An Enhanced Faster-RCNN Based Deep Learning Model for Crop Diseases Detection and Classification
Mallarapu Harish1, A V L N Sujith2, K. Santhi3
1Mallarapu Harish*, Computer Science and Engineering, S V College of Engineering, Tirupati, (Andhra Pradesh), India.
2A V L N Sujith, Computer Science and Engineering, S V College of Engineering, Tirupati, (Andhra Pradesh), India.
3K. SANTHI, Computer Science and Engineering, S V College of Engineering, Tirupati, (Andhra Pradesh), India.
Manuscript received on July 20, 2019. | Revised Manuscript received on August 30, 2019. | Manuscript published on August 30, 2019. | PP: 4714-4719 | Volume-8 Issue-6, August 2019. | Retrieval Number: F9212088619/2019©BEIESP | DOI: 10.35940/ijeat.F9212.088619
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Recently Plant phenotyping has gained the attention of many researchers such that it plays a vital role in the context of enhancing agricultural productivity. Indian economy highly depends on agriculture and this factor elevates the importance of early disease detection of the crops within the agricultural fields. Addressing this problem several researchers have proposed Computer Vision and Pattern recognition based mechanisms through which they have attempted to identify the infected crops in the early stages.in this scenario, CNN convolution neural network-based architecture has demonstrated exceptional performance when compared with state-of-art mechanisms. This paper introduces an enhanced RCNN recurrent convolution neural network-based architecture that enhances the prediction accuracy while detecting the crop diseases in early stages. Based on the simulative studies is observed that the proposed model outperforms when compared with CNN and other state-of-art mechanisms.
Keywords: CNN, RCNN, Deep Learning, Plant diseases.