Development of Enhanced Weed Detection System with Adaptive Thresholding and Support Vector Machine

This paper proposes a sophisticated classification process to segment the leaves of carrots from weeds. In the early stages of the plants' development, the color of both the plants and the weeds are similar, making it difficult to differentiate between the two. The process becomes even harder if the weeds and plants overlap. The proposed system addresses this problem by creating a sophisticated mean for weed identification. The major components of this system are composed of three processes: image segmentation, feature extraction and decision-making. In the image segmentation process, the input images are processed into lower units where the relevant features are extracted. In the decision-making process, the system makes use of the Support Vector Machine to analyze and segregate the weeds from the plants. Afterward, the findings are used to dictate which plants receive herbicides and which do not. The main priority for the image segmentation process is on the overlapping images where weeds need to be isolated from plants so that they can be used for cultivation purpose. The evaluation of the approach is done using an open dataset of images consisting of carrot plants. The system is able to achieve 88.99% accuracy for weed classification using this dataset. This methodology will help to reduce the use of herbicides while improving the performance and costs of precision agriculture.