Identification of Plant Species Using Deep Learning
Classification of plant species using machine learning is an automated task for recognizing the unknown plant species. Classification is very challenging due to the morphological similarity of different species of plants. In this paper, we have proposed a
- PDF / 340,562 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 97 Downloads / 243 Views
Abstract Classification of plant species using machine learning is an automated task for recognizing the unknown plant species. Classification is very challenging due to the morphological similarity of different species of plants. In this paper, we have proposed a deep convolutional neural network (CNN) based model for the identification of plant species using plant leaf images. The main intuition of using CNN is learning the leaf features directly from the input images/data. Furthermore, it is observed that CNN-based techniques significantly increase the performances in case of different plant species having identical shape and sizes of leaves. The proposed model is compared with other existing techniques in the same domain. It is found that our model improves the recognition accuracy significantly. Keywords Machine learning · Convolutional neural network(cnn) · Feature extraction
1 Introduction Plants are the richest property of the earth and very essential for human life and the environment as well. Recognition of plant species provides collectable information in plant research and also can be useful for protecting useful plant species. To collect the information about plants, one needs to visit either to a botanist, or to a nursery, or needs to collect information from the Internet, which is very time consuming [19]. Therefore, automated identification of plant species is very effective and speed up the process. For classification/Identification of plants, extraction of feature is the essential task and leaves are the main visual organ that can be exploited by computer vision and pattern recognition [1]. Nowadays, convolutional neural network(CNN) S. K. Mahmudul Hassan (B) · A. Kumar Maji Department of Information Technology, NEHU, Shillong 793022, India e-mail: [email protected] A. Kumar Maji e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 D. Bhattacharjee et al. (eds.), Proceedings of International Conference on Frontiers in Computing and Systems, Advances in Intelligent Systems and Computing 1255, https://doi.org/10.1007/978-981-15-7834-2_11
115
116
S. K. Mahmudul Hassan and A. Kumar Maji
Fig. 1 Some Sample leaf dataset [3] images
has achieved remarkable results in the field of image classification and pattern recognition. The plant leaf image feature representation is a crucial component of a plant leaf recognition algorithm. There are different methods to represent and describe the feature in machine learning for any classification problem. Among them, the most prominent methods are traditional hand-crafted features and deep learning (DL) based features. In hand-crafted features, we need to extract the different features like color, shape, and texture and then apply the classifier for identification. However, in DL-based methods, features are extracted and learned automatically as it is superior in providing deeper information about the image. For this reason, DL-based classification is so popular and widely used. This paper begins with the introduction, which states
Data Loading...
