Please use this identifier to cite or link to this item:
http://nopr.niscpr.res.in/handle/123456789/34349| Title: | Enhancing the Diagnosis of Corn Pests using Gabor Wavelet Features and SVM Classification |
| Authors: | Mousavi, S A Hanifeloo, Z Sumari, P Arshad, M R M |
| Keywords: | Plant diseases;Neural network;Gabor;Laplacian;Support vector machine |
| Issue Date: | Jun-2016 |
| Publisher: | NISCAIR-CSIR, India |
| Abstract: | One factor of quality and quantity reduction in crop yields is plant disease. Consultation with plant pathologists to diagnose plant diseases is time consuming, considering that time is an important factor in disease control. Therefore, it is necessary to present a simple, fast, cheap, and precise method for diagnosing plant diseases. In this article, five corn leaf diseases (Southern Leaf Blight, Southern Rust, Gray Leaf Spot, Holcus Spot, and Stewart's Wilt) are diagnosed using image-processing techniques. After collecting and transferring the diseased leaves to the laboratory, the images of leaves are produced under controlled conditions of light. In the following, four different methods are used to diagnose the disease and then the results are analyzed and compared. In the first proposed method, the damaged areas are separated using histogram equalization method and then a two-layer perceptron neural network classification system is used to categorize the final results and diagnose the disease. In the second method, Laplacian and Canny filters are used to separate the damaged areas. In the third method, the principal components analysis method is used and finally in the fourth method, a combination of Gabor filter and visual features is used to diagnose the disease. The results demonstrate that the combination of Gabor filter and visual features could successfully classify the specified disease spots by means of neural network classification system with the average accuracy of 90.04%. |
| Page(s): | 349-354 |
| ISSN: | 0975-1084 (Online); 0022-4456 (Print) |
| Appears in Collections: | JSIR Vol.75(06) [June 2016] |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| JSIR 75(6) 349-354.pdf | 369.74 kB | Adobe PDF | View/Open |
Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.