Transactions on Computational Science XXXI

A novel methodology for matching of illumination-invariant and heterogeneous faces is proposed here. We present a novel image representation called local extremum logarithm difference (LELD). Theoretical analysis proves that LELD is an illumination-invariant edge feature in coarse level. Since edges are invariant in different modalities, more importance is given on edges. Finally, a novel local zigzag binary pattern LZZBP is presented to capture the local variation of LELD, and we call it a zigzag pattern of local extremum logarithm difference (ZZPLELD). For refinement of ZZPLELD, a model based weight value learning is suggested. We tested the proposed methodology on different illumination variations, sketch-photo and NIR-VIS benchmark databases. Rank-1 recognition of 96.93% on CMU-PIE database and 95.81% on Extended Yale B database under varying illumination, show that ZZPLELD is an efficient method for illumination invariant face recognition. In the case of viewed sketches, the rank-1 recognition accuracy of 98.05% is achieved on CUFSF database. In the case of NIR-VIS matching, the rank-1 accuracy of 99.69% is achieved and which is superior to other state-of-the-art methods.

Tabla is the most common rhythmic instrument in Indian Classical music. A bol the fundamental unit of tabla play and it is produced by striking either or both of the two drums of tabla. Tala (rhythm) is formed with a basic sequence of bols that appears in a cyclic pattern. In this work, bols are automatically segmented from tabla signal following Attack-Decay-Sustain-Release (ADSR) model. Subsequently segmented bols are recognized using low level spectral descriptors and support vector machine (SVM). The identified bol sequence generates transcript of tabla play. A template based matching approach is used to identify tala from the transcript. Proposed system tested successfully with a variety of collection of tabla signal of different talas and it can be utilized in rhythm analysis of music. Moreover, for the learners also the system can help in analyzing their performance.

OCT is a promising technology that allows getting a lot of data in each sample. Authors hope that it is possible to create a system that would automatically diagnose various retinal diseases basing on OCT images with the accuracy of 95% which may revolutionize and shorten diagnostic pathway. At the beginning authors focus on automatic distinguishing the healthy images from pathological retinas. In this paper a novel approach has been presented. The algorithm has been described and results have been revealed and discussed. OCT is a way for detecting many various diseases. However, the amount of information to be processed is much more numerous so the task seems to be more difficult than it is in fundus imaging. In this paper some advanced diseases with the macular oedema detection algorithm basing on OCT images are presented.

Quantum circuits exhibit several features of large-scale distributed systems. They have a concise design formalism but behavior that is challenging to represent let alone predict. Issues of scalability—both in the yet-to-be-engineered quantum hardware and in classical simulators—are paramount. They require sparse representations for efficient modeling. Whereas simulators represent both the system’s current state and its operations directly, emulators manipulate the images of system states under a mapping to a different formalism. We describe three such formalisms for quantum circuits. The first two extend the polynomial construction of Dawson et al. [1] to (i) work for any set of quantum gates obeying a certain “balance” condition and (ii) produce a single polynomial over any sufficiently structured field or ring. The third appears novel and employs only simple Boolean formulas, optionally limited to a form we call “parity-of-AND” equations. Especially the third can combine with off-the-shelf state-of-the-art third-party software, namely model counters and \(\mathrm {\#SAT}\) solvers, that we show capable of vast improvements in the emulation time in natural instances. We have programmed all three constructions to proof-of-concept level and report some preliminary tests and applications. These include algebraic analysis of special quantum circuits and the possibility of a new classical attack on the factoring problem. Preliminary comparisons are made with the libquantum simulator [2–4].

Power grid and energy theft has an eternal relationship. Though we moved towards Smart Grid, with an expectation for a more efficient, reliable and secure service, so does the attackers. Smart Grid and AMI systems incorporate a good number of security measures, still it is open to various threats. Recent attacks on Smart Grids in U.S., Gulf State and Ukraine proved that the attacks on the grid have become more sophisticated. In this paper we have introduced a new, distributed and intelligent energy theft: DIET attack and proposed an advanced Intrusion Detection System to protect AMI system. The proposed IDS can perform a passive monitoring on the system as well as detect attackers. This features make this IDS more robust and reliable.

We introduce an abstract domain for information-flow analysis of software. The proposal combines variable dependency analysis with numerical abstractions, yielding to accuracy and efficiency improvements. We apply the full power of the proposal to the case of database query languages as well. Finally, we present an implementation of the analysis, called \(\mathsf {Sails}\), as an instance of a generic static analyzer. Keeping the modular construction of the analysis, the tool allows one to tune the granularity of heap analysis and to choose the numerical domain involved in the reduced product. This way the user can tune the information leakage analysis at different levels of precision and efficiency.

Image quality improvement is not bounded within the application of different types of filtering. Resolution improvement is also essential and it solely depends on the estimation of the unknown pixel value that involves a lot of computation. Here a resolution enhancement technique is proposed to reduce the aliasing effects from the text documented image with a reduced amount of computational time. The proposed hybrid method provides better resolution at most informative regions. Here, the unknown pixel value is estimated based on their local informative region. This technique finds the most informative areas, discontinuity at the edges and less informative areas separately. The foreground regions are segmented at the first phase. The unknown pixels values of the foreground regions are calculated in the second step. All-of-these separated images are combined together to construct the high-resolution image at the third phase. The proposed method is mainly verified on aliasing affected text documented images. A distinct advantage of the proposed method over other conventional approaches is that it requires lower computational time to construct a high-resolution image from a single low-resolution one.