Weed Detection Using AlexN et in Sesame Crops

Farming is the most essential part of humans for food. Rapid population growth increases the demand for more prominent functional capacity in agriculture. Yet, some issues, such as weeds, pose challenges to harvesting good crop yield, like absorbing nutrients, water, and light to minimize gain productively. To increase crop yield, humans spray weed-destructive toxins over the crop to destroy weeds. However, this also dilapidates crops and brings uncertainty to human food production. Recently, computer vision-based methods have been applied to automatically recognize weeds in the crop field by analyzing filed images for properties such as color, size, shape, shading, and texture. This paper utilizes an AlexN et-8 architecture to detect weeds in a field of Sesame crops. AlexN et architecture consists of8layers in a Convolutional Neural Network (CNN). The experimental analysis on 1300 Sesame crop images showed that the presented AlexN et architecture-based model achieved 97.6
^%
accuracy on RGB images.