Maize Leaf Images Segmentation Using Color Threshold and K-means Clustering Methods to Identify the Percentage of the Affected Areas

Abstract

Control of diseases affecting plants is very important as it relates to the issue of food security, which is a very serious threat on human life. According to the International Maize and Wheat Improvement Center (CIMMYT), most of the maize diseases are caused by mildew. There are more than fifty different mildew diseases affecting maize. The diseases that infect plants go through different stages of the degree of infection. Therefore, determining the degree of injury helps decision-makers spend money on controlling the disease properly. The common method used to determine the degree of injury is visual examination. It’s known that visual examination leaves a large area for error. It is unable to determine the degree of injury with high accuracy as a percentage for example. In this study, we used two methods of segmentation to determine the injury ratio in maize leaves. The methodology of segmentation depends on the Color Threshold Application (CTA) and K-mean clustering. For comparison between methods of segmentation, different color spaces (RGB, HSV, YCbCr, and L*a*b*) have been used. The results obtained indicate that the color threshold segmentation is more efficient than K-means clustering in terms of determining the injury ratio, especially with HSV color space. While it contrary to the RGB color model because it gives the worst result.

An Improved CNN-Based Apple Appearance Quality Classification Method With Small Samples

“…The best clustering number is determined by the number of gray histogram peaks, and the clustering center data is filtered by comparing the average value with the threshold value determined by Otsu. Then the open and close operation of mathematical morphology is used to get the outline of the fish's body, and a more accurate and stable algorithm model is obtained. Abbas et al (2020) [13] firstly classify the image data of diseased corn by using k-means clustering, and then use the color threshold segmentation to obtain a threshold segmentation method, which is more effective than the single use of K-means clustering after inspection. Kour et al(2019) [14] firstly preprocess the fruit images such as guava and mango, and then extract different features according to the color and texture.…”

Enhancement of Wheat Leaf Images Using Fuzzy-Logic Based Histogram Equalization to Recognize Diseases

“…Their approach relied on several pre-processing steps, such as image cropping, image resizing, and image converting to threshold value, reaching a recognition accuracy of 91%. Abbas et al [11] used the segmentation methodology to determine the percentage of affected areas in maize leaves. Their methodology relied on image classification by K-Means clustering along with image segmentation using Color Threshold, and then estimating the affected area by calculating the number of white pixels and dividing it by the number of total pixels.…”