New Supervised Approach for Plant Leaves Classification using Artificial Social Bees

New Supervised Approach for Plant Leaves Classification using Artificial Social Bees

Mohamed Elhadi Rahmani, Hadj Ahmed Bouarara, Abdelmalek Amine, Reda Mohamed Hamou, Hanane Menad
DOI: 10.4018/IJOCI.2016010102
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Life is based on plants, They are the major source of oxygen, food, and medicines. And biology is one of the major research in last years, but scientists don't stop on studying the biological life and understanding different mechanisms in life, they go further by inspiring from it, as organization on bee colony. This last is very impressive, especially in workplace. This work presents a new approach of supervised plant leaves classification using a meta-heuristic algorithm based on social bees. First, the authors used to represent leaves using three different features extracted from images: a fine-scale margin feature histogram, a Centroid Contour Distance Curve shape signature, or an interior texture feature histogram. Then the authors classified vectors by artificial social bees, and they evaluated the classification by its accuracy and error.
Article Preview
Top

Introduction

For all forms of life, plants form the basic food staples, and this is just one reason why plants are important. They are the major source of oxygen and food on earth since no animal is able to supply the components necessary without plants. The fish we eat consume algae and the cattle we eat as beef feed on grass, so even if you are not a fan of salads, your food source relies on plants. Plants also provide animals with shelter, produce clothing material, medicines, paper products, reduce noise levels and wind speed, reduce water runoff and soil erosion. Coal is also produced from plant materials that were once alive. All that gives plants its important role in life on earth. For example, as natural resource managers, they must understand what they manage, and plant identification is a key component of that understanding. The ability to know, or identify plants allows them to assess many important rangeland or pasture variables that are critical to proper management: range condition, proper stocking rates, forage production, wildlife habitat quality, and rangeland trend, either upward or downward. Natural resource managers, especially those interested in grazing and wildlife management must be able to evaluate the presence or absence of many plant species in order to assess these variables.

In nature, plant leaves are two dimensional containing important features that can be useful for classification of various plant species such as shapes, colours, textures and structures of their leaf, bark, flower, seedling and morph. According to Bhardwaj and al (Bhardwaj, 2013), if the plant classification is based on only two dimensional images, it is very difficult to study the shapes of flowers, seedling and morph of plants because of their complex three dimensional structures.

Biology is a very large domain, it contains a lot of sub-disciplines, and physiology of human body is one, with the big number of complex mechanisms that help life keep going. Understanding of these mechanisms is a principal source to inspire different algorithms and solution for problems in technology era.

the presented work proposes a bio-inspired algorithm for supervised classification, it applied for a multi-class problem to classify plant leaves, with different representations of it based on its Margin, Shape, and Textures; The organization of this paper is given as follows: section 2 provides a stat of the art speaking first about bio inspired algorithms, section 3 gives details about the biological aspect of social bees colony, section 4 describe the artificial aspect of that. Used dataset, and discussion of the results will be shown in section 5 and comparison with classical methods, and finally in section 6 gives the overall conclusion and the scope for future research.

Complete Article List

Search this Journal:
Reset
Volume 14: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 13: 1 Issue (2023)
Volume 12: 4 Issues (2022)
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 4 Issues (2016)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing