The use of hidden conditional random fields （HCRFs） for tone modeling is explored. The tone recognition performance is improved using HCRFs by taking advantage of intra-syllable dynamic, inter-syllable dynamic and d...The use of hidden conditional random fields （HCRFs） for tone modeling is explored. The tone recognition performance is improved using HCRFs by taking advantage of intra-syllable dynamic, inter-syllable dynamic and duration features. When the tone model is integrated into continuous speech recognition, the discriminative model weight training （DMWT） is proposed. Acoustic and tone scores are scaled by model weights discriminatively trained by the minimum phone error （MPE） criterion. Two schemes of weight training are evaluated and a smoothing technique is used to make training robust to overtraining problem. Experiments show that the accuracies of tone recognition and large vocabulary continuous speech recognition （LVCSR） can be improved by the HCRFs based tone model. Compared with the global weight scheme, continuous speech recognition can be improved by the discriminative trained weight combinations.展开更多
To reduce the computation cost of a combined probabilistic graphical model and a deep neural network in semantic segmentation, the local region condition random field (LRCRF) model is investigated which selectively ap...To reduce the computation cost of a combined probabilistic graphical model and a deep neural network in semantic segmentation, the local region condition random field (LRCRF) model is investigated which selectively applies the condition random field (CRF) to the most active region in the image. The full convolutional network structure is optimized with the ResNet-18 structure and dilated convolution to expand the receptive field. The tracking networks are also improved based on SiameseFC by considering the frame relations in consecutive-frame traffic scene maps. Moreover, the segmentation results of the greyscale input data sets are more stable and effective than using the RGB images for deep neural network feature extraction. The experimental results show that the proposed method takes advantage of the image features directly and achieves good real-time performance and high segmentation accuracy.展开更多
Markov random fields(MRF) have potential for predicting and simulating petroleum reservoir facies more accurately from sample data such as logging, core data and seismic data because they can incorporate interclass re...Markov random fields(MRF) have potential for predicting and simulating petroleum reservoir facies more accurately from sample data such as logging, core data and seismic data because they can incorporate interclass relationships. While, many relative studies were based on Markov chain, not MRF, and using Markov chain model for 3D reservoir stochastic simulation has always been the difficulty in reservoir stochastic simulation. MRF was proposed to simulate type variables(for example lithofacies) in this work. Firstly, a Gibbs distribution was proposed to characterize reservoir heterogeneity for building 3-D(three-dimensional) MRF. Secondly, maximum likelihood approaches of model parameters on well data and training image were considered. Compared with the simulation results of MC(Markov chain), the MRF can better reflect the spatial distribution characteristics of sand body.展开更多
针对畜禽疫病文本语料匮乏、文本内包含大量疫病名称及短语等未登录词问题,提出了一种结合词典匹配的BERT-BiLSTM-CRF畜禽疫病文本分词模型。以羊疫病为研究对象,构建了常见疫病文本数据集,将其与通用语料PKU结合,利用BERT(Bidirectiona...针对畜禽疫病文本语料匮乏、文本内包含大量疫病名称及短语等未登录词问题,提出了一种结合词典匹配的BERT-BiLSTM-CRF畜禽疫病文本分词模型。以羊疫病为研究对象,构建了常见疫病文本数据集,将其与通用语料PKU结合,利用BERT(Bidirectional encoder representation from transformers)预训练语言模型进行文本向量化表示;通过双向长短时记忆网络(Bidirectional long short-term memory network,BiLSTM)获取上下文语义特征;由条件随机场(Conditional random field,CRF)输出全局最优标签序列。基于此,在CRF层后加入畜禽疫病领域词典进行分词匹配修正,减少在分词过程中出现的疫病名称及短语等造成的歧义切分,进一步提高了分词准确率。实验结果表明,结合词典匹配的BERT-BiLSTM-CRF模型在羊常见疫病文本数据集上的F1值为96.38%,与jieba分词器、BiLSTM-Softmax模型、BiLSTM-CRF模型、未结合词典匹配的本文模型相比,分别提升11.01、10.62、8.3、0.72个百分点,验证了方法的有效性。与单一语料相比,通用语料PKU和羊常见疫病文本数据集结合的混合语料,能够同时对畜禽疫病专业术语及疫病文本中常用词进行准确切分,在通用语料及疫病文本数据集上F1值都达到95%以上,具有较好的模型泛化能力。该方法可用于畜禽疫病文本分词。展开更多
针对油气领域知识图谱构建过程中命名实体识别使用传统方法存在实体特征信息提取不准确、识别效率低的问题,提出了一种基于BERT-BiLSTM-CRF模型的命名实体识别研究方法。该方法首先利用BERT(bidirectional encoder representations from...针对油气领域知识图谱构建过程中命名实体识别使用传统方法存在实体特征信息提取不准确、识别效率低的问题,提出了一种基于BERT-BiLSTM-CRF模型的命名实体识别研究方法。该方法首先利用BERT(bidirectional encoder representations from transformers)预训练模型得到输入序列语义的词向量;然后将训练后的词向量输入双向长短期记忆网络(bi-directional long short-term memory,BiLSTM)模型进一步获取上下文特征;最后根据条件随机场(conditional random fields,CRF)的标注规则和序列解码能力输出最大概率序列标注结果,构建油气领域命名实体识别模型框架。将BERT-BiLSTM-CRF模型与其他2种命名实体识别模型(BiLSTM-CRF、BiLSTM-Attention-CRF)在包括3万多条文本语料数据、4类实体的自建数据集上进行了对比实验。实验结果表明,BERT-BiLSTM-CRF模型的准确率(P)、召回率(R)和F_(1)值分别达到91.3%、94.5%和92.9%,实体识别效果优于其他2种模型。展开更多
We propose a 3D model feature line extraction method using templates for guidance. The 3D model is first projected into a depth map, and a set of candidate feature points are extracted. Then, a conditional random fiel...We propose a 3D model feature line extraction method using templates for guidance. The 3D model is first projected into a depth map, and a set of candidate feature points are extracted. Then, a conditional random fields (CRF) model is established to match the sketch points and the candidate feature points. Using sketch strokes, the candidate feature points can then be connected to obtain the feature lines, and using a CRF-matching model, the 2D image shape similarity features and 3D model geometric features can be effectively integrated. Finally, a relational metric based on shape and topological similarity is proposed to evaluate the matching results, and an iterative matching process is applied to obtain the globally optimized model feature lines. Experimental results showed that the proposed method can extract sound 3D model feature lines which correspond to the initial sketch template.展开更多
With the widespread application of deep learning in the field of computer vision,gradually allowing medical image technology to assist doctors in making diagnoses has great practical and research significance.Aiming a...With the widespread application of deep learning in the field of computer vision,gradually allowing medical image technology to assist doctors in making diagnoses has great practical and research significance.Aiming at the shortcomings of the traditional U-Net model in 3D spatial information extraction,model over-fitting,and low degree of semantic information fusion,an improved medical image segmentation model has been used to achieve more accurate segmentation of medical images.In this model,we make full use of the residual network(ResNet)to solve the over-fitting problem.In order to process and aggregate data at different scales,the inception network is used instead of the traditional convolutional layer,and the dilated convolution is used to increase the receptive field.The conditional random field(CRF)can complete the contour refinement work.Compared with the traditional 3D U-Net network,the segmentation accuracy of the improved liver and tumor images increases by 2.89%and 7.66%,respectively.As a part of the image processing process,the method in this paper not only can be used for medical image segmentation,but also can lay the foundation for subsequent image 3D reconstruction work.展开更多
文摘The use of hidden conditional random fields （HCRFs） for tone modeling is explored. The tone recognition performance is improved using HCRFs by taking advantage of intra-syllable dynamic, inter-syllable dynamic and duration features. When the tone model is integrated into continuous speech recognition, the discriminative model weight training （DMWT） is proposed. Acoustic and tone scores are scaled by model weights discriminatively trained by the minimum phone error （MPE） criterion. Two schemes of weight training are evaluated and a smoothing technique is used to make training robust to overtraining problem. Experiments show that the accuracies of tone recognition and large vocabulary continuous speech recognition （LVCSR） can be improved by the HCRFs based tone model. Compared with the global weight scheme, continuous speech recognition can be improved by the discriminative trained weight combinations.
文摘To reduce the computation cost of a combined probabilistic graphical model and a deep neural network in semantic segmentation, the local region condition random field (LRCRF) model is investigated which selectively applies the condition random field (CRF) to the most active region in the image. The full convolutional network structure is optimized with the ResNet-18 structure and dilated convolution to expand the receptive field. The tracking networks are also improved based on SiameseFC by considering the frame relations in consecutive-frame traffic scene maps. Moreover, the segmentation results of the greyscale input data sets are more stable and effective than using the RGB images for deep neural network feature extraction. The experimental results show that the proposed method takes advantage of the image features directly and achieves good real-time performance and high segmentation accuracy.
基金Project(2011ZX05002-005-006)supported by the National "Twelveth Five Year" Science and Technology Major Research Program,China
文摘Markov random fields(MRF) have potential for predicting and simulating petroleum reservoir facies more accurately from sample data such as logging, core data and seismic data because they can incorporate interclass relationships. While, many relative studies were based on Markov chain, not MRF, and using Markov chain model for 3D reservoir stochastic simulation has always been the difficulty in reservoir stochastic simulation. MRF was proposed to simulate type variables(for example lithofacies) in this work. Firstly, a Gibbs distribution was proposed to characterize reservoir heterogeneity for building 3-D(three-dimensional) MRF. Secondly, maximum likelihood approaches of model parameters on well data and training image were considered. Compared with the simulation results of MC(Markov chain), the MRF can better reflect the spatial distribution characteristics of sand body.
文摘针对畜禽疫病文本语料匮乏、文本内包含大量疫病名称及短语等未登录词问题,提出了一种结合词典匹配的BERT-BiLSTM-CRF畜禽疫病文本分词模型。以羊疫病为研究对象,构建了常见疫病文本数据集,将其与通用语料PKU结合,利用BERT(Bidirectional encoder representation from transformers)预训练语言模型进行文本向量化表示;通过双向长短时记忆网络(Bidirectional long short-term memory network,BiLSTM)获取上下文语义特征;由条件随机场(Conditional random field,CRF)输出全局最优标签序列。基于此,在CRF层后加入畜禽疫病领域词典进行分词匹配修正,减少在分词过程中出现的疫病名称及短语等造成的歧义切分,进一步提高了分词准确率。实验结果表明,结合词典匹配的BERT-BiLSTM-CRF模型在羊常见疫病文本数据集上的F1值为96.38%,与jieba分词器、BiLSTM-Softmax模型、BiLSTM-CRF模型、未结合词典匹配的本文模型相比,分别提升11.01、10.62、8.3、0.72个百分点,验证了方法的有效性。与单一语料相比,通用语料PKU和羊常见疫病文本数据集结合的混合语料,能够同时对畜禽疫病专业术语及疫病文本中常用词进行准确切分,在通用语料及疫病文本数据集上F1值都达到95%以上,具有较好的模型泛化能力。该方法可用于畜禽疫病文本分词。
文摘针对油气领域知识图谱构建过程中命名实体识别使用传统方法存在实体特征信息提取不准确、识别效率低的问题,提出了一种基于BERT-BiLSTM-CRF模型的命名实体识别研究方法。该方法首先利用BERT(bidirectional encoder representations from transformers)预训练模型得到输入序列语义的词向量;然后将训练后的词向量输入双向长短期记忆网络(bi-directional long short-term memory,BiLSTM)模型进一步获取上下文特征;最后根据条件随机场(conditional random fields,CRF)的标注规则和序列解码能力输出最大概率序列标注结果,构建油气领域命名实体识别模型框架。将BERT-BiLSTM-CRF模型与其他2种命名实体识别模型(BiLSTM-CRF、BiLSTM-Attention-CRF)在包括3万多条文本语料数据、4类实体的自建数据集上进行了对比实验。实验结果表明,BERT-BiLSTM-CRF模型的准确率(P)、召回率(R)和F_(1)值分别达到91.3%、94.5%和92.9%,实体识别效果优于其他2种模型。
基金supported by the National Natural Science Foundation of China (Nos. 61272219, 61100110, and 61021062)the National High-Tech R&D Program (863) of China (No. 2007AA01Z334)+1 种基金the Program for New Century Excellent Talents in University (No. NCET-0404605)the Science and Technology Program of Jiangsu Province, China (Nos. BE2010072, BE2011058, and BY2012190)
文摘We propose a 3D model feature line extraction method using templates for guidance. The 3D model is first projected into a depth map, and a set of candidate feature points are extracted. Then, a conditional random fields (CRF) model is established to match the sketch points and the candidate feature points. Using sketch strokes, the candidate feature points can then be connected to obtain the feature lines, and using a CRF-matching model, the 2D image shape similarity features and 3D model geometric features can be effectively integrated. Finally, a relational metric based on shape and topological similarity is proposed to evaluate the matching results, and an iterative matching process is applied to obtain the globally optimized model feature lines. Experimental results showed that the proposed method can extract sound 3D model feature lines which correspond to the initial sketch template.
文摘With the widespread application of deep learning in the field of computer vision,gradually allowing medical image technology to assist doctors in making diagnoses has great practical and research significance.Aiming at the shortcomings of the traditional U-Net model in 3D spatial information extraction,model over-fitting,and low degree of semantic information fusion,an improved medical image segmentation model has been used to achieve more accurate segmentation of medical images.In this model,we make full use of the residual network(ResNet)to solve the over-fitting problem.In order to process and aggregate data at different scales,the inception network is used instead of the traditional convolutional layer,and the dilated convolution is used to increase the receptive field.The conditional random field(CRF)can complete the contour refinement work.Compared with the traditional 3D U-Net network,the segmentation accuracy of the improved liver and tumor images increases by 2.89%and 7.66%,respectively.As a part of the image processing process,the method in this paper not only can be used for medical image segmentation,but also can lay the foundation for subsequent image 3D reconstruction work.
