Unmanned aerial vehicle(UAV)photography has become the main power system inspection method;however,automated fault detection remains a major challenge.Conventional algorithms encounter difficulty in processing all the...Unmanned aerial vehicle(UAV)photography has become the main power system inspection method;however,automated fault detection remains a major challenge.Conventional algorithms encounter difficulty in processing all the detected objects in the power transmission lines simultaneously.The object detection method involving deep learning provides a new method for fault detection.However,the traditional non-maximum suppression(NMS)algorithm fails to delete redundant annotations when dealing with objects having two labels such as insulators and dampers.In this study,we propose an area-based non-maximum suppression(A-NMS)algorithm to solve the problem of one object having multiple labels.The A-NMS algorithm is used in the fusion stage of cropping detection to detect small objects.Experiments prove that A-NMS and cropping detection achieve a mean average precision and recall of 88.58%and 91.23%,respectively,in case of the aerial image datasets and realize multi-object fault detection in aerial images.展开更多
When the attributes of unknown targets are not just numerical attributes,but hybrid attributes containing linguistic attributes,the existing recognition methods are not effective.In addition,it is more difficult to id...When the attributes of unknown targets are not just numerical attributes,but hybrid attributes containing linguistic attributes,the existing recognition methods are not effective.In addition,it is more difficult to identify the unknown targets densely distributed in the feature space,especially when there is interval overlap between attribute measurements of different target classes.To address these problems,a novel method based on intuitionistic fuzzy comprehensive evaluation model(IFCEM)is proposed.For numerical attributes,targets in the database are divided into individual classes and overlapping classes,and for linguistic attributes,continuous interval-valued linguistic term set(CIVLTS)is used to describe target characteristic.A cloud modelbased method and an area-based method are proposed to obtain intuitionistic fuzzy decision information of query target on numerical attributes and linguistic attributes respectively.An improved inverse weighted kernel fuzzy c-means(IWK-FCM)algorithm is proposed for solution of attribute weight vector.The possibility matrix is applied to determine the identity and category of query target.Finally,a case study composed of parameter sensitivity analysis,recognition accuracy analysis.and comparison with other methods,is taken to verify the superiority of the proposed method.展开更多
Background: Tree species recognition is the main bottleneck in remote sensing based inventories aiming to produce an input for species-specific growth and yield models. We hypothesized that a stratification of the ta...Background: Tree species recognition is the main bottleneck in remote sensing based inventories aiming to produce an input for species-specific growth and yield models. We hypothesized that a stratification of the target data according to the dominant species could improve the subsequent predictions of species-specific attributes in particular in study areas strongly dominated by certain species. Methods: We tested this hypothesis and an operational potential to improve the predictions of timber volumes, stratified to Scots pine, Norway spruce and deciduous trees, in a conifer forest dominated by the pine species. We derived predictor features from airborne laser scanning (ALS) data and used Most Similar Neighbor (MSN) and Seemingly Unrelated Regression (SUR) as examples of non-parametric and parametric prediction methods, respectively Results: The relationships between the ALS features and the volumes of the aforementioned species were considerably different depending on the dominant species. Incorporating the observed dominant species inthe predictions improved the root mean squared errors by 13.3-16.4 % and 12.6-28.9 % based on MSN and SUR, respectively, depending on the species. Predicting the dominant species based on a linear discriminant analysis had an overall accuracy of only 76 % at best, which degraded the accuracies of the predicted volumes. Consequently, the predictions that did not consider the dominant species were more accurate than those refined with the predicted species. The MSN method gave slightly better results than models fitted with SUR. Conclusions: According to our results, incorporating information on the dominant species has a clear potential to improve the subsequent predictions of species-specific forest attributes. Determining the dominant species based solely on ALS data is deemed challenging, but important in particular in areas where the species composition is otherwise seemingly homogeneous except being dominated by certain species.展开更多
In this study,310 destructively sampled plots were used to develop two equation systems for the three main pine species in NW Spain(P.pinaster;P.radiata and P.sylvestris):one for estimating loads of understorey fuel c...In this study,310 destructively sampled plots were used to develop two equation systems for the three main pine species in NW Spain(P.pinaster;P.radiata and P.sylvestris):one for estimating loads of understorey fuel components by size and condition(live and dead)and another one for forest floor fuels.Additive systems of equations were simultaneously fitted for estimating fuel loads using overstorey,understorey and forest floor variables as regressors.The systems of equations included both the effect of pine species and the effect of understorey compositions dominated by ferns-brambles or by woody species,due to their obvious structural and physiological differences.In general,the goodness-of-fit statistics indicated that the estimates were reasonably robust and accurate for all of the fuel fractions.The best results were obtained for total understorey vegetation,total forest floor and raw humus fuel loads,with more than 76%of the observed variability explained,whereas the poorest results were obtained for coarse fuel loads of understory vegetation with a 53%of observed variability explained.To reduce the overall costs associated with the field inventories necessary for operational use of the models,the additive systems were fitted again using only overstorey variables as potential regressors.Only relationships for fine(<6 mm)and total understorey vegetation and total forest floor fuel loads were obtained,indicating the complexity of the forest overstorey-understorey and overstorey-forest floor relationships.Nevertheless,these models explained around 52%of the observed variability.Finally,equations estimating the total understorey vegetation and the total forest floor fuel loads based only on canopy cover were fitted.These models explained only 26%-32%of the observed variability;however,their main advantage is that although understorey vegetation in forested landscapes is largely invisible to remote sensing,canopy cover can be estimated with moderate accuracy,allowing for landscape-scale estimates of total fuel loads.The equations represent an appreciable advance in understorey and forest floor fuel load assessment in the region and areas with similar characteristics and may be instrumental in generating fuel maps,fire management improvement and better C storage assessment by vegetation type,among many other uses.展开更多
Objective: To identify the patterns of tuberculosis (TB) notification rates in Phnom Penh and examine their relationships with the population density, socioeconomic, residential and occupational characteristics. Metho...Objective: To identify the patterns of tuberculosis (TB) notification rates in Phnom Penh and examine their relationships with the population density, socioeconomic, residential and occupational characteristics. Methods: The numbers of total TB and smear-positive pulmonary TB cases reported between January 1, 2010 and December 31, 2012 in Phnom Penh were counted for 76 communes in Cambodia according to TB registration records filed under the national TB programme. Population, socioeconomic, residential and occupational characteristics for the communes were obtained from the 2008 General Population Census of Cambodia. The following indicators were developed for individual communes: smear-positive pulmonary TB notification rate (SPTB-NR) (per 100,000 population, in 36 months), population density (per km2), socioeconomic indicators, residential characteristics and occupational characteristics. Geographic patterns of these indicators and characteristics were analysed using ArcGIS. Associations between SPTB-NR and characteristics were analysed. Results: A total of 4102 TB cases were reported in 36 months, including 2046 SPTB cases. SPTB-NR for Phnom Penh was 135 cases per 100,000;median SPTB-NR by commune was 100. SPTB-NR was higher in outlying areas than in city centre communes;population density was high in the centre and low in the outlying areas. SPTB-NR was associated with larger percentage of household members per room (PR: 2.81, 95%CI: 2.68 - 2.93), percentage of population resident in the same commune Conclusions: The SPTB-NR in Phnom Penh did not follow the pattern of population density. Socioeconomic, residential and occupational characteristics by commune were associated with SPTB-NR. Development of prevention and control programmes by considering commune level characteristics is encouraged.展开更多
Background:The human immunodeficiency virus/acquired immunodeficiency syndrome(HIV/AIDS)epidemic is a typical global health concern.The impact of HIV/AIDS is global,and we cannot effectively solve the problem without ...Background:The human immunodeficiency virus/acquired immunodeficiency syndrome(HIV/AIDS)epidemic is a typical global health concern.The impact of HIV/AIDS is global,and we cannot effectively solve the problem without a global effort.In this study,we report our research on global HIV/AIDS control with an innovative fourdimensional approach.Methods:Countries(n=148)with data available on area size,total population,and the total number of persons living with HIV(PLWH)were included.The HIV epidemic across the globe was described using 4 indicators,including the total count,population-based P rate,geographic area-based G rate,and population and geographic area-based PG rate.Results:A total of 35,426,911 PLWH were included,with a global prevalence rate of 0.51 per 1,000 population.The total PLWH count provided data on resource allocation in individual countries to improve HIV/AIDS care;and the top five countries with the highest PLWH counts were South Africa(7,000),Nigeria(3,500),India(2,100),Kenya(1,500),and Mozambique(1,500).The other three indicators provide a measure to assess the global risk profile of HIV transmission and to provide information on HIV/AIDS prevention strategies.The top five countries with the highest P rates(per 1,000 persons)were Swaziland(170.9),Botswana(154.7),Lesotho(145.2),South Africa(127.4),and Zimbabwe(89.7);the top five countries with the highest G rates(per 100 km2)were Swaziland(1,279.1),Malawi(1,039.5),Lesotho(1,021.1),Rwanda(810.7),and Uganda(748.1);and the top five countries with highest PG rates(per 1,000,000 persons per 100 km2)were Barbados(2,127.9),Swaziland(993.8),Lesotho(478.3),Malta(375.0),and Mauritius(319.7).With PG rate,we detected countries in two hotspots(south and middle Africa and the Caribbean region)and one belt across the Euro-Asian region with high risks of HIV transmission.Conclusions:This study expanded the conventional measures by adding two new indicators,thus forming a new four-dimensional framework to quantify the global HIV epidemic.In addition to gaining a better insight into the epidemic than before,study findings provide new data on country-level and global efforts to end the AIDS epidemic by 2030.展开更多
基金the National Grid Corporation Headquarters Science and Technology Project:Key Technology Research,Equipment Development and Engineering Demonstration of Artificial Smart Drived Electric Vehicle Smart Travel Service(No.52020118000G).
文摘Unmanned aerial vehicle(UAV)photography has become the main power system inspection method;however,automated fault detection remains a major challenge.Conventional algorithms encounter difficulty in processing all the detected objects in the power transmission lines simultaneously.The object detection method involving deep learning provides a new method for fault detection.However,the traditional non-maximum suppression(NMS)algorithm fails to delete redundant annotations when dealing with objects having two labels such as insulators and dampers.In this study,we propose an area-based non-maximum suppression(A-NMS)algorithm to solve the problem of one object having multiple labels.The A-NMS algorithm is used in the fusion stage of cropping detection to detect small objects.Experiments prove that A-NMS and cropping detection achieve a mean average precision and recall of 88.58%and 91.23%,respectively,in case of the aerial image datasets and realize multi-object fault detection in aerial images.
基金supported by the Youth Foundation of the National Science Foundation of China(62001503)the Excellent Youth Scholar of the National Defense Science and Technology Foundation of China(2017-JCJQ-ZQ-003)the Special Fund for Taishan Scholar Project(ts201712072).
文摘When the attributes of unknown targets are not just numerical attributes,but hybrid attributes containing linguistic attributes,the existing recognition methods are not effective.In addition,it is more difficult to identify the unknown targets densely distributed in the feature space,especially when there is interval overlap between attribute measurements of different target classes.To address these problems,a novel method based on intuitionistic fuzzy comprehensive evaluation model(IFCEM)is proposed.For numerical attributes,targets in the database are divided into individual classes and overlapping classes,and for linguistic attributes,continuous interval-valued linguistic term set(CIVLTS)is used to describe target characteristic.A cloud modelbased method and an area-based method are proposed to obtain intuitionistic fuzzy decision information of query target on numerical attributes and linguistic attributes respectively.An improved inverse weighted kernel fuzzy c-means(IWK-FCM)algorithm is proposed for solution of attribute weight vector.The possibility matrix is applied to determine the identity and category of query target.Finally,a case study composed of parameter sensitivity analysis,recognition accuracy analysis.and comparison with other methods,is taken to verify the superiority of the proposed method.
基金financed by the Finnish Funding Agency for Innovation(Tekes) and its business and research partners
文摘Background: Tree species recognition is the main bottleneck in remote sensing based inventories aiming to produce an input for species-specific growth and yield models. We hypothesized that a stratification of the target data according to the dominant species could improve the subsequent predictions of species-specific attributes in particular in study areas strongly dominated by certain species. Methods: We tested this hypothesis and an operational potential to improve the predictions of timber volumes, stratified to Scots pine, Norway spruce and deciduous trees, in a conifer forest dominated by the pine species. We derived predictor features from airborne laser scanning (ALS) data and used Most Similar Neighbor (MSN) and Seemingly Unrelated Regression (SUR) as examples of non-parametric and parametric prediction methods, respectively Results: The relationships between the ALS features and the volumes of the aforementioned species were considerably different depending on the dominant species. Incorporating the observed dominant species inthe predictions improved the root mean squared errors by 13.3-16.4 % and 12.6-28.9 % based on MSN and SUR, respectively, depending on the species. Predicting the dominant species based on a linear discriminant analysis had an overall accuracy of only 76 % at best, which degraded the accuracies of the predicted volumes. Consequently, the predictions that did not consider the dominant species were more accurate than those refined with the predicted species. The MSN method gave slightly better results than models fitted with SUR. Conclusions: According to our results, incorporating information on the dominant species has a clear potential to improve the subsequent predictions of species-specific forest attributes. Determining the dominant species based solely on ALS data is deemed challenging, but important in particular in areas where the species composition is otherwise seemingly homogeneous except being dominated by certain species.
基金funded by following projects:INIA p5608,INIA p7613,INIA p8038,INIA 9130 and INIA SC96-034 of the Sectorial Research Program of the INIA(Spanish National Institute of Agrarian Research,Ministry of Agriculture),INIA-RTA 2009-00153-C03(INFOCOPAS),INIA-RTA 2014-00011-C06(GEPRIF)and INIA-RTA2017-00042-C05(VIS4FIRE)of the Spanish National Program of Research,Development and Innovation co-funded by the ERDF Program of the European Unionby project CTYO-0087 of the Science and Technology for Environmental Protection Program and projects ENV5V-CT94-0473,ENV4CT98-0701(SALTUS),ENV-CT97-0715(FIRE TORCH),EVG1-CT200100041(FIRESTAR),EVR1-CT-2002-4002(EUFIRELAB)and CTFP6018505(FIRE PARADOX)+1 种基金funded by the Environment Program of the Directorate-General for Research and Innovation,of the European Commission of the European Unionby project PGIDITOSRF050202PR of the Xunta de Galicia。
文摘In this study,310 destructively sampled plots were used to develop two equation systems for the three main pine species in NW Spain(P.pinaster;P.radiata and P.sylvestris):one for estimating loads of understorey fuel components by size and condition(live and dead)and another one for forest floor fuels.Additive systems of equations were simultaneously fitted for estimating fuel loads using overstorey,understorey and forest floor variables as regressors.The systems of equations included both the effect of pine species and the effect of understorey compositions dominated by ferns-brambles or by woody species,due to their obvious structural and physiological differences.In general,the goodness-of-fit statistics indicated that the estimates were reasonably robust and accurate for all of the fuel fractions.The best results were obtained for total understorey vegetation,total forest floor and raw humus fuel loads,with more than 76%of the observed variability explained,whereas the poorest results were obtained for coarse fuel loads of understory vegetation with a 53%of observed variability explained.To reduce the overall costs associated with the field inventories necessary for operational use of the models,the additive systems were fitted again using only overstorey variables as potential regressors.Only relationships for fine(<6 mm)and total understorey vegetation and total forest floor fuel loads were obtained,indicating the complexity of the forest overstorey-understorey and overstorey-forest floor relationships.Nevertheless,these models explained around 52%of the observed variability.Finally,equations estimating the total understorey vegetation and the total forest floor fuel loads based only on canopy cover were fitted.These models explained only 26%-32%of the observed variability;however,their main advantage is that although understorey vegetation in forested landscapes is largely invisible to remote sensing,canopy cover can be estimated with moderate accuracy,allowing for landscape-scale estimates of total fuel loads.The equations represent an appreciable advance in understorey and forest floor fuel load assessment in the region and areas with similar characteristics and may be instrumental in generating fuel maps,fire management improvement and better C storage assessment by vegetation type,among many other uses.
文摘Objective: To identify the patterns of tuberculosis (TB) notification rates in Phnom Penh and examine their relationships with the population density, socioeconomic, residential and occupational characteristics. Methods: The numbers of total TB and smear-positive pulmonary TB cases reported between January 1, 2010 and December 31, 2012 in Phnom Penh were counted for 76 communes in Cambodia according to TB registration records filed under the national TB programme. Population, socioeconomic, residential and occupational characteristics for the communes were obtained from the 2008 General Population Census of Cambodia. The following indicators were developed for individual communes: smear-positive pulmonary TB notification rate (SPTB-NR) (per 100,000 population, in 36 months), population density (per km2), socioeconomic indicators, residential characteristics and occupational characteristics. Geographic patterns of these indicators and characteristics were analysed using ArcGIS. Associations between SPTB-NR and characteristics were analysed. Results: A total of 4102 TB cases were reported in 36 months, including 2046 SPTB cases. SPTB-NR for Phnom Penh was 135 cases per 100,000;median SPTB-NR by commune was 100. SPTB-NR was higher in outlying areas than in city centre communes;population density was high in the centre and low in the outlying areas. SPTB-NR was associated with larger percentage of household members per room (PR: 2.81, 95%CI: 2.68 - 2.93), percentage of population resident in the same commune Conclusions: The SPTB-NR in Phnom Penh did not follow the pattern of population density. Socioeconomic, residential and occupational characteristics by commune were associated with SPTB-NR. Development of prevention and control programmes by considering commune level characteristics is encouraged.
文摘Background:The human immunodeficiency virus/acquired immunodeficiency syndrome(HIV/AIDS)epidemic is a typical global health concern.The impact of HIV/AIDS is global,and we cannot effectively solve the problem without a global effort.In this study,we report our research on global HIV/AIDS control with an innovative fourdimensional approach.Methods:Countries(n=148)with data available on area size,total population,and the total number of persons living with HIV(PLWH)were included.The HIV epidemic across the globe was described using 4 indicators,including the total count,population-based P rate,geographic area-based G rate,and population and geographic area-based PG rate.Results:A total of 35,426,911 PLWH were included,with a global prevalence rate of 0.51 per 1,000 population.The total PLWH count provided data on resource allocation in individual countries to improve HIV/AIDS care;and the top five countries with the highest PLWH counts were South Africa(7,000),Nigeria(3,500),India(2,100),Kenya(1,500),and Mozambique(1,500).The other three indicators provide a measure to assess the global risk profile of HIV transmission and to provide information on HIV/AIDS prevention strategies.The top five countries with the highest P rates(per 1,000 persons)were Swaziland(170.9),Botswana(154.7),Lesotho(145.2),South Africa(127.4),and Zimbabwe(89.7);the top five countries with the highest G rates(per 100 km2)were Swaziland(1,279.1),Malawi(1,039.5),Lesotho(1,021.1),Rwanda(810.7),and Uganda(748.1);and the top five countries with highest PG rates(per 1,000,000 persons per 100 km2)were Barbados(2,127.9),Swaziland(993.8),Lesotho(478.3),Malta(375.0),and Mauritius(319.7).With PG rate,we detected countries in two hotspots(south and middle Africa and the Caribbean region)and one belt across the Euro-Asian region with high risks of HIV transmission.Conclusions:This study expanded the conventional measures by adding two new indicators,thus forming a new four-dimensional framework to quantify the global HIV epidemic.In addition to gaining a better insight into the epidemic than before,study findings provide new data on country-level and global efforts to end the AIDS epidemic by 2030.