High-resolution images captured by Unmanned Aerial Vehicles (UAVs) play a vital role in precision agriculture, particularly in evaluating crop health and detecting weeds. However, the detailed pixel information in these images makes classification a time-consuming and resource-intensive process. Despite these challenges, UAV imagery is increasingly utilized for various agricultural classification tasks. This study introduces an automatic classification method designed to streamline the process, specifically targeting cassava plants, weeds, and soil classification. The approach combines K-means unsupervised classification with spectral trend-based labeling, significantly reducing the need for manual intervention. The method ensures reliable and accurate classification results by leveraging color indices derived from RGB data and applying mean-shift filtering parameters. Key findings reveal that the combination of the blue (B) channel, Visible Atmospherically Resistant Index (VARI), and color index (CI) with filtering parameters, including a spatial radius (sp) = 5 and a color radius (sr) = 10, effectively differentiates soil from vegetation. Notably, using the green (G) channel, excess red (ExR), and excess green (ExG) with filtering parameters (sp = 10, sr = 20) successfully distinguishes cassava from weeds. The classification maps generated by this method achieved high kappa coefficients of 0.96, with accuracy levels comparable to supervised methods like Random Forest classification. This technique offers significant reductions in processing time compared to traditional methods and does not require training data, making it adaptable to different cassava fields captured by various UAV-mounted optical sensors. Ultimately, the proposed classification process minimizes manual intervention by incorporating efficient pre-processing steps into the classification workflow, making it a valuable tool for precision agriculture.

Using the high-resolution image from UAV is a major challenge for crop classification. The lighting conditions play an essential role in altering spectral reflectance, affecting the classification accuracy. With this barrier, it is difficult to use a single training data set in machine learning and apply to the images taken on different dates and times. This study aims to demonstrate the application of vegetation indices combined with the red, green, or blue dataset to enhance the image classification by using the Unsupervised classification method with the K-mean technic. This study shows the experiments of using three same image layers in three different windows of mean shift filter and presents the experiments of using three different image layers in three different windows. The optimal dataset shows Thailand's best performance of eight traditional cassava fields. As a result, three of the same BI and Blue layers presented the optimal dataset, Kappa coefficient = 0.864938 and 0.729271 for soil and plant classification, respectively. Alternatively, using a different dataset as the input to three windows of the mean shift filters was considered in this study to compare the results with the single-type input. The criterion of one primary dataset and two indices showed the optimal dataset among those experimental results (Kappa coefficient > 0.948681) in soil classification, as well as the optimal dataset for plant classification (Kappa coefficient > 0.773652). It can be concluded that combining complementary spectral features captured by different indices makes the classification model more robust and capable of capturing a broader range of soil characteristics, resulting in higher accuracy and reliability in delineating soil boundaries and plants in the cassava fields.