Statistical analysis of texture in trunk images for biometric identification of tree species
- 27 March 2015
- journal article
- Published by Springer Nature in Environmental Monitoring and Assessment
- Vol. 187 (4) , 1-9
Abstract
The identification of tree species is a key step for sustainable management plans of forest resources, as well as for several other applications that are based on such surveys. However, the present available techniques are dependent on the presence of tree structures, such as flowers, fruits, and leaves, limiting the identification process to certain periods of the year. Therefore, this article introduces a study on the application of statistical parameters for texture classification of tree trunk images. For that, 540 samples from five Brazilian native deciduous species were acquired and measures of entropy, uniformity, smoothness, asymmetry (third moment), mean, and standard deviation were obtained from the presented textures. Using a decision tree, a biometric species identification system was constructed and resulted to a 0.84 average precision rate for species classification with 0.83accuracy and 0.79 agreement. Thus, it can be considered that the use of texture presented in trunk images can represent an important advance in tree identification, since the limitations of the current techniques can be overcome.Keywords
This publication has 16 references indexed in Scilit:
- Partial differential equations and fractal analysis to plant leaf identificationJournal of Physics: Conference Series, 2013
- Applications of Image Processing in Agriculture: A SurveyInternational Journal of Computer Applications, 2012
- Mapping forest composition from the Canadian National Forest Inventory and land cover classification mapsEnvironmental Monitoring and Assessment, 2011
- Measuring and analyzing color and texture information in anatomical leaf cross sections: an approach using computer vision to aid plant species identificationBotany, 2011
- PLANT LEAF IDENTIFICATION BASED ON VOLUMETRIC FRACTAL DIMENSIONInternational Journal of Pattern Recognition and Artificial Intelligence, 2009
- Plant leaf identification using Gabor waveletsInternational Journal of Imaging Systems and Technology, 2009
- AUTOMATIC LEAF STRUCTURE BIOMETRY: COMPUTER VISION TECHNIQUES AND THEIR APPLICATIONS IN PLANT TAXONOMYInternational Journal of Pattern Recognition and Artificial Intelligence, 2009
- Texture Analysis for Mapping Tamarix parviflora Using Aerial Photographs along the Cache Creek, CaliforniaEnvironmental Monitoring and Assessment, 2006
- Multiparameter analysis of vertical vegetation structure based on digital image processingFlora, 2003
- A Coefficient of Agreement for Nominal ScalesEducational and Psychological Measurement, 1960