Tree Detection Research Articles

Fast Tree Detection and Counting on UAVs for Sequential Aerial Images with Generating Orthophoto Mosaicing

Individual tree counting (ITC) is a popular topic in the remote sensing application field. The number and planting density of trees are significant for estimating the yield and for futher planing, etc. Although existing studies have already achieved great performance on tree detection with satellite imagery, the quality is often negatively affected by clouds and heavy fog, which limits the application of high-frequency inventory. Nowadays, with ultra high spatial resolution and convenient usage, Unmanned Aerial Vehicles (UAVs) have become promising tools for obtaining statistics from plantations. However, for large scale areas, a UAV cannot capture the whole region of interest in one photo session. In this paper, a real-time orthophoto mosaicing-based tree counting framework is proposed to detect trees using sequential aerial images, which is very effective for fast detection of large areas. Firstly, to guarantee the speed and accuracy, a multi-planar assumption constrained graph optimization algorithm is proposed to estimate the camera pose and generate orthophoto mosaicing simultaneously. Secondly, to avoid time-consuming box or mask annotations, a point supervised method is designed for tree counting task, which greatly speeds up the entire workflow. We demonstrate the effectiveness of our method by performing extensive experiments on oil-palm and acacia trees. To avoid the delay between data acquisition and processing, the proposed framework algorithm is embedded into the UAV for completing tree counting tasks, which also reduces the quantity of data transmission from the UAV system to the ground station. We evaluate the proposed pipeline using sequential UAV images captured in Indonesia. The proposed pipeline achieves an F1-score of 98.2% for acacia tree detection and 96.3% for oil-palm tree detection with online orthophoto mosaicing generation.

Pomelo Tree Detection Method Based on Attention Mechanism and Cross-Layer Feature Fusion

Deep learning is the subject of increasing research for fruit tree detection. Previously developed deep-learning-based models are either too large to perform real-time tasks or too small to extract good enough features. Moreover, there has been scarce research on the detection of pomelo trees. This paper proposes a pomelo tree-detection method that introduces the attention mechanism and a Ghost module into the lightweight model network, as well as a feature-fusion module to improve the feature-extraction ability and reduce computation. The proposed method was experimentally validated and showed better detection performance and fewer parameters than some state-of-the-art target-detection algorithms. The results indicate that our method is more suitable for pomelo tree detection.

Effects of Flight and Smoothing Parameters on the Detection of Taxus and Olive Trees with UAV-Borne Imagery

Recent technical and jurisdictional advances, together with the availability of low-cost platforms, have facilitated the implementation of unmanned aerial vehicles (UAVs) in individual tree detection (ITD) applications. UAV-based photogrammetry or structure from motion is an example of such a low-cost technique, but requires detailed pre-flight planning in order to generate the desired 3D-products needed for ITD. In this study, we aimed to find the most optimal flight parameters (flight altitude and image overlap) and processing options (smoothing window size) for the detection of taxus trees in Belgium. Next, we tested the transferability of the developed marker-controlled segmentation algorithm by applying it to the delineation of olive trees in an orchard in Greece. We found that the processing parameters had a larger effect on the accuracy and precision of ITD than the flight parameters. In particular, a smoothing window of 3 × 3 pixels performed best (F-scores of 0.99) compared to no smoothing (F-scores between 0.88 and 0.90) or a window size of 5 (F-scores between 0.90 and 0.94). Furthermore, the results show that model transferability can still be a bottleneck as it does not capture management induced characteristics such as the typical crown shape of olive trees (F-scores between 0.55 and 0.61).

Using Airborne LiDAR to Monitor Spatial Patterns in South Central Oregon Dry Mixed-Conifer Forest

Abstract A common forest restoration goal is to achieve a spatial distribution of trees consistent with historical forest structure, which can be characterized by the distribution of individuals, clumps, and openings (ICO). With the stated goal of restoring historical spatial patterns comes a need for effectiveness monitoring at appropriate spatial scales. Airborne light detection and ranging (LiDAR) can be used to identify individual tree locations and collect data at landscape scales, offering a method of analyzing tree spatial distributions over the scales at which forest restoration is conducted. In this study, we investigated whether tree locations identified by airborne LiDAR data can be used with existing spatial analysis methods to quantify ICO distributions for use in restoration effectiveness monitoring. Results showed fewer large clumps and large openings, and more small clumps and small openings relative to historical spatial patterns, suggesting that the methods investigated in this study can be used to monitor whether restoration efforts are successful at achieving desired tree spatial patterns. Study Implications: Achieving a desired spatial pattern is often a goal of forest restoration. Monitoring for spatial pattern, however, can be complex and time-consuming in the field. LiDAR technology offers the ability to analyze spatial pattern at landscape scales. Preexisting methods for evaluation of the distribution of individuals, clumps, and openings were used in this study along with LiDAR individual tree detection methodology to assess whether a forest restoration project implemented in a Southern Oregon landscape achieved desired spatial patterns.

Comparative Evaluation of a Newly Developed Trunk-Based Tree Detection/Localization Strategy on Leaf-Off LiDAR Point Clouds with Varying Characteristics

LiDAR data acquired by various platforms provide unprecedented data for forest inventory and management. Among its applications, individual tree detection and segmentation are critical and prerequisite steps for deriving forest structural metrics, especially at the stand level. Although there are various tree detection and localization approaches, a comparative analysis of their performance on LiDAR data with different characteristics remains to be explored. In this study, a new trunk-based tree detection and localization approach (namely, height-difference-based) is proposed and compared to two state-of-the-art strategies—DBSCAN-based and height/density-based approaches. Leaf-off LiDAR data from two unmanned aerial vehicles (UAVs) and Geiger mode system with different point densities, geometric accuracies, and environmental complexities were used to evaluate the performance of these approaches in a forest plantation. The results from the UAV datasets suggest that DBSCAN-based and height/density-based approaches perform well in tree detection (F1 score > 0.99) and localization (with an accuracy of 0.1 m for point clouds with high geometric accuracy) after fine-tuning the model thresholds; however, the processing time of the latter is much shorter. Even though our new height-difference-based approach introduces more false positives, it obtains a high tree detection rate from UAV datasets without fine-tuning model thresholds. However, due to the limitations of the algorithm, the tree localization accuracy is worse than that of the other two approaches. On the other hand, the results from the Geiger mode dataset with low point density show that the performance of all approaches dramatically deteriorates. Among them, the proposed height-difference-based approach results in the greatest number of true positives and highest F1 score, making it the most suitable approach for low-density point clouds without the need for parameter/threshold fine-tuning.

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

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.

Estimation of forest stand characteristics using individual tree detection, stochastic geometry and a sequential spatial point process model

Airborne Laser Scanning (ALS) results in point-wise measurements of canopy height, which can further be used for Individual Tree Detection (ITD). However, ITD cannot find all trees because small trees can hide below larger tree crowns. Here we discuss methods where the plot totals and means of tree-level characteristics are estimated in such context. The starting point is a previously presented Horvitz–Thompson-like (HT-like) estimator, where the detectability is based on the larger tree crowns and a tuning parameter α that models the detection condition. We propose a new method which is based on modeling the spatial pattern of hidden tree locations using a sequential spatial point process model, with a tuning parameter θ. We also explore whether the variability of the tuning parameters α and θ can be predicted using ALS features to improve the predictions. The accuracy of stand density, dominant height and mean height is used as comparison criteria in a cross-validation procedure. The HT-like estimator with empirically estimated tuning parameter α performed the best. The overall performance of the new method was comparable. The new method was computationally less demanding, which makes it attractive for practical use.

Cross-Comparison of Individual Tree Detection Methods Using Low and High Pulse Density Airborne Laser Scanning Data

Numerous individual tree detection (ITD) methods have been developed for use with airborne laser scanning (ALS) data to provide tree-scale forest inventories across large spatial extents. Despite the growing number of methods, relatively few have been comparatively assessed using a single benchmark forest inventory validation dataset, limiting their operational application. In this study, we assessed seven ITD methods, representing three common approaches (point-cloud-based, raster-based, hybrid), across coniferous forest stands with diverse structure and composition to understand how ITD and height measurement accuracy vary with method, input parameters and data, and stand density. There was little variability in accuracy between the ITD methods where the average F-score and standard deviation (±SD) were 0.47 ± 0.03 using a lower pulse density ALS dataset with an average of 8 pulses per square meter (ppm2) and 0.50 ± 0.02 using a higher pulse density ALS dataset with an average of 22 ppm2. Using higher ALS pulse density data produced higher ITD accuracies (F-score increase of 10–13%) in some of the methods versus more modest gains in other methods (F-score increase of 1–3%). Omission errors were strongly related with stand density and largely consisted of suppressed trees underneath the dominant canopy. Simple canopy height model (CHM)-based methods that utilized fixed-size local maximum filters had the lowest omission errors for trees across all canopy positions. ITD accuracy had large intra-method variation depending on input parameters; however, the highest accuracies were obtained when parameters such as search window size and spacing thresholds were equal to or less than the average crown diameter of trees in the study area. All ITD methods produced height measurements for the detected trees that had low RMSE (<1.1 m) and bias (<0.5 m). Overall, the results from this study may help guide end-users with ITD method application and highlight future ITD method improvements.

Simulation-driven 3D forest growth forecasting based on airborne topographic LiDAR data and shading

Reliable forest growth forecasting requires detailed tree data for forest simulation, while manual on-site collection of relevant data is work-intensive and unfeasible in larger forests. This paper proposes a complete methodology for fully automated forest growth simulation that relies primarily on airborne topographic Light Detection And Ranging (LiDAR) point clouds of individual trees. The proposed method estimates tree parameters and performs growth of individual trees based on an individual-based forest growth simulator, named BWINPro. In addition, competition and detailed asymmetric tree crown growth are modeled regarding the shading of tree crowns, which is estimated from the surrounding environment and neighbor trees. The result of the proposed approach is a new point cloud for subsequent analyses. The proposed method was validated by comparing canopy height models derived from the point clouds of the simulated trees with canopy height models derived from more recent ground truth point clouds. The results demonstrate the efficacy of the proposed method which achieves a 9.4% higher accuracy than the averaged linear regression model and, in the case of datasets with more distinct self-standing trees, where a tree crown boundary plays major role, a 4.1% higher accuracy than the directly fitted linear regression model.

Tree growth at gap edges. Insights from long term research plots in mixed mountain forests

The structure and dynamics of Central Europe’s forests are increasingly characterized by canopy gaps which either result from disturbances or from planned silvicultural actions. It remains, however, unclear, whether the neighborhood to a gap has effects on tree growth which cannot be sufficiently covered by existing standard models so far. In order to test for such effects, we used data from a series of long term experiments in Southern German mixed mountain forests. In parallel, we developed a method for an automatized detection of canopy gaps and gap edge trees given such data. We found that the basal area growth of such trees amounted to about 15–30% more compared to what could be covered by a classic spatial competition index, with a plausible ranking of the main species. Our results suggest, in addition, that an exposition to a gap has a longer lasting effect on tree growth, even after the gap has closed again. With regard to such long-term effects, we found that tree size at the first exposition matters, with strong species-specific differentiation. We concluded that gap exposition effects on tree growth, given their order of magnitude, require being included in tree growth models which are used for planning purposes.

Evaluation The Accuracy of Oil Palm Tree Detection Using Deep Learning and Support Vector Machine Classifiers

Oil palm is one of the important sources of vegetable oil that is mostly consumed by the Malaysian citizen. Because the demand of vegetable oil is high in Malaysia, the expansion of oil palm tree plantation has been increasing rapidly. Remote sensing application in monitoring and detecting oil palm trees has become a very useful tool to help minimize the human energy to monitor in large plantations. The aim of this study is to evaluate the accuracy of oil palm tree detection using deep learning and support vector machine (SVM) approaches using high-resolution remote sensing images. Deep learning is one of the base frameworks for oil palm tree detection using high-resolution remote sensing images. Deep learning also is a counting tool provided in newest technology software such as ArcGIS Pro, where the tools use the pattern recognition concept as a template in detecting objects in a high-resolution image. In machine learning, support vector machines are supervised learning models with associated machine learning algorithms that analyse data for classification analysis. Based on this study, 91% of the oil palm trees detected using deep learning approach gained higher accuracy than SVM classifier with 86%. The accuracy of oil palm tree detection using deep learning is higher than the accuracy of support vector machine classifier. Based on the findings, deep learning approach create better object interpretation than SVM. The oil palm tree detection using both classification approaches had also been displayed by using a spatial map distribution for easier analysed and observation.

ReforesTree: A Dataset for Estimating Tropical Forest Carbon Stock with Deep Learning and Aerial Imagery

Forest biomass is a key influence for future climate, and the world urgently needs highly scalable financing schemes, such as carbon offsetting certifications, to protect and restore forests. Current manual forest carbon stock inventory methods of measuring single trees by hand are time, labour, and cost intensive and have been shown to be subjective. They can lead to substantial overestimation of the carbon stock and ultimately distrust in forest financing. The potential for impact and scale of leveraging advancements in machine learning and remote sensing technologies is promising, but needs to be of high quality in order to replace the current forest stock protocols for certifications. In this paper, we present ReforesTree, a benchmark dataset of forest carbon stock in six agro-forestry carbon offsetting sites in Ecuador. Furthermore, we show that a deep learning-based end-to-end model using individual tree detection from low cost RGB-only drone imagery is accurately estimating forest carbon stock within official carbon offsetting certification standards. Additionally, our baseline CNN model outperforms state-of-the-art satellite-based forest biomass and carbon stock estimates for this type of small-scale, tropical agro-forestry sites. We present this dataset to encourage machine learning research in this area to increase accountability and transparency of monitoring, verification and reporting (MVR) in carbon offsetting projects, as well as scaling global reforestation financing through accurate remote sensing.

基于嵌入式YOLO轻量级网络的树木检测算法

基于嵌入式YOLO轻量级网络的树木检测算法

Information Perception Method for Fruit Trees Based on 2D LiDAR Sensor

To solve the problem of orchard environmental perception, a 2D LiDAR sensor was used to scan fruit trees on both sides of a test platform to obtain their position. Firstly, the two-dimensional iterative closest point (2D-ICP) algorithm was used to obtain the complete point cloud data of fruit trees on both sides. Then, combining the lightning connection algorithm (LAPO) and the density-based clustering algorithm (DBSCAN), a fruit tree detection method based on density-based lightning connection clustering (LAPO-DBSCAN) was proposed. After obtaining the point cloud data of fruit trees on both sides of the test platform using the 2D-ICP algorithm, the LAPO-DBSCAN algorithm was used to obtain the position of fruit trees. The experimental results show that the positive detection rate was 96.69%, the false detection rate was 3.31%, and the average processing time was 1.14 s, verifying the reliability of the algorithm. Therefore, this algorithm can be used to accurately find the position of fruit trees, meaning that it can be applied to orchard navigation in a later stage.

An automatic approach for tree species detection and profile estimation of urban street trees using deep learning and Google street view images

Tree species and canopy structural profile (‘tree profile’) are among the most critical environmental factors in determining urban ecosystem services such as climate and air quality control from urban trees. To accurately characterize a tree profile, the tree diameter, height, crown width, and height to the lowest live branch must be all measured, which is an expensive and time-consuming procedure. Recent advances in artificial intelligence aids to efficiently and accurately measure the aforementioned tree profile parameters. This can be particularly helpful if spatially extensive and accurate street-level images provided by Google (‘streetview’) or Kakao (‘roadview’) are utilized. We focused on street trees in Seoul, the capital city of South Korea, and suggested a novel approach to create a tree profile and inventory based on deep learning algorithms. We classified urban tree species using the YOLO (You Only Look Once), one of the most popular deep learning object detection algorithms, which provides an uncomplicated method of creating datasets with custom classes. We further utilized semantic segmentation algorithm and graphical analysis to estimate tree profile parameters by determining the relative location of the interface of tree and ground surface. We evaluated the performance of the model by comparing the estimated tree heights, diameters, and locations from the model with the field measurements as ground truth. The results are promising and demonstrate the potential of the method for creating urban street tree profile inventory. In terms of tree species classification, the method showed the mean average precision (mAP) of 0.564. When we used the ideal tree images, the method also reported the normalized root mean squared error (NRMSE) for the tree height, diameter at breast height (DBH), and distances from the camera to the trees as 0.24, 0.44, and 0.41.

Detection of Pine Wilt Nematode from Drone Images Using UAV

Pine wilt nematode disease is a devastating forest disease that spreads rapidly. Using drone remote sensing to monitor pine wilt nematode trees promptly is an effective way to control the spread of pine wilt nematode disease. In this study, the YOLOv4 algorithm was used to automatically identify abnormally discolored wilt from pine wilt nematode disease on UAV remote sensing images. Because the network structure of YOLOv4 is too complex, although the detection accuracy is high, the detection speed is relatively low. To solve this problem, the lightweight deep learning network MobileNetv2 is used to optimize the backbone feature extraction network. Furthermore, the YOLOv4 algorithm was improved by improving the backbone network part, adding CBAM attention, and adding the Inceptionv2 structure to reduce the number of model parameters and improve the accuracy and efficiency of identification. The speed and accuracy of the Faster R-CNN, YOLOv4, SSD, YOLOv5, and the improved MobileNetv2-YOLOv4 algorithm were compared, and the detection effects of the Faster R-CNN, YOLOv4, SSD, YOLOv5 and the improved MobileNetv2-YOLOv4 algorithm on trees with pine wilt nematode were analyzed. The experimental results show that the average precision of the improved MobileNetv2-YOLOv4 algorithm is 86.85%, the training time of each iteration cycle is 156 s, the parameter size is 39.23 MB, and the test time of a single image is 15 ms, which is better than Faster R-CNN, YOLOv4, and SSD, but comparable to YOLOv5. Compared with the advantages and disadvantages, comprehensively comparing these four indicators, the improved algorithm has a more balanced performance in the detection speed, the parameter size, and the average precision. The score of the improved algorithm (95.60%) was higher than that of Faster R-CNN (90.80%), YOLOv4 (94.56%), and SSD (92.14%), which met the monitoring requirements of pine wilt nematode trees. Faster R-CNN and SSD pine-wilt-nematode tree detection models are not ideal in practical applications. Compared with the YOLOv4 pine-wilt-nematode tree detection model, the improved MobileNetv2-YOLOv4 algorithm satisfies the condition of maintaining a lower model parameter quantity to obtain higher detection accuracy; therefore, it is more suitable for practical application scenarios of embedded devices. It can be used for the rapid detection of pine wilt nematode diseased trees.

Sustainable Oil Palm Resource Assessment Based on an Enhanced Deep Learning Method

Knowledge of the number and distribution of oil palm trees during the crop cycle is vital for sustainable management and predicting yields. The accuracy of the conventional image processing method is limited for the hand-crafted feature extraction method and the overfitting problem occurs due to the insufficient dataset. We propose a modification of the Faster Region-based Convolutional Neural Network (FRCNN) for palm tree detection to reduce the overfitting problem and improve the detection accuracy. The enhanced FRCNN (EFRCNN) leads to improved performance for detecting objects (in the same image) when they are of multiple sizes by using a feature concatenation method. Transfer learning based on a ResNet50 model is used to extract the features of the input image. High-resolution images of oil palm trees from a drone are used to form the data set, containing mature, young, and mixed oil palm tree regions. We train and test the EFRCNN, the FRCNN, a CNN used recently for oil palm image detection, and two standard methods, namely, the support vector machine (SVM) and template matching (TM). The results reveal an overall accuracy of ≥96.8% for the EFRCNN on the three test sets. The accuracy is higher than the CNN and FRCNN and substantially higher than SVM and TM. For large-scale plantations, the accuracy improvement is significant. This research provides a method for automatically counting the oil palm trees in large-scale plantations.

Mapping the urban forest in detail: From LiDAR point clouds to 3D tree models

Trees are an integral component of the urban environment and important for human well-being, adaption measures to climate change and sustainable urban transformation. Understanding the small-scale impacts of urban trees and strategically managing the ecosystem services they provide requires high-resolution information on urban forest structure, which is still scarce. In contrast, there is an abundance of data portraying urban areas and an associated trend towards smart cities and digital twins as analysis platforms. A GIS workflow is presented in this paper that may close this data gap by classifying the urban forest from LiDAR point clouds, detecting and reconstructing individual crowns, and enabling a tree representation within semantic 3D city models. The workflow is designed to provide robust results for point clouds with a density of at least 4 pts/m2 that are widely available. Evaluation was conducted by mapping the urban forest of Dresden (Germany) using a point cloud with 4 pts/m². An object-based data fusion approach is implemented for the classification of the urban forest. A classification accuracy of 95 % for different urban settings is achieved by combining LiDAR with multispectral imagery and a 3D building model. Individual trees are detected by local maxima filtering and crowns are segmented using marker-controlled watershed segmentation. Evaluation highlights the influences of both urban and forest structure on individual tree detection. Substantial differences in detection accuracies are evident between trees along streets (72 %) and structurally more complex tree stands in green areas (31 %), as well as dependencies on tree height and crown diameter. Furthermore, an approach for parameterized reconstruction of tree crowns is presented, which enables efficient and realistic city-wide modeling. The suitability of LiDAR to measure individual tree metrics is illustrated as well as a framework for modeling individual tree crowns via geometric primitives.

UAV-Based Characterization of Tree-Attributes and Multispectral Indices in an Uneven-Aged Mixed Conifer-Broadleaf Forest

Unmanned aerial vehicles (UAVs) have contributed considerably to forest monitoring. However, gaps in the knowledge still remain, particularly for natural forests. Species diversity, stand heterogeneity, and the irregular spatial arrangement of trees provide unique opportunities to improve our perspective of forest stands and the ecological processes that occur therein. In this study, we calculated individual tree metrics, including several multispectral indices, in order to discern the spectral reflectance of a natural stand as a pioneer area in Mexican forests. Using data obtained by UAV DJI 4, and in the free software environments OpenDroneMap and QGIS, we calculated tree height, crown area, number of trees and multispectral indices. Digital photogrammetric procedures, such as the ForestTools, Structure from Motion and Multi-View Stereo algorithms, yielded results that improved stand mapping and the estimation of stand attributes. Automated tree detection and quantification were limited by the presence of overlapping crowns but compensated by the novel stand density mapping and estimates of crown attributes. Height estimation was in line with expectations (R2 = 0.91, RMSE = 0.36) and is therefore a useful parameter with which to complement forest inventories. The diverse spectral indices applied yielded differential results regarding the potential vegetation activity present and were found to be complementary to each other. However, seasonal monitoring and careful estimation of photosynthetic activity are recommended in order to determine the seasonality of plant response. This research contributes to the monitoring of natural forest stands and, coupled with accurate in situ measurements, could refine forest productivity parameters as a strategy for the validity of results. The metrics are reliable and rapid and could serve as model inputs in modern inventories. Nevertheless, increased efforts in the configuration of new technologies and algorithms are required, including full consideration of the costs implied by their adoption.

COMPUTER VISION SYSTEM FOR DETECTING ORCHARD TREES FROM UAV IMAGES

Abstract. Orchard tree inventory plays an important role in acquiring up-to-date information on planted trees for effective treatments and crop insurance purposes. Determining tree damage could help assess orchards’ health faster and cheaper. Having accurate information on the tree’s status could also help managers to plan necessary fieldwork and predict productivity. Traditional orchard inventory is often performed manually, and thus is time-consuming, costly, and subject to error. An alternative is computer vision algorithms that could automatically detect orchard trees based on UAV imagery. The objective of this study is to develop a method using advanced computer vision algorithms to automatically detect apple trees on UAV multispectral images. This task is challenging since apple trees are overlapping over the UAV images, and hence distinguishing different crowns could be difficult. Motivated by the latest advances in UAV imagery and deep-learning models, addressed the tree detection problem by exploring the two CNN models YOLO (You Only Look Once) and DeepForest for detecting apple trees on UAV images. We first constructed a labelled dataset by dividing the study area into equally sized patches. Then we manually annotated all apple trees seen in RGB images. The annotated dataset was then randomly divided into three subsets (training, validation, and testing), for training and testing machine learning models. The performed experiments demonstrate the efficiency and validity of the proposed approach for orchard tree inventory. In particular, the proposed framework achieved a precision of 91% and an F1-score of 87% by adopting the DeepForest model for tree detection.