A novel method to reduce time investment when processing videos from camera trap studies.
ABSTRACT: Camera traps have proven very useful in ecological, conservation and behavioral research. Camera traps non-invasively record presence and behavior of animals in their natural environment. Since the introduction of digital cameras, large amounts of data can be stored. Unfortunately, processing protocols did not evolve as fast as the technical capabilities of the cameras. We used camera traps to record videos of Eurasian beavers (Castor fiber). However, a large number of recordings did not contain the target species, but instead empty recordings or other species (together non-target recordings), making the removal of these recordings unacceptably time consuming. In this paper we propose a method to partially eliminate non-target recordings without having to watch the recordings, in order to reduce workload. Discrimination between recordings of target species and non-target recordings was based on detecting variation (changes in pixel values from frame to frame) in the recordings. Because of the size of the target species, we supposed that recordings with the target species contain on average much more movements than non-target recordings. Two different filter methods were tested and compared. We show that a partial discrimination can be made between target and non-target recordings based on variation in pixel values and that environmental conditions and filter methods influence the amount of non-target recordings that can be identified and discarded. By allowing a loss of 5% to 20% of recordings containing the target species, in ideal circumstances, 53% to 76% of non-target recordings can be identified and discarded. We conclude that adding an extra processing step in the camera trap protocol can result in large time savings. Since we are convinced that the use of camera traps will become increasingly important in the future, this filter method can benefit many researchers, using it in different contexts across the globe, on both videos and photographs.
Project description:BACKGROUND:Carrizo Plain National Monument (San Joaquin Desert, California, USA) is home to many threatened and endangered species including the blunt-nosed leopard lizard (Gambelia sila). Vegetation is dominated by annual grasses, and shrubs such as Mormon tea (Ephedra californica), which is of relevance to our target species, the federally listed blunt-nosed leopard lizard, and likely also provides key ecosystem services. We used relatively nonintrusive camera traps, or trail cameras, to capture interactions between animals and these shrubs using a paired shrub-open deployment. Cameras were placed within the shrub understory and in open microhabitats at ground level to estimate animal activity and determine species presence. FINDINGS:Twenty cameras were deployed from April 1st, 2015 to July 5th, 2015 at paired shrub-open microsites at three locations. Over 425,000 pictures were taken during this time, of which 0.4 % detected mammals, birds, insects, and reptiles including the blunt-nosed leopard lizard. Trigger rate was very high on the medium sensitivity camera setting in this desert ecosystem, and rates did not differ between microsites. CONCLUSIONS:Camera traps are an effective, less invasive survey method for collecting data on the presence or absence of desert animals in shrub and open microhabitats. A more extensive array of cameras within an arid region would thus be an effective tool to estimate the presence of desert animals and potentially detect habitat use patterns.
Project description:Camera traps is an important wildlife inventory tool for estimating species diversity at a site. Knowing what minimum trapping effort is needed to detect target species is also important to designing efficient studies, considering both the number of camera locations, and survey length. Here, we take advantage of a two-year camera trapping dataset from a small (24-ha) study plot in Gutianshan National Nature Reserve, eastern China to estimate the minimum trapping effort actually needed to sample the wildlife community. We also evaluated the relative value of adding new camera sites or running cameras for a longer period at one site. The full dataset includes 1727 independent photographs captured during 13,824 camera days, documenting 10 resident terrestrial species of birds and mammals. Our rarefaction analysis shows that a minimum of 931 camera days would be needed to detect the resident species sufficiently in the plot, and c. 8700 camera days to detect all 10 resident species. In terms of detecting a diversity of species, the optimal sampling period for one camera site was c. 40, or long enough to record about 20 independent photographs. Our analysis of evaluating the increasing number of additional camera sites shows that rotating cameras to new sites would be more efficient for measuring species richness than leaving cameras at fewer sites for a longer period.
Project description:Identification of pollen vectors is a fundamental objective of pollination biology. The foraging and social behavior of these pollinators has profound effects on plant mating, making quantification of their behavior critical for understanding the ecological and evolutionary consequences of different pollinators for the plants they visit. However, accurate quantification of visitation may be problematic, especially for shy animals and/or when the temporal and spatial scale of observation desired is large. Sophisticated heat- and movement-triggered motion-sensor cameras ("camera trapping") provide new, underutilized tools to address these challenges. However, to date, there has been no rigorous evaluation of the sampling considerations needed for using camera trapping in pollination research.We measured the effectiveness of camera trapping for identifying vertebrate visitors and quantifying their visitation rates and foraging behavior on Banksia menziesii (Proteaceae). Multiple still cameras (Reconyx HC 500) and a video camera (Little Acorn LTL5210A) were deployed.From 2,753 recorded visits by vertebrates, we identified five species of nectarivorous honeyeater (Meliphagidae) and the honey possum (Tarsipedidae), with significant variation in the species composition of visitors among inflorescences. Species of floral visitor showed significant variation in their time of peak activity, duration of visits, and numbers of flowers probed per visit. Where multiple cameras were deployed on individual inflorescences, effectiveness of individual still cameras varied from 15% to 86% of all recorded visits. Methodological issues and solutions, and the future uses of camera traps in pollination biology, are discussed. Conclusions and wider implications: Motion-triggered cameras are promising tools for the quantification of vertebrate visitation and some aspects of behavior on flowers. However, researchers need to be mindful of the variation in effectiveness of individual camera traps in detecting animals. Pollinator studies using camera traps are in their infancy, and the full potential of this developing technology is yet to be realized.
Project description:The use of camera traps as a tool for studying wildlife populations is commonplace. However, few have considered how the number of detections of wildlife differ depending upon the number of camera traps placed at cameras-sites, and how this impacts estimates of occupancy and community composition. During December 2015-February 2016, we deployed four camera traps per camera-site, separated into treatment groups of one, two, and four camera traps, in southern Illinois to compare whether estimates of wildlife community metrics and occupancy probabilities differed among survey methods. The overall number of species detected per camera-site was greatest with the four-camera survey method (P<0.0184). The four-camera survey method detected 1.25 additional species per camera-site than the one-camera survey method, and was the only survey method to completely detect the ground-dwelling silvicolous community. The four-camera survey method recorded individual species at 3.57 additional camera-sites (P = 0.003) and nearly doubled the number of camera-sites where white-tailed deer (Odocoileus virginianus) were detected compared to one- and two-camera survey methods. We also compared occupancy rates estimated by survey methods; as the number of cameras deployed per camera-site increased, occupancy estimates were closer to naïve estimates, detection probabilities increased, and standard errors of detection probabilities decreased. Additionally, each survey method resulted in differing top-ranked, species-specific occupancy models when habitat covariates were included. Underestimates of occurrence and misrepresented community metrics can have significant impacts on species of conservation concern, particularly in areas where habitat manipulation is likely. Having multiple camera traps per site revealed significant shortcomings with the common one-camera trap survey method. While we realize survey design is often constrained logistically, we suggest increasing effort to at least two camera traps facing opposite directions per camera-site in habitat association studies, and to utilize camera-trap arrays when restricted by equipment availability.
Project description:Conventional cameras rely upon a pixelated sensor to provide spatial resolution. An alternative approach replaces the sensor with a pixelated transmission mask encoded with a series of binary patterns. Combining knowledge of the series of patterns and the associated filtered intensities, measured by single-pixel detectors, allows an image to be deduced through data inversion. In this work we extend the concept of a 'single-pixel camera' to provide continuous real-time video at 10?Hz , simultaneously in the visible and short-wave infrared, using an efficient computer algorithm. We demonstrate our camera for imaging through smoke, through a tinted screen, whilst performing compressive sampling and recovering high-resolution detail by arbitrarily controlling the pixel-binning of the masks. We anticipate real-time single-pixel video cameras to have considerable importance where pixelated sensors are limited, allowing for low-cost, non-visible imaging systems in applications such as night-vision, gas sensing and medical diagnostics.
Project description:Motion triggered camera traps are an increasingly popular tool for wildlife research and can be used to survey for multiple species simultaneously. As with all survey techniques, it is crucial to conduct camera trapping research following study designs that include adequate spatial and temporal replication, and sufficient probability of detecting species presence. The use and configuration of multiple camera traps within a single survey site are understudied considerations that could have a substantial impact on detection probability. Our objective was to test the role that camera number (one, two or three units), and spacing along a linear transect (100 m or 150 m), have on the probability of detecting a species given it is present. From January to March, 2017 we collected data on six mammal species in Maine, USA: coyote (Canis latrans), fisher (Pekania pennanti), American marten (Martes americana), short-tailed weasel (Mustela erminea), snowshoe hare (Lepus americanus), and American red squirrel (Tamiasciurus hudsonicus). We used multi-scale occupancy modelling to compare pooled detection histories of different configuration of five cameras deployed at the same survey site (n = 32), and how the configuration would influence the probability of detecting a species given it was available at the site. Across all six species, we found substantial increases in probability of detection as the number of cameras increased from one to two (22 to 400 percent increase), regardless of the spacing between cameras. For most species the magnitude of the increase was less substantial when adding a third camera (4 to 85 percent increase), with coyote and snowshoe hare showing a pronounced effect. The influence of survey station features also varied by species. We suggest that using pooled data from two or three cameras at a survey site is a cost effective approach to increase detection success over a single camera.
Project description:Practical techniques are required to monitor invasive animals, which are often cryptic and occur at low density. Camera traps have potential for this purpose, but may have problems detecting and identifying small species. A further challenge is how to standardise the size of each camera's field of view so capture rates are comparable between different places and times. We investigated the optimal specifications for a low-cost camera trap for small mammals. The factors tested were 1) trigger speed, 2) passive infrared vs. microwave sensor, 3) white vs. infrared flash, and 4) still photographs vs. video. We also tested a new approach to standardise each camera's field of view. We compared the success rates of four camera trap designs in detecting and taking recognisable photographs of captive stoats (Mustelaerminea), feral cats (Felis catus) and hedgehogs (Erinaceuseuropaeus). Trigger speeds of 0.2-2.1 s captured photographs of all three target species unless the animal was running at high speed. The camera with a microwave sensor was prone to false triggers, and often failed to trigger when an animal moved in front of it. A white flash produced photographs that were more readily identified to species than those obtained under infrared light. However, a white flash may be more likely to frighten target animals, potentially affecting detection probabilities. Video footage achieved similar success rates to still cameras but required more processing time and computer memory. Placing two camera traps side by side achieved a higher success rate than using a single camera. Camera traps show considerable promise for monitoring invasive mammal control operations. Further research should address how best to standardise the size of each camera's field of view, maximise the probability that an animal encountering a camera trap will be detected, and eliminate visible or audible cues emitted by camera traps.
Project description:Camera trapping has become an increasingly widespread tool for wildlife ecologists, with large numbers of studies relying on photo capture rates or presence/absence information. It is increasingly clear that camera placement can directly impact this kind of data, yet these biases are poorly understood. We used a paired camera design to investigate the effect of small-scale habitat features on species richness estimates, and capture rate and detection probability of several mammal species in the Shenandoah Valley of Virginia, USA. Cameras were deployed at either log features or on game trails with a paired camera at a nearby random location. Overall capture rates were significantly higher at trail and log cameras compared to their paired random cameras, and some species showed capture rates as much as 9.7 times greater at feature-based cameras. We recorded more species at both log (17) and trail features (15) than at their paired control cameras (13 and 12 species, respectively), yet richness estimates were indistinguishable after 659 and 385 camera nights of survey effort, respectively. We detected significant increases (ranging from 11-33%) in detection probability for five species resulting from the presence of game trails. For six species detection probability was also influenced by the presence of a log feature. This bias was most pronounced for the three rodents investigated, where in all cases detection probability was substantially higher (24.9-38.2%) at log cameras. Our results indicate that small-scale factors, including the presence of game trails and other features, can have significant impacts on species detection when camera traps are employed. Significant biases may result if the presence and quality of these features are not documented and either incorporated into analytical procedures, or controlled for in study design.
Project description:Camera traps are used increasingly to estimate population density for elusive and difficult to observe species. A standard practice for mammalian surveys is to place cameras on roads, trails, and paths to maximize detections and/or increase efficiency in the field. However, for many species it is unclear whether track-based camera surveys provide reliable estimates of population density.Understanding how the spatial arrangement of camera traps affects population density estimates is of key interest to contemporary conservationists and managers given the rapid increase in camera-based wildlife surveys.We evaluated the effect of camera-trap placement, using several survey designs, on density estimates of a widespread mesopredator, the red fox Vulpes vulpes, over a two-year period in a semi-arid conservation reserve in south-eastern Australia. Further, we used the certainty in the identity and whereabouts of individuals (via GPS collars) to assess how resighting rates of marked foxes affect density estimates using maximum likelihood spatially explicit mark-resight methods.Fox detection rates were much higher at cameras placed on tracks compared with off-track cameras, yet in the majority of sessions, camera placement had relatively little effect on point estimates of density. However, for each survey design, the precision of density estimates varied considerably across sessions, influenced heavily by the absolute number of marked foxes detected, the number of times marked foxes was resighted, and the number of detection events of unmarked foxes.Our research demonstrates that the precision of population density estimates using spatially explicit mark-resight models is sensitive to resighting rates of identifiable individuals. Nonetheless, camera surveys based either on- or off-track can provide reliable estimates of population density using spatially explicit mark-resight models. This underscores the importance of incorporating information on the spatial behavior of the subject species when planning camera-trap surveys.
Project description:Motion-activated wildlife cameras (or "camera traps") are frequently used to remotely and noninvasively observe animals. The vast number of images collected from camera trap projects has prompted some biologists to employ machine learning algorithms to automatically recognize species in these images, or at least filter-out images that do not contain animals. These approaches are often limited by model transferability, as a model trained to recognize species from one location might not work as well for the same species in different locations. Furthermore, these methods often require advanced computational skills, making them inaccessible to many biologists. We used 3 million camera trap images from 18 studies in 10 states across the United States of America to train two deep neural networks, one that recognizes 58 species, the "species model," and one that determines if an image is empty or if it contains an animal, the "empty-animal model." Our species model and empty-animal model had accuracies of 96.8% and 97.3%, respectively. Furthermore, the models performed well on some out-of-sample datasets, as the species model had 91% accuracy on species from Canada (accuracy range 36%-91% across all out-of-sample datasets) and the empty-animal model achieved an accuracy of 91%-94% on out-of-sample datasets from different continents. Our software addresses some of the limitations of using machine learning to classify images from camera traps. By including many species from several locations, our species model is potentially applicable to many camera trap studies in North America. We also found that our empty-animal model can facilitate removal of images without animals globally. We provide the trained models in an R package (MLWIC2: Machine Learning for Wildlife Image Classification in R), which contains Shiny Applications that allow scientists with minimal programming experience to use trained models and train new models in six neural network architectures with varying depths.