AIM:To analyze ultrasound biomicroscopy(UBM)images using random forest network to find new features to make predictions about vault after implantable collamer lens(ICL)implantation.METHODS:A total of 450 UBM images we...AIM:To analyze ultrasound biomicroscopy(UBM)images using random forest network to find new features to make predictions about vault after implantable collamer lens(ICL)implantation.METHODS:A total of 450 UBM images were collected from the Lixiang Eye Hospital to provide the patient’s preoperative parameters as well as the vault of the ICL after implantation.The vault was set as the prediction target,and the input elements were mainly ciliary sulcus shape parameters,which included 6 angular parameters,2 area parameters,and 2 parameters,distance between ciliary sulci,and anterior chamber height.A random forest regression model was applied to predict the vault,with the number of base estimators(n_estimators)of 2000,the maximum tree depth(max_depth)of 17,the number of tree features(max_features)of Auto,and the random state(random_state)of 40.0.RESULTS:Among the parameters selected in this study,the distance between ciliary sulci had a greater importance proportion,reaching 52%before parameter optimization is performed,and other features had less influence,with an importance proportion of about 5%.The importance of the distance between the ciliary sulci increased to 53% after parameter optimization,and the importance of angle 3 and area 1 increased to 5% and 8%respectively,while the importance of the other parameters remained unchanged,and the distance between the ciliary sulci was considered the most important feature.Other features,although they accounted for a relatively small proportion,also had an impact on the vault prediction.After parameter optimization,the best prediction results were obtained,with a predicted mean value of 763.688μm and an actual mean value of 776.9304μm.The R²was 0.4456 and the root mean square error was 201.5166.CONCLUSION:A study based on UBM images using random forest network can be performed for prediction of the vault after ICL implantation and can provide some reference for ICL size selection.展开更多
We investigated forest road networks and forestry operations before and after mechanization on aggregated forestry operation sites. We developed equations to estimate densities of road networks with average slope angl...We investigated forest road networks and forestry operations before and after mechanization on aggregated forestry operation sites. We developed equations to estimate densities of road networks with average slope angles, operational efficiency of bunching operations with road network density, and average forwarding distances with operation site areas. Subsequently, we analyzed the effects of aggregating forests, establishing forest road networks, and mechanization on operational efficiency and costs. Six ha proved to be an appropriate operation site area with minimum operation expenses. The operation site areas of the forest owners＇ cooperative in this region aggregated approximately 6 ha and the cooperative conducted forestry operations on aggregated sites. Therefore, 6 ha would be an appropriate operation site area in this region. Regarding road network density, higher-density road networks increased operational expenses due to the higher direct operational expenses of strip road establishment. Therefore, road network density should be reduced to approximately 200 m.展开更多
In this paper, sixty-eight research articles published between 2000 and 2017 as well as textbooks which employed four classification algorithms: K-Nearest-Neighbor (KNN), Support Vector Machines (SVM), Random Forest (...In this paper, sixty-eight research articles published between 2000 and 2017 as well as textbooks which employed four classification algorithms: K-Nearest-Neighbor (KNN), Support Vector Machines (SVM), Random Forest (RF) and Neural Network (NN) as the main statistical tools were reviewed. The aim was to examine and compare these nonparametric classification methods on the following attributes: robustness to training data, sensitivity to changes, data fitting, stability, ability to handle large data sizes, sensitivity to noise, time invested in parameter tuning, and accuracy. The performances, strengths and shortcomings of each of the algorithms were examined, and finally, a conclusion was arrived at on which one has higher performance. It was evident from the literature reviewed that RF is too sensitive to small changes in the training dataset and is occasionally unstable and tends to overfit in the model. KNN is easy to implement and understand but has a major drawback of becoming significantly slow as the size of the data in use grows, while the ideal value of K for the KNN classifier is difficult to set. SVM and RF are insensitive to noise or overtraining, which shows their ability in dealing with unbalanced data. Larger input datasets will lengthen classification times for NN and KNN more than for SVM and RF. Among these nonparametric classification methods, NN has the potential to become a more widely used classification algorithm, but because of their time-consuming parameter tuning procedure, high level of complexity in computational processing, the numerous types of NN architectures to choose from and the high number of algorithms used for training, most researchers recommend SVM and RF as easier and wieldy used methods which repeatedly achieve results with high accuracies and are often faster to implement.展开更多
The process of transformation of rainfall into runoff over a catchment is very complex and highly nonlinear and exhibits both tempor al and spatial variabilities. In this article, a rainfall-runoff model using th e ar...The process of transformation of rainfall into runoff over a catchment is very complex and highly nonlinear and exhibits both tempor al and spatial variabilities. In this article, a rainfall-runoff model using th e artificial neural networks (ANN) is proposed for simula ting the runoff in storm events. The study uses the data from a coa stal forest catchment located in Seto Inland Sea, Japan. This article studies the accuracy of the short-term rainfall forecast obta ined by ANN time-series analysis techniques and using antecedent rainfa ll depths and stream flow as the input information. The verification results from the proposed model indicate that the approach of ANN rai nfall-runoff model presented in this paper shows a reasonable agreement in rainfall-runoff modeling with high accuracy.展开更多
A forest fire is a severe threat to forest resources and human life, In this paper, we propose a forest-fire detection system that has an artificial neural network algorithm implemented in a wireless sensor network （...A forest fire is a severe threat to forest resources and human life, In this paper, we propose a forest-fire detection system that has an artificial neural network algorithm implemented in a wireless sensor network （WSN）. The proposed detection system mitigates the threat of forest fires by provide accurate fire alarm with low maintenance cost. The accuracy is increased by the novel multi- criteria detection, referred to as an alarm decision depends on multiple attributes of a forest fire. The multi-criteria detection is implemented by the artificial neural network algorithm. Meanwhile, we have developed a prototype of the proposed system consisting of the solar batter module, the fire detection module and the user interface module.展开更多
The construction of a pest detection algorithm is an important step to couple"ground-space"characteristics,which is also the basis for rapid and accurate monitoring and detection of pest damage.In four exper...The construction of a pest detection algorithm is an important step to couple"ground-space"characteristics,which is also the basis for rapid and accurate monitoring and detection of pest damage.In four experimental areas in Sanming City,Jiangle County,Sha County and Yanping District in Fujian Province,sample data on pest damage in 182 sets of Dendrolimus punctatus were collected.The data were randomly divided into a training set and testing set,and five duplicate tests and one eliminating-indicator test were done.Based on the characterization analysis of the host for D.punctatus damage,seven characteristic indicators of ground and remote sensing including leaf area index,standard error of leaf area index(SEL)of pine forest,normalized difference vegetation index(NDVI),wetness from tasseled cap transformation(WET),green band(B2),red band(B3),near-infrared band(B4)of remote sensing image are obtained to construct BP neural networks and random forest models of pest levels.The detection results of these two algorithms were comprehensively compared from the aspects of detection precision,kappa coefficient,receiver operating characteristic curve,and a paired t test.The results showed that the seven indicators all were responsive to pest damage,and NDVI was relatively weak;the average pest damage detection precision of six tests by BP neural networks was 77.29%,the kappa coefficient was 0.6869 and after the RF algorithm,the respective values were 79.30%and 0.7151,showing that the latter is more optimized,but there was no significant difference(p>0.05);the detection precision,kappa coefficient and AUC of the RF algorithm was higher than the BP neural networks for three pest levels(no damage,moderate damage and severe damage).The detection precision and AUC of BP neural networks were a little higher for mild damage,but the difference was not significant(p>0.05)except for the kappa coefficient for the no damage level(p<0.05).An"over-fitting"phenomenon tends to occur in BP neural networks,while RF method is more robust,providing a detection effect that is better than the BP neural networks.Thus,the application of the random forest algorithm for pest damage and multilevel dispersed variables is thus feasible and suggests that attention to the proportionality of sample data from various categories is needed when collecting data.展开更多
This paper proposes a wildfire monitoring and detection system based on wireless sensor network. This system detects fire by monitoring surrounding temperature, humidity and smoke. Once fire is detected, a warning mes...This paper proposes a wildfire monitoring and detection system based on wireless sensor network. This system detects fire by monitoring surrounding temperature, humidity and smoke. Once fire is detected, a warning message containing probable location of that fire is immediately sent to the responsible authority over cellular network. In order for the system to be more effective, communities living near forests or national parks can send warning messages through the same system to the responsible authority using their mobile handsets once they witness wildfire or illegal activities. For the system to be fully functional, the only requirement is the availability of cellular network coverage in forests or national parks to enable short message services to take place. The system prototype is developed using Arduino microcontroller, several sensors to detect temperature, relative humidity and smoke as well as wireless network connection modules. At the control center Telerivet messaging platform is used to design the messaging service. The experimental results justify the capability of the proposed system in detecting wildfire in real time.展开更多
The nitrate-nitrogen（NO 3-N） concentrations from shallow groundwater wells situated in 29 of the Chinese Ecosystem Research Network field stations,representing typical agroand forest ecosystems,were assessed using m...The nitrate-nitrogen（NO 3-N） concentrations from shallow groundwater wells situated in 29 of the Chinese Ecosystem Research Network field stations,representing typical agroand forest ecosystems,were assessed using monitoring data collected between 2004 and 2010.Results from this assessment permit a national scale assessment of nitrate concentrations in shallow groundwater,and allow linkages between nitrate concentrations in groundwater and broad land use categories to be made.Results indicated that most of the NO 3--N concentrations in groundwater from the agroand forest ecosystems were below the Class 3 drinking water standard stated in the Chinese National Standard：Quality Standard for Ground Water（≤ 20 mg/L）.Over the study period,the average NO 3--N concentrations were significantly higher in agro-ecosystems（4.1 ± 0.33 mg/L） than in forest ecosystems（0.5 ± 0.04 mg/L）.NO 3-N concentrations were relatively higher（〉 10 mg N /L） in 10 of the 43 wells sampled in the agricultural ecosystems.These elevated concentrations occurred mainly in the Ansai,Yucheng,Linze,Fukang,Akesu,and Cele field sites,which were located in arid and semiarid areas where irrigation rates are high.We suggest that improvements in N fertilizer application and irrigation management practices in the arid and semi-arid agricultural ecosystems of China are the key to managing groundwater nitrate concentrations.展开更多
One of the basic parameters in forest management planning is detailed knowledge of growing stock,information collected by forest inventory.Sampling methods must be accurate,inexpensive,and be easy to implement in the ...One of the basic parameters in forest management planning is detailed knowledge of growing stock,information collected by forest inventory.Sampling methods must be accurate,inexpensive,and be easy to implement in the field.This study presents a new sampling method called branching transect for use in the Iranian Zagros forests and similar forests.Features of the new method include greater accuracy,easy implementation in nature,simplicity of statistical calculations,and low cost.In this method,transect is used,which includes some subtransects(side branches).The length of the main transect,side branches,number of trees measured in each side branch,and the number of sub-branches in this method are changeable based on homogeneity,heterogeneity,and density of a forest.In this study,based on the density and heterogeneity of the forest area studied,20-m transects with four and eight side branches were used.Sampling plots(Transects)in four inventory networks(100 m×100 m,100 m×150 m,150 m×150 m and 100 m×200 m)were implemented in the GIS environment.The results of this sampling method were compared to the results of total inventory(100%count)in terms of accuracy,precision(t-test),and inventory error percentage.Branching transect results were statistially similar to total inventory counts in all cases.The results show that this method of estimating density and canopy per hectare can be used in Zagros forests and similar forests.展开更多
Considering the environmental protection, forest fire becomes a more and more serious problem and requires more concerns. This paper provides an efficient method for fire monitoring and detection in forests using wire...Considering the environmental protection, forest fire becomes a more and more serious problem and requires more concerns. This paper provides an efficient method for fire monitoring and detection in forests using wireless sensor network technology. The proposed technique estimates the location of a sensor node based on the current set of hop-count values, which are collected through the anchor nodes’ broadcast. Our algorithm incorporates two salient features;grid-based output and event-triggering mechanism, to improve the accuracy while reducing the power consumption. Through the computer simulation, the output region obtained from our algorithm can always cover the target node. In addition, the algorithm was implemented and tested with a set of Crossbow sensors. Experimental results demonstrated the high feasibility and worked well in real environment.展开更多
文摘AIM:To analyze ultrasound biomicroscopy(UBM)images using random forest network to find new features to make predictions about vault after implantable collamer lens(ICL)implantation.METHODS:A total of 450 UBM images were collected from the Lixiang Eye Hospital to provide the patient’s preoperative parameters as well as the vault of the ICL after implantation.The vault was set as the prediction target,and the input elements were mainly ciliary sulcus shape parameters,which included 6 angular parameters,2 area parameters,and 2 parameters,distance between ciliary sulci,and anterior chamber height.A random forest regression model was applied to predict the vault,with the number of base estimators(n_estimators)of 2000,the maximum tree depth(max_depth)of 17,the number of tree features(max_features)of Auto,and the random state(random_state)of 40.0.RESULTS:Among the parameters selected in this study,the distance between ciliary sulci had a greater importance proportion,reaching 52%before parameter optimization is performed,and other features had less influence,with an importance proportion of about 5%.The importance of the distance between the ciliary sulci increased to 53% after parameter optimization,and the importance of angle 3 and area 1 increased to 5% and 8%respectively,while the importance of the other parameters remained unchanged,and the distance between the ciliary sulci was considered the most important feature.Other features,although they accounted for a relatively small proportion,also had an impact on the vault prediction.After parameter optimization,the best prediction results were obtained,with a predicted mean value of 763.688μm and an actual mean value of 776.9304μm.The R²was 0.4456 and the root mean square error was 201.5166.CONCLUSION:A study based on UBM images using random forest network can be performed for prediction of the vault after ICL implantation and can provide some reference for ICL size selection.
文摘We investigated forest road networks and forestry operations before and after mechanization on aggregated forestry operation sites. We developed equations to estimate densities of road networks with average slope angles, operational efficiency of bunching operations with road network density, and average forwarding distances with operation site areas. Subsequently, we analyzed the effects of aggregating forests, establishing forest road networks, and mechanization on operational efficiency and costs. Six ha proved to be an appropriate operation site area with minimum operation expenses. The operation site areas of the forest owners＇ cooperative in this region aggregated approximately 6 ha and the cooperative conducted forestry operations on aggregated sites. Therefore, 6 ha would be an appropriate operation site area in this region. Regarding road network density, higher-density road networks increased operational expenses due to the higher direct operational expenses of strip road establishment. Therefore, road network density should be reduced to approximately 200 m.
文摘In this paper, sixty-eight research articles published between 2000 and 2017 as well as textbooks which employed four classification algorithms: K-Nearest-Neighbor (KNN), Support Vector Machines (SVM), Random Forest (RF) and Neural Network (NN) as the main statistical tools were reviewed. The aim was to examine and compare these nonparametric classification methods on the following attributes: robustness to training data, sensitivity to changes, data fitting, stability, ability to handle large data sizes, sensitivity to noise, time invested in parameter tuning, and accuracy. The performances, strengths and shortcomings of each of the algorithms were examined, and finally, a conclusion was arrived at on which one has higher performance. It was evident from the literature reviewed that RF is too sensitive to small changes in the training dataset and is occasionally unstable and tends to overfit in the model. KNN is easy to implement and understand but has a major drawback of becoming significantly slow as the size of the data in use grows, while the ideal value of K for the KNN classifier is difficult to set. SVM and RF are insensitive to noise or overtraining, which shows their ability in dealing with unbalanced data. Larger input datasets will lengthen classification times for NN and KNN more than for SVM and RF. Among these nonparametric classification methods, NN has the potential to become a more widely used classification algorithm, but because of their time-consuming parameter tuning procedure, high level of complexity in computational processing, the numerous types of NN architectures to choose from and the high number of algorithms used for training, most researchers recommend SVM and RF as easier and wieldy used methods which repeatedly achieve results with high accuracies and are often faster to implement.
文摘The process of transformation of rainfall into runoff over a catchment is very complex and highly nonlinear and exhibits both tempor al and spatial variabilities. In this article, a rainfall-runoff model using th e artificial neural networks (ANN) is proposed for simula ting the runoff in storm events. The study uses the data from a coa stal forest catchment located in Seto Inland Sea, Japan. This article studies the accuracy of the short-term rainfall forecast obta ined by ANN time-series analysis techniques and using antecedent rainfa ll depths and stream flow as the input information. The verification results from the proposed model indicate that the approach of ANN rai nfall-runoff model presented in this paper shows a reasonable agreement in rainfall-runoff modeling with high accuracy.
文摘A forest fire is a severe threat to forest resources and human life, In this paper, we propose a forest-fire detection system that has an artificial neural network algorithm implemented in a wireless sensor network （WSN）. The proposed detection system mitigates the threat of forest fires by provide accurate fire alarm with low maintenance cost. The accuracy is increased by the novel multi- criteria detection, referred to as an alarm decision depends on multiple attributes of a forest fire. The multi-criteria detection is implemented by the artificial neural network algorithm. Meanwhile, we have developed a prototype of the proposed system consisting of the solar batter module, the fire detection module and the user interface module.
基金supported by the National Natural Science Foundation of China(Grant Nos.41501361,41401385,30871965)the China Postdoctoral Science Foundation(No.2018M630728)+2 种基金the Open Fund of Fujian Provincial Key Laboratory of Resources and Environment Monitoring&Sustainable Management and Utilization(No.ZD1403)the Open Fund of Fujian Mine Ecological Restoration Engineering Technology Research Center(No.KS2018005)the Scientific Research Foundation of Fuzhou University(No.XRC1345)
文摘The construction of a pest detection algorithm is an important step to couple"ground-space"characteristics,which is also the basis for rapid and accurate monitoring and detection of pest damage.In four experimental areas in Sanming City,Jiangle County,Sha County and Yanping District in Fujian Province,sample data on pest damage in 182 sets of Dendrolimus punctatus were collected.The data were randomly divided into a training set and testing set,and five duplicate tests and one eliminating-indicator test were done.Based on the characterization analysis of the host for D.punctatus damage,seven characteristic indicators of ground and remote sensing including leaf area index,standard error of leaf area index(SEL)of pine forest,normalized difference vegetation index(NDVI),wetness from tasseled cap transformation(WET),green band(B2),red band(B3),near-infrared band(B4)of remote sensing image are obtained to construct BP neural networks and random forest models of pest levels.The detection results of these two algorithms were comprehensively compared from the aspects of detection precision,kappa coefficient,receiver operating characteristic curve,and a paired t test.The results showed that the seven indicators all were responsive to pest damage,and NDVI was relatively weak;the average pest damage detection precision of six tests by BP neural networks was 77.29%,the kappa coefficient was 0.6869 and after the RF algorithm,the respective values were 79.30%and 0.7151,showing that the latter is more optimized,but there was no significant difference(p>0.05);the detection precision,kappa coefficient and AUC of the RF algorithm was higher than the BP neural networks for three pest levels(no damage,moderate damage and severe damage).The detection precision and AUC of BP neural networks were a little higher for mild damage,but the difference was not significant(p>0.05)except for the kappa coefficient for the no damage level(p<0.05).An"over-fitting"phenomenon tends to occur in BP neural networks,while RF method is more robust,providing a detection effect that is better than the BP neural networks.Thus,the application of the random forest algorithm for pest damage and multilevel dispersed variables is thus feasible and suggests that attention to the proportionality of sample data from various categories is needed when collecting data.
文摘This paper proposes a wildfire monitoring and detection system based on wireless sensor network. This system detects fire by monitoring surrounding temperature, humidity and smoke. Once fire is detected, a warning message containing probable location of that fire is immediately sent to the responsible authority over cellular network. In order for the system to be more effective, communities living near forests or national parks can send warning messages through the same system to the responsible authority using their mobile handsets once they witness wildfire or illegal activities. For the system to be fully functional, the only requirement is the availability of cellular network coverage in forests or national parks to enable short message services to take place. The system prototype is developed using Arduino microcontroller, several sensors to detect temperature, relative humidity and smoke as well as wireless network connection modules. At the control center Telerivet messaging platform is used to design the messaging service. The experimental results justify the capability of the proposed system in detecting wildfire in real time.
基金supported by the Key Direction in Knowledge Innovation Program of Chinese Academy of Sciences (No. KZCX2-EW-310)the National Natural Science Foundation of China (No. 41171153)
文摘The nitrate-nitrogen（NO 3-N） concentrations from shallow groundwater wells situated in 29 of the Chinese Ecosystem Research Network field stations,representing typical agroand forest ecosystems,were assessed using monitoring data collected between 2004 and 2010.Results from this assessment permit a national scale assessment of nitrate concentrations in shallow groundwater,and allow linkages between nitrate concentrations in groundwater and broad land use categories to be made.Results indicated that most of the NO 3--N concentrations in groundwater from the agroand forest ecosystems were below the Class 3 drinking water standard stated in the Chinese National Standard：Quality Standard for Ground Water（≤ 20 mg/L）.Over the study period,the average NO 3--N concentrations were significantly higher in agro-ecosystems（4.1 ± 0.33 mg/L） than in forest ecosystems（0.5 ± 0.04 mg/L）.NO 3-N concentrations were relatively higher（〉 10 mg N /L） in 10 of the 43 wells sampled in the agricultural ecosystems.These elevated concentrations occurred mainly in the Ansai,Yucheng,Linze,Fukang,Akesu,and Cele field sites,which were located in arid and semiarid areas where irrigation rates are high.We suggest that improvements in N fertilizer application and irrigation management practices in the arid and semi-arid agricultural ecosystems of China are the key to managing groundwater nitrate concentrations.
文摘One of the basic parameters in forest management planning is detailed knowledge of growing stock,information collected by forest inventory.Sampling methods must be accurate,inexpensive,and be easy to implement in the field.This study presents a new sampling method called branching transect for use in the Iranian Zagros forests and similar forests.Features of the new method include greater accuracy,easy implementation in nature,simplicity of statistical calculations,and low cost.In this method,transect is used,which includes some subtransects(side branches).The length of the main transect,side branches,number of trees measured in each side branch,and the number of sub-branches in this method are changeable based on homogeneity,heterogeneity,and density of a forest.In this study,based on the density and heterogeneity of the forest area studied,20-m transects with four and eight side branches were used.Sampling plots(Transects)in four inventory networks(100 m×100 m,100 m×150 m,150 m×150 m and 100 m×200 m)were implemented in the GIS environment.The results of this sampling method were compared to the results of total inventory(100%count)in terms of accuracy,precision(t-test),and inventory error percentage.Branching transect results were statistially similar to total inventory counts in all cases.The results show that this method of estimating density and canopy per hectare can be used in Zagros forests and similar forests.
文摘Considering the environmental protection, forest fire becomes a more and more serious problem and requires more concerns. This paper provides an efficient method for fire monitoring and detection in forests using wireless sensor network technology. The proposed technique estimates the location of a sensor node based on the current set of hop-count values, which are collected through the anchor nodes’ broadcast. Our algorithm incorporates two salient features;grid-based output and event-triggering mechanism, to improve the accuracy while reducing the power consumption. Through the computer simulation, the output region obtained from our algorithm can always cover the target node. In addition, the algorithm was implemented and tested with a set of Crossbow sensors. Experimental results demonstrated the high feasibility and worked well in real environment.
