Action Camera Research Articles

Measuring Docility, Boldness, and Exploration in the Free-Living Yellow Ground Squirrel Spermophilus fulvus Licht. (Sciuridae)

Studying intraspecific variation in animal behavior is one of the key trends in recent biology. Comparative behavioral studies of different species require developing an adequate methodology for every new species. We performed three types of behavioral tests with juveniles of the yellow ground squirrel, a large hibernating rodent. Docility tests with an estimation of the time and intensity of the animal locomotor activity in a bag showed low repeatability and appeared to be inappropriate for this species since most of the pups demonstrated freezing behavior. Boldness tests with an action camera installed near the squirrel’s burrow were more promising, but the juveniles still did not emerge aboveground for a long time after they escaped to their burrows, remained vigilant, and did not explore the new object. Test results indicate that yellow ground squirrels are very shy and cautious, which may be a persistent behavioral trait of this species.

360-degree virtual reality video to teach neonatal resuscitation: an exploratory development study

Simulation is an effective training method for neonatal resuscitation (NR). However, the limitations brought about by the COVID-19 pandemic, and other resource constraints, have necessitated exploring alternatives. Virtual reality (VR), particularly 360-degree VR videos, have gained attraction in medical training due to their immersive qualities. The primary objectives of the study were to produce a high quality 360-degree virtual reality (VR) video capturing NR simulation and to determine if it could be an acceptable adjunct to teach NR. The secondary objective was to determine which aspects of NR could benefit from the incorporation of such a video in training. This was an exploratory development study. The first part consisted of producing the video using a GoPro action camera, Adobe Premiere Pro, and Unity Editor. In the second part participants were recruited, based on level of experience, to watch the video and answer questionnaires to determine acceptability (user experience and cognitive load) and aspects of NR which could benefit from the video. The video was successfully developed. Forty-six participants showed a strong general appreciation. User experience revealed high means (> 6) in the positive subscales and low means (< 4) for immersion side effect, with no difference between groups. Cognitive load was higher than anticipated. Participants indicated that this video could be effective for teaching crisis resource management principles, human and environment interactions, and procedural skills. The 360-degree VR video could be a potential new simulation adjunct for NR. Future studies are needed to evaluate learning outcomes of such videos.

QUANTIFICATION OF KINEMATIC AND MOTOR VARIABLES IN PARACANOEING THROUGH THE ACTION CAMERA: A STUDY EXPLORATORY

Paracanoeing is a speed boating sport, practiced by men and women, who seek to cover a distance of 200 meters in the shortest possible time. To generate the propulsion of the vessel and overcome the aerodynamic and hydrodynamic drags, the athletes perform the stroke motor gesture, which consists of bilateral cyclic and symmetrical movements with a two-blade paddle. Since, the motor performance in the modality can be influenced by motor aspects related to rowing and the athlete's ability to produce and maintain high speed indexes. Thus, the objective of the present study was to explore the different ways of quantifying variables through GPS and action camera technologies, as well as programs that systematize this information, being able to transform it into graphs through commonly used statistical programs. With this, it is possible to perform the interpretation and quantification of the variables through technological equipment, GPS and action camera, the data of a volunteer, a beginner athlete of paracanoeing, were analyzed, the speed, paddling index, paddling phases were analyzed ( water and air) and paddling frequency. In this way, the study opens the possibility to use the technique in paracanoeing training programs, with the objective of evaluating the athlete, since through the quantification of the variables it is possible to map the athlete's performance in each variable. Received: 2024/02/17 Reformulated: 2024/04/23 Accepted: 2024/04/23

Investigation of Underwater Photogrammetry Method with Cost-Effective Action Cameras and Comparative Analysis between Reconstructed 3D Point Clouds

Currently, digital cameras and equipment used underwater are often inaccessible to the general public due to their professional-grade quality and high cost. Therefore alternative solutions have been sought that are both cost-effective and suitable for nonprofessional use. A review of the literature shows that researchers primarily use GoPro action cameras, while other action cameras with similar capabilities are rarely used. This study thus examines underwater photogrammetry methods using a widely recognized action camera as a reference and compares it with another camera of similar characteristics as a potential alternative. For a comprehensive temporal analysis in underwater studies, both cameras were used to capture photographic and video imagery, and the resulting 3D point clouds were compared. Comparison criteria included data collection and processing times, point cloud densities, cloud-to-cloud analysis, and assessments of surface density and roughness. Having analysed, the study concluded that the proposed alternative action camera can feasibly be used in underwater photogrammetry.

Immersive drawing: Mark-making beneath the surface

Drawing evokes notions of mark-making on a surface. This paper explores an immersive and fluid kind of mark-making that originates beneath the surface, gradually emerging to reveal what is going on at a deeper level more specifically under the sea. Immersive mark-making arises from below in the form of undercurrents, generating ripples that progressively and delicately make their way up as they glimmer and reflect the light penetrating the surface of the water. Hand gestures contribute to fluid movements that form tiny, frothy bubbles, each containing a reflective space awaiting to be explored and delved into. The fleeting bubbles are so delicate and elusive that their elegant movement can hardly be absorbed by the naked eye. A camera is needed to catch these special moments lasting only a few seconds and which can never be replicated. The current set of immersive charcoal drawings documents the result of swift mark-making gestures below the surface of the Mediterranean Sea captured by a submerged action camera. What the collection of photographs reveals is a universe made of miniscule worlds, drifting and rolling in a playful unchoreographed manner before vanishing into thin air. The poetic passage, both agitated and graceful, allows for drawing in water. It comprises shapes and rhythms, light and darkness, shifts and stillness, aspects we rarely have the time to reflect upon. The ensuing discussion presents the findings emanating from beneath by bringing to the surface of the paper the liquid cosmos generated by drawing in water.

Heuristics and semantic spaces for the analysis of students’ work in mathematical problem solving

In this research, our objective is to characterize the problem-solving procedures of primary and lower secondary students when they solve problems in real class conditions. To do so, we rely first on the concept of heuristics. As this term is very polysemic, we exploit the definition proposed by Rott (2014) to develop a coding manual and thus analyze students’ procedures. Then, we interpret the results of these analyses in a qualitative way by mobilizing the concept of semantic space (Poitrenaud, 1998). This detailed analysis of students’ procedures is made possible by collecting audiovisual data as close as possible to the students’ work using an action camera mounted on the students’ heads. We thus succeed in highlighting three different investigation profiles that we have named explorer, butterfly, and prospector. Our first results tend to show a correlation with these profiles and the success in problem-solving, yet this would need more investigation.

A wearable three-axis force sensor based on deep learning technology for plantar measurement

A challenge in the measurement of plantar shear stress measurement is the absence of wearable and comfortable monitoring devices. In this paper, a wearable three-axis force sensor based on end-to-end deep learning technology is developed for plantar shear stress monitoring. The sensor can convert the three-axis stress it experiences into image signals through the deformation of the pattern layer embedded in the elastomeric probe. These images and three-axis stresses are fed into a convolution neural network for training, establishing a mapping relationship between images and three-axis forces. The prepared sensor has the advantages of small size (1.62 cm3), light weight (1.68 g), and flexibility, which make it possible to form a sensor array in the shoe without affecting the comfort of the wearer. By synchronizing with an action camera, the sensor array can provide real-time three-axis stress in different plantar regions for each frame of motions such as walking, running, playing football or basketball. This work contributes to the advancement of gait analysis, the clinical diagnosis of foot diseases, and the evaluation of athletic training.

Two-Dimensional Analysis of Lower Extremities to Predict Best Initial Condition on Marching Movement

Himpunan Mahasiswa Mesin ITB has a specific type of marching that follows a march song’s rhythm called “derap”. In practice, derap was a hard move to maintain and unify in a group. This difficulty was observed by noting the differences that occur between the subjects' initial and middle march motion conditions. The most noticeable difference was the height of the group. This study intends to observe and find how the differences vary in this motion. The differences that occur corresponded to the relative knee and ankle angle. This study utilized five active markers to reconstruct the lower extremities' activity during motion, one action camera at 60 fps, and a workstation. This study was also constrained by using three cycles of march song at 115 bpm. Direct Linear Transformation was used to obtain the intrinsic factor of the camera to reconstruct the motion. Evaluating the angular kinematic parameters of relative knee and ankle angle, the authors found the relative knee angle has increased from 110° to around 130°, which happened at the second song cycle and kept stable at the rest of the song. This result brought the conclusion that more extended derap motion would tend to a steady condition of relative knee angle and suggested that the subject should begin with 130° of relative knee angle to give less effort on marching.

Benefit of an action camera in endoscopy education for medical students under COVID-19

BackgroundEndoscopy is an important form of clinical gastroenterology education because it gives students the opportunity to learn about diagnosis procedures and even treatment. During the COVID-19 pandemic, medical students were observed from outside the endoscopy room due to the risk of airborne infection. In this study, we investigated the efficacy of combining endoscopy education with doctor’s-eye-view videos of the procedure obtained using live-action cameras (GoPro®).MethodsFrom February to May 2021, endoscopists wore GoPro Hero8 cameras on their heads to display a doctor’s-eye view video outside the room. The efficacy of the GoPro videos in combination with endoscopic monitoring was evaluated by 15 participating medical students. The participants rated the efficacy on a 5-point scale and commented on the positive and negative points.ResultsA total of 78.6% of participants evaluated the GoPro as good; 57.2% answered that it increased their understanding, with 71.4% stating that it increased their understanding of procedures in particular. A total of 85.7% of the students answered that their interest in endoscopy had increased, and 85.7% evaluated the benefit of the GoPro videos as good. In addition, 64.3% answered that the method was effective in preventing COVID-19 infection. Education using GoPro videos enabled students to feel as if they were conducting the endoscopy themselves and enabled them to concentrate on learning.ConclusionsPractical endoscopic education using a GoPro is an effective educational tool that not only increases understanding of endoscopic practice but also stimulates students’ interest and awareness of their future as doctors.

Faster R-CNN Structure for Computer Vision-based Road Pavement Distress Detection

Smart cities can be controlled in all aspects and it is desired to have a structure that is planned to have controllable feedback. Asphalt is generally used as pavement material on roads that provide transportation of vehicles such as cars and buses on the highway. Asphalt material is deformed due to weather conditions, heavy vehicle passage. In the smart city structure, similar deformations should be reported to the relevant unit. In this article, it was tried to determine the deteriorations on the asphalt by selecting the data set obtained from a region with image processing methods and deep learning technique. With the action camera placed in an automobile, a total of 4315 asphalt images with various distortions and without any deterioration were used as dataset. The dataset was classified using a pixel-based Faster Region-based Convolutional Neural Network. Accuracy, precision and sensitivity values were used to make the performance result obtained as a result of classification meaningful. With this proposed method, the average accuracy rate was 93.2%. With these results, an approach that can automatically detect asphalt deterioration in smart city structures has been developed.

Energy Consumption and Cutting Performance of Battery-Powered Chainsaws

The aim of this research is to measure the energy consumption and compare the cutting performance of three types of battery-powered chainsaws (Stihl MSA 200, Stihl MSA 220, and Stihl MSA 300). Tested chainsaws were powered by two different battery packs (Stihl AP 300 S and Stihl AP 500 S) with different energy capacity and output current when cutting two wooden beams, European beech (Fagus sylvatica L.) and black locust (Robinia pseudoacacia L.) with different densities and moisture content. Energy consumption was measured using a multimeter built-in battery charger housing, and consumed energy was noted in watt-hours. Each individual cut was recorded with an action camera at 60 frames per second, and the duration of the cuts recorded in centiseconds was observed on the obtained recordings. Results show significant differences in energy consumption and cutting duration between individual chainsaw/battery combinations. Energy consumption ranged from 2.45 to 4.50 Wh per cut for beech (Stihl MSA300E_AP500S and Stihl MSA220_AP300S, respectively) and from 3.10 to 5.00 Wh per cut for black locust (Stihl MSA300M_AP500S and Stihl MSA220_AP500S, respectively). Duration of the cut ranged from 3.48 to 9.24 s per cut for beech (Stihl MSA300M/H_AP500S and Stihl MSA220_AP300S, respectively) and from 3.74 to 9.35 s per cut for the black locust (Stihl MSA300M_AP500S and Stihl MSA220_AP300S, respectively). In general, it can be concluded that more powerful chainsaw/battery combinations (Stihl MSA300_AP500S) consumed less energy per cut and had better cutting performance (shorter cut). In that regard, more powerful chainsaws had greater cutting efficiency, i.e., more cuts on one charge (battery) but shorter cutting time on one charge. The effect of tree species (fresh-sawed beech/air-dried black locust) on energy consumption and cutting performance is absent on the most powerful chainsaw/battery combinations. Today, lithium-ion batteries still have 70 times less energy density than petrol. However, if there is a significant increase in the energy density of the batteries in the coming times, battery chainsaws will most certainly displace petrol chainsaws from use in the future.

A Drone Study of Sociality in the Finless Porpoise Neophocaena asiaeorientalis in the Ariake Sound, Japan

The social structure of animal populations is a fundamental component of their biology, influencing gene flow, habitat use, competition and co-operation around resources, and communication. However, ecological and social relationships can be challenging to describe in most marine mammals, who spend the majority of their lives underwater. The finless porpoise (Neophocaena asiaeorientalis) is one such cetacean species with a largely unknown social structure. Recent advances in drone technology enable more systematic surveys, photogrammetry, and photo-identification for diverse animal species. The present study aimed to validate new survey methods and provide a preliminary description of the spatiotemporal distribution of free-ranging finless porpoises in the coastal open-sea area of Ariake Sound, Japan. A vertical take-off and landing (VTOL) drone equipped with an action camera yielded GPS location datasets through line and area surveys, covering a total sea area of 120 km2. The results suggest highly flexible and varied aggregation sizes in finless porpoises. Distance analysis across individuals and aggregations revealed a cohesive tendency among groups, compared to solitaries and in pairs. Therefore, the present VTOL drone surveys both elucidated some social aspects of the study population and confirmed the efficacy of these standardized research protocols involving automated, programmed, and repeatable flight missions.

Tree Trunk Detection of Eastern Red Cedar in Rangeland Environment with Deep Learning Technique

Uncontrolled spread of eastern red cedar invades the United States Great Plains prairie ecosystems and lowers biodiversity across native grasslands. The eastern red cedar (ERC) infestations cause significant challenges for ranchers and landowners, including the high costs of removing mature red cedars, reduced livestock forage feed, and reduced revenue from hunting leases. Therefore, a fleet of autonomous ground vehicles (AGV) is proposed to address the ERC infestation. However, detecting the target tree or trunk in a rangeland environment is critical in automating an ERC cutting operation. A tree trunk detection method was developed in this study for ERC trees trained in natural rangeland environments using a deep learning-based YOLOv5 model. An action camera acquired RGB images in a natural rangeland environment. A transfer learning method was adopted, and the YOLOv5 was trained to detect the varying size of the ERC tree trunk. A trained model precision, recall, and average precision were 87.8%, 84.3%, and 88.9%. The model accurately predicted the varying tree trunk sizes and differentiated between trunk and branches. This study demonstrated the potential for using pretrained deep learning models for tree trunk detection with RGB images. The developed machine vision system could be effectively integrated with a fleet of AGVs for ERC cutting. The proposed ERC tree trunk detection models would serve as a fundamental element for the AGV fleet, which would assist in effective rangeland management to maintain the ecological balance of grassland systems.

Vibration quality and ride comfort investigation with transient excitation in a ground vehicle simulator environment

Ground vehicle ride comfort is closely associated with vibration quality as perceived by the human occupants of a vehicle. While outstanding vibration quality is a desirable outcome of the vehicle design process, it is challenging to relate customer perceptions of vibration quality to customer-facing vehicle responses measured inside of a vehicle. Previous work with steady-state vibration has shown that human perception of vibration can be compared to how humans perceive sound, where both the amplitude and frequency of the excitation will alter its perception. Thus, a simple measurement of vibration level is insufficient to predict comfort. In this work, an analysis of human perception of transient vibration is undertaken using a ground vehicle simulator. Physical measurements from inside four different production vehicles are presented including accelerometer, microphone, and action camera recordings. The test event performed is travel over a cleat, with varying vehicle speeds and tire inflation pressures. The physical measurements are used as inputs to the simulator. A comparison is made between the perceived vibration quality of the vehicles as reported by an expert human jury and the physically measured vehicle responses. A ranking of 36 cleat test events for vibratory harshness from best to worst is presented.

Driver Behavior Prediction Based on Environmental Observation Using Fuzzy Hidden Markov Model

The development of autonomous vehicle systems has progressed rapidly in recent years. One challenge that persists is the capability of the autonomous system to respond to human drivers. Human behavior is an integral part of driving; thus, driver behavior determines changing lanes and speed adjustments. However, human behavior is unpredictable and immeasurable. Some traffic accidents are caused due to the erratic behavior of the driver. Although, traffic laws, such as in Indonesia, regulate the use of lanes concerning the vehicle’s speed. The drivers’ behavior in the lane is more likely to be influenced by the regulation. This paper proposes a novel method of predicting drivers’ behavior by utilizing the concept of fuzzy Hidden Markov Model (fuzzy HMM). HMM has been proven reliable in predicting human behavior by observing measurable states to determine unmeasurable hidden states. The use of fuzzy logic is to mimic the way that humans perceive the speeds of other vehicles. The fuzzy logic determines the relative observed state of other vehicles according to the measured velocity of an ego vehicle and the observed state of observed vehicles. Observation data is obtained by equipping an ego vehicle with an action camera. The observed data, in the form of a video, is then discretized every 2 seconds. The resulting sequence of images is processed to determine several variables: speed and state of the observed vehicles (lane position and speed) and the time instance of the observation. The fuzzy HMM is generated based on observational data. A predictor created using fuzzy HMM equipped with a training and prediction algorithm successfully predicts the behavior of other drivers on the road.

Accuracy assessment of high-resolution terrain data produced from UAV images georeferenced with on-board PPK positioning

In recent years, unmanned aerial vehicles (UAVs) have increased rapidly and successfully established themselves as a tool for the rapid collection of high-resolution images as baseline data in land cover studies and topographic mapping. In photogrammetry using the SfM-MVS method of processing captured images, indirect georeferencing of the digital data through ground control points (GCPs) is usually applied. But selecting, marking, and coordinating GCPs in hardto-reach terrains is time-consuming and sometimes dangerous or impossible. The main objective of this study is to evaluate the accuracy of high-resolution topographic data (HRTD) products of photogrammetric processing of PPK-directly georeferenced images by SfM-MVS workflow. Direct and indirect methods of georeferencing digital products are compared. The planimetric and vertical root mean square error (RMSE) in the position of the validation points were calculated by the differences between measured coordinates in dense point clouds, orthophoto mosaics, and terrain surfaces (DSM), and precisely measured coordinates of the validation points by GNSSRTK receivers. The analysis is based on a statistical evaluation of experimental data obtained from a TAROT X6-based hexacopter equipped with two different image sensor configurations: 1) Sony RX0 action camera and 2) Sony A6000 mirrorless camera, and 3) DJI Phantom 4 Pro quadcopter with integrated additional L1-GNSS module for direct georeferencing by PPKmethod. HRTD generation was performed with three block control configurations for each UAV: 1) Indirect georeferencing via GCP only, 2) PPK direct georeferencing without GCP, and 3) PPK georeferencing using one GCP. Our research showed that when using L1-GNSS onboard receivers for PPK-georeferencing without any GCPs, the point cloud&amp;#39;s planimetric accuracy (RMSExy) was from 0.125 to 0.231 cm, depending on the UAV/camera configuration. However, two flight missions produced significant vertical offsets, most likely due to ionospheric disturbances affecting the resolution of phase cycle ambiguities in the single-frequency receivers used. When adding one control point in the PPK georeferencing method, the planimetric and vertical accuracy of the data is comparable to the indirect GCP referencing method. Furthermore, our results show that camera properties (i.e., focal length, resolution, sensor quality) affect the quality and accuracy of digital products. The HRTDs were also evaluated according to the ASPRS (American Photogrammetry and Remote Sensing Society) Standards for Accuracy of Digital Geospatial Data. Analyzing the accuracy of the HRTDs obtained with the experimental UAV/camera configurations for the test area, the present study shows that the PPK-SfM-MVS workflow can provide quality data with a centimeters accuracy of the photogrammetric products.

Impact Evaluation of Score Classes and Annotation Regions in Deep Learning-Based Dairy Cow Body Condition Prediction.

Body condition scoring is a simple method to estimate the energy supply of dairy cattle. Our study aims to investigate the accuracy with which supervised machine learning, specifically a deep convolutional neural network (CNN), can be used to retrieve body condition score (BCS) classes estimated by an expert. We recorded images of animals' rumps in three large-scale farms using a simple action camera. The images were annotated with classes and three different-sized bounding boxes by an expert. A CNN pretrained model was fine-tuned on 12 and 3 BCS classes. Training in 12 classes with a 0 error range, the Cohen's kappa value yielded minimal agreement between the model predictions and ground truth. Allowing an error range of 0.25, we obtained minimum or weak agreement. With an error range of 0.5, we had strong or almost perfect agreement. The kappa values for the approach trained on three classes show that we can classify all animals into BCS categories with at least moderate agreement. Furthermore, CNNs trained on 3 BCS classes showed a remarkably higher proportion of strong agreement than those trained in 12 classes. The prediction precision when training with various annotation region sizes showed no meaningful differences. The weights of our trained CNNs are freely available, supporting similar works.

Spying the dog: Wearable action camera as a tool to understand dog’s behaviour during homing (Carnivora: Canidae)

Automated data collection methods, such as using GPS collars and animal-borne cameras, represent an efficient way of data collection and may be instrumental in the research of animal orientation, including magnetoreception. In this study, we designed a wearable dog action camera (DAC) system for hunting dogs consisting of a Garmin Virb Elite camera housed in the stainless mount attached to the dog harness. We evaluated the DAC’s reliability and potential to capture various behaviour, and we tested the effect of the DAC on the dog’s activity and well-being. We found no significant impact of the DAC on the dogs’ average speed. The tested system was reliable, efficient and safe for dogs. We recorded various behaviours connected to orientation, such as olfactory behaviour and head scanning, the latter described for the first time in domestic dogs. Furthermore, we observed other important behaviours such as hunting, exploration and comfort behaviour. Using wearable action cameras for studying domestic and free-roaming tame animals can bring new opportunities for future behavioural and sensory ecology research.

Online clinical skills education using a wearable action camera for medical students.

The purpose of this study is to analyze medical students' satisfaction and educational effectiveness after real-time online clinical skills education using a wearable action camera. One hundred eighteen third-year medical students participated in this study. The education program consisted of clinical skills (wound management and vascular access procedures) and a medical interview with a simulated patient. The author wore the head-mounted action camera throughout the class. A systematic questionnaire survey was issued to the students after the education program to analyze the educational satisfaction and effectiveness. This education program showed their high educational satisfaction. The online environment was also mostly positive except for some negative comments about video quality. In the educational satisfaction, though the comparative satisfaction with the face-to-face skill class was slightly lower, the rest of the questionnaire items also showed high relative satisfaction. Online video education using a wearable camera could be a good modality that can replace a face-to-face class when it is impossible to open the class.

Benchmarking laser scanning and terrestrial photogrammetry to extract forest inventory parameters in a complex temperate forest

National forest inventories (NFI) are important for the assessment of the state and development of forests. Traditional NFIs often rely on statistical sampling approaches as well as expert assessment which may suffer from observer bias and may lack robustness for time series analysis. Over the course of the last decade, close-range remote sensing techniques such as terrestrial and mobile laser scanning became ever more established for the assessment of three-dimensional (3D) forest structure. With the ongoing trend to make the systems smaller, easier to use and more efficient, the pathway is being opened for an operational inclusion of such devices within the framework of an NFI to support the traditional field assessment. Close-range remote sensing could potentially speed up field inventory work as well as increase the area in which certain parameters are assessed. Benchmarks are needed to evaluate the performance of different close-range remote sensing devices and approaches, both in terms of efficiency as well as accuracy. In this study we evaluate the performance of two terrestrial (TLS), one handheld mobile (PLS) and two drone based (UAVLS) laser scanning systems to detect trees and extract the diameter at breast height (DBH) in three plots with a steep gradient in tree and understorey vegetation density. As a novelty, we also tested the acquisition of 3D point-clouds using a low-cost action camera (GoPro) in conjunction with the Structure from Motion (SfM) technique and compared its performance with those of the more costly LiDAR devices. Among the many parameters evaluated in traditional NFIs, the focus of the performance evaluation of this study is set on the automatic tree detection and DBH extraction. The results showed that TLS delivers the highest tree detection rate (TDR) of up to 94.6% under leaf-off and up to 82% under leaf-on conditions and a relative RMSE (rRMSE) for the DBH extraction between 2.5 and 9%, depending on the undergrowth complexity. The tested PLS system (leaf-on) achieved a TDR of up to 80% with an rRMSE between 3.7 and 5.8%. The tested UAVLS systems showed lowest TDR of less than 77% under leaf-off and less than 37% under leaf-on conditions. The novel GoPro approach achieved a TDR of up to 53% under leaf-on conditions. The reduced TDR can be explained by the reduced area coverage due to the chosen circular acquisition path taken with the GoPro approach. The DBH extraction performance on the other hand is comparable to those of the LiDAR devices with an rRMSE between 2 and 9%. Further benchmarks are needed in order to fully assess the applicability of these systems in the framework of an NFI. Especially the robustness under varying forest conditions (seasonality) and over a broader range of forest types and canopy structure has to be evaluated.