Leaf Recognition Based on BGP Texture Matching
Automatic leaf recognition has been a hot research topic as digital leaf images capturing becomes more and more convenient and popular, which is also essential for plant education. However, fast and robust automatic recognition for leafs remains a challen
- PDF / 631,211 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 45 Downloads / 208 Views
Abstract Automatic leaf recognition has been a hot research topic as digital leaf images capturing becomes more and more convenient and popular, which is also essential for plant education. However, fast and robust automatic recognition for leafs remains a challenging problem. In this paper, we present a novel method for leaf recognition based on texture matching. To measure the similarity of two leaves which normally have different color distributions, lighting distributions, and viewing angles, we use binary Gabor pattern (BGP) matching to efficiently extract the texture feature by transforming an image into a pattern histogram. Support vector machine (SVM) classifier is then used to determine the final recognition results. Due to the robustness of combination of BGP and SVM, our method achieves an average recognition rate of up to 95.2 %. Keywords Leaf recognition
Binary Gabor pattern Rotation invariance
1 Introduction Plants are the most important resources for human survival and development. There are more than 350,000 kinds of higher plants known around the world. A botanist is impossible to manually classify every kind of plant faced with such a H. Wu (&) P. Pu B. Zhang College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, China e-mail: [email protected] G. He Landscape Department, Shenzhen University, Shenzhen, China F. Zhao Shenzhen Guoyipark Development Co., Ltd, Shenzhen, China
Z. Wen and T. Li (eds.), Foundations of Intelligent Systems, Advances in Intelligent Systems and Computing 277, DOI: 10.1007/978-3-642-54924-3_13, Springer-Verlag Berlin Heidelberg 2014
135
136
H. Wu et al.
large plant world, so it is even difficult for a normal person to identify the leaves come across in the wild. Moreover, there still exist many kinds of plants on the edge of extinction because of the development of human society. So it becomes an emergency to find an efficient and convenient manner for leaf recognition and plant classification. Because of the development of computer science, artificial intelligence becomes a hot topic in recent years. Thus, we have the opportunity to solve the difficult plant classification problem using methodology in the area of artificial intelligence. Because leaves can be easily captured by digital imaging devices, we may come across a plant in any growing stage all around a year, which makes the leaf recognition a challenging problem. In addition, leaves can be captured with different lightings, colors, scales, and viewing angles, which are also difficult issues to be solved in automatic leaf recognition. Many researchers have proposed several methods to automatically classify plant species through recognizing leaves, but most of their approaches focused on the outside shape of leaf and ignored the inside texture distribution, which is the key distinctive features when two leaves have similar shape [1–6]. In this paper, we apply the efficient binary Gabor pattern (BGP) to explicitly extract the texture features inside a leaf object. Support
Data Loading...
