Audio steganalysis using deep belief networks

This paper presents a new steganalysis method that uses a deep belief network (DBN) as a classifier for audio files. It has been tested on three steganographic techniques: StegHide, Hide4PGP and FreqSteg. The results were compared to two other existing robust steganalysis methods based on support vector machines (SVMs) and Gaussian mixture models (GMMs). Afterwards, another classification task aiming at identifying the type of steganographic applied or not to the speech signal was carried out. The results of this four-way classification show that in most cases, the proposed DBN-based steganalysis method gives higher classification rates than the two other steganalysis methods based on SVMs and GMMs.     

Routing automorphisms of the hypercube

We present an online algorithm for routing the automorphisms (BPC permutations) of the queueless MIMD hypercube. The routing algorithm has the virtue of being executed by each node of the hypercube without knowing the state of the others nodes. The algorithm is also vertex and link-contention free. We show, using the proposed algorithm, that BPC permutations are arbitrarily routable in the considered communication model.     

Neural Network-Based Long-Term Hydropower Forecasting System

Artificial neural networks are alternatives to stochastic models even if the optimization of their architectures remains a tricky problem. Two different approaches in long-term forecasting of potential energy inflows using a feedforward neural network (FNN) and a recurrent neural network (RNN) are proposed. The problem of overfitting, particularly critical for limited hydrologic data records, is addressed using a new approach entitled optimal weight estimate procedure (OWEP). The efficiency of the two models using OWEP is assessed through multistep forecasts. The experiment results show that, in general, OWEP improves the models' performance and significantly reduces the training time on the order of 60 percent. The RNN outperforms the FNN but costs about a factor of 2 longer in training time. Furthermore, the neural network-based models provide more accurate forecasts than traditional stochastic models. Overall, the RNN appears to be the best suited for potential energy inflows forecasting and therefore for hydropower systems management and planning.     

Characterization of psychrotrophic bacterial communities in modified atmosphere-packed meat with terminal restriction fragment length polymorphism

Characterization of psychrotrophic lactic acid bacteria (LAB) and Brochothrix thermosphacta communities is needed to understand the microbial ecology of spoilage of modified atmosphere-packed (MAP) meats. To overcome the limitations of the currently used methods for the characterization of psychrotrophic bacterial communities in meat, we developed a culture-independent, 16S rRNA gene-targeted terminal restriction fragment length polymorphism (T-RFLP) method. An identification library consisting of 100 Gram-positive and 30 Gram-negative meat-associated bacterial strains was set up to identify the terminal restriction fragments derived from the communities. The taxonomic resolution level of the T-RFLP method was in between genus and species within the investigated LAB strains and within family and genus within the investigated Gram-negative strains. The established library was applied to identify the members of bacterial communities in MAP minced meat at the end of the shelf life. The T-RFLP results and plate counts on Man-Rogosa-Sharpe, Violet Red Bile Glucose, and Streptomycin sulfate thallium acetate actidione agars indicated that LAB and B. thermosphacta predominated in meat. The bacterial taxa associated with the T-RFLP results were compared to those identified among plate-grown LAB isolates by numerical ribopattern analysis. Both methods agreed that Leuconostoc spp. and Carnobacterium spp. prevailed in the LAB community in minced meat followed by Lactobacillus algidus, Lactococcus spp. and Weissella spp. Colony identification revealed that Leuconostoc gasicomitatum, L. gelidum, Carnobacterium divergens and C. maltaromaticum were the predominant LAB species. The T-RFLP results were shown to correlate with viable counts of Leuconostoc spp. and B. thermosphacta. The T-RFLP method was found to be a useful tool enabling rapid and high-throughput characterization of psychrotrophic bacteria prevailing in MAP meat.     

Detection of streptomycetes in house dust--comparison of culture and PCR methods

Streptomycetes are gram-positive, spore producing, filamentous bacteria common in soil, but also present in indoor environments. They are potent producers of secondary metabolites and inducers of inflammatory responses in vitro. Polymerase chain reaction (PCR)- and culture-based detection methods for streptomycetes in house dust samples were compared. A total of 47 dust samples were investigated, and the presence of streptomycetes was determined by cultivation on tryptone-yeast-extract-glucose agar and PCR. The 16S rRNA gene of actinomycete isolates from house dust was partially sequenced to investigate if they belong to the genus Streptomyces. Both PCR and culture showed more frequent occurrence of streptomycetes in moisture-damaged homes, although the results did not correlate well. The occurrence of streptomycetes in house dust was associated with moisture damage of the home. The amount of Streptomyces-specific PCR amplification product was significantly higher in dust from moisture-damaged homes than in homes with no moisture damage (P < 0.05, Mann-Whitney U test). A correlation between streptomycetes and moisture damage, although not statistically significant, was also observed when using binary data, e.g. presence or absence of streptomycetes or moisture damage (P = 0.054 for PCR, and P = 0.127 for culture, Fisher's exact test). Altogether, the presence of streptomycetes in house dust seems to indicate the presence of moisture damage in the building.     

“自动化工作室”——不仅仅是一款编程工具

“Automation Studio的开发是以支持机械制造商将一个概念变为现实的目标为导向的,覆盖了机器生命周期中的各个阶段。”     

Combinatorial Bidirectional Path‐Tracing for Efficient Hybrid CPU/GPU Rendering

This paper presents a reformulation of bidirectional path‐tracing that adequately divides the algorithm into processes efficiently executed in parallel on both the CPU and the GPU. We thus benefit from high‐level optimization techniques such as double buffering, batch processing, and asyncronous execution, as well as from the exploitation of most of the CPU, GPU, and memory bus capabilities. Our approach, while avoiding pure GPU implementation limitations (such as limited complexity of shaders, light or camera models, and processed scene data sets), is more than ten times faster than standard bidirectional path‐tracing implementations, leading to performance suitable for production‐oriented rendering engines.     

Dynamic surfel set refinement for high-quality rendering

Splatting-based rendering techniques are currently the best choice for efficient high-quality rendering of point-based geometries. However, such techniques are not suitable for large magnification, especially when the object is under-sampled. This paper improves the rendering quality of pure splatting techniques using a fast dynamic up-sampling algorithm for point-based geometry. Our algorithm is inspired by interpolatory subdivision surfaces where the geometry is refined iteratively. At each step the refined geometry is that from the previous step enriched by a new set of points. The point insertion procedure uses three operators: a local neighborhood selection operator, a refinement operator (adding new points) and a smoothing operator. Even though our insertion procedure makes the analysis of the limit surface complicated and it does not guarantee its G¹ continuity, it remains very efficient for high-quality real-time point rendering. Indeed, while providing an increased rendering quality, especially for large magnification, our algorithm needs no other preprocessing nor any additional information beyond that used by any splatting technique. This extended version (Real-time point cloud refinement, in: Proceedings of Eurographics Symposium on Point-Based Graphic, 2004, pp. 41.) contains details on creases handling and more comparison to other smoothing operators.