Pest and Disease Identification in Paddy by Symptomatic Assessment of The Leaf using Hybrid CNN-LSTM Algorithm
A. Pushpa Athisaya Sakila Rani1, N. Suresh Singh2
1A. Pushpa Athisaya Sakila Rani*, Research Scholar, Department of Computer Science, Malankara Catholic College, Kaliyakkavilai (Tamil Nadu), India
2Dr. N. Suresh Singh, Associate Professor and Head, Department of Computer Applications, Malankara Catholic College, Kaliyakkavilai (Tamil Nadu), India
Manuscript received on January 28, 2022. | Revised Manuscript received on February 04, 2022. | Manuscript published on March 30, 2022. | PP: 7-14 | Volume-10 Issue-6, March 2022. | Retrieval Number: 10.35940/ijrte.F6795.0310622 | DOI: 100.1/ijrte.F67950310622
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© 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: The crop damage is caused by various types of pests that feed on the leaf, stem, roots or entire part of the plants and also by fungal, bacterial and viral infections. In most cases, the diseases are transmitted from one plant to another by vectors. The pests act as vectors in spreading most of the viral infections. It is necessary to identify the disease incidence or pest infestation in the early stages itself and contains its spread before it causes any damage to plants. Several machine and deep learning approaches are involved in rice disease and pest identification. In the preceding works Long Short-Term Memory (LSTM) and CNN algorithms respectively were used in identification and classification of the disease and pest that affects paddy. Here, a Hybrid CNN-LSTM method is applied for rice disease and pest identification using the various symptoms exhibited in paddy leaves. The accuracy of 97.8% in pest and disease identification proves the superiority of this method over the existing methods. 
Keywords: Disease Classification, Hybrid CNN-LSTM, Leaf Disease Detection, Pest Identification
Scope of the Article: 
Algorithm Engineering