A novel system for automatic detection and classification of animal

Matúška, Slavomír; Hudec, Róbert; Benčo, Miroslav; Kamencay, Patrik; Zachariášová, Martina

doi:10.1109/elektro.2014.6847875

Cited by 19 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An earlier work using machine learning methods using edge devices to classify and detect animals automatically was proposed by Matuska et al [62]. They proposed an Automatic System for Animal Recognition (ASFAR).…”

Section: Using Edge Devices For Camera Trap Animal Classificationmentioning

confidence: 99%

An IoT System Using Deep Learning to Classify Camera Trap Images on the Edge

Zualkernan

Dhou

Judas³

et al. 2022

Computers

View full text Add to dashboard Cite

Camera traps deployed in remote locations provide an effective method for ecologists to monitor and study wildlife in a non-invasive way. However, current camera traps suffer from two problems. First, the images are manually classified and counted, which is expensive. Second, due to manual coding, the results are often stale by the time they get to the ecologists. Using the Internet of Things (IoT) combined with deep learning represents a good solution for both these problems, as the images can be classified automatically, and the results immediately made available to ecologists. This paper proposes an IoT architecture that uses deep learning on edge devices to convey animal classification results to a mobile app using the LoRaWAN low-power, wide-area network. The primary goal of the proposed approach is to reduce the cost of the wildlife monitoring process for ecologists, and to provide real-time animal sightings data from the camera traps in the field. Camera trap image data consisting of 66,400 images were used to train the InceptionV3, MobileNetV2, ResNet18, EfficientNetB1, DenseNet121, and Xception neural network models. While performance of the trained models was statistically different (Kruskal–Wallis: Accuracy H(5) = 22.34, p < 0.05; F1-score H(5) = 13.82, p = 0.0168), there was only a 3% difference in the F1-score between the worst (MobileNet V2) and the best model (Xception). Moreover, the models made similar errors (Adjusted Rand Index (ARI) > 0.88 and Adjusted Mutual Information (AMU) > 0.82). Subsequently, the best model, Xception (Accuracy = 96.1%; F1-score = 0.87; F1-Score = 0.97 with oversampling), was optimized and deployed on the Raspberry Pi, Google Coral, and Nvidia Jetson edge devices using both TenorFlow Lite and TensorRT frameworks. Optimizing the models to run on edge devices reduced the average macro F1-Score to 0.7, and adversely affected the minority classes, reducing their F1-score to as low as 0.18. Upon stress testing, by processing 1000 images consecutively, Jetson Nano, running a TensorRT model, outperformed others with a latency of 0.276 s/image (s.d. = 0.002) while consuming an average current of 1665.21 mA. Raspberry Pi consumed the least average current (838.99 mA) with a ten times worse latency of 2.83 s/image (s.d. = 0.036). Nano was the only reasonable option as an edge device because it could capture most animals whose maximum speeds were below 80 km/h, including goats, lions, ostriches, etc. While the proposed architecture is viable, unbalanced data remain a challenge and the results can potentially be improved by using object detection to reduce imbalances and by exploring semi-supervised learning.

show abstract

Section: Using Edge Devices For Camera Trap Animal Classificationmentioning

confidence: 99%

An IoT System Using Deep Learning to Classify Camera Trap Images on the Edge

Zualkernan

Dhou

Judas³

et al. 2022

Computers

View full text Add to dashboard Cite

show abstract

“…We will use our existing system ASFAR [5] and we will extend this system with NIR camera and also we will add new block for object detection and feature extraction in NIR domain mainly used for night capturing and it is illustrated on Figure 3. This extension is highlighted in red area.…”

Section: System Overviewmentioning

confidence: 99%

Local Feature Extraction in the Near Infra-Red Domain for Wildlife Mammals Tracking Purpose

Trnovszký

Hudec

2016

Civil and Environmental Engineering

View full text Add to dashboard Cite

Nowadays, research in area of intelligent transportation system is focused on analysis of traffic flow by computer vision techniques. Keypoint analysis represents detection, modelling and recognition of objects in traffic flow and object tracking as well. The main goal of this paper is to propose new approach to detect and modelling 3D objects that move on road surface. At the beginning, basic methods of object recognition and modelling of 3D object are shortly described. The modified algorithm based on background subtraction and creation of 3D model by depth map is proposed. Finally, the results of developed algorithm are depicted through the last part of this paper.

show abstract

“…Matuska et al [4] proposed an Automatic System For Animal Recognition (ASFAR) making use of Bag of Visual Words for animal detection. SIFT and SURF features of the image were used as key point detectors.…”

Section: Literature Reviewmentioning

confidence: 99%

A vision based human - elepahant collision detection system

Dua

Shukla

Mittal

2015

2015 Third International Conference on Image Information Processing (ICIIP)

View full text Add to dashboard Cite

Elephant intrusion in areas with high human movement can prove lethal for both human beings and elephants.The proposed system seeks to identify elephants with the aid of a Video Camera. The suggested methodology was applied to zones having high intervention of human beings and elephants. Regions with higher human movements like roads were extracted from the initial video frames. This process is followed by detecting motion in the video frame. The objects in the area of motion are then identified as elephant or non-elephant with the help of PHOG features and Support Vector Machines (SVM) classifiers.A dataset constituting of images of elephants and other objects was used for training the proposed algorithm. An overall accuracy of 85.29% was attained when static images containing elephants and other objects were classified. The same approach for detection was applied to identify moving elephants in the video frames. The proposed system endeavors to prevent casualties that occur in areas having high human elephant interaction.

show abstract

A novel system for automatic detection and classification of animal

Cited by 19 publications

References 7 publications

An IoT System Using Deep Learning to Classify Camera Trap Images on the Edge

An IoT System Using Deep Learning to Classify Camera Trap Images on the Edge

Local Feature Extraction in the Near Infra-Red Domain for Wildlife Mammals Tracking Purpose

A vision based human - elepahant collision detection system

Contact Info

Product

Resources

About