Font Size: a A A

Wood Microscopic Structure Image And Identification Characteristic Research Based On Structure5.0 And SVM

Posted on:2017-11-14Degree:MasterType:Thesis
Country:ChinaCandidate:Z Z WangFull Text:PDF
GTID:2348330488474820Subject:Wood science and technology
Abstract/Summary:PDF Full Text Request
With the constant development of computer technology, wood identification points to automation and intelligence. The wood species can be accurately distinguished by observing their structure characteristics. In this study, Structure 5.0 v (NatureGene Life Sciences Company, Hong Kong) was first applied to wood identification by measuring and analyzing the wood microscopic images. By extracting wood samples of wood fiber and its measure, further to verify our recognition results. Finally to Structure5.0 measured wood microstructure characteristic value to establish the database, using the SVM for wood identification, microscopic image get wood microscopic image recognition rate under different parameters.The main results of this study as follows:1. Structure 5.0 software was applied to wood identification by measuring and analyzing the wood microscopic images. Some structure characteristics of Sample A and C were observed and measured through using Structure 5.0 software, including their pore type and arrangement, porosity, tangential diameter of vessel, and wood ray density, type and height. Comparing with the wood structure characteristics in the references, the identification results of Sample A and C were finally obtained.2. Some structure characteristics of Sample B were observed and measured through using Structure 5.0 software, including the type and arrangement of their wood ray, longitudinal parenchyma, tracheid, and resin canal. Comparing with the wood structure characteristics in the references, the identification results of Sample B were finally obtained.3. To extract the sample of wood fiber. Measured the length and width of the sample wood fiber by Structure 5.0 and Calculated amount of wood fibre etc..Further verification experiment wood sample identification results.4. Using polynomial kernel functions and radial basis function (RBF) of SVM. Measure wood ray height, wood ray width, the density of wood ray, wood ray cells number base on Structure 5.0 as the parameter to recdgnit the microscopic image.5. Compared with the DVProEM software of microscope, Structure 5.0 software had an evident advantage in the aspect of data measurement and statistical analysis.
Keywords/Search Tags:image processing, Timber identification, Structure5.0, SVM, Wood microstructure
PDF Full Text Request
Related items