Two classification and identification methods based on pattern discrimination models and the majority-vote technique were investigated for implementing a World Wide Web-based system for the identification of rice dise...Two classification and identification methods based on pattern discrimination models and the majority-vote technique were investigated for implementing a World Wide Web-based system for the identification of rice diseases. The experiment was carried out using color and shape patterns in 425 images of three rice diseases, which were classified into four classes： two classes of leaf blast, and one class each of sheath blight and brown spot. A method consisting of two discrimination steps involving application of multiple discrimination models of a support vector machine gave the best result because of its capacity to evaluate the similarity of disease types. This accuracy of the method was 88% for leaf blast （A-type）, 94% for sheath blight, and 80% for leaf blast （B-type） and brown spot; on average, the accuracy of this method was 5% greater than that of the other method when three classes were used in the model. Although the accuracy of both methods was inadequate, the results of this study show that it is possible to estimate the least number of possible or similar diseases from a large number of diseases. Therefore, we conclude that there is merit in grouping classes into subgroups rather than attempting to discriminate between all classes simultaneously and that these methods are effective in identifying diseases for web-based diagnosis.展开更多
Differences are found in the attributes of microseismic events caused by coal seam rupture,underground structure activation,and groundwater movement in coal mine production.Based on these differences,accurate classific...Differences are found in the attributes of microseismic events caused by coal seam rupture,underground structure activation,and groundwater movement in coal mine production.Based on these differences,accurate classification and analysis of microseismic events are important for the water inrush warning of the coal mine working facefloor.Cluster analysis,which classifies samples according to data similarity,has remarkable advantages in nonlinear classification.A water inrush early warning method for coal minefloors is proposed in this paper.First,the short time average over long time average(STA/LTA)method is used to identify effective events from continuous microseismic records to realize the identification of microseismic events in coal mines.Then,ten attributes of microseismic events are extracted,and cluster analysis is conducted in the attribute domain to realize unsupervised classification of microseismic events.Clustering results of synthetic andfield data demonstrate the effectiveness of the proposed method.The analysis offield data clustering results shows that thefirst kind of events with time change rules is of considerable importance to the early warning of water inrush from the coal mine working facefloor.展开更多
文摘Two classification and identification methods based on pattern discrimination models and the majority-vote technique were investigated for implementing a World Wide Web-based system for the identification of rice diseases. The experiment was carried out using color and shape patterns in 425 images of three rice diseases, which were classified into four classes： two classes of leaf blast, and one class each of sheath blight and brown spot. A method consisting of two discrimination steps involving application of multiple discrimination models of a support vector machine gave the best result because of its capacity to evaluate the similarity of disease types. This accuracy of the method was 88% for leaf blast （A-type）, 94% for sheath blight, and 80% for leaf blast （B-type） and brown spot; on average, the accuracy of this method was 5% greater than that of the other method when three classes were used in the model. Although the accuracy of both methods was inadequate, the results of this study show that it is possible to estimate the least number of possible or similar diseases from a large number of diseases. Therefore, we conclude that there is merit in grouping classes into subgroups rather than attempting to discriminate between all classes simultaneously and that these methods are effective in identifying diseases for web-based diagnosis.
基金supported in part by the National Natural Science Foundation of China under Grant 41904098in part by the Beijing Nova Program under Grant 2022056in part by the National Natural Science Foundation of China (52174218)。
文摘Differences are found in the attributes of microseismic events caused by coal seam rupture,underground structure activation,and groundwater movement in coal mine production.Based on these differences,accurate classification and analysis of microseismic events are important for the water inrush warning of the coal mine working facefloor.Cluster analysis,which classifies samples according to data similarity,has remarkable advantages in nonlinear classification.A water inrush early warning method for coal minefloors is proposed in this paper.First,the short time average over long time average(STA/LTA)method is used to identify effective events from continuous microseismic records to realize the identification of microseismic events in coal mines.Then,ten attributes of microseismic events are extracted,and cluster analysis is conducted in the attribute domain to realize unsupervised classification of microseismic events.Clustering results of synthetic andfield data demonstrate the effectiveness of the proposed method.The analysis offield data clustering results shows that thefirst kind of events with time change rules is of considerable importance to the early warning of water inrush from the coal mine working facefloor.
