Using the double-difference earthquake location algorithm,the deterministic method (PTD method) and the CAP seismic moment tensor inversion method,the paper selects the primary waveform data of 78 earthquakes recorded...Using the double-difference earthquake location algorithm,the deterministic method (PTD method) and the CAP seismic moment tensor inversion method,the paper selects the primary waveform data of 78 earthquakes recorded by the "China Earthquake Science Array Probe Project in the Northern Part of North South Seismic Belt ",the "China Earthquake Scientific Exploration Array Data Center"of Institute of Geophysics,China Earthquake Administration,and the Inner Mongolia Digital Seismic Network to calculate the focal depths of the mainshock and the seismic sequence of the M_S5.8 Alxa Left Banner earthquake in Inner Mongolia. The results show that the focal depth of the main shock is 20. 6 km,determined by the double-difference earthquake location method,18. 1 km by the PTD method,and 19. 2 km by the CAP method. The focal depth of the earthquake sequence calculated by the double-difference location method is larger. The deterministic method (PTD method) and double-difference location method are the methods that fit the tectonic characteristics of the seismic source area,and the CAP method is suitable for larger earthquakes.展开更多
In the paper, an approach is proposed for the problem of consistency in depth maps estimation from binocular stereo video sequence. The consistent method includes temporal consistency and spatial consistency to elimin...In the paper, an approach is proposed for the problem of consistency in depth maps estimation from binocular stereo video sequence. The consistent method includes temporal consistency and spatial consistency to eliminate the flickering artifacts and smooth inaccuracy in depth recovery. So the improved global stereo matching based on graph cut and energy optimization is implemented. In temporal domain, the penalty function with coherence factor is introduced for temporal consistency, and the factor is determined by Lucas-Kanade optical flow weighted histogram similarity constraint(LKWHSC). In spatial domain, the joint bilateral truncated absolute difference(JBTAD) is proposed for segmentation smoothing. The method can smooth naturally and uniformly in low-gradient region and avoid over-smoothing as well as keep edge sharpness in high-gradient discontinuities to realize spatial consistency. The experimental results show that the algorithm can obtain better spatial and temporal consistent depth maps compared with the existing algorithms.展开更多
In this paper, based on the results of tomographic image of Tangshan and Xingtai areas, the relations between thecharacteristics of the two strong earthquake sequences and their three-dimensional velocity structures a...In this paper, based on the results of tomographic image of Tangshan and Xingtai areas, the relations between thecharacteristics of the two strong earthquake sequences and their three-dimensional velocity structures are studied.The research results indicate that:① Mosaic distribution of low-velocity bodies and high-velocity bodies, especially the existence of high-velocity bodies with large size in crust are the common basis of development of thetwo earthquake sequences. ② Scale, depth, and heterogeneity of high-velocity and low-velocity bodies are theimportant factors to effect the characteristic of earthquake sequences. ③ The depth of the high-velocity body inTangshan area is less than that in Xingtai area, which is the principal reason why the dominant focal depth and thebiggest focal depth of Tangshan earthquake sequence are less than Xingtai's. ④ The depth of the high-velocitybodies in Ninghe area is more than that in Tangshan-Luanxian area, which lead to the biggest magnitude and epicentral intensity are lower. These results could be helpful for predicting the main shock of strong swarm-typeearthquakes and later strong aftershocks.展开更多
Based on 294 earthquake sequences with magnitude greater than or equal to 5.0 occurred in Chinese mainland since 1970, the spatial distribution features of sequence types have been studied. In southwestern China, it t...Based on 294 earthquake sequences with magnitude greater than or equal to 5.0 occurred in Chinese mainland since 1970, the spatial distribution features of sequence types have been studied. In southwestern China, it takes mainshock-aftershock sequence type （MAT） as the major in Chuan-Dian rhombic block and concerned Xianshuihe-Anninghe-Xiaojiang seismic belt, as well as in Jinshajiang-Honghe seismic belt. Multiple mainshock type （MMT） mainly distributes in western Yunnan, and Longlin and Lancang areas in Tengchong-Baoshan block in west of Nujiang-Lancangjiang fault zone. A few isolated earthquake type （IET） mainly occurred in northwestern Sichuan and there is no IET occurred in Yunnan region. In northwestern China, it takes mainshock-aftershock sequence type （MAT） as the major in west segment of South Tianshan in Xinjiang region. Some MMT also occurred in this area in the intersection of Kalpin block and the Puchang fault zone. It takes IET as the major in middle Tianshan in Xinjiang. Along the Qilianshan seismic belt, most of sequences are MAT. In Qinghai region, it takes MAT as the major, but the regional feature of the spatial distribution of sequence types is not very clear. In North China, it takes MAT as the major in Yinshan-Yanshan-Bohai seismic belt, north edge of North China, and in Hebei plain seismic belt, as well as in sub-plate of lower river area of Yangtze River. In intersection of north segment of Shanxi seismic belt and the NW-trending Yinshan-Yanshan-Bohai seismic belt, there are several moderate or strong MMT with magnitude from 5.0 to 6.0 occurred. In south of North China around the latitude line of 35°N, it takes IET as the major. The spatial distribution of sequence types is relevant to the patterns of tectonic movements. MAT is mostly produced by the ruptures of locked units or asperities or the neonatal separating segments inside the fault zones. MMT is generally relevant to the conjugate structures or intersection of many tectonic settings. Further extension of simple fault often produces IET. Spatial distribution of sequence types is also correlative to the regional and deep environment of crustal medium to some extent. MAT mainly distributes in high velocity area in upper crust or in the transition zone between high velocity area and low velocity area, MMT mostly occurred in the low velocity area in upper crust.展开更多
In order to solve the problems of the fine division of sedimentary sequence cycles and their change in two-dimensional space as well as lateral extension contrast, we developed a method of wavelet depth-frequency anal...In order to solve the problems of the fine division of sedimentary sequence cycles and their change in two-dimensional space as well as lateral extension contrast, we developed a method of wavelet depth-frequency analysis. The single signal and composite signal of different Milankovitch cycles are obtained by numerical simulation. The simulated composite signal can be separated into single signals of a single frequency cycle. We also develop a well-seismic calibration insertion technology which helps to realize the calibration from the spectrum characteristics of a single well to the seismic profile. And then we determine the change and distribution characteristics of spectrum cycles in the two-dimensional space. It points out the direction in determining the variations of the regional sedimentary sequence cycles, underground strata structure and the contact relationship.展开更多
The measuring of the depth profile and electrical activity of implantation impurity in the top nanometer range of silicon encounters various difficulties and limitations, though it is known to be critical in fabricati...The measuring of the depth profile and electrical activity of implantation impurity in the top nanometer range of silicon encounters various difficulties and limitations, though it is known to be critical in fabrication of silicon complementary metal–oxide–semiconductor(CMOS) devices. In the present work, SRIM program and photocarrier radiometry(PCR)are employed to monitor the boron implantation in industrial-grade silicon in an ultra-low implantation energy range from 0.5 keV to 5 keV. The differential PCR technique, which is improved by greatly shortening the measurement time through the simplification of reference sample, is used to investigate the effects of implantation energy on the frequency behavior of the PCR signal for ultra-shallow junction. The transport parameters and thickness of shallow junction, extracted via multi-parameter fitting the dependence of differential PCR signal on modulation frequency to the corresponding theoretical model, well explain the energy dependence of PCR signal and further quantitatively characterize the recovery degree of structure damage induced by ion implantation and the electrical activation degree of impurities. The monitoring of nmlevel thickness and electronic properties exhibits high sensitivity and apparent monotonicity over the industrially relevant implantation energy range. The depth profiles of implantation boron in silicon with the typical electrical damage threshold(YED) of 5.3×10^(15)cm^(-3) are evaluated by the SRIM program, and the determined thickness values are consistent well with those extracted by the differential PCR. It is demonstrated that the SRIM and the PCR are both effective tools to characterize ultra-low energy ion implantation in silicon.展开更多
In this paper,a test or alternative scheme for studying large earthquake sequences through the study of small earthquake sequences is suggested,and a small earthquake sequence,the Lima earthquake sequence for which an...In this paper,a test or alternative scheme for studying large earthquake sequences through the study of small earthquake sequences is suggested,and a small earthquake sequence,the Lima earthquake sequence for which analogue records have been turned into digital data,is used here.In order to provide the deep construction background and the spatial distribution of structure for generating earthquakes,the P-wave and S-wave layered velocity models in this area are obtained by using mine explosion and earthquake observed records; then,the hypocenter locations and focal depths of the Lima earthquake sequence are determined adopting the velocity models given above and using a location method with numerical properties for a microseismic monitoring network(Zhao et al.,1994)and a new method for determining focal depth from data of a local seismographic network(Zhao,1992); finally,based on this,the variation of quality factor Q of the crustal medium during the period of the sequence is estimated.The obtained展开更多
Temporal and spatial variations in accommodation (i.e., paleo-water depth) and sediment accumulation (amount of deposition) in the intra-arc Osaka Basin, Japan, were reconstructed from the post-glacial transgression t...Temporal and spatial variations in accommodation (i.e., paleo-water depth) and sediment accumulation (amount of deposition) in the intra-arc Osaka Basin, Japan, were reconstructed from the post-glacial transgression through the sea-level highstand, a total of 9000 years. At the beginning of the marine transgressive stage (about 11,000 cal y BP), paleo-water depths were shallow and the sediment accumulation was large. The area occupied by Osaka Bay gradually extended and sediment deposition decreased from 11,000 to 6000 cal y BP. During the period of maximum transgression (6000 - 5000 cal y BP), an inner bay, Kawachi Bay with a water depth of 5 - 10 m, was expanded in the inland eastern Osaka area, and paleo-water depths reached a maximum and depositional rates reached a minimum. During the subsequent highstand and small regression (about 5000 cal y BP to the present), however, deposition increased rapidly as a result of river delta and shoreline progradations. Regional differences were observed in accommodation and accumulation between the outer bay area and the inner bay area. During both the transgressive and regressive stages, deposition decreased in the inner bay area. In contrast, in the outer bay area and in the basin overall, deposition was high during the first part of the transgressive stage but it decreased during the maximum transgression, before reaching a maximum during the subsequent highstand and regression. During the regressive stage, fluvial delta progradation led to the formation of a thick sequence of delta body sediments. Sediment accumulation was 30% - 40% higher during the regressive stage than that during the transgressive stage.展开更多
Next-generation sequencing(NGS) technologies have revolutionized the field of genomics and provided unprecedented opportunities for high-throughput analysis at the levels of genomics,transcriptomics and epigenetics.Ho...Next-generation sequencing(NGS) technologies have revolutionized the field of genomics and provided unprecedented opportunities for high-throughput analysis at the levels of genomics,transcriptomics and epigenetics.However,the cost of NGS is still prohibitive for many laboratories.It is imperative to address the trade-off between the sequencing depth and cost.In this review,we will discuss the effects of sequencing depth on the detection of genes,quantification of gene expression and discovering of gene structural variants.This will provide readers information on choosing appropriate sequencing depth that best meet the needs of their particular project.展开更多
多聚焦图像3维形貌重建旨在利用不同聚焦水平的图像序列恢复场景的3维结构信息.现有的3维形貌重建方法大多从单一尺度对图像序列的聚焦水平进行评价,通过引入正则化或后处理方法引导重建过程,由于深度信息选择空间的局限性往往导致重建...多聚焦图像3维形貌重建旨在利用不同聚焦水平的图像序列恢复场景的3维结构信息.现有的3维形貌重建方法大多从单一尺度对图像序列的聚焦水平进行评价,通过引入正则化或后处理方法引导重建过程,由于深度信息选择空间的局限性往往导致重建结果无法有效收敛.针对上述问题,提出一种多尺度代价聚合的多聚焦图像3维形貌重建框架(multi-scale cost aggregation framework for 3D shape reconstruction from multi-focus images,MSCAS),该框架首先引入非降采样的多尺度变换增加输入图像序列的深度信息选择空间,然后联合尺度内序列关联与尺度间信息约束进行代价聚合,通过这种扩张-聚合模式实现了场景深度表征信息的倍增与跨尺度和跨序列表征信息的有效融合.作为一种通用框架,MSCAS框架可实现已有模型设计类方法和深度学习类方法的嵌入进而实现性能提升.实验结果表明:MSCAS框架在嵌入模型设计类SFF方法后4组数据集中的均方根误差(root mean squared error,RMSE)平均下降14.91个百分点,结构相似度(structural similarity index measure,SSIM)平均提升56.69个百分点,嵌入深度学习类SFF方法后4组数据集中的RMSE平均下降1.55个百分点,SSIM平均提升1.61个百分点.验证了MSCAS框架的有效性和通用性.展开更多
Quantitative gene expression analysis plays an important role in identifying differentially expressed genes in various pathological states, gene expression regulation and co-regulation, shedding light on gene function...Quantitative gene expression analysis plays an important role in identifying differentially expressed genes in various pathological states, gene expression regulation and co-regulation, shedding light on gene functions. Although microarray is widely used as a powerful tool in this regard, it is suboptimal quantitatively and unable to detect unknown gene variants. Here we demonstrated effective detection of differential expression and co-regulation of certain genes by expressed sequence tag analysis using a selected subset of cDNA libraries. We discussed the issues of sequencing depth and library preparation, and propose that increased sequencing depth and improved preparation procedures may allow detection of many expression features for less abundant gene variants. With the reduction of sequencing cost and the emerging of new generation sequencing technology, in-depth sequencing of cDNA pools or libraries may represent a better and powerful tool in gene expression profiling and cancer biomarker detection. We also propose using sequence-specific subtraction to remove hundreds of the most abundant housekeeping genes to increase sequencing depth without affecting relative expression ratio of other genes, as transcripts from as few as 300 most abundantly expressed genes constitute about 20% of the total transcriptome. In-depth sequencing also represents a unique advantage of detecting unknown forms of transcripts, such as alternative splicing variants, fusion genes, and regulatory RNAs, as well as detecting mutations and polymorphisms that may play important roles in disease pathogenesis.展开更多
基金funded by the “Three-in-one Task of Monitoring,Prediction and Research” of China Earthquake Administration(CEA-JC/3JH-1605031)Major Science and Technology Projects in Inner Mongolia Autonomous Region:Research,Development,Popularization and Demonstration of Earthquake Prediction and Early Warning Technology in Key Areas
文摘Using the double-difference earthquake location algorithm,the deterministic method (PTD method) and the CAP seismic moment tensor inversion method,the paper selects the primary waveform data of 78 earthquakes recorded by the "China Earthquake Science Array Probe Project in the Northern Part of North South Seismic Belt ",the "China Earthquake Scientific Exploration Array Data Center"of Institute of Geophysics,China Earthquake Administration,and the Inner Mongolia Digital Seismic Network to calculate the focal depths of the mainshock and the seismic sequence of the M_S5.8 Alxa Left Banner earthquake in Inner Mongolia. The results show that the focal depth of the main shock is 20. 6 km,determined by the double-difference earthquake location method,18. 1 km by the PTD method,and 19. 2 km by the CAP method. The focal depth of the earthquake sequence calculated by the double-difference location method is larger. The deterministic method (PTD method) and double-difference location method are the methods that fit the tectonic characteristics of the seismic source area,and the CAP method is suitable for larger earthquakes.
基金the Science and Technology Innovation Project of Ministry of Culture of China(No.2014KJCXXM08)the National Key Technology Research and Development Program of the Ministry of Science and Technology of China(No.2012BAH37F02)the National High Technology Research and Development Program(863)of China(No.2011AA01A107)
文摘In the paper, an approach is proposed for the problem of consistency in depth maps estimation from binocular stereo video sequence. The consistent method includes temporal consistency and spatial consistency to eliminate the flickering artifacts and smooth inaccuracy in depth recovery. So the improved global stereo matching based on graph cut and energy optimization is implemented. In temporal domain, the penalty function with coherence factor is introduced for temporal consistency, and the factor is determined by Lucas-Kanade optical flow weighted histogram similarity constraint(LKWHSC). In spatial domain, the joint bilateral truncated absolute difference(JBTAD) is proposed for segmentation smoothing. The method can smooth naturally and uniformly in low-gradient region and avoid over-smoothing as well as keep edge sharpness in high-gradient discontinuities to realize spatial consistency. The experimental results show that the algorithm can obtain better spatial and temporal consistent depth maps compared with the existing algorithms.
文摘In this paper, based on the results of tomographic image of Tangshan and Xingtai areas, the relations between thecharacteristics of the two strong earthquake sequences and their three-dimensional velocity structures are studied.The research results indicate that:① Mosaic distribution of low-velocity bodies and high-velocity bodies, especially the existence of high-velocity bodies with large size in crust are the common basis of development of thetwo earthquake sequences. ② Scale, depth, and heterogeneity of high-velocity and low-velocity bodies are theimportant factors to effect the characteristic of earthquake sequences. ③ The depth of the high-velocity body inTangshan area is less than that in Xingtai area, which is the principal reason why the dominant focal depth and thebiggest focal depth of Tangshan earthquake sequence are less than Xingtai's. ④ The depth of the high-velocitybodies in Ninghe area is more than that in Tangshan-Luanxian area, which lead to the biggest magnitude and epicentral intensity are lower. These results could be helpful for predicting the main shock of strong swarm-typeearthquakes and later strong aftershocks.
基金Joint Seismological Science Foundation of China (105076) and continued subject ″Statistic Features of Aftershock Sequences and Forecast of the Large Aftershocks″ (2004BA601B01-04-02), Ministry of Science and Technology of China in the 10th Five-Year Plan.
文摘Based on 294 earthquake sequences with magnitude greater than or equal to 5.0 occurred in Chinese mainland since 1970, the spatial distribution features of sequence types have been studied. In southwestern China, it takes mainshock-aftershock sequence type （MAT） as the major in Chuan-Dian rhombic block and concerned Xianshuihe-Anninghe-Xiaojiang seismic belt, as well as in Jinshajiang-Honghe seismic belt. Multiple mainshock type （MMT） mainly distributes in western Yunnan, and Longlin and Lancang areas in Tengchong-Baoshan block in west of Nujiang-Lancangjiang fault zone. A few isolated earthquake type （IET） mainly occurred in northwestern Sichuan and there is no IET occurred in Yunnan region. In northwestern China, it takes mainshock-aftershock sequence type （MAT） as the major in west segment of South Tianshan in Xinjiang region. Some MMT also occurred in this area in the intersection of Kalpin block and the Puchang fault zone. It takes IET as the major in middle Tianshan in Xinjiang. Along the Qilianshan seismic belt, most of sequences are MAT. In Qinghai region, it takes MAT as the major, but the regional feature of the spatial distribution of sequence types is not very clear. In North China, it takes MAT as the major in Yinshan-Yanshan-Bohai seismic belt, north edge of North China, and in Hebei plain seismic belt, as well as in sub-plate of lower river area of Yangtze River. In intersection of north segment of Shanxi seismic belt and the NW-trending Yinshan-Yanshan-Bohai seismic belt, there are several moderate or strong MMT with magnitude from 5.0 to 6.0 occurred. In south of North China around the latitude line of 35°N, it takes IET as the major. The spatial distribution of sequence types is relevant to the patterns of tectonic movements. MAT is mostly produced by the ruptures of locked units or asperities or the neonatal separating segments inside the fault zones. MMT is generally relevant to the conjugate structures or intersection of many tectonic settings. Further extension of simple fault often produces IET. Spatial distribution of sequence types is also correlative to the regional and deep environment of crustal medium to some extent. MAT mainly distributes in high velocity area in upper crust or in the transition zone between high velocity area and low velocity area, MMT mostly occurred in the low velocity area in upper crust.
基金supported by the Fundamental Research Funds for the Central Universities(Grant No.2011YYL128)the CNPC Innovation Foundation(GrantNo.2012D-5006-0103)the Ministry of Land and Resources special funds for scientific research on public cause(Grant No.201311107)
文摘In order to solve the problems of the fine division of sedimentary sequence cycles and their change in two-dimensional space as well as lateral extension contrast, we developed a method of wavelet depth-frequency analysis. The single signal and composite signal of different Milankovitch cycles are obtained by numerical simulation. The simulated composite signal can be separated into single signals of a single frequency cycle. We also develop a well-seismic calibration insertion technology which helps to realize the calibration from the spectrum characteristics of a single well to the seismic profile. And then we determine the change and distribution characteristics of spectrum cycles in the two-dimensional space. It points out the direction in determining the variations of the regional sedimentary sequence cycles, underground strata structure and the contact relationship.
基金Project supported by the National Natural Science Foundation of China (Grant Nos. 61771103, 61704023, and 61601092)。
文摘The measuring of the depth profile and electrical activity of implantation impurity in the top nanometer range of silicon encounters various difficulties and limitations, though it is known to be critical in fabrication of silicon complementary metal–oxide–semiconductor(CMOS) devices. In the present work, SRIM program and photocarrier radiometry(PCR)are employed to monitor the boron implantation in industrial-grade silicon in an ultra-low implantation energy range from 0.5 keV to 5 keV. The differential PCR technique, which is improved by greatly shortening the measurement time through the simplification of reference sample, is used to investigate the effects of implantation energy on the frequency behavior of the PCR signal for ultra-shallow junction. The transport parameters and thickness of shallow junction, extracted via multi-parameter fitting the dependence of differential PCR signal on modulation frequency to the corresponding theoretical model, well explain the energy dependence of PCR signal and further quantitatively characterize the recovery degree of structure damage induced by ion implantation and the electrical activation degree of impurities. The monitoring of nmlevel thickness and electronic properties exhibits high sensitivity and apparent monotonicity over the industrially relevant implantation energy range. The depth profiles of implantation boron in silicon with the typical electrical damage threshold(YED) of 5.3×10^(15)cm^(-3) are evaluated by the SRIM program, and the determined thickness values are consistent well with those extracted by the differential PCR. It is demonstrated that the SRIM and the PCR are both effective tools to characterize ultra-low energy ion implantation in silicon.
文摘In this paper,a test or alternative scheme for studying large earthquake sequences through the study of small earthquake sequences is suggested,and a small earthquake sequence,the Lima earthquake sequence for which analogue records have been turned into digital data,is used here.In order to provide the deep construction background and the spatial distribution of structure for generating earthquakes,the P-wave and S-wave layered velocity models in this area are obtained by using mine explosion and earthquake observed records; then,the hypocenter locations and focal depths of the Lima earthquake sequence are determined adopting the velocity models given above and using a location method with numerical properties for a microseismic monitoring network(Zhao et al.,1994)and a new method for determining focal depth from data of a local seismographic network(Zhao,1992); finally,based on this,the variation of quality factor Q of the crustal medium during the period of the sequence is estimated.The obtained
文摘Temporal and spatial variations in accommodation (i.e., paleo-water depth) and sediment accumulation (amount of deposition) in the intra-arc Osaka Basin, Japan, were reconstructed from the post-glacial transgression through the sea-level highstand, a total of 9000 years. At the beginning of the marine transgressive stage (about 11,000 cal y BP), paleo-water depths were shallow and the sediment accumulation was large. The area occupied by Osaka Bay gradually extended and sediment deposition decreased from 11,000 to 6000 cal y BP. During the period of maximum transgression (6000 - 5000 cal y BP), an inner bay, Kawachi Bay with a water depth of 5 - 10 m, was expanded in the inland eastern Osaka area, and paleo-water depths reached a maximum and depositional rates reached a minimum. During the subsequent highstand and small regression (about 5000 cal y BP to the present), however, deposition increased rapidly as a result of river delta and shoreline progradations. Regional differences were observed in accommodation and accumulation between the outer bay area and the inner bay area. During both the transgressive and regressive stages, deposition decreased in the inner bay area. In contrast, in the outer bay area and in the basin overall, deposition was high during the first part of the transgressive stage but it decreased during the maximum transgression, before reaching a maximum during the subsequent highstand and regression. During the regressive stage, fluvial delta progradation led to the formation of a thick sequence of delta body sediments. Sediment accumulation was 30% - 40% higher during the regressive stage than that during the transgressive stage.
文摘Next-generation sequencing(NGS) technologies have revolutionized the field of genomics and provided unprecedented opportunities for high-throughput analysis at the levels of genomics,transcriptomics and epigenetics.However,the cost of NGS is still prohibitive for many laboratories.It is imperative to address the trade-off between the sequencing depth and cost.In this review,we will discuss the effects of sequencing depth on the detection of genes,quantification of gene expression and discovering of gene structural variants.This will provide readers information on choosing appropriate sequencing depth that best meet the needs of their particular project.
文摘多聚焦图像3维形貌重建旨在利用不同聚焦水平的图像序列恢复场景的3维结构信息.现有的3维形貌重建方法大多从单一尺度对图像序列的聚焦水平进行评价,通过引入正则化或后处理方法引导重建过程,由于深度信息选择空间的局限性往往导致重建结果无法有效收敛.针对上述问题,提出一种多尺度代价聚合的多聚焦图像3维形貌重建框架(multi-scale cost aggregation framework for 3D shape reconstruction from multi-focus images,MSCAS),该框架首先引入非降采样的多尺度变换增加输入图像序列的深度信息选择空间,然后联合尺度内序列关联与尺度间信息约束进行代价聚合,通过这种扩张-聚合模式实现了场景深度表征信息的倍增与跨尺度和跨序列表征信息的有效融合.作为一种通用框架,MSCAS框架可实现已有模型设计类方法和深度学习类方法的嵌入进而实现性能提升.实验结果表明:MSCAS框架在嵌入模型设计类SFF方法后4组数据集中的均方根误差(root mean squared error,RMSE)平均下降14.91个百分点,结构相似度(structural similarity index measure,SSIM)平均提升56.69个百分点,嵌入深度学习类SFF方法后4组数据集中的RMSE平均下降1.55个百分点,SSIM平均提升1.61个百分点.验证了MSCAS框架的有效性和通用性.
文摘Quantitative gene expression analysis plays an important role in identifying differentially expressed genes in various pathological states, gene expression regulation and co-regulation, shedding light on gene functions. Although microarray is widely used as a powerful tool in this regard, it is suboptimal quantitatively and unable to detect unknown gene variants. Here we demonstrated effective detection of differential expression and co-regulation of certain genes by expressed sequence tag analysis using a selected subset of cDNA libraries. We discussed the issues of sequencing depth and library preparation, and propose that increased sequencing depth and improved preparation procedures may allow detection of many expression features for less abundant gene variants. With the reduction of sequencing cost and the emerging of new generation sequencing technology, in-depth sequencing of cDNA pools or libraries may represent a better and powerful tool in gene expression profiling and cancer biomarker detection. We also propose using sequence-specific subtraction to remove hundreds of the most abundant housekeeping genes to increase sequencing depth without affecting relative expression ratio of other genes, as transcripts from as few as 300 most abundantly expressed genes constitute about 20% of the total transcriptome. In-depth sequencing also represents a unique advantage of detecting unknown forms of transcripts, such as alternative splicing variants, fusion genes, and regulatory RNAs, as well as detecting mutations and polymorphisms that may play important roles in disease pathogenesis.
