Harumanis mango leaf disease recognition system using image processing technique

Abstract

Current Harumanismango farming technique in Malaysia still mostlydepends on the farmers' own expertise to monitor the crops from the attack ofpests and insects. This approach is susceptible to human errors, and thosewho do not possess this skill may not be able to detect the disease at the righttime. As leaf diseases seriously affect the crop's growth and the quality of theyield, this study aims to develop a recognition system that detects thepresence of disease in the mango leaf using image processing technique.First, the image is acquired through a smartphone camera; once it has beenpre-processed, it is then segmented in which the RGB image is converted toan HSI image, then the features are extracted. Lastly, the classification ofdisease is done to determine thetype of leaf disease. The proposed systemeffectively detects and classify the disease with an accuracy of 68.89%. Thefindings of this project will contribute to farmers and society's benefit, andresearchers can use the approach to address similar issues in future works.

Potato leaf disease detection through ensemble average deep learning model and classifying the disease severity

“…When compared to the latest approach, it was found that the suggested deep learning model performed well. CNN 98.7% [22] Image processing 68.8% [23] CNN 97.2% [24] fuzzy set ---- [25] Image processing 94.5%…”

“…Gining et al colleagues. used image processing techniques in [22] to categorize the images of mango leafs. They employ the feature extraction technique to identify the images after converting RGB to HSI.…”

A Deep Learning Based Smartphone Application for Detecting Mango Diseases and Pesticide Suggestions

“…They may increase the region of interest's healthy portion's accuracy and adjust the shadow in the images of detection of disease in mango trees using color features to account for different lighting conditions. Abdullah et al [8] illustrated SVM-based image processing which was used to demonstrate a strategy for recognizing mango leaf disease. They got an accuracy percentage of 68.89%.…”

Detection of the patient with COVID-19 relying on ML technology and FAST algorithms to extract the features

“…However, you should not sacrifice computational efficiency to get high performance. In the absence of ML advancements, N is often set at 12 k. A high-speed testing policy may be utilized to eliminate non-corner sites [31]- [35]. Figure 1 shows the detector of segment tests.…”