Assuming seismic data in a suitable domain is low rank while missing traces or noises increase the rank of the data matrix,the rank⁃reduced methods have been applied successfully for seismic interpolation and denoisin...Assuming seismic data in a suitable domain is low rank while missing traces or noises increase the rank of the data matrix,the rank⁃reduced methods have been applied successfully for seismic interpolation and denoising.These rank⁃reduced methods mainly include Cadzow reconstruction that uses eigen decomposition of the Hankel matrix in the f⁃x(frequency⁃spatial)domain,and nuclear⁃norm minimization(NNM)based on rigorous optimization theory on matrix completion(MC).In this paper,a low patch⁃rank MC is proposed with a random⁃overlapped texture⁃patch mapping for interpolation of regularly missing traces in a three⁃dimensional(3D)seismic volume.The random overlap plays a simple but important role to make the low⁃rank method effective for aliased data.It shifts the regular column missing of data matrix to random point missing in the mapped matrix,where the missing data increase the rank thus the classic low⁃rank MC theory works.Unlike the Hankel matrix based rank⁃reduced method,the proposed method does not assume a superposition of linear events,but assumes the data have repeated texture patterns.Such data lead to a low⁃rank matrix after the proposed texture⁃patch mapping.Thus the methods can interpolate the waveforms with varying dips in space.A fast low⁃rank factorization method and an orthogonal rank⁃one matrix pursuit method are applied to solve the presented interpolation model.The former avoids the singular value decomposition(SVD)computation and the latter only needs to compute the large singular values during iterations.The two fast algorithms are suitable for large⁃scale data.Simple averaging realizations of several results from different random⁃overlapped texture⁃patch mappings can further increase the reconstructed signal⁃to⁃noise ratio(SNR).Examples on synthetic data and field data are provided to show successful performance of the presented method.展开更多
One way to improve practicability of automatic program repair（APR） techniques is to build prediction models which can predict whether an application of a APR technique on a bug is effective or not. Existing predicti...One way to improve practicability of automatic program repair（APR） techniques is to build prediction models which can predict whether an application of a APR technique on a bug is effective or not. Existing prediction models have some limitations. First, the prediction models are built with hand crafted features which usually fail to capture the semantic characteristics of program repair task. Second, the performance of the prediction models is only evaluated on Genprog, a genetic-programming based APR technique. This paper develops prediction models, i.e., random forest prediction models for SPR, another kind of generate-and-validate APR technique, which can distinguish ineffective repair instances from effective repair instances. Rather than handcrafted features, we use features automatically learned by deep belief network（DBN） to train the prediction models. The empirical results show that compared to the baseline models, that is, all effective models, our proposed models can at least improve the F1 by 9% and AUC（area under the receiver operating characteristics curve） by 19%. At the same time, the prediction model using learned features at least outperforms the one using hand-crafted features in terms of F1 by 11%.展开更多
Product storage policy, single picking volume and picking routing are the three factors of vital importance that affect the efficiency of a crane to pick goods in automated storage and retrieval systems(AS/RS). Compar...Product storage policy, single picking volume and picking routing are the three factors of vital importance that affect the efficiency of a crane to pick goods in automated storage and retrieval systems(AS/RS). Comparative experiments on picking efficiency were conducted targeting picking operation with order of 1 to 20. Based on dedicated and random storage policies, 4 picking methods of patching-based, S-type, return-type and optimized-type routes were used and compared in the experiments. The results show that either the dedicated policy or the random policy was applied, crane worked most efficiently with optimizedtype route, followed by S-type path, patching-based path, and return-type path. When the number of orders in a single picking is larger(more than 5), the random storage policy is preferable to the dedicated policy.展开更多
A novel frequency selective surface （FSS） for reducing radar cross section （RCS） is proposed in this paper. This FSS is based on the random distribution method, so it can be called random surface. In this paper, t...A novel frequency selective surface （FSS） for reducing radar cross section （RCS） is proposed in this paper. This FSS is based on the random distribution method, so it can be called random surface. In this paper, the stacked patches serving as periodic elements are employed for RCS reduction. Previous work has demonstrated the efficiency by utilizing the microstrip patches, especially for the reflectarray. First, the relevant theory of the method is described. Then a sample of a three-layer variable-sized stacked patch random surface with a dimension of 260 mm x 260 mm is simulated, fabricated, and measured in order to demonstrate the validity of the proposed design. For the normal incidence, the 8-dB RCS reduction can be achieved both by the simulation and the measurement in 8 GHz-13 GHz. The oblique incidence of 30° is also investigated, in which the 7-dB RCS reduction can be obtained in a frequency range of 8 GHz-14 GHz.展开更多
目的:观察益气温经穴位贴贴敷联合西药口服治疗绝经后骨质疏松症(postmenopausal osteoporosis,PMOP)肾虚血瘀证的临床疗效。方法:将符合要求的120例PMOP肾虚血瘀证患者随机分为2组,每组60例,分别采用单纯西药(阿仑膦酸钠片、钙尔奇D片...目的:观察益气温经穴位贴贴敷联合西药口服治疗绝经后骨质疏松症(postmenopausal osteoporosis,PMOP)肾虚血瘀证的临床疗效。方法:将符合要求的120例PMOP肾虚血瘀证患者随机分为2组,每组60例,分别采用单纯西药(阿仑膦酸钠片、钙尔奇D片、骨化三醇胶丸)口服治疗(西药治疗组)与益气温经穴位贴贴敷联合西药口服治疗(联合治疗组)。3个月为1个疗程,每个疗程之间隔1个月,共治疗3个疗程。分别于治疗前、治疗开始后6个月、治疗开始后12个月,比较2组患者的骨密度T值、血清雌二醇(estradiol,E2)水平、血清卵泡刺激素(follicle stimulating hormone,FSH)水平、血清Ⅰ型前胶原氨基端前肽(N-terminal propeptide of typeⅠprecollagen,PⅠNP)水平、血清Ⅰ型胶原羧基端肽β特殊序列(β-C-terminal telopeptide of typeⅠcollagen,β-CTX)水平、腰背部疼痛视觉模拟量表(visual analogue scale,VAS)评分、简明健康状况调查表(short form 36 health survey questionnaire,SF-36)评分。结果:①一般情况。共5例患者退出试验,其中西药治疗组3例、联合治疗组2例。最终西药治疗组纳入57例患者,联合治疗组纳入58例患者。②骨密度T值。治疗前后骨密度T值总体上随时间推移呈上升趋势(F=109.349,P=0.000),2组间差异无统计学意义(F=0.014,P=0.905)。③血清E2水平。治疗前后西药治疗组患者血清E2水平随时间推移无明显变化(F=0.790,P=0.456),联合治疗组患者血清E2水平随时间推移呈上升趋势(F=6.559,P=0.002)。治疗开始后6个月、12个月,联合治疗组患者血清E2水平均高于西药治疗组(t=-3.983,P=0.000;t=-4.177,P=0.000)。④血清FSH水平。治疗前后血清FSH水平总体上随时间推移呈下降趋势(F=17.933,P=0.000),2组间差异无统计学意义(F=1.123,P=0.291)。⑤血清PⅠNP水平。治疗前后西药治疗组患者血清PⅠNP水平随时间推移无明显变化(F=0.368,P=0.693),联合治疗组患者血清PⅠNP水平随时间推移呈上升趋势(F=11.724,P=0.000)。治疗开始后6个月、12个月,联合治疗组患者血清PⅠNP水平均高于西药治疗组(t=-5.844,P=0.000;t=-5.619,P=0.000)。⑥血清β-CTX水平。治疗前后血清β-CTX水平总体上随时间推移呈下降趋势(F=6.371,P=0.002),2组间差异无统计学意义(F=1.898,P=0.171)。⑦腰背部疼痛VAS评分。治疗前后2组患者腰背部疼痛VAS评分随时间推移均呈下降趋势(F=35.668,P=0.000;F=147.005,P=0.000)。治疗开始后6个月、12个月,联合治疗组患者腰背部疼痛VAS评分均低于西药治疗组(t=12.848,P=0.000;t=13.291,P=0.000)。⑧SF-36评分。治疗前后2组患者SF-36评分随时间推移均呈上升趋势(F=54.622,P=0.000;F=105.957,P=0.000)。治疗开始后6个月、12个月,联合治疗组患者SF-36评分均高于西药治疗组(t=-10.331,P=0.000;t=-10.323,P=0.000)。结论:益气温经穴位贴贴敷联合西药口服治疗PMOP肾虚血瘀证,在提高血清E2水平、促进骨形成、减轻腰背部疼痛、改善患者生活质量方面优于单纯西药口服治疗;但两者在降低血清FSH水平、抑制骨吸收、提高骨密度方面的效果相当。展开更多
针对如何充分利用空间特征来达到较高的高光谱图像分类精度的问题,提出了一种基于三维离散小波变换(3D-DWT)与随机补丁网络(RPNet)结合的高光谱图像的地物属性分类算法。在分类过程中,综合3D-DWT提取的特征和RPNet深度学习框架提取的特...针对如何充分利用空间特征来达到较高的高光谱图像分类精度的问题,提出了一种基于三维离散小波变换(3D-DWT)与随机补丁网络(RPNet)结合的高光谱图像的地物属性分类算法。在分类过程中,综合3D-DWT提取的特征和RPNet深度学习框架提取的特征,利用支持向量机(SVM)对特征向量进行分类。所提出的方法在Indian Pines和University of Pavia两个数据集上进行测试,结果表明该方法比现有方法有显著的分类性能的提高。展开更多
文摘Assuming seismic data in a suitable domain is low rank while missing traces or noises increase the rank of the data matrix,the rank⁃reduced methods have been applied successfully for seismic interpolation and denoising.These rank⁃reduced methods mainly include Cadzow reconstruction that uses eigen decomposition of the Hankel matrix in the f⁃x(frequency⁃spatial)domain,and nuclear⁃norm minimization(NNM)based on rigorous optimization theory on matrix completion(MC).In this paper,a low patch⁃rank MC is proposed with a random⁃overlapped texture⁃patch mapping for interpolation of regularly missing traces in a three⁃dimensional(3D)seismic volume.The random overlap plays a simple but important role to make the low⁃rank method effective for aliased data.It shifts the regular column missing of data matrix to random point missing in the mapped matrix,where the missing data increase the rank thus the classic low⁃rank MC theory works.Unlike the Hankel matrix based rank⁃reduced method,the proposed method does not assume a superposition of linear events,but assumes the data have repeated texture patterns.Such data lead to a low⁃rank matrix after the proposed texture⁃patch mapping.Thus the methods can interpolate the waveforms with varying dips in space.A fast low⁃rank factorization method and an orthogonal rank⁃one matrix pursuit method are applied to solve the presented interpolation model.The former avoids the singular value decomposition(SVD)computation and the latter only needs to compute the large singular values during iterations.The two fast algorithms are suitable for large⁃scale data.Simple averaging realizations of several results from different random⁃overlapped texture⁃patch mappings can further increase the reconstructed signal⁃to⁃noise ratio(SNR).Examples on synthetic data and field data are provided to show successful performance of the presented method.
基金Supported by the National Natural Science Foundation of China(61603242)Opening Project of Collaborative Innovation Center for Economics Crime Investigation and Prevention Technology(JXJZXTCX-030)+1 种基金the Scientific Research Fund of Zhaoqing Univeristy(201734)Innovative Guidance Fund of Zhaoqing City(201704030409)
文摘One way to improve practicability of automatic program repair（APR） techniques is to build prediction models which can predict whether an application of a APR technique on a bug is effective or not. Existing prediction models have some limitations. First, the prediction models are built with hand crafted features which usually fail to capture the semantic characteristics of program repair task. Second, the performance of the prediction models is only evaluated on Genprog, a genetic-programming based APR technique. This paper develops prediction models, i.e., random forest prediction models for SPR, another kind of generate-and-validate APR technique, which can distinguish ineffective repair instances from effective repair instances. Rather than handcrafted features, we use features automatically learned by deep belief network（DBN） to train the prediction models. The empirical results show that compared to the baseline models, that is, all effective models, our proposed models can at least improve the F1 by 9% and AUC（area under the receiver operating characteristics curve） by 19%. At the same time, the prediction model using learned features at least outperforms the one using hand-crafted features in terms of F1 by 11%.
基金Funded by National Social Science Foundation of China(16CGL018)the Soft Science Research Funds for Chengdu Science and Technology Project(2015-RK00-00206-ZF)the National United Engineering Laboratory of Integrated and Intelligent Transportation,Southwest Jiaotong University,China
文摘Product storage policy, single picking volume and picking routing are the three factors of vital importance that affect the efficiency of a crane to pick goods in automated storage and retrieval systems(AS/RS). Comparative experiments on picking efficiency were conducted targeting picking operation with order of 1 to 20. Based on dedicated and random storage policies, 4 picking methods of patching-based, S-type, return-type and optimized-type routes were used and compared in the experiments. The results show that either the dedicated policy or the random policy was applied, crane worked most efficiently with optimizedtype route, followed by S-type path, patching-based path, and return-type path. When the number of orders in a single picking is larger(more than 5), the random storage policy is preferable to the dedicated policy.
文摘A novel frequency selective surface （FSS） for reducing radar cross section （RCS） is proposed in this paper. This FSS is based on the random distribution method, so it can be called random surface. In this paper, the stacked patches serving as periodic elements are employed for RCS reduction. Previous work has demonstrated the efficiency by utilizing the microstrip patches, especially for the reflectarray. First, the relevant theory of the method is described. Then a sample of a three-layer variable-sized stacked patch random surface with a dimension of 260 mm x 260 mm is simulated, fabricated, and measured in order to demonstrate the validity of the proposed design. For the normal incidence, the 8-dB RCS reduction can be achieved both by the simulation and the measurement in 8 GHz-13 GHz. The oblique incidence of 30° is also investigated, in which the 7-dB RCS reduction can be obtained in a frequency range of 8 GHz-14 GHz.
文摘目的:观察益气温经穴位贴贴敷联合西药口服治疗绝经后骨质疏松症(postmenopausal osteoporosis,PMOP)肾虚血瘀证的临床疗效。方法:将符合要求的120例PMOP肾虚血瘀证患者随机分为2组,每组60例,分别采用单纯西药(阿仑膦酸钠片、钙尔奇D片、骨化三醇胶丸)口服治疗(西药治疗组)与益气温经穴位贴贴敷联合西药口服治疗(联合治疗组)。3个月为1个疗程,每个疗程之间隔1个月,共治疗3个疗程。分别于治疗前、治疗开始后6个月、治疗开始后12个月,比较2组患者的骨密度T值、血清雌二醇(estradiol,E2)水平、血清卵泡刺激素(follicle stimulating hormone,FSH)水平、血清Ⅰ型前胶原氨基端前肽(N-terminal propeptide of typeⅠprecollagen,PⅠNP)水平、血清Ⅰ型胶原羧基端肽β特殊序列(β-C-terminal telopeptide of typeⅠcollagen,β-CTX)水平、腰背部疼痛视觉模拟量表(visual analogue scale,VAS)评分、简明健康状况调查表(short form 36 health survey questionnaire,SF-36)评分。结果:①一般情况。共5例患者退出试验,其中西药治疗组3例、联合治疗组2例。最终西药治疗组纳入57例患者,联合治疗组纳入58例患者。②骨密度T值。治疗前后骨密度T值总体上随时间推移呈上升趋势(F=109.349,P=0.000),2组间差异无统计学意义(F=0.014,P=0.905)。③血清E2水平。治疗前后西药治疗组患者血清E2水平随时间推移无明显变化(F=0.790,P=0.456),联合治疗组患者血清E2水平随时间推移呈上升趋势(F=6.559,P=0.002)。治疗开始后6个月、12个月,联合治疗组患者血清E2水平均高于西药治疗组(t=-3.983,P=0.000;t=-4.177,P=0.000)。④血清FSH水平。治疗前后血清FSH水平总体上随时间推移呈下降趋势(F=17.933,P=0.000),2组间差异无统计学意义(F=1.123,P=0.291)。⑤血清PⅠNP水平。治疗前后西药治疗组患者血清PⅠNP水平随时间推移无明显变化(F=0.368,P=0.693),联合治疗组患者血清PⅠNP水平随时间推移呈上升趋势(F=11.724,P=0.000)。治疗开始后6个月、12个月,联合治疗组患者血清PⅠNP水平均高于西药治疗组(t=-5.844,P=0.000;t=-5.619,P=0.000)。⑥血清β-CTX水平。治疗前后血清β-CTX水平总体上随时间推移呈下降趋势(F=6.371,P=0.002),2组间差异无统计学意义(F=1.898,P=0.171)。⑦腰背部疼痛VAS评分。治疗前后2组患者腰背部疼痛VAS评分随时间推移均呈下降趋势(F=35.668,P=0.000;F=147.005,P=0.000)。治疗开始后6个月、12个月,联合治疗组患者腰背部疼痛VAS评分均低于西药治疗组(t=12.848,P=0.000;t=13.291,P=0.000)。⑧SF-36评分。治疗前后2组患者SF-36评分随时间推移均呈上升趋势(F=54.622,P=0.000;F=105.957,P=0.000)。治疗开始后6个月、12个月,联合治疗组患者SF-36评分均高于西药治疗组(t=-10.331,P=0.000;t=-10.323,P=0.000)。结论:益气温经穴位贴贴敷联合西药口服治疗PMOP肾虚血瘀证,在提高血清E2水平、促进骨形成、减轻腰背部疼痛、改善患者生活质量方面优于单纯西药口服治疗;但两者在降低血清FSH水平、抑制骨吸收、提高骨密度方面的效果相当。
文摘针对如何充分利用空间特征来达到较高的高光谱图像分类精度的问题,提出了一种基于三维离散小波变换(3D-DWT)与随机补丁网络(RPNet)结合的高光谱图像的地物属性分类算法。在分类过程中,综合3D-DWT提取的特征和RPNet深度学习框架提取的特征,利用支持向量机(SVM)对特征向量进行分类。所提出的方法在Indian Pines和University of Pavia两个数据集上进行测试,结果表明该方法比现有方法有显著的分类性能的提高。
