A novel deep learning method for detection and classification of plant diseases

Momina Masood,Waleed Albattah,Ali Javed,Marriam Nawaz,Saleh Albahli

doi:10.1007/s40747-021-00536-1

Abstract

The agricultural production rate plays a pivotal role in the economic development of a country. However, plant diseases are the most significant impediment to the production and quality of food. The identification of plant diseases at an early stage is crucial for global health and wellbeing. The traditional diagnosis process involves visual assessment of an individual plant by a pathologist through on-site visits. However, manual examination for crop diseases is restricted because of less accuracy and the small accessibility of human resources. To tackle such issues, there is a demand to design automated approaches capable of efficiently detecting and categorizing numerous plant diseases. Precise identification and classification of plant diseases is a tedious job due because of the occurrence of low-intensity information in the image background and foreground, the huge color resemblance in the healthy and diseased plant areas, the occurrence of noise in the samples, and changes in the position, chrominance, structure, and size of plant leaves. To tackle the above-mentioned problems, we have introduced a robust plant disease classification system by introducing a Custom CenterNet framework with DenseNet-77 as a base network. The presented method follows three steps. In the first step, annotations are developed to get the region of interest. Secondly, an improved CenterNet is introduced in which DenseNet-77 is proposed for deep keypoints extraction. Finally, the one-stage detector CenterNet is used to detect and categorize several plant diseases. To conduct the performance analysis, we have used the PlantVillage Kaggle database, which is the standard dataset for plant diseases and challenges in terms of intensity variations, color changes, and differences found in the shapes and sizes of leaves. Both the qualitative and quantitative analysis confirms that the presented method is more proficient and reliable to identify and classify plant diseases than other latest approaches.

Highlights

According to the Food and Agriculture Organization (FAO) of the United Nations, the population of the world will rise to 9.1 billion by 2050
The major cause of the better performance of the DenseNet-77 framework is due to its shallow network structure that makes effective reuse of model parameters without employing redundant keypoint maps
It can be concluded from the discussed results that our improved DenseNet-77 based CenterNet model shows robust accuracy in comparison to other deep learning (DL)-based approaches both in terms of classification performance and execution time

Summary

Introduction

According to the Food and Agriculture Organization (FAO) of the United Nations, the population of the world will rise to 9.1 billion by 2050. Techniques proposed in molecular biology and immunology were utilized to detect crop diseases at the earliest stage [4, 5]. These approaches required human experts, huge resources, and cost to be established. According to FOA, the majority of cultivation areas are small and run by people in under-developed nations having low income [6] Such expensive solutions are impractical for them and researchers need to propose efficient and effective approaches that are accessible to all farmers [7]. Due to the advancement of digital methods, a huge amount of information is being gathered in real-time on which ML-based approaches are applied to make an optimized decision. Still, there exists a need for performance improvements, for decision-support frameworks that assist in converting the huge amount of data into valuable recommendations

Objectives

Results

Conclusion