Tropical Wood Species Recognition Based on Gabor Filter

Abstract

Tropical timber woods have more than 1,000 species. Some of the species have similar patterns with others and some have different patterns even though they are of the same species. One of the main problems in wood species recognition system is the lack of discriminative features of the texture images. Gabor filter has been extensively used as feature extractor for various applications such as face detection, face recognition, image retrieval and font type extraction. In our work, we propose the use of Gabor filter to generate multiple processed images from a single image so that more features can be extracted and will be trained by neural network. The use of Gabor filters will optimally localized the properties of the images in both spatial and frequency domain. The features of the filtered images are extracted using co- occurrence matrix approach, known as grey level co- occurrence matrix (GLCM). A multi-layer neural network based on the popular BP (back propagation) algorithm is used for classification. The results show that increasing the number of features by means of Gabor filters as well as the right combination of Gabor filters increases the accuracy rate of the system.