An Efficient AdaBoost and CNN Hybrid Model for Weed Detection and Removal

With the aid of AI-based methods like convolution neural networks, this article seeks to enhance weed detection and removal. A plant that is not wanted and is growing wild, especially one that does so on cultivated land and produces a low crop yield, is referred to as weed. In order to increase production in agriculture, it is essential to have a strong and positive weed identification and removal. This paper suggested a weed control programme for wheat fields indicated on the RGB combination of image elements with essential attributes due to the low weed control rate and availability of inefficient weed removal approaches in the wheat sector. AdaBoost estimation and the CNN model are combined to enhance the performance of the weed distinguishing proof. The approach tests the bounds of the 2D topographical from RGB images while seeing wheat-like grass weeds. The accuracy of the test results displayed by the algorithm during the process was 94.7%. In comparison to previously employed methods, the combination of both algorithms used provides a high degree of accuracy in the estimation of various types of weeds.