LLAM-MDCNet for Detecting Remote Sensing Images of Dead Tree Clusters

Abstract

Clusters of dead trees are forest fires-prone. To maintain ecological balance and realize its protection, timely detection of dead trees in forest remote sensing images using existing computer vision methods is of great significance. Remote sensing images captured by Unmanned aerial vehicles (UAVs) typically have several issues, e.g., mixed distribution of adjacent but different tree classes, interference of redundant information, and high differences in scales of dead tree clusters, making the detection of dead tree clusters much more challenging. Therefore, based on the Multipath dense composite network (MDCN), an object detection method called LLAM-MDCNet is proposed in this paper. First, a feature extraction network called Multipath dense composite network is designed. The network’s multipath structure can substantially increase the extraction of underlying and semantic features to enhance its extraction capability for rich-information regions. Following that, in the row, column, and diagonal directions, the Longitude Latitude Attention Mechanism (LLAM) is presented and incorporated into the feature extraction network. The multi-directional LLAM facilitates the suppression of irrelevant and redundant information and improves the representation of high-level semantic feature information. Lastly, an AugFPN is employed for down-sampling, yielding a more comprehensive representation of image features with the combination of low-level texture features and high-level semantic information. Consequently, the network’s detection effect for dead tree cluster targets with high-scale differences is improved. Furthermore, we make the collected high-quality aerial dead tree cluster dataset containing 19,517 images shot by drones publicly available for other researchers to improve the work in this paper. Our proposed method achieved 87.25% mAP with an FPS of 66 on our dataset, demonstrating the effectiveness of the LLAM-MDCNet for detecting dead tree cluster targets in forest remote sensing images.

Automatic detection of standing dead trees based on improved YOLOv7 from airborne remote sensing imagery

“…Li et al. pro-posed the LLAM-MDCNet method based on the MDCN network to detect clusters of dead trees in aerial images, aiming to reduce the interference from complex back-grounds and variable target scales by introducing the LIAM attention module ( Li et al., 2022 ). However, their methods did not achieve the recognition of individual dead trees.…”

“…In similar studies of dead tree detection, Chiang et al applied transfer learning to the Mask RCNN network for automated detection of SDTs in aerial images (Chiang et al, 2020). Li et al pro-posed the LLAM-MDCNet method based on the MDCN network to detect clusters of dead trees in aerial images, aiming to reduce the interference from complex back-grounds and variable target scales by introducing the LIAM attention module (Li et al, 2022). However, their methods did not achieve the recognition of individual dead trees.…”

“…In the field of forestry, deep learning has been widely used for tasks such as forest resource management, soil analysis, tree detection and classification ( Srivastava et al., 2021 ; Wang et al., 2021 ). Scholars have conducted research on dead tree detection based on deep learning ( Chiang et al., 2020 ; Li et al., 2022 ; Wang et al., 2022 ). They primarily focus on detecting larger or densely packed targets, with limited studies addressing the detection of individual dead trees at a smaller scale.…”

Edge AI-Based Tree Trunk Detection for Forestry Monitoring Robotics

“…Some are to prevent and detect diseases in forest trees using Deep Learning (DL) and Unmanned Aerial Vehicle (UAV) imagery [2][3][4] or satellite high-resolution imagery [5]. The use of UAVs as a remote sensing platform is also a common way of detecting dead trees individually [6] or in clusters [7], as this can help prevent the occurrence of wildfires. The early detection of wildfires can help prevent their progress over forest lands, hence reducing their ecological and societal impact.…”

Determination of the state of artificial plantings of the green belt of Astana using remote sensing data of the earth

“…В мире широко используются методы дистанционного зондирования Земли (ДЗЗ), с помощью которых можно выполнить множественные исследования в различных областях лесного хозяйства -почвенные, физиологические, таксационные и др. С помощью данных ДЗЗ выявляют очаги возбудителей болезней и вредителей леса [Zhang et al, 2019;Zhu et al, 2021;Garza et al, 2020;Barta et al, 2022;Zhang et al, 2023;Poblete-Echeverria et al, 2023], оценивают биомассы древостоев и их породный состав [Wan et al, 2021;Li et al, 2022;Zhang et al, 2022;Mielczarek et al, 2022;Qiao et al, 2023;, определяют влияние засухи на состояние растений [Moreno-Fernandez et al, 2022] и жизненное состояние лесов [Laze, 2022].…”