Narrow Results

Clear images are crucial for the optimal performance of various high-level vision-based tasks. However, some inevitable causes, such as bad weather and underwater conditions degrade scene visibility. The tiny particles present in the air absorb and scatter light, causing severe attenuation that results in unclear, low-brightness, and poor-contrast images. Several techniques have been introduced to restore the degradation. However, no model exists to date that can restore multiple degradations using a single model. Therefore, to improve the scene visibility, a unified model called a Multidomain Contextual Conditional Generative Adversarial Network (MCCGAN) is designed, which uses the same parameters across the domains to restore multiple degradations such as fog, haze, rain streaks, snowflakes, smoke, shadows, underwater, and muddy underwater. The proposed model has a novel addition of multiple $1\times 1$ convolutional context encoding bottleneck layers between a simple lightweight eight-block encoder and decoder with skip connections which learns the context of each input domain thoroughly, thus generating better-restored images. The MCCGAN is qualitatively and quantitatively compared to various state-of-the-art image-to-image translation models and tested on a few real unseen image domains such as smog, dust, and lightning, and the obtained results successfully improved scene visibility, proving the generalizability of MCCGAN. Moreover, the MS-COCO 2017 validation dataset is used for comparing the performance of object detection, instance segmentation, and image captioning on (1) weather-degraded images, (2) restored images by MCCGAN, and (3) ground truth images, and the results demonstrated the success of our model. An ablation study is also carried out to check the significance of the discriminator, skip connections, and bottleneck layers in MCCGAN, and the analysis suggests that MCCGAN performs better by adding a discriminator, skip connections, and four bottleneck layers in the generator architecture.

ATM networks are high speed networks and the volume of information carried by them at any instant of time is enormous. Since this information is likely to be of high commercial value, the effect of a transmission link or node failure would be catastrophic. Therefore ATM networks need to be equipped with mechanisms by which the failure of a network entity can be detected and service restored by changing over to alternate entities. Restoration mechanisms based on the self healing concept have been successful in achieving restoration times well below the two seconds limit to prevent any interruption of service to the users. This paper studies two self healing protocols for ATM networks. One is the Backup Virtual Path Self Healing protocol, and the second is the Broadcast Self Healing protocol, which is an extension of the protocol proposed for STM networks. Simulation studies were carried out to compare their restoration speeds, spare resource requirements and maximum restoration ratio. The results suggest that the backup virtual path protocol has an edge over the broadcast protocol.

This paper considers quality of service in terms of end-to-end delay in planning backup routes for multicast communications. The problem of preplanning backup routes considering both cost minimization and end-to-end delay guarantee for multicast communications in the case a single link failure is investigated. Two delay labels, the limited label and the tolerable label, are defined to evaluate the end-to-end delay requirement. Four heuristic algorithms, including two tree-based algorithms, one subtree-based algorithm, and one link-based algorithm, are proposed to determine the delay-constrained backup routes having the minimum costs. Two procedures to determine a node selection sequence, Random and Minimum-Cost, are used in the tree-based algorithm. Experimental results show that the tree-based algorithm by the Minimum-Cost sequence yields the best performance in cost minimization with guarantee of end-to-end delay.

The article proposes an adaptive, noise-exclusive and assessment-based fuzzy switching median filter for high intensity impulse noise. The filter adaptively changes the size of sliding window in accordance with noise intensity and noise-exclusive operation is performed only on noise-free pixels. These steps thus preserve the image details effectively. Judicious assessment of the previously restored pixels is made in order to use them in replacing the noisy pixels and further improve the efficiency of filter. With these possibilities, fuzzy switching median filter-based approach is also incorporated for improving the performance. Experimental results justified the efficacy of the proposed fuzzy-based filtering method qualitatively and quantitatively for twelve different highly corrupted noisy images, and showed significant improvement to existing algorithms.

Nowadays, Data Sanitization is considered as a highly demanded area for solving the issue of privacy preservation in Data mining. Data Sanitization, means that the sensitive rules given by the users with the specific modifications and then releases the modified database so that, the unauthorized users cannot access the sensitive rules. Promisingly, the confidentiality of data is ensured against the data mining methods. The ultimate goal of this paper is to build an effective sanitization algorithm for hiding the sensitive rules given by users/experts. Meanwhile, this paper concentrates on minimizing the four sanitization research challenges namely, rate of hiding failure, rate of Information loss, rate of false rule generation and degree of modification. Moreover, this paper proposes a heuristic optimization algorithm named Self-Adaptive Firefly (SAFF) algorithm to generate the small length key for data sanitization and also to adopt lossless data sanitization and restoration. The generated optimized key is used for both data sanitation as well as the data restoration process. The proposed SAFF-based algorithm is compared and examined against the other existing sanitizing algorithms like Fire Fly (FF), Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Differential Evolution algorithm (DE) algorithms and the results have shown the excellent performance of proposed algorithm. The proposed algorithm is implemented in JAVA. The data set used are Chess, Retail, T10, and T40.

Handling huge amount of data from different sources more so in the images is the latest challenge. One of the solutions to this is sparse representation. The idea of sparsity has been receiving much attention recently from many researchers in the areas such as satellite image processing, signal processing, medical image processing, microscopy image processing, pattern recognition, neuroscience, seismic imaging, etc. Many algorithms have been developed for various areas of sparse representation. The main objective of this paper is to provide a comprehensive study and highlight the challenges in the area of sparse representation which will be helpful for researchers. Also, the current challenges and opportunities of applying sparsity to image reconstruction, namely, image super-resolution, image denoising and image restoration are discussed. This survey on sparse representation categorizes the existing methods into three groups: dictionary learning approach, greedy strategy approximation approach and deep learning approach.

The quality of the underwater image is impairing with the atmospheric conditions. In this, one of the most significant issues in recent days is due to the motion blur induced by the imaging device or by the movement of the object in underwater image quality degradation. The various parameters of the blurred image must be identified to fix the effect of blurring in post-imaging. Therefore, spectrum-based parameter estimation method is proposed. Initially, to estimate the point spread function (PSF), the angle and the length is measured from image spectrum using radon transform. Then, for the accurate estimation of PSF, Optimized Polynomial Lagrange Interpolation (OPLI) is proposed. The data were collected and analyzed in various natural and structured water bodies in Chennai without affecting the real environment. It is observed that for the underwater images collected, the proposed OPLI approach outperforms compared to few existing traditional estimation methods like cepstral, hough, and radon. Then this veracious interpolated measure of angle and length (VIMAL) is restored using modified Lucy algorithm and is evaluated which results in high performance than the existing classical state-of-the-art methods.

This study used a newly developed proximal contact strength (PCS) device to evaluate the tightness of proximal tooth contact for Class II cavity size restoration with different materials using an auxiliary separation ring system. A measurement device based on the equilibrium of forces acted on the clamp rod converts a pull-out force between interdental spaces on a force sensing resistor to express the PCS. This device was designed using dental floss as the test end and can be moved with constant speed during measurement through a bevel gear that transforms the rotation of motor shaft into linear movement of clamp rod. A manikin model was used with 60 artificial first molars in which an mesial occlusal (MO) preparation was ground. Samples were divided into six groups (each n = 10) for simulating amalgam and resin composite restoration with three different cavity sizes. The different cavities were defined using the ratio of the actual isthmus width to the intercuspal width (W) to 1/3, 2/3 and 1. The PCS value in each sample was measured after restoration. The result showed that the mean PCS value and standard deviation were 2283.1 ± 216.5 gf, 2419.1 ± 375 gf and 1737.6 ± 372.7 g for 1/3 W, 2/3 W and W cavities of the amalgam restoration, respectively. The corresponding PCS values were 1178.0 ± 230.4 gf, 1205.8 ± 249.1 gf and 1247.0 ± 157.5 gf for 1/3 W, 2/3 W and W cavities of the resin composite restoration. PCS values with amalgam restoration were larger than those for resin composite restorations under the same cavity size. Large cavity (W) PCS might be lost with amalgam restoration. No significant difference was found in resin composite restoration PCS among the different cavity sizes.

Historic buildings and monuments are an important part of our cultural heritage that must be protected and their sustainability ensured, especially when earthquakes occur. In this paper, a technique that uses structural steel frames is proposed as one way of strengthening unreinforced masonry (URM) in historical buildings. The idea underpinning this technique is to reduce the earthquake displacement demand on non-ductile URM walls by attaching steel frames to the building floors from inside. These frames run parallel to the structural system of the building and are fixed at their base to the existing foundation of the building. Furthermore, they are constructed rapidly, do not occupy architectural space, save the building’s historic fabric, and can be easily replaced after an earthquake if some minor damage ensues. The proposed technique was applied to a five-story historical masonry building in Istanbul. The results of seismic performance analysis indicate that even though the building has plan irregularities, the proposed steel frames are able to effectively enhance the building’s seismic performance by reducing inter-story drifts and increasing lateral stiffness and strength.

This paper explains the important conceptual and technical differences between the method of p-cycles and two other recent advances involving a cyclic orientation to protection. These are enhanced rings and cycle double covers. The most fundamental difference that is unique to p-cycles is the aspect of straddling span failure protection. This enables mesh-like efficiency levels at well under 100% redundancy. In contrast enhanced rings and advanced cycle cover methods are both seeking to reduce span overlaps in what is otherwise a purely ring-like logical paradigm in which 100% redundancy remains the best that can possibly be achieved.

In the modern business environment for optical mesh networking, there is considerable interest in being able to support a range of service classes, and to charge accordingly. In this work we consider the capacity design problem for different combinations of four protection priority classes in a span-restorable mesh network. Most significantly we have test-case findings that 15% to 30% of all demand can be in the gold class and enjoy 100% restorability, solely through preemption of economy class service capacity. This indicates the potential to design and operate mesh-based re-storable networks which have no spare capacity at all in the conventional sense: all capacity is bearing service of some paying type. The methods provided could be the basis for future mesh-based optical transport networks that simultaneously support multiple QoP services with optimal capacity.

Current link-state routing protocols such as OSPF or IS-IS react to link failures through link-state advertisement and re-computation of the routing table. This involves a convergence process. The convergence process increases router overload, transient loops and slows down reaction to failures. We propose a new routing technique called Source-Based Tree Failure Recovery (STFR), which allows a router to discover a new working path under failure without waiting for completely state consistency in all routers. The aim of STFR is to eliminate the convergence process and achieve fast recovery from link failures. Our simulation shows that compared to previous work of local recovery methods, STFR can provide better resilience guarantee under failures with low overheads.