Use of Texture Information to Classify Vegetation in River Bank by SVM

Abstract

In order to manage river bank, we propose a method to classify vegetation in river band by Support Vector Machine (SVM) on the basis of texture information. The method comprises three steps: detection of vegetation, initial classification by SVM, and reclassification to reduce noises. The results for 38 images obtained by the Akita Office of River and National Highway show that the proposed method using both texture information and RGB components can accurately classify grasses and harmful vegetation in comparison with a case using the RGB original components.

Generating Learning Data for Hierarchical Vegetation Classification Methods using Support Vector Machine

“…To improve the versatility of the hierarchical vegetation classifi cation method, we compare these with the previous method [4]. Furthermore, the pixel matching rate is calculated as follows.…”

“…The difference in size following removal is a result of the character string size difference of the given point. Next, we extract the target area for vegetation classifi cation based on the previous method [4]. The equation for extracting the vegetation area is as follows.…”

“…Step 4. Secondary classification processing: Our previous study clarified that the R-value was the most suitable value for classifi cation [4], so we adopted it for the secondary classifi cation area. Thus, learning by SVM is performed using R-values and homogeneity, with a total of four features, and the secondary classification region pixels are reclassified.…”

“…In recent years, classification methods using machine learning have been developed in order to classify target vegetation with high accuracy [2] To date, we have proposed a ratio estimation algorithm for harmful vegetation to reduce the burden on river management workers [3]. In addition, we proposed a method for classifying hierarchical vegetation [4], focusing on color and texture information [5] obtained from images, using a support vector machine (SVM) [6]. The SVM is a pattern recognition model, which uses supervised learning obtained from the Omonogawa River bank imagery.…”