The presence of weeds in agricultural crops has been one of the problems of greatest interest in recent years as they consume natural resources and negatively affect the agricultural process. For this purpose, a model has been implemented to segment weed in aerial images. The proposed model relies on DeepLabv3 architecture trained upon patches extracted from high-resolution aerial imagery. The dataset employed consisted in 5 high-resolution images that describes a sugar beet agricultural field in Germany. SegNet and U-Net architectures were selected for comparison purposes. Our results demonstrate that balancing of data, together with a greater spatial context leads better results with DeepLabv3 achieving up to 0.89 and 0.81 in terms of AUC and F1-score, respectively.
18 March 2025 in International Journal of Combinatorial Optimization Problems and Informatics
This research aims to apply architectural photogrammetry for the documentation and conservation of the Monumental Clock of Pachuca de Soto, in the State of Hidalgo. To this end, an unmanned aerial vehicle (UAV) of the quadcopter type was constructed, enabling the capture of high-resolution photographic images, which were processed to generate a detailed three-dimensional model of the structure. The results demonstrate that this technique provides an accurate representation of the monument, facilitating both its conservation and the planning of future interventions. This study highlights the importance of model quality, which is influenced by climatic variables, as image capture may be limited on cloudy or rainy days. The research emphasizes the innovative application of UAV-based photogrammetry in the field of architectural heritage preservation, offering an efficient and non-invasive tool for the documentation of historical monuments. The findings confirm that this methodology holds significant potential for implementation in the digital conservation of cultural heritage.