Matches in SemOpenAlex for { <https://semopenalex.org/work/W2961983701> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W2961983701 abstract "In order to explore the identification of tea plant diseases based on Convolutional Neural Networks (CNN), a convolutional neural network model is constructed by combining K-means clustering algorithm with CNN through collecting tea leaves from natural growth field in xx experimental field, and the running environment is designed to simulate the model. The obtained data are analyzed. In the analysis of the influence of K value selection on accuracy, it is found that when K value increases gradually and the accuracy increases slowly to K=128, the accuracy of convolutional neural network model tends to be stable and the amplitude decreases slowly. Therefore, the value of K is 128 from the comprehensive point of view of calculation. When comparing the size of different patch image blocks, it is found that the accuracy of the three K values is similar near 9*9, 11*11 and 13*13. Finally, from the comprehensive point of view, when the patch image block is 11*11, the effect will be better. Comparing with the traditional neural network algorithm, it is found that the recognition rate of the convolution neural network model for tea plant diseases is much higher than that of the traditional algorithm, and the recognition rate of the convolution neural network for disease categories is as high as 96.65%, while the recognition rate of the traditional neural network algorithm is lower than that of the CNN method. When the number of iterations is analyzed, it is found that when the number of iterations of the convolutional neural network model in this study is 100, the average correct rate is higher and the training time is basically appropriate. Therefore, through the study of tea plant disease identification in this study, it can be found that the application of convolutional neural network to tea plant disease identification accuracy will be greatly improved, with higher robustness, and meet the experimental expectations. Although there are some shortcomings in the experimental process, it can still provide a reference for the later identification of tea plant diseases." @default.
- W2961983701 created "2019-07-23" @default.
- W2961983701 creator A5054712826 @default.
- W2961983701 date "2019-06-16" @default.
- W2961983701 modified "2023-09-26" @default.
- W2961983701 title "Tea Plant Disease Recognition Based on Convolutional Neural Network" @default.
- W2961983701 hasPublicationYear "2019" @default.
- W2961983701 type Work @default.
- W2961983701 sameAs 2961983701 @default.
- W2961983701 citedByCount "0" @default.
- W2961983701 crossrefType "journal-article" @default.
- W2961983701 hasAuthorship W2961983701A5054712826 @default.
- W2961983701 hasConcept C11413529 @default.
- W2961983701 hasConcept C153180895 @default.
- W2961983701 hasConcept C154945302 @default.
- W2961983701 hasConcept C202444582 @default.
- W2961983701 hasConcept C2524010 @default.
- W2961983701 hasConcept C2777210771 @default.
- W2961983701 hasConcept C33923547 @default.
- W2961983701 hasConcept C41008148 @default.
- W2961983701 hasConcept C45347329 @default.
- W2961983701 hasConcept C50644808 @default.
- W2961983701 hasConcept C73555534 @default.
- W2961983701 hasConcept C81363708 @default.
- W2961983701 hasConcept C9652623 @default.
- W2961983701 hasConceptScore W2961983701C11413529 @default.
- W2961983701 hasConceptScore W2961983701C153180895 @default.
- W2961983701 hasConceptScore W2961983701C154945302 @default.
- W2961983701 hasConceptScore W2961983701C202444582 @default.
- W2961983701 hasConceptScore W2961983701C2524010 @default.
- W2961983701 hasConceptScore W2961983701C2777210771 @default.
- W2961983701 hasConceptScore W2961983701C33923547 @default.
- W2961983701 hasConceptScore W2961983701C41008148 @default.
- W2961983701 hasConceptScore W2961983701C45347329 @default.
- W2961983701 hasConceptScore W2961983701C50644808 @default.
- W2961983701 hasConceptScore W2961983701C73555534 @default.
- W2961983701 hasConceptScore W2961983701C81363708 @default.
- W2961983701 hasConceptScore W2961983701C9652623 @default.
- W2961983701 hasIssue "3" @default.
- W2961983701 hasLocation W29619837011 @default.
- W2961983701 hasOpenAccess W2961983701 @default.
- W2961983701 hasPrimaryLocation W29619837011 @default.
- W2961983701 hasRelatedWork W1978342095 @default.
- W2961983701 hasRelatedWork W2392821168 @default.
- W2961983701 hasRelatedWork W2586062600 @default.
- W2961983701 hasRelatedWork W2765381040 @default.
- W2961983701 hasRelatedWork W2810294686 @default.
- W2961983701 hasRelatedWork W2891300275 @default.
- W2961983701 hasRelatedWork W2901950140 @default.
- W2961983701 hasRelatedWork W2913024596 @default.
- W2961983701 hasRelatedWork W2942604505 @default.
- W2961983701 hasRelatedWork W2966554828 @default.
- W2961983701 hasRelatedWork W2968713876 @default.
- W2961983701 hasRelatedWork W3000291594 @default.
- W2961983701 hasRelatedWork W3004227095 @default.
- W2961983701 hasRelatedWork W3018925616 @default.
- W2961983701 hasRelatedWork W3028061992 @default.
- W2961983701 hasRelatedWork W3118327638 @default.
- W2961983701 hasRelatedWork W3177439627 @default.
- W2961983701 hasRelatedWork W2956620726 @default.
- W2961983701 hasRelatedWork W3019005981 @default.
- W2961983701 hasRelatedWork W3051732164 @default.
- W2961983701 hasVolume "36" @default.
- W2961983701 isParatext "false" @default.
- W2961983701 isRetracted "false" @default.
- W2961983701 magId "2961983701" @default.
- W2961983701 workType "article" @default.