Classification of Soybean Diseases Using Pre-trained Deep Convolutional Neural Networks

In this work, a novel soybean leaf disease classification technique related to pre-trained GoogleNet deep convolutional neural networks (CNN) architecture proposed. The proposed GoogleNet architecture trained on a database of 550 image samples of unhealth

  • PDF / 1,939,737 Bytes
  • 11 Pages / 439.37 x 666.142 pts Page_size
  • 101 Downloads / 296 Views

DOWNLOAD

REPORT


3

Bharati Vidyapeeths College of Engineering Kolhapur, Kolhapur, India [email protected] 2 Maratha Mandal Engineering College Belagavi, Belagavi, India [email protected] Rajgad Shri. Chattrapati Shivajiraje College of Engineering, Bhor, Pune, India [email protected]

Abstract. In this work, a novel soybean leaf disease classification technique related to pre-trained GoogleNet deep convolutional neural networks (CNN) architecture proposed. The proposed GoogleNet architecture trained on a database of 550 image samples of unhealthy and healthy soybean leaflets with 3 symptoms of an unhealthy class particularly, septoria brown spot, bacterial blight, frogeye leaf spot, and 1 healthy class using a deep transfer learning approach. As specified 3 unhealthy class and 1 healthy class identification, we have used the 5-fold cross-validation approach, the intended pre-trained GoogleNet-CNN architecture attains an accuracy of 96.25%. It was found that the accuracy of our proposed CNN architecture is enormously more precise than the formal machine learning models. The results of performance analysis to the recognition of soybean diseases exhibit the expediency and highest success rate using the proposed GoogleNet CNN model. Keywords: CNN

 Machine learning  GoogleNet  Cassification  Diseases

1 Introduction The recent development in convolutional neural networks (CNNs) has attained remarkable outcomes in the domain of pattern recognition [1]. This work is consisting of novel access to the improvement of plant infection diagnosis system, based on soybean diseased leaf image classification, using pre-trained GoogleNet and AlexNet deep learning architecture models. The 3 common soybean diseases include brown spot, frogeye spot and blight were considered for classification of disease task, these diseases cause significant crop yield loss and its consequence is on the livelihood and economy of a farmer. An automated system designed using recent technology like machine learning for identification of diseases related on the appearance of visible symptoms occurred on the leaf could be of numerous aid to farmers in the agriculture manner and also plant pathologists as a precise and reliable disease investigation system [2, 3].

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 J. I.-Z. Chen et al. (Eds.): ICIPCN 2020, AISC 1200, pp. 746–756, 2021. https://doi.org/10.1007/978-3-030-51859-2_68

Classification of Soybean Diseases

747

To enhance the precision and promptness of the investigation outcomes, numerous researchers have developed the computerized plant disease investigation system related to image processing and computer vision recognition systems applying support vector machine (SVM) [9], digital image processing methods [2] related on color analysis and thresholding based segmentation techniques [14] and computer vision. Also, artificial neural networks ANNs based strategies are presently applied for recognizing plant diseases [4] these imply fused with variou