nine-noded QPEs and to behave almost the same with other QPEs. The layered transition elements proposed by V. Murti and S.Valliapan [Engineering Fracture Mechanics 25, 237–258] areformulated correctly. The effect of layered transition elements is shown by two numerical examples.     

Blind spatial unmixing of multispectral images: New methods combining sparse component analysis,clustering and non-negativity constraints

Remote sensing has become an unavoidable tool for better managing our environment, generally by realizing maps of land cover using classification techniques. Traditional classification techniques assign only one class (e.g., water, soil, grass) to each pixel of remote sensing images. However, the area covered by one pixel contains more than one surface component and results in the mixture of these surface components. In such situations, classical classification is not acceptable for many major applications, such as environmental monitoring, agriculture, mineral exploration and mining, etc. Most methods proposed for treating this problem have been developed for hyperspectral images. On the contrary, there are very few automatic techniques suited to multispectral images. In this paper, we propose new unsupervised spatial methods (called 2D-Corr-NLS and 2D-Corr-NMF) in order to unmix each pixel of a multispectral image for better recognizing the surface components constituting the observed scene. These methods are related to the blind source separation (BSS) problem, and are based on sparse component analysis (SCA), clustering and non-negativity constraints. Our approach consists in first identifying the mixing matrix involved in this BSS problem, by using the first stage of a spatial correlation-based SCA method with very limited source sparsity constraints, combined with clustering. Non-negative least squares (NLS) or non-negative matrix factorization (NMF) methods are then used to extract spatial sources. An important advantage of our proposed methods is their applicability to the possibly globally underdetermined, but locally (over)determined BSS model in multispectral remote sensing images. Experiments based on realistic synthetic mixtures and real multispectral images collected by the Landsat ETM+ and the Formosat-2 sensors are performed to evaluate the performance of the proposed approach. We also show that our methods significantly outperform the sequential maximum angle convex cone (SMACC) method.     

Selecting effective features on prediction of delay in servicing ships arriving to ports using a combination of Clonal Selection and Grey Wolf Optimization algorithms—Case study: Shahid Rajaee port in Bandar Abbas

Predicting the delay in servicing incoming ships to ports is crucial for maritime transportation. In this study, we use support vector regression (SVR) in order to accurately predict this delay for ships arriving to the terminal No. 1 of Shahid Rajaee's port in Bandar Abbas. To achieve this goal, a combination of Clonal Selection and Grey Wolf Optimization algorithms (named as CLOGWO) is used for two purposes: (i) selecting the most important features among the features that affect prediction of this delay and (ii) optimizing SVR parameters for a more accurate prediction. Performance of the proposed method was compared with Genetic Algorithm (GA), Clonal Selection (CS), Grey Wolf Optimization (GWO), and Particle Swarm Optimization (PSO) algorithms on the following metrics: correlation, rate of feature reduction, root mean square error (RMSE), and normalized RMSE (NRMSE). Evaluations on Shahid Rajaee dataset showed that the mean value of these metrics in 10 independent runs of the proposed method were 0.867, 74.45%, 0.080, and 9.02, respectively. These results and evaluations on standard datasets indicate that the proposed method provides competitive results with other evolutionary algorithms.     

Toward the development of interactive virtual dissection with haptic feedback

Traditional, hands-on dissection of an animal is common practice in many classrooms to aid in the study of anatomy and biology. More specifically, virtual dissection environments have been developed making it possible to study the inner workings of animals without cutting them up. In this paper, we present a novel virtual reality dissection simulator, where a user can dissect an animal (i.e. frog) and its organs using a 3D force feedback haptic device. The simulator uses force feedback as part of a multimodal cue to provide guidance and performance feedback to the user. This paper highlights methodologies which are used for addressing some of the key challenges involved in designing and developing simulators, such as: modelling and mechanics of deformation, collision detection between multiple deformable bodies, and haptic feedback.     

Interactive multi-modal suturing

We present a mechanics-based interactive multi-modal environment designed as part of a serious gaming platform. The specific objectives are to teach basic suturing and knotting techniques for simple skin or soft tissue wound closure. The pre-wound suturing target, skin, or deformable tissue is modeled as a modified mass-spring system. The suturing material is designed as a mechanics-based deformable linear object. Tools involved in a typical suturing procedures are also simulated. Collision management modules between the soft tissue and the needle, the soft tissue and the suture are analyzed. In addition to modeling the interactive environment of a typical suturing procedure, basics of the modeling approaches on the evaluation of a stitch formed by the user are also discussed. For example, if needle insertion points are too close from each other or to the edge of the wound, when the suture is pulled, the suture will tear the soft tissue instead of suturing the incision together. Experiment results show that our simulator can run on a standard personal computer and allow users to perform different suturing patterns with smooth graphics and haptic feedback.     

An efficient optimal algorithm for the quantity discount problem in material requirement planning

An optimal algorithm based on branch-and-bound approach is presented in this paper to determine lot sizes for a single item in material requirement planning environments with deterministic time-phased demand and constant ordering cost with zero lead time, where all-units discounts are available from vendors and backlog is not permitted. On the basis of the proven properties of optimal order policy, a tree-search procedure is presented to construct the sequence of optimal orders. Some useful fathom rules have been proven, which make the algorithm very efficient. To compare the performance of this algorithm with the other existing optimal algorithms, an experimental design with various environments has been developed. Experimental results show that the performance of our optimal algorithm is much better than the performance of other existing optimal algorithms. Considering computational time as the performance measure, this algorithm is considered the best among the existing optimal algorithms for real problems with large dimensions (i.e. large number of periods and discount levels).     

An improved algorithm for layout design in cellular manufacturing systems

Layout has a significant role on the efficiency of manufacturing systems, but it has not received attention of researchers in comparison to cell formation in cellular manufacturing systems. In this paper, a mathematical model for facility layout in a cellular manufacturing system is proposed that minimizes both inter-cell and intra-cell material handling costs. A variant of simulated annealing algorithm is developed to solve the model. The developed algorithm produces solutions with better quality and less computation time in comparison with the benchmarked algorithm. The superiority of the algorithm in computation time is considerable especially when the size of the problem increases.     

Counting eosinophils in bronchoalveolar lavage fluid images with fuzzy methodology

This paper proposes a new fuzzy approach to count eosinophils, as a measure of inflammation, in bronchoalveolar lavage fluid images, provided by digital camera through microscope. We use fuzzy cluster analysis and fuzzy classification algorithm to determine the number of objects in an image. For this purpose, a fuzzy image processing procedure consisting of five main stages is presented. The first stage is pre-highlighting the objects in the images by using an image pre-processing method for enhancement, which is sharpening the image with the Laplaian high pass filter in order to have acceptable contrast in the image. The second stage is segmentation by clustering with fuzzy c-mean algorithm for portioning. In this stage the clustered data are the rough symbols of objects in the image containing noise. In the third step, first, a Gaussian low pass filter is used for noise reduction. Then, a contrast adoption in the image is done by modifying the membership functions in the image [H.R. Tizhoosh, G. Krell, B. Michaelis, Knowledge-based enhancement of megavoltage images in radiation therapy using a hybrid neuro-fuzzy system, Image and Vision Computing 19(July) (2000) 217–233]. Object recognition, the fourth stage, will be done by using fuzzy labeling for the objects in the image, using a fuzzy classification method. The number of labeled images shows the number of eosinophils in an image which is an index for diagnosing inflammation. The last stage is tuning parameters and verification of the system performance by using a feed forward Neural Network.