A Hybridized Auto-encoder and Convolution Neural Network-Based Model for Plant Identification

Plant plays an important role in earth’s atmosphere as well as in industry and medicinal field. Therefore, it is important to analyze and correctly identify the plant species using plant leaf. In this paper, we have implemented auto-encoder and auto-encoder with convolution neural network (AE-CNN) for plant identification from leaf. Experimental result demonstrates that more accurate results can be obtained in classification using AE-CNN. The proposed technique is also compared with other state-of-art existing techniques, and it is found our methods are having an accuracy of 96%. In this work, we have also used several types of data augmentation techniques, namely sample-wise and feature-wise pixel standardization, ZCA transform, rotation, random flip and random shift. The proposed method is very effective in identification of leaves with varying size and orientation.

[1]  Heba A. Elnemr.  Feature selection for texture-based plant leaves classification , 2017, 2017 Intl Conf on Advanced Control Circuits Systems (ACCS) Systems & 2017 Intl Conf on New Paradigms in Electronics & Information Technology (PEIT).

[2]  A. Ugur,et al.  Recognition of leaves based on morphological features derived from two half-regions , 2012, 2012 International Symposium on Innovations in Intelligent Systems and Applications.

[3]  Yeni Herdiyeni,et al.  Fusion of Local Binary Patterns features for tropical medicinal plants identification , 2013, 2013 International Conference on Advanced Computer Science and Information Systems (ICACSIS).

[4]  Li Shang,et al.  Hybrid Deep Learning for Plant Leaves Classification , 2015, ICIC.

[5]  Alvin B. Culaba,et al.  Identification of philippine herbal medicine plant leaf using artificial neural network , 2017, 2017IEEE 9th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM).

[6]  Milan Tuba,et al.  Leaf recognition algorithm using support vector machine with Hu moments and local binary patterns , 2017, 2017 IEEE 15th International Symposium on Applied Machine Intelligence and Informatics (SAMI).

[7]  Y. Herdiyeni,et al.  Combination of morphological, Local Binary Pattern Variance and color moments features for Indonesian medicinal plants identification , 2012, 2012 International Conference on Advanced Computer Science and Information Systems (ICACSIS).

[8]  W. Sinthupinyo,et al.  Leaf classification using structure features and Support Vector Machines , 2012, 2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (ISSDM2012).

[9]  Paulus Insap Santosa,et al.  Performance Improvement of Leaf Identification SystemUsing Principal Component Analysis , 2012 .

[10]  Jyotismita Chaki,et al.  Plant leaf recognition using texture and shape features with neural classifiers , 2015, Pattern Recognit. Lett..

[11]  Paolo Remagnino,et al.  Deep-plant: Plant identification with convolutional neural networks , 2015, 2015 IEEE International Conference on Image Processing (ICIP).

[12]  Hulya Yalcin,et al.  Plant classification using convolutional neural networks , 2016, 2016 Fifth International Conference on Agro-Geoinformatics (Agro-Geoinformatics).

[13]  K. Sreekumar,et al.  Literature review of image features and classifiers used in leaf based plant recognition through image analysis approach , 2017, 2017 International Conference on Inventive Communication and Computational Technologies (ICICCT).