Fisheye Image Detection of Trees Using Improved YOLOX for Tree Height Estimation

Abstract

Tree height is an essential indicator in forestry research. This indicator is difficult to measure directly, as well as wind disturbance adds to the measurement difficulty. Therefore, tree height measurement has always been an issue that experts and scholars strive to improve. We propose a tree height measurement method based on tree fisheye images to improve the accuracy of tree height measurements. Our aim is to extract tree height extreme points in fisheye images by proposing an improved lightweight target detection network YOLOX-tiny. We added CBAM attention mechanism, transfer learning, and data enhancement methods to improve the recall rate, F1 score, AP, and other indicators of YOLOX-tiny. This study improves the detection performance of YOLOX-tiny. The use of deep learning can improve measurement efficiency while ensuring measurement accuracy and stability. The results showed that the highest relative error of tree measurements was 4.06% and the average relative error was 1.62%. The analysis showed that the method performed better at all stages than in previous studies.

Plot-Level Maize Early Stage Stand Counting and Spacing Detection Using Advanced Deep Learning Algorithms Based on UAV Imagery

“…Zhang et al [37] adopted YOLOX for the fruit-counting area and achieved a value of 99.519% with respect to the best detection rate. Additionally, Song et al [38] used the improved YOLOX-tiny for tree height estimation based on fisheye images, and the highest relative error of the tree measurements was 4.06%. Zhong et al [39] successfully adopted YOLOR to detect real-time marine animals in a coral reef ecosystem, and it achieved an AP above 0.79 in both the fish and turtle categories.…”

Optimization of unmanned aerial vehicle application for measuring in complex urban green space with rich landscape value

“…These tree attributes further quantify benefits such as 3 carbon sequestration, improved air quality and stormwater retention (Nyelele et al, 2019;Song et al, 2020). Through the basic information, various evaluations and quantifications can be obtained, which are useful for urban road planning (Song et al, 2022), old tree protection (Qiu et al, 2018), tree maintenance in streets and parks (Cimburova and Barton, 2020), and assisting urban planners, landscape architects and policy makers in optimizing the structure and composition of urban green spaces (Song et al, 2020). When the main tree variables are measured by field measurement, DBH is relatively easy to obtain, while tree height and crown diameter are relatively difficult and time-consuming to measure (Avsar, 2004).…”

“…With the help of electronic theodolite, it has extremely high accuracy in measurement (Yao et al, 2022). However, there are problems such as low efficiency and high cost (Song et al, 2022).…”

Optimization of unmanned aerial vehicle application for measuring in complex urban green space

“…The complexity of urban parks also limits the access of MLS to measure under trees. In addition, the high cost of ALS and the difficulty of carrying TLS will be taken into account (Song et al, 2022). UAVs have the advantages of flexibility, low cost and high image spatial resolution (Zhou and Zhang, 2020).…”