High-resolution(HR) image reconstruction from single low-resolution(LR) image is one of the important vision applications. Despite numerous algorithms have been successfully proposed in recent years, efficient and rob...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
High-resolution(HR) image reconstruction from single low-resolution(LR) image is one of the important vision applications. Despite numerous algorithms have been successfully proposed in recent years, efficient and robust single-image superresolution(SR) reconstruction is still challenging by several factors, such as inherent ambiguous mapping between the HRLR images, necessary huge exemplar images, and computational load. In this paper, we proposed a new learning-based method of single-image SR. Inspired by simple mapping functions method, a mapping matrix table of HR-LR feature patches is calculated in the training phase. Each atom of dictionary learned from LR feature patches is corresponding to a mapping matrix in the mapping matrix table. Combining this mapping table with sparse coding, high quality and HR images are reconstructed in reconstruction phase. The effectiveness and efficiency of this method is validated with experiments on the training datasets. Compared with state-of-art methods, jagged and blurred artifacts are depressed effectively and high reconstruction quality is acquired with less exemplar images.
Word segmentation is in most cases a base for text analysis and absolutely vital to the accuracy of subsequent natural language processing(NLP) tasks. While word segmentation for normal text has been intensively studi...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
Word segmentation is in most cases a base for text analysis and absolutely vital to the accuracy of subsequent natural language processing(NLP) tasks. While word segmentation for normal text has been intensively studied and quite a few algorithms have been proposed, these algorithms however do not work well in special fields, e.g., in clinical text analysis. Besides, most state-of-the-art methods have difficulties in identifying out-of-vocabulary(OOV) words. For these two reasons, in this paper, we propose a semi-supervised CRF(semiCRF) algorithm for Chinese clinical text word ***-CRF is implemented by modifying the learning objective so as to adapt for partial labeled data. Training data are obtained by applying a bidirectional lexicon matching scheme. A modified Viterbi algorithm using lexicon matching scheme is also proposed for word segmentation on raw sentences. Experiments show that our model has a precision of 93.88% on test data and outperforms two popular open source Chinese word segmentation tools i.e.,Han LP and THULAC. By using lexicon, our model is able to be adapted for other domain text word segmentation.
In this paper, we deal with the method of grid generation. From the perspective of deformation mechanics, we propose a new method to generate grid with boundary constrain. Based on the stress balance equation on the n...
详细信息
The intelligent home old-age service was an important way to solve the problem of current worsening old-age care reality in *** intelligent old-age home endowment monitoring system based on internet of things was prop...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
The intelligent home old-age service was an important way to solve the problem of current worsening old-age care reality in *** intelligent old-age home endowment monitoring system based on internet of things was proposed in this *** system has higher practical value,old-age health condition could be real-time monitored,the alerts could be sent when the unforeseen situation happened to the old man.
Short texts matching problem is a special issue in natural language matching. Different from common natural language, short texts have their own characteristices, such as casual expressions and limited lengths, especi...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
Short texts matching problem is a special issue in natural language matching. Different from common natural language, short texts have their own characteristices, such as casual expressions and limited lengths, especially in the sentences from social media. Previous works usually use rule-based model and retrieval-based model to match short texts. These models merely focus on word-level similarity between short texts and can not capture deep matching relation of them. To boost the performance of short texts matching, we investigate a basic convolutional neural network model to learn the sentence-level deep matching relation between short texts. Subsequently, we propose a hybrid model to merge sentence-level deep matching relation with shallow features to generate the final matching score. We evaluate our model on a dataset of short-text conversation based on real-world instances from Sina Weibo. The experimental results show that our model outperforms the previous state-of-art work on this task.
Hot topic detection has always been a hot research field, and there are a large number of the applications of this technology in real life. Most of the previous work, however, focused only on the textual information o...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
Hot topic detection has always been a hot research field, and there are a large number of the applications of this technology in real life. Most of the previous work, however, focused only on the textual information of the news itself, while ignoring the other attributes of the news, such as the time the news was published, which can also tell the topic described in its perspective. And others use only one certain method to calculate the text similarity, which all have their disadvantages. To solve these problems, we proposed our own topic detection algorithm, which takes into account the information difference between the title and the text, combines several methods to calculate text similarity, and combines text and time similarity together. We tested the combined similarity calculation methods, and tested the effect of several time similarity equations. Then we took three different models to calculate the combined similarity which are linear model, quadratic polynomial model and neural network model. Finally, we give out the results and analysis of our experiments.
In the field of information retrieval, with the rapid growth of the amount of questions and answers, automatic question-answering system comes up to be a hot research direction, which consists of three procedures: que...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
In the field of information retrieval, with the rapid growth of the amount of questions and answers, automatic question-answering system comes up to be a hot research direction, which consists of three procedures: question classification, information retrieval and answer extraction. Question classification is the first and most important part of the whole task. Currently, two kinds of algorithms are employed, rule-based algorithms and statisticalmodel-based algorithms. Rule-based algorithms have good performance in accuracy and pertinence with the shortcoming of relying on professional knowledge and poor scalability. Statistical-model-based algorithms get classification models from training dataset, these methods extract syntax features heuristically and provide better scalability and thus most question classification algorithms are based on statisticalmodel. However, semantic features have largely been overlooked in existing statistical-model-based question classification algorithms. In this paper, we propose a contextaware hybrid model based on a statistical-model PGM and a semantic language model word2 vec. The experimental evaluations demonstrate the capability of the proposed model.
In order to enhance the white box security of software, we proposed a reduplicate code obfuscation algorithm to protect the source code. Firstly, we apply the parameter decomposition tree to formalize the code, and th...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
In order to enhance the white box security of software, we proposed a reduplicate code obfuscation algorithm to protect the source code. Firstly, we apply the parameter decomposition tree to formalize the code, and then we utilize flattening control flow system to decompose the source code into a multi-branch WHILE-SWITCH loop structure. Finally, we apply opaque predicates to obfuscate the flattened code for the secondary obfuscation. In this paper, opaque predicate code representation and different methods of inserting opaque predicates into program braches and sequence blocks were given. Experiments has been made to compare time-space cost of source code and obfuscated code. The results demonstrate that the proposed algorithm can improve code's anti-attack ability, increasing the difficulty of reverse engineering as well.
this paper delineates a case study analyzing and forecasting of the outpatient visits frequency of a hospital in Zhengzhou, China. By evaluating the annual out-patient data throughout the year of 2015, this paper appl...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
this paper delineates a case study analyzing and forecasting of the outpatient visits frequency of a hospital in Zhengzhou, China. By evaluating the annual out-patient data throughout the year of 2015, this paper applies the D as timescale and carries out the experiment so as to forecast the number of visiting patients with the impact of the W taken into consideration. Two models are used separately: the Autoregressive Integrated Moving Average(ARIMA) with seasonal index and the Seasonal Autoregressive Integrated Moving Average(SARIMA). Based on the empirical findings from the comparison of the fitting effect and forecasting effect of the above two models, it is clear that SARIMA reaches a satisfactory outcome: it displays optimum indexes. Therefore it is preferable to deploy the SARIMA model to proceed a forecasting of outpatient visits for medical institutions. Meanwhile the paper also aims to provide management of medical institution with theory grounds of working and personnel arrangement and insight so as to make a prompt and reasonable contingency plan when it comes to sudden disease.
With the advance of time and scholars pay close attention to prediction-error expansion in reversible data hiding, a large number of adaptive prediction-error expansion algorithms are emerging. Previous methods often ...
详细信息
ISBN:
(纸本)9781538619797;9781538619780
With the advance of time and scholars pay close attention to prediction-error expansion in reversible data hiding, a large number of adaptive prediction-error expansion algorithms are emerging. Previous methods often use closed pixel correlation to predict pixels, but the prediction accuracy is low in the image texture region. In this paper, we sum a reversible data hiding framework based on prediction-error expansion at first. Depending on this framework, we proposed an iterative regularization method to predict pixels by applying a first order difference edge preserving operator predictor. The continuous iterative algorithm is used to modify the prediction results to obtain the optimal and stable prediction results. In this way, the overall prediction effect of the image is improved, especially in the texture region of the image. Moreover, the first order difference sum is used to sort the order of the embedded information, so as to improve the quality of the stego image. The experimental results show the mathod proposed is better than some state-of-the-art methods.
暂无评论