Historical documents are important sources of information. With the information revolution and the advancement of science and technology, the digitalization of historical documents will change the protection, collection, transmission, and utilization of historical documents. However, historical documents often suffer from varying degrees of degradation, such as stains, creases, uneven illumination, and seepage of ink. Compared with the general image, the document image has many characteristics. For example, the background pixels are more concentrated than the foreground pixels, and the foreground and background pixels are relatively concentrated. Based on these problems and characteristics, this paper introduces three commonly used historical document image enhancement algorithms, and improves the Sauvola algorithm for a new historical document image enhancement algorithm that does not have to calculate a special threshold for each pixel, but uses the same threshold within a fixed area to reduce the cost of time while maintaining image quality. Experimental results prove that the improved algorithm can significantly reduce the time required while ensuring the quality of the image.

We propose a half-draining round robin matching (HDRRM) algorithm for scheduling of high performance input queued switches. HDRRM uses round robin arbiters to find a matching between the input ports and the output ports. Once an input- output pair is matched, HDRRM continuously services half number of the cells in its virtual output queue. The implementation of HDRRM is also discussed. Simulation results show that HDRRM with a single iteration achieves high throughput and low average delay and delay variation under both uniform and nonuniform traffic patterns.

As music databases grow in size and number dramatically with the development of the Internet. Automatic music emotion recognition becomes very important to bridge the semantic gap between acoustic feature of pop music and emotion. In this paper, we got the pop songs from music web site, QQ Music, via crawling technology. We asked the subjects to annotate the emotion tag to every song and got a collection of music, in which all songs have social tags consistent with that of the subjects in four major classes. Base on the music collection, we evaluated the performance of different classifiers for the music emotion recognition and an integrated classifying scheme was propose to achieve more stable performance.

In order to improve the recognition efficiency of classical DTW algorithm in speech recognition of confusable word, a new method for recognition of confusable word is proposed in this paper. The distance between the measured speech and the template speech is calculated by the general distance matrix and the ideal state distance matrix. The Offset-Loss is introduced and it is used to evaluate the similarity between the measured speech and the template speech. The experimental results show that it improves the matching speed and the accuracy of the improved algorithm is almost the same as the classical DTW algorithm.

Broadcast monitoring is a very useful method in advertisement broadcasting, copyright management and testing of illegal broadcasting etc of broadcast station. In this paper, the histogram pruning algorithm is applied to broadcast monitoring. Meanwhile, an adaptive algorithm based on frequency spectral analysis is proposed to solve the problem of low retrieval rate caused by noise interference in compressed audio applications.

There is a constantly growing interest in evaluating music information retrieval (MIR) systems that can provide effective management of the music resources. The crucial characteristic of music is its emotion, which reflect the human's perception. To do the automatic classification of Chinese music emotions more effective, we use the lyrics of music to analysis and classify music based on emotion. There are many algorithms to achieve text classification, and one of the most popular algorithms is Naive Bayes algorithm. Although it is simple, it can classify text effectively. In this paper, we crawl the music lyrics and their labels from a popular website named Baidu music and make our four different datasets. We also train four classifiers with different datasets and report their performance. We evaluate the classifiers trained by four different datasets, and the final accuracy we get is approximately 68%.

Automatic extraction of popular music ringtones have become an important and useful area for communications and telecommunications industry. Quick and batch extraction of music ringtones increases the convenience in practical application. In this paper, we propose an automatic technology to extract the ringtones from popular music based on the musical structural analysis. This is a meaning attempt to use the theory of musical structural analysis to solve a practical problem. Experiments show the feasibility of the process, several sets of comparative experiments reflect that various stages of threshold may have different effects. The experiments also reflected some problems which inspired us to optimize the processing. On our testing database composed of 186 popular songs, the best accuracy of boundary detection with tolerance ±3 seconds achieves up to 79.9%. We also try to invite strangers to evaluate the result by using voting mechanism.

In this paper, a text similarity computation method named VSM-Cilin which is based on semantic vector space model is proposed in the background of radio station. VSM-Cilin improved the traditional VSM in the following areas. First, consider the semantic relations between words. Second, use semantic resources to reduce dimension. Third, use inverted index to filter out candidate document set. Forth, take the weight of the feature item into consideration when compute the similarity. The experiments show that the accuracy of VSM-Cilin is significantly improved compared with the traditional vector space model and the method of bidirectional mapping based on HITIR-Lab Tongyici Cilin.

Singing voice detection for musical structural analysis is an important but difficult area, which makes a great contribution to music segment. This paper proposes an efficient singing voice detection system by using beat tracking technique and SVM classification. We do the experiment using beat as a unit of classification instead of frame. Compared with a conventional frame-based classification, the run time of beat-based classification system reduces about 20%, and achieve 72.8% of precision. Meanwhile, in practice, the beat-based classification system achieves about 88.3% of precision.

As scalable routers being a promising way to scale to higher capacity, scalable switch fabric as its key component has received a great deal of attention. However, the reliability calculation method for general switch fabric does not suit scalable switch fabrics with special features. In this paper, we study the features of scalable switch fabrics, and propose a novel reliability measure called Failure Influence which focuses on the robustness of scalable switch fabric. Using our improved heuristic algorithms, we analyze the Failure Influence properties of some classical topologies and compare them with Plus 2^i (P2i) of our prior work. Experimental results show that P2i has better Failure Influence properties. Since the Peak Point Problem (PPP) of P2i, Add Edge P2i (AE-P2i) and Failure Influence Optimal P2i (FO-P2i) are also proposed to solve this problem.

Simulation is crucial to performance evaluation of switches. Currently stochastic simulation is the predominant approach, which has two drawbacks: few workload patterns and long simulation times. In this paper, we propose a novel switch simulation method that is based on two techniques: exhaustive workload pattern generation and smoothed periodic input generation. The exhaustive workload pattern generation can produce a huge number of nonuniform workload patterns, which far exceeds the traditional few ones; conclusions based on such truly extensive simulations are much more convincing. The periodic input or cell arrival pattern outperforms the stochastic counterpart in terms of easy repetition and fast convergence of simulation; in particular, among periodic inputs, the smoothed periodic input is most favorable to switch scheduling, and hence switches performing poor with it probably performs worse under others. Combining these two techniques together can systematically identify lots of stuck states at which some switches show poor performances such as low throughput. Specifically, this method discovers that the throughput of iSLIP, FIRM and DRRM, each with one iteration, may be lower than 60% under certain nonuniform and smoothed periodic traffic pattern.

Input queued switches have been well studied in the recent past. A scheduling algorithm is required to schedule the transfer of packets through the switch fabric (e.g. crossbar) at every time slot. The maximum weight matching (MWM) algorithm is known to deliver 100% throughput under any admissible traffic. However, MWM is not practical for its high computational complexity O(N/sup 3/). In this paper, we study a class of approximation algorithms to MWM from the point of view of local search. Notably we observe that: (a) Local search is well suitable for parallel computation. (b) Each line card of high performance router has at least one processor. Based on the two important observations, a parallelized greedy scheduling algorithm is proposed. The proposed algorithm is proved to be rate stable under any admissible traffic. Simulation results show that our algorithm outperforms other randomized approximations to MWM.

Scalability considerations drive the switch fabric design to evolve from output queueing to input queueing and further to combined input and crosspoint queueing (CICQ). However, few CICQ switches are known with guaranteed quality of service, and credit-based flow control induces a scalability bottleneck. In this paper, we propose a novel CICQ switch called the smoothed buffered crossbar or sBUX, based on a new design objective of smoothness and on a new rate-based flow control scheme called the smoothed multiplexer or sMUX. It is proved that with a buffer of just four cells at each crosspoint, sBUX can utilize 100% of the switch capacity to provide deterministic guarantees of bandwidth and fairness, delay and jitter bounds for each flow. In particular, neither credit-based flow control nor speedup is used, and arbitrary fabric-internal latency is allowed between line cards and the switch core.

Input queued switching architectures have become predominant in high speed switches and routers. In this paper, we change the point of view from weight-based matching to weight-based service, and propose a distributed load adaptive scheduling (DLAS) algorithm. In DLAS, the round robin arbiters are used to find a matching between the input ports and output ports. Once the matching between an input-output pair is established, the scheduler will keep it for a certain period, which is a function of the number of cells queued in the corresponding VOQ. Simulation results show that our scheme achieves high throughput and low delay under admissible traffic. For uniform Bernoulli i.i.d. traffic, it achieves 100% throughput, and for nonuniform traffic, its throughput is almost 100%.

Some wireless communication systems, such as mobile ad hoc network, operate on a limited battery supply, and energy efficiency is therefore crucial to the network survival time. This paper proposes a multiscale load adaptive scheduling (MLAS) scheme to determine the transmission time of the packets. In MLAS, traffic is predicted over the scale of time window, and then a transmission time stretching algorithm is proposed to adjust the packet transmission time packet by packet. Compared with the fixed transmission time scheme, MLAS can reduce the energy consumption of data transmission by up to 90% corresponding to the load factor and the burst characteristic. MLAS also outperforms the look ahead based scheme in the performance of energy efficiency with the same average delay.