Intelligent System for Detection of Wild Animals Using HOG and CNN in Automobile Applications

Munian, Yuvaraj; Martínez-Molina, Antonio; Alamaniotis, Miltiadis

doi:10.1109/iisa50023.2020.9284365

Cited by 21 publications

(14 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Modern approaches use deep learning techniques such as Convolutional Neural Networks (CNN) that extract features from images, which were subsequently classified using several dense layers. There are also solutions that combine traditional features such as HoG with CNN [ 50 , 51 ]. An extensive review of the object detection techniques using deep learning is provided in [ 52 ].…”

Section: Related Workmentioning

confidence: 99%

Vision-Based Module for Herding with a Sheepdog Robot

Castillo

Sánchez-González

Campazas-Vega

et al. 2022

Sensors

View full text Add to dashboard Cite

Livestock farming is assisted more and more by technological solutions, such as robots. One of the main problems for shepherds is the control and care of livestock in areas difficult to access where grazing animals are attacked by predators such as the Iberian wolf in the northwest of the Iberian Peninsula. In this paper, we propose a system to automatically generate benchmarks of animal images of different species from iNaturalist API, which is coupled with a vision-based module that allows us to automatically detect predators and distinguish them from other animals. We tested multiple existing object detection models to determine the best one in terms of efficiency and speed, as it is conceived for real-time environments. YOLOv5m achieves the best performance as it can process 64 FPS, achieving an mAP (with IoU of 50%) of 99.49% for a dataset where wolves (predator) or dogs (prey) have to be detected and distinguished. This result meets the requirements of pasture-based livestock farms.

show abstract

Section: Related Workmentioning

confidence: 99%

Vision-Based Module for Herding with a Sheepdog Robot

Castillo

Sánchez-González

Campazas-Vega

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Munian et al [5] designed an intelligence technique for monitoring wild creatures in an automotive application utilizing HOG and CNN. This research adds a novel intensity to wild animal auto-detection throughout dynamic midnight periods, utilizing infrared thermal interpretation via a camera car attachment to reduce automobile and wildlife collisions.…”

Section: Literature Reviewmentioning

confidence: 99%

Development of Wild Animal Ward-Off System

Shivaramu¹,

Yaseen²,

Noronha³

2022

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN ELECTRICAL, ELE

View full text Add to dashboard Cite

Farmers suffer losses as a result of crop damage caused by animals and birds during the stages of sowing, seeding, ripening, and harvesting. Traditional methods of controlling bird and animal attacks, such as beating drums, screaming, lighting fireworks, and using reflective materials, are less effective when used on daytime attacks by birds and animals. Surveillance, as is well known, is widely used in a variety of settings, including homes, hospitals, schools, public spaces, farms, and so on. It allows us to keep an eye on a specific area, prevent theft, and provide proof of evidence. Monitoring is critical in the case of farmlands or agricultural grounds to both protect the environment and prevent unauthorized individuals from entering the region area from animals. Image processing techniques, night vision cameras, and IoT technologies will undoubtedly solve these problems. This system is capable of repelling animals from agricultural fields when they attempt to intrude; the system is outfitted with different triggering systems based on animal type. It has a loud noise generator for larger animals like elephants and a rotten egg sprayer for smaller animals like deer and boar. At first, these animal movements will be detected by passive infrared and microwave sensors, which will turn on the camera for image processing following successful animal detection. The primary goal of this work is to increase food production rates, prevent economic losses for farmers due to animal intrusions, and bring a smile to the farmer community.

show abstract

“…Object detection algorithms for AVs focus primarily on road signs, pedestrians, cyclists, or other vehicles [e.g., 52-58], with comparatively fewer methods designed for animal detection [59][60][61][62].…”

Section: Obstacle Detection and Motion Trackingmentioning

confidence: 99%

“…The high levels of morphological variation across animal species, along with a wide range of sensory perception processes, behavioral responses, and means of locomotion, introduce several obstacles to automated animal detection methods. Munian et al [59] employed thermal imaging and a convolutional neural network (CNN) with the Histogram of Oriented Gradient (HOG) transform, to reach an average accuracy of 89%. This particular method experiences limitations with coldblooded species, as it is based on thermal images, or for higher vehicle speeds, as the processing time is between 1 to 3 seconds.…”

Section: Obstacle Detection and Motion Trackingmentioning

confidence: 99%

“…The development of more appropriate animal detection methods is also necessary. Relying only on algorithms tailored for human detection may lead to inaccurate interpretations of animal behavior or their impending motion, and current animal-specific methods still face many obstacles: relatively high response times only applicable at low vehicle speeds [59,62], the need for clear lane demarcation [61], no motion tracking [60], or limited training datasets [62,65].…”

Section: Infrastructure and Technical Limitationsmentioning

confidence: 99%

See 1 more Smart Citation

Emerging opportunities for wildlife with sustainable autonomous transportation

Silva¹,

Calabrese²

2021

Preprint

View full text Add to dashboard Cite

Autonomous vehicles (AV) are expected to play a key role in the future of transportation, introducing a disruptive yet potentially beneficial change for vehicle-wildlife interactions. However, this assumption has not been critically examined. Here, we introduce a new conceptual framework covering the intersection between AV technological innovation and wildlife conservation to reduce wildlife-vehicle collisions. We suggest future research within this framework to focus on developing robust warning systems and animal detection methods for AV systems, and to incorporate wildlife-vehicle interactions into decision-making algorithms. With large-scale deployment a looming reality, it is vital to incorporate conservation and sustainability into the societal, ethical, and legal implications of AV technology. We appeal for further debate and interdisciplinary collaborations between scientists, developers, and policymakers.

show abstract

Intelligent System for Detection of Wild Animals Using HOG and CNN in Automobile Applications

Cited by 21 publications

References 9 publications

Vision-Based Module for Herding with a Sheepdog Robot

Vision-Based Module for Herding with a Sheepdog Robot

Development of Wild Animal Ward-Off System

Emerging opportunities for wildlife with sustainable autonomous transportation

Contact Info

Product

Resources

About