Accuracy of identifications of mammal species from camera trap images: A northern Australian case study

Camera traps are a powerful and increasingly popular tool for mammal research, but like all survey methods, they have limitations. Identifying animal species from images is a critical component of camera trap studies, yet while researchers recognize constraints with experimental design or camera technology, image misidentification is still not well understood. We evaluated the effects of a species’ attributes (body mass and distinctiveness) and individual observer variables (experience and confidence) on the accuracy of mammal identifications from camera trap images. We conducted an Internet‐based survey containing 20 questions about observer experience and 60 camera trap images to identify. Images were sourced from surveys in northern Australia and included 25 species, ranging in body mass from the delicate mouse (Pseudomys delicatulus, 10 g) to the agile wallaby (Macropus agilis, >10 kg). There was a weak relationship between the accuracy of mammal identifications and observer experience. However, accuracy was highest (100%) for distinctive species (e.g. Short‐beaked echidna [Tachyglossus aculeatus]) and lowest (36%) for superficially non‐distinctive mammals (e.g. rodents like the Pale field‐rat [Rattus tunneyi]). There was a positive relationship between the accuracy of identifications and body mass. Participant confidence was highest for large and distinctive mammals, but was not related to participant experience level. Identifications made with greater confidence were more likely to be accurate. Unreliability in identifications of mammal species is a significant limitation to camera trap studies, particularly where small mammals are the focus, or where similar‐looking species co‐occur. Integration of camera traps with conventional survey techniques (e.g. live‐trapping), use of a reference library or computer‐automated programs are likely to aid positive identifications, while employing a confidence rating system and/or multiple observers may lead to a collection of more robust data. Although our study focussed on Australian species, our findings apply to camera trap studies globally.     

Determining the efficacy of camera traps,live capture traps,and detection dogs for locating cryptic small mammal species

Metal box (e.g., Elliott, Sherman) traps and remote cameras are two of the most commonly employed methods presently used to survey terrestrial mammals. However, their relative efficacy at accurately detecting cryptic small mammals has not been adequately assessed. The present study therefore compared the effectiveness of metal box (Elliott) traps and vertically oriented, close range, white flash camera traps in detecting small mammals occurring in the Scenic Rim of eastern Australia. We also conducted a preliminary survey to determine effectiveness of a conservation detection dog (CDD) for identifying presence of a threatened carnivorous marsupial, Antechinus arktos, in present‐day and historical locations, using camera traps to corroborate detections. 200 Elliott traps and 20 white flash camera traps were set for four deployments per method, across a site where the target small mammals, including A. arktos, are known to occur. Camera traps produced higher detection probabilities than Elliott traps for all four species. Thus, vertically mounted white flash cameras were preferable for detecting the presence of cryptic small mammals in our survey. The CDD, which had been trained to detect A. arktos scat, indicated in total 31 times when deployed in the field survey area, with subsequent camera trap deployments specifically corroborating A. arktos presence at 100% (3) indication locations. Importantly, the dog indicated twice within Border Ranges National Park, where historical (1980s–1990s) specimen‐based records indicate the species was present, but extensive Elliott and camera trapping over the last 5–10 years have resulted in zero A. arktos captures. Camera traps subsequently corroborated A. arktos presence at these sites. This demonstrates that detection dogs can be a highly effective means of locating threatened, cryptic species, especially when traditional methods are unable to detect low‐density mammal populations.     

An approach to rapid processing of camera trap images with minimal human input

1. Camera traps have become an extensively utilized tool in ecological research, but the manual processing of images created by a network of camera traps rapidly becomes an overwhelming task, even for small camera trap studies.
2. We used transfer learning to create convolutional neural network (CNN) models for identification and classification. By utilizing a small dataset with an average of 275 labeled images per species class, the model was able to distinguish between species and remove false triggers.
3. We trained the model to detect 17 object classes with individual species identification, reaching an accuracy up to 92% and an average F1 score of 85%. Previous studies have suggested the need for thousands of images of each object class to reach results comparable to those achieved by human observers; however, we show that such accuracy can be achieved with fewer images.
4. With transfer learning and an ongoing camera trap study, a deep learning model can be successfully created by a small camera trap study. A generalizable model produced from an unbalanced class set can be utilized to extract trap events that can later be confirmed by human processors.

     

Software for minimalistic data management in large camera trap studies

The use of camera traps is now widespread and their importance in wildlife studies is well understood. Camera trap studies can produce millions of photographs and there is a need for a software to help manage photographs efficiently. In this paper, we describe a software system that was built to successfully manage a large behavioral camera trap study that produced more than a million photographs. We describe the software architecture and the design decisions that shaped the evolution of the program over the study's three year period. The software system has the ability to automatically extract metadata from images, and add customized metadata to the images in a standardized format. The software system can be installed as a standalone application on popular operating systems. It is minimalistic, scalable and extendable so that it can be used by small teams or individual researchers for a broad variety of camera trap studies.     

Community structure and diversity of tropical forest mammals: data from a global camera trap network

Terrestrial mammals are a key component of tropical forest communities as indicators of ecosystem health and providers of important ecosystem services. However, there is little quantitative information about how they change with local, regional and global threats. In this paper, the first standardized pantropical forest terrestrial mammal community study, we examine several aspects of terrestrial mammal species and community diversity (species richness, species diversity, evenness, dominance, functional diversity and community structure) at seven sites around the globe using a single standardized camera trapping methodology approach. The sites-located in Uganda, Tanzania, Indonesia, Lao PDR, Suriname, Brazil and Costa Rica-are surrounded by different landscape configurations, from continuous forests to highly fragmented forests. We obtained more than 51 000 images and detected 105 species of mammals with a total sampling effort of 12 687 camera trap days. We find that mammal communities from highly fragmented sites have lower species richness, species diversity, functional diversity and higher dominance when compared with sites in partially fragmented and continuous forest. We emphasize the importance of standardized camera trapping approaches for obtaining baselines for monitoring forest mammal communities so as to adequately understand the effect of global, regional and local threats and appropriately inform conservation actions.     

A comparison of camera trap and permanent recording video camera efficiency in wildlife underpasses

In the current context of biodiversity loss through habitat fragmentation, the effectiveness of wildlife crossings, installed at great expense as compensatory measures, is of vital importance for ecological and socio‐economic actors. The evaluation of these structures is directly impacted by the efficiency of monitoring tools (camera traps…), which are used to assess the effectiveness of these crossings by observing the animals that use them. The aim of this study was to quantify the efficiency of camera traps in a wildlife crossing evaluation. Six permanent recording video systems sharing the same field of view as six Reconyx HC600 camera traps installed in three wildlife underpasses were used to assess the exact proportion of missed events (event being the presence of an animal within the field of view), and the error rate concerning underpass crossing behavior (defined as either Entry or Refusal). A sequence of photographs was triggered by either animals (true trigger) or artefacts (false trigger). We quantified the number of false triggers that had actually been caused by animals that were not visible on the images (“false” false triggers). Camera traps failed to record 43.6% of small mammal events (voles, mice, shrews, etc.) and 17% of medium‐sized mammal events. The type of crossing behavior (Entry or Refusal) was incorrectly assessed in 40.1% of events, with a higher error rate for entries than for refusals. Among the 3.8% of false triggers, 85% of them were “false” false triggers. This study indicates a global underestimation of the effectiveness of wildlife crossings for small mammals. Means to improve the efficiency are discussed.     

不同相机布放模式在古田山兽类资源监测中的比较

2010年6月-2011年8月,利用一字型、十字型、随机型和海拔垂直型4种相机布放模式在古田山国家级自然保护区拍摄兽类独立照片(independent photograph,IP)共2551张,能识别物种的兽类IP有1240张,占整体IP的48.6%,不能识别物种的兽类(松鼠科以外的啮齿动物)IP1167张,占45.8%;无效IP144张,占5.6%。所拍能识别物种的兽类共有14种,隶属4目9科,其中黑麂(Muntiacus crinifrons)为国家Ⅰ级保护动物,鬛羚(Naemo-rhedus sumatraensis)和黑熊(Selenarctos thibetanus)为国家Ⅱ级保护动物。在种-监测日曲线当中,4种监测模式随监测日的不断延长,兽类种类大体呈现半年左右趋于稳定的规律,建议监测时间以半年为一个周期较为合适。研究发现,兽类IP和单相机拍摄率在一字型、十字型和随机型之间不存在显著差异(P>0.05),但与海拔垂直型存在极显著差异(P<0.01)。海拔垂直型由于相机分布均匀,在拍摄种类、拍摄率等各项监测指标上都显著高于其他3种模式。     

Studying Large Mammals With Imperfect Detection: Status and Habitat Preferences of Wild Cattle and Large Carnivores in Eastern Cambodia

Studying large mammal species in tropical forests is a conservation challenge with species’ behavior and ecology often increasing the probability of non‐detection during surveys. Consequently, knowledge of the distribution, status, and natural history of many large mammal species in Southeast Asia is limited. I developed occupancy models from camera‐trapping data, thereby accounting for imperfect detection at sampling sites, to clarify the status and habitat requirements of four globally threatened or near threatened large mammals (banteng Bos javanicus, gaur Bos gaurus, dhole Cuon alpinus, and leopard Panthera pardus) in Mondulkiri Protected Forest, eastern Cambodia. Camera traps were operational for >3500 trap nights with 202 photographic encounters of the four study species. Model averaged occupancy estimates were between 5 percent (leopard) and 140 percent (gaur) higher than naive estimates (i.e., proportion of camera‐trap sites species recorded from) thus highlighting the importance of accounting for detectability during conservation surveys. I recommend the use of an occupancy framework when using camera‐trap data to study the status, ecology, and habitat preferences of poorly known and elusive species. The results highlight the importance of mixed deciduous and semi‐evergreen forest for wild cattle in eastern Cambodia and I emphasize that these habitats must be considered in conservation planning across the Lower Mekong Dry Forest Ecoregion.     

Monitoring small and arboreal mammals by camera traps: effectiveness and applications

Camera trapping has been widely applied to studies of medium to large terrestrial mammals, but its application to small arboreal mammals has hardly been tested. We employed camera trapping and other conventional monitoring methods during a mammal survey in a Site of Community Importance located within the Adda North Regional Park (Lombardy, Italy). Camera trapping was particularly successful for monitoring arboreal mammals, allowing the first detection of presence of the invasive grey squirrel (Sciurus carolinensis) in an area occupied by indigenous red squirrels (Sciurus vulgaris) and the collection of a large amount of data on squirrels and common dormice (Muscardinus avellanarius). When triggered, cameras were set to record short video clips (10 to 40 s). More than 400 events were recorded and analysed, mainly from the autumn and winter months. The daily activity pattern of both species displayed a trend from two to three activity peaks in summer to a unimodal pattern in winter, with the peaks of the two species temporally separated. Camera trapping could be a useful method also when applied to monitoring small mammals, particularly endangered arboreal or invasive alien species. For instance, the monitoring of the spread of S. carolinensis is particularly important, where the early detection of new population can be crucial for the conservation of indigenous European species. Camera trapping can be an effective addition to traditional survey methods. It provides a simple non-invasive technique for collecting a large amount of data per device with limited human effort.     

Take Only Photographs,Leave Only Footprints: Novel Applications of Non-Invasive Survey Methods for Rapid Detection of Small,Arboreal Animals

The development of appropriate wildlife survey techniques is essential to promote effective and efficient monitoring of species of conservation concern. Here, we demonstrate the utility of two rapid-assessment, non-invasive methods to detect the presence of elusive, small, arboreal animals. We use the hazel dormouse, Muscardinus avellanarius, a rodent of conservation concern, as our focal species. Prevailing hazel dormouse survey methods are prolonged (often taking months to years to detect dormice), dependent on season and habitat, and/or have low detection rates. Alternatives would be of great use to ecologists who undertake dormouse surveys, especially those assessing the need for mitigation measures, as legally required for building development projects. Camera traps and footprint tracking are well-established tools for monitoring elusive large terrestrial mammals, but are rarely used for small species such as rodents, or in arboreal habitats. In trials of these adapted methods, hazel dormice visited bait stations and were successfully detected by both camera traps and tracking equipment at each of two woodland study sites, within days to weeks of installation. Camera trap images and footprints were of adequate quality to allow discrimination between two sympatric small mammal species (hazel dormouse and wood mouse, Apodemus sylvaticus). We discuss the relative merits of these methods with respect to research aims, funds, time available and habitat.     

Data acquisition and management software for camera trap data: A case study from the TEAM Network

Camera traps and the images they generate are becoming an essential tool for field biologists studying and monitoring terrestrial animals, in particular medium to large terrestrial mammals and birds. In the last five years, camera traps have made the transition to digital technology, where these devices now produce hundreds of instantly available images per month and a large amount of ancillary metadata (e.g., date, time, temperature, image size, etc.). Despite this accelerated pace in the development of digital image capture, field biologists still lack adequate software solutions to process and manage the increasing amount of information in a cost efficient way. In this paper we describe a software system that we have developed, called DeskTEAM, to address this issue. DeskTEAM has been developed in the context of the Tropical Ecology Assessment and Monitoring Network (TEAM), a global network that monitors terrestrial vertebrates. We describe the software architecture and functionality and its utility in managing and processing large amounts of digital camera trap data collected throughout the global TEAM network. DeskTEAM incorporates software features and functionality that make it relevant to the broad camera trapping community. These include the ability to run the application locally on a laptop or desktop computer, without requiring an Internet connection, as well as the ability to run on multiple operating systems; an intuitive navigational user interface with multiple levels of detail (from individual images, to whole groups of images) which allows users to easily manage hundreds or thousands of images; ability to automatically extract EXIF and custom metadata information from digital images to increase standardization; availability of embedded taxonomic lists to allow users to easily tag images with species identities; and the ability to export data packages consisting of data, metadata and images in standardized formats so that they can be transferred to online data warehouses for easy archiving and dissemination. Lastly, building these software tools for wildlife scientists provides valuable lessons for the ecoinformatics community.     

利用红外相机对四川白水河国家级自然保护区鸟兽资源的初步调查

2017年5月至2018年5月, 我们在四川白水河国家级自然保护区内设置红外相机对地面活动鸟兽进行了初步调查。布设在24个位点的24台相机累计工作3,832天, 共获得可识别物种的独立有效照片535张。经鉴定, 兽类有4目10科17种, 鸟类有2目4科10种。其中, 国家I级重点保护野生动物5种, 国家II级重点保护野生动物8种, 中国豪猪(Hystrix hodgsoni)、宝兴歌鸫(Turdus mupinensis)和黑顶噪鹛(Trochalopteron affine)为保护区新记录种, 而大熊猫(Ailuropoda melanoleuca)为汶川地震后首次拍到。兽类中, 花面狸(Paguma larvata)、黄喉貂(Martes flavigula)和中华斑羚(Naemorhedus griseus) 3种动物的独立有效照片总数占全部兽类独立有效照片数的50.2%。鸟类中, 血雉(Ithaginis cruentus)和红腹角雉(Tragopan temminckii)的独立有效照片总数占全部鸟类独立有效照片数的91.6%。本研究为白水河国家级自然保护区野生动物资源管理和保护提供了参考依据。     

Linking camera‐trap data to taxonomy: Identifying photographs of morphologically similar chipmunks

Remote cameras are a common method for surveying wildlife and recently have been promoted for implementing large‐scale regional biodiversity monitoring programs. The use of camera‐trap data depends on the correct identification of animals captured in the photographs, yet misidentification rates can be high, especially when morphologically similar species co‐occur, and this can lead to faulty inferences and hinder conservation efforts. Correct identification is dependent on diagnosable taxonomic characters, photograph quality, and the experience and training of the observer. However, keys rooted in taxonomy are rarely used for the identification of camera‐trap images and error rates are rarely assessed, even when morphologically similar species are present in the study area. We tested a method for ensuring high identification accuracy using two sympatric and morphologically similar chipmunk (Neotamias) species as a case study. We hypothesized that the identification accuracy would improve with use of the identification key and with observer training, resulting in higher levels of observer confidence and higher levels of agreement among observers. We developed an identification key and tested identification accuracy based on photographs of verified museum specimens. Our results supported predictions for each of these hypotheses. In addition, we validated the method in the field by comparing remote‐camera data with live‐trapping data. We recommend use of these methods to evaluate error rates and to exclude ambiguous records in camera‐trap datasets. We urge that ensuring correct and scientifically defensible species identifications is incumbent on researchers and should be incorporated into the camera‐trap workflow.     

Camera trapping mammals in the scrubland’s of the Cape Floristic Kingdom—the importance of effort,spacing and trap placement

As a non-invasive monitoring method camera traps are noted as being an effective, accurate and rapid means of compiling species richness estimates of medium to large terrestrial mammals. However, crucial elements of camera trap survey design are rarely empirically addressed, which has raised the need for both a standardised and optimised camera trapping protocol. Our study confirms that an appropriate camera placement buffer and targeting areas of animal activity, contributes to more complete species richness estimates as well as significantly reducing the rate of false trigger events. However, attaining the required survey effort in terms of camera days was the most important factor in providing accurate species richness estimates. Our results suggest that reliable estimates of species richness can be achieved in open scrubland when cameras are spaced 1 × 1 km apart and left in the targeted area until a survey effort of a 1000 camera days is realised.     

广西恩城国家级自然保护区兽类和鸟类多样性红外相机监测初报

2019年12月—2021年1月,利用红外相机技术在广西恩城国家级自然保护区内布设了58个红外相机监测位点,对保护区内的兽类及地栖性鸟类多样性开展监测。本次监测共完成17 332个相机工作日,累计获得独立有效照片6 680张。共鉴定出兽类5目13科20种,鸟类9目20科61种,其中黑叶猴(Trachypithecus francoisi)为国家一级重点保护野生动物,26种为国家二级重点保护野生动物。被中国脊椎动物红色名录评估为濒危(EN)的有2种,易危(VU) 6种,近危(NT) 19种。物种相对多度指数(RAI)最高的兽类是小泡巨鼠(Leopoldamys edwardsi),鸟类是白鹇(Lophura nycthemera)。本次监测初步掌握了广西恩城国家级自然保护区内的兽类及地栖性鸟类的种类组成和相对多度指数,为后续开展野生动物研究与保护管理工作提供基础资料。     

Animal Scanner: Software for classifying humans,animals, and empty frames in camera trap images

Camera traps are a popular tool to sample animal populations because they are noninvasive, detect a variety of species, and can record many thousands of animal detections per deployment. Cameras are typically set to take bursts of multiple photographs for each detection and are deployed in arrays of dozens or hundreds of sites, often resulting in millions of photographs per study. The task of converting photographs to animal detection records from such large image collections is daunting, and made worse by situations that generate copious empty pictures from false triggers (e.g., camera malfunction or moving vegetation) or pictures of humans. We developed computer vision algorithms to detect and classify moving objects to aid the first step of camera trap image filtering—separating the animal detections from the empty frames and pictures of humans. Our new work couples foreground object segmentation through background subtraction with deep learning classification to provide a fast and accurate scheme for human–animal detection. We provide these programs as both Matlab GUI and command prompt developed with C++. The software reads folders of camera trap images and outputs images annotated with bounding boxes around moving objects and a text file summary of results. This software maintains high accuracy while reducing the execution time by 14 times. It takes about 6 seconds to process a sequence of ten frames (on a 2.6 GHZ CPU computer). For those cameras with excessive empty frames due to camera malfunction or blowing vegetation automatically removes 54% of the false‐triggers sequences without influencing the human/animal sequences. We achieve 99.58% on image‐level empty versus object classification of Serengeti dataset. We offer the first computer vision tool for processing camera trap images providing substantial time savings for processing large image datasets, thus improving our ability to monitor wildlife across large scales with camera traps.     

Comparing methods for sampling large- and medium-sized mammals: camera traps and track plots

Activities involving fauna monitoring are usually limited by the lack of resources; therefore, the choice of a proper and efficient methodology is fundamental to maximize the cost–benefit ratio. Both direct and indirect methods can be used to survey mammals, but the latter are preferred due to the difficulty to come in sight of and/or to capture the individuals, besides being cheaper. We compared the performance of two methods to survey medium and large-sized mammal: track plot recording and camera trapping, and their costs were assessed. At Jataí Ecological Station (S21°31'15"–W47°34'42"-Brazil) we installed ten camera traps along a dirt road directly in front of ten track plots, and monitored them for 10 days. We cleaned the plots, adjusted the cameras, and noted down the recorded species daily. Records taken by both methods showed they sample the local richness in different ways (Wilcoxon, T = 231; p ;; 0.01). The track plot method performed better on registering individuals whereas camera trapping provided records which permitted more accurate species identification. The type of infra-red sensor camera used showed a strong bias towards individual body mass (R ² = 0.70; p = 0.017), and the variable expenses of this method in a 10-day survey were estimated about 2.04 times higher compared to track plot method; however, in a long run camera trapping becomes cheaper than track plot recording. Concluding, track plot recording is good enough for quick surveys under a limited budget, and camera trapping is best for precise species identification and the investigation of species details, performing better for large animals. When used together, these methods can be complementary.     

Studying the quantity component of seed dispersal effectiveness from exclosure treatments and camera trapping

The quantity component of effectiveness of seed dispersal by animals is determined by two events: fruit removal (intensity of the interaction) and animal visitation to the plant (frequency of interactions). Considering dispersal of Prosopis flexuosa seeds as case study, this work aimed at investigating the strengths and weaknesses of the two methods for assessing the quantity component of seed dispersal effectiveness: exclosures and camera traps. Prosopis fruits were offered for 48 hr. Exclosure treatments were performed using two types of wire‐screen cages, allowing access to ants (“closed exclosure”) and to small mammals up to 100 g (“open to small mammals”), and a treatment without exclosure (“open to all removers”). The camera trapping experiment was carried out using vertically oriented cameras placed at approximately 1.80 m height and focused on the fruits. The cameras were set in “motion detect mode,” taking series of three consecutive photographs. The exclosures largely allowed estimation of fruit removal by size‐based groups of animals, but did not provide information on species identity. In contrast, camera traps were able to identify all visitors to species level and could not only determine the number of visits by each species but also the proportion of visits, which resulted in removal of fruits. Camera trapping allowed discriminating among small mammals playing different roles, without underestimating fruit removal by scatter‐hoarding species. The quality of estimation of the quantity component of seed dispersal is remarkably better when the camera trapping method is applied. Additional information obtained, such as activity patterns of visitors, can contribute to a better understanding of the seed dispersal process.     

Camera Traps Can Be Heard and Seen by Animals

Camera traps are electrical instruments that emit sounds and light. In recent decades they have become a tool of choice in wildlife research and monitoring. The variability between camera trap models and the methods used are considerable, and little is known about how animals respond to camera trap emissions. It has been reported that some animals show a response to camera traps, and in research this is often undesirable so it is important to understand why the animals are disturbed. We conducted laboratory based investigations to test the audio and infrared optical outputs of 12 camera trap models. Camera traps were measured for audio outputs in an anechoic chamber; we also measured ultrasonic (n = 5) and infrared illumination outputs (n = 7) of a subset of the camera trap models. We then compared the perceptive hearing range (n = 21) and assessed the vision ranges (n = 3) of mammals species (where data existed) to determine if animals can see and hear camera traps. We report that camera traps produce sounds that are well within the perceptive range of most mammals’ hearing and produce illumination that can be seen by many species.