Auto-tuning Similarity Search Algorithms on Multi-core Architectures

In recent times, large high-dimensional datasets have become ubiquitous. Video and image repositories, financial, and sensor data are just a few examples of such datasets in practice. Many applications that use such datasets require the retrieval of data items similar to a given query item, or the nearest neighbors (NN or $k$ -NN) of a given item. Another common query is the retrieval of multiple sets of nearest neighbors, i.e., multi $k$ -NN, for different query items on the same data. With commodity multi-core CPUs becoming more and more widespread at lower costs, developing parallel algorithms for these search problems has become increasingly important. While the core nearest neighbor search problem is relatively easy to parallelize, it is challenging to tune it for optimality. This is due to the fact that the various performance-specific algorithmic parameters, or “tuning knobs”, are inter-related and also depend on the data and query workloads. In this paper, we present (1) a detailed study of the various tuning knobs and their contributions on increasing the query throughput for parallelized versions of the two most common classes of high-dimensional multi-NN search algorithms: linear scan and tree traversal, and (2) an offline auto-tuner for setting these knobs by iteratively measuring actual query execution times for a given workload and dataset. We show experimentally that our auto-tuner reaches near-optimal performance and significantly outperforms un-tuned versions of parallel multi-NN algorithms for real video repository data on a variety of multi-core platforms.     

Coarse‐to‐Fine Isometric Shape Correspondence by Tracking Symmetric Flips

We address the symmetric flip problem that is inherent to multi‐resolution isometric shape matching algorithms. To this effect, we extend our previous work which handles the dense isometric correspondence problem in the original 3D Euclidean space via coarse‐to‐fine combinatorial matching. The key idea is based on keeping track of all optimal solutions, which may be more than one due to symmetry especially at coarse levels, throughout denser levels of the shape matching process. We compare the resulting dense correspondence algorithm with state‐of‐the‐art techniques over several 3D shape benchmark datasets. The experiments show that our method, which is fast and scalable, is performance‐wise better than or on a par with the best performant algorithms existing in the literature for isometric (or nearly isometric) shape correspondence. Our key idea of tracking symmetric flips can be considered as a meta‐approach that can be applied to other multi‐resolution shape matching algorithms, as we also demonstrate by experiments.     

River: an intermediate language for stream processing

This paper presents both a calculus for stream processing, named Brooklet, and its realization as an intermediate language, named River. Because River is based on Brooklet, it has a formal semantics that enables reasoning about the correctness of source translations and optimizations. River builds on Brooklet by addressing the real‐world details that the calculus elides. We evaluated our system by implementing front‐ends for three streaming languages, and three important optimizations, and a back‐end for the System S distributed streaming runtime. Overall, we significantly lower the barrier to entry for new stream‐processing languages and thus grow the ecosystem of this crucial style of programming. Copyright © 2015 John Wiley & Sons, Ltd.     

人脸检测系统的研究

人脸检测系统采用的是一种基于模板匹配的快速人脸定位算法。该算法利用积分投影法先预测出人脸可能的范围与位置,然后在限定的范围内利用模板匹配的方法定位人脸,从而实现了快速而且可靠的人脸定位。     

AODV的本地修复算法

提出了一种新的路由本地修复算法,将修复尽量限制在断链的局部范围内,减少了修复时间和降低路由维护的开销, 且不用考虑失效链路在整个路由上的相对位置,均可对其进行修复,提高了处理失效链路的能力,有利于提高网络的可扩展性.仿真表明,这种新的路由本地修复算法明显地提高了网络多方面的性能.     

Synthesis and electrochemical,in situ spectroelectrochemical,electrical and gas sensing properties of ball-type homo- and hetero-dinuclear phthalocyanines with four [1a,8b-dihydronaphtho[b]naphthofuro[3,2-d]furan-7,10-diyl] bridges

The new mono-nuclear Fe^II 2 and ball-type homo-dinuclear Fe^II–Fe^II 3 phthalocyanines have been synthesized from the corresponding 4,4′-(1a,8b-dihydronaphtho[b]naphthofuro[3,2-d]-furan-7,10-diyl)bis(oxy)diphthalonitrile 1 while ball-type hetero-dinuclear Fe^II–Co^II phthalocyanine 4 was synthesized from 2. The compounds have been characterized by elemental analysis, UV–vis, IR and MALDI-TOF-mass spectroscopies. The redox behaviours of the complexes were identified by controlled-potential coulometry, cyclic voltammetry and differential pulse voltammetry measurements on Pt in dimethylsulfoxide containing tetrabutylammonium perchlorate. The assignments of the redox processes and the understanding of the interactions between the metal phthalocyanine units in 3 and 4 were achieved by the combined evaluation of the voltammetric and in situ spectroelectrochemical outcomes. Complex 3 showed ring-based mixed-valence behaviour as a result of the considerable interaction between the phthalocyanine rings. On the other hand, the interactions between the two metal phthalocyanine units of 4 were found to be much weaker than those in 3. The potential application of molecular organic semiconductors needs the control adjustment of conductivity. Ac and dc conductivity measurements were performed with the applied external electric filed. At high frequency, the conduction follows the universal power law and conduction mechanism can be explained by classical hopping barriers mechanism for the system.     

Neural extension of experimental data to investigate using phosphogypsume in light brick production

In this study, usability of wastes produced in phosphoric acid plants in structural brick manufacture has been investigated. There are several parameters involved in using these wastes in brick production namely the rate of added waste, firing speed and firing temperature. The performance of these parameters can be measured by several criteria such as natural drying shortening, water absorption and weight loss. Therefore, so many experiments are needed to investigate the effects of these parameters on the bricks produced with these wastes. The result of a series of experiments were utilized to achieve this end. The results have shown that the industrial wastes considered improve the performance of the bricks in terms of the criteria mentioned above. However, the results have also shown that further investigations are needed to explore the effects of interim values on the performance of the bricks. To achieve that end, a neural experimental study is adopted. For this purpose, the results of the experiments conducted were used to construct an artificial neural network. The trained and tested network was then used to check the effects of 280 different combinations for each type of material mixtures mentioned. The outcome of these artificial tests have provided the optimal values for the waste addition rate, firing speed and firing temperature based on the four criteria mentioned previously.     

Series resistance determination of Au/Polypyrrole/p-Si/Al structure by current–voltage measurements at low temperatures

The current–voltage (I–V) characteristics of Polypyrrole(PPy)/p-Si/Al structure have been investigated as a function of temperature. A modified Norde function combined with conventional forward I–V method has been used to extract the junction parameters including the ideality factor, barrier height and series resistance. Norde function has been compared with the Cheung functions and it has been seen that there is a good agreement with both method for the barrier height values. However, the values of series resistance have very different especially towards to the lower temperatures. This is attributed to non-ideal I–V characteristics of the Au/PPy/p-Si/Al structure and non-pure Thermionic emission theory due to the low temperature effects.     

Combine Effects of Essential Oils and Packaging on Smoked Rainbow Trout Stored 4°C

In this study, the effect of essential oils (thyme and clove) and packaging treatments (vacuum and modified atmosphere) on the storage of hot smoked rainbow trout (Oncorhynchus mykiss) fillets over a period of 120 days at 4°C was investigated. Fillets were subjected to microbiological (total aerobic mesophilic bacteria, psychrotrophic bacteria, lactic acid bacteria, yeast and mould), chemical analyses (pH, thiobarbituric acid reactive substances and total volatile base nitrogen) and quality testing throughout the storage period. Difference in microbiological and chemical changes between samples was found to be significant (p < 0.05) during storage period. Sensory evaluation showed significant differences between the samples, and V2 and M2 samples were assessed as the most acceptable products by the panellists. In conclusion, essential oils treatment to hot smoking had a positive effect on shelf life of rainbow trout, especially clove oil showed more positive effect when compared with thyme oil. Copyright © 2015 John Wiley & Sons, Ltd.