We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
A Novel Approach: Pests Detection Using Artificial Neural Network and Image Processing.
- Authors
Yadav, Ravindra; Yadav, Ashok Kumar
- Abstract
The need for effective and precise detection technologies to lessen the impact of pests on agricultural output is a global concern. This research introduces a new method for detecting pests using image processing techniques and artificial neural networks (ANNs). A strong solution that can identify pests in different agricultural contexts is offered by our suggested technique, which uses learning and image analysis. Artificial neural networks (ANNs) allow for the training of specialised models that can identify different types of pests using visual features collected from digital photos. In addition, picture preparation methods are used to boost model performance and feature extraction. Extensive studies on real-world datasets reveal that the suggested method outperforms standard pest detection approaches in terms of efficiency and accuracy, proving its efficacy. A dependable and automated approach for early pest identification is offered by this study, which advances the area of precision agriculture. This allows for earlier intervention and minimises crop loss. We use artificial neural networks (ANN) because of their capacity to learn complicated patterns straight from picture data; this method is based on the latest developments in deep learning as well as image processing. Additionally, input pictures are preprocessed using feature extraction methods such morphological operations and histogram equalisation to successfully decrease noise and improve the recovered features' discriminative strength. In order to test how well our method works, we use a wide range of datasets that include different types of pests and their environments. Achieving excellent accuracy rates in pest identification tasks, the results show that our technology is resilient and scalable across multiple agricultural situations. Additionally, we contrast our technique with more conventional ways to pest identification, such as chemical-based methods and human inspection, to demonstrate the benefits of automation and efficiency that our system provides. Our research highlights the incredible power of AI and image processing to transform pest control approaches. This might lead to more resilient and sustainable agricultural systems that can withstand new threats.
- Subjects
ARTIFICIAL neural networks; IMAGE processing; PESTS; AGRICULTURAL pests; FEATURE extraction; PRECISION farming
- Publication
Journal of Advanced Zoology, 2024, Vol 45, p224
- ISSN
0253-7214
- Publication type
Article