A new method for crop image segmentation based on 2D histogram using multi-strategy shuffled frog leaping algorithm

Abstract

The economy of a country is directly impacted by agricultural productivity. If proper precautions are not taken in this area, plants suffer serious consequences, which have an impact on the quality, quantity, or productivity of the corresponding products. In this context, the farmers need an agriculture expert for the examination of the plant diseases, which takes a lot of time as well as continuous monitoring of plants. Hence, multilevel thresholding is a possible solution, which is useful for identifying the diseases in the crops by changing the color variation of a segmented image. However, it has a large range of applications in various domains such as remote sensing, the medical domain, the biometric domain. In this paper, an improved technique using horizontal and vertical crossover shuffled frog leaping algorithm (HVSFLA) is proposed for multilevel thresholding for the crop image based on a 2D histogram. The 2D histogram uses the grayscale value and non-local mean (NLM), whereas the one-dimension (1D) histogram uses only grayscale value. Therefore, in the proposed method, 2D Kapur’s entropy integrating with non-local mean 2D histogram is exploited for multilevel thresholding of the crop images. To investigate the efficacy, the proposed method is compared with well-known optimization techniques like beta differential evolution, artificial bee colony, bacterial foraging optimization, and particle swarm optimization. It is evident from the experimental results that the proposed technique yields better results than the 1D histogram technique in terms of root-mean-square error, structural similarity index, peak signal-to-noise ratio, feature similarity index, coefficient of variation, and fitness function. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.