In the last decade, small unmanned aerial vehicles (UAVs/drones) have become increasingly popular in the airborne observation of large areas for many purposes, such as the monitoring of agricultural areas, the tracking of wild animals in their natural habitats, and the counting of livestock. Coupled with deep learning, they allow for automatic image processing and recognition. The aim of this work was to detect and count the deer population in northwestern Serbia from such images using deep neural networks, a tedious process that otherwise requires a lot of time and effort. In this paper, we present and compare the performance of several state-of-the-art network architectures, trained on a manually annotated set of images, and use it to predict the presence of objects in the rest of the dataset. We implemented three versions of the You Only Look Once (YOLO) architecture and a Single Shot Multibox Detector (SSD) to detect deer in a dense forest environment and measured their performance based on mean average precision (mAP), precision, recall, and F1 score. Moreover, we also evaluated the models based on their real-time performance. The results showed that the selected models were able to detect deer with a mean average precision of up to 70.45% and a confidence score of up to a 99%. The highest precision was achieved by the fourth version of YOLO with 86%, as well as the highest recall value of 75%. Its compressed version achieved slightly lower results, with 83% mAP in its best case, but it demonstrated four times better real-time performance. The counting function was applied on the best-performing models, providing us with the exact distribution of deer over all images. Yolov4 obtained an error of 8.3% in counting, while Yolov4-tiny mistook 12 deer, which accounted for an error of 7.1%.

This paper presents the standardized precipitation evapotranspiration index (SPEI)-based approach to agricultural drought monitoring (ADM-SPEI approach) combining well-known methods, expert’ critical opinions, and local agro-climatic specificities. The proposed approach has been described in detail in three phases. This allows its application in any region and modification according to different agro-climatic conditions. The application of the ADM-SPEI approach has resulted in obtaining a modified SPEI for different crops (agricultural drought SPEI (AD-SPEI_crop)) in the Vojvodina region. In the first phase of the proposed approach, analytical hierarchy process (AHP) was used to obtain an experts’ group decision regarding the most suitable method for calculating evapotranspiration for a particular analyzed region. In the second phase, SPEI was modified and adjusted to the conditions in Vojvodina, where ET₀ was replaced by ET_c. In the validation phase, the results of the application of AD-SPEI_crop were compared to crop yields and well-known indices and evaluated by the experts’ feedback. The statistically significant correlations were achieved between AD-SPEI_crop and crop yields. The highest correlations were achieved in the months when the analyzed crops were in the developmental stages when they are most sensitive to drought. The AD-SPEI_crop better correlates to the crop yields compared to SPEI. The comparison of AD-SPEI_crop to the standardized precipitation index (SPI), SPEI, and self-calibrated Palmer drought severity index (SC-PDSI) shows that it can successfully detect dry and wet periods. The results have indicated that the proposed approach can be successfully applied, and AD-SPEI_crop has shown a good performance for agricultural drought monitoring.