Publication Type:

Conference Paper

Source:

2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI) (2017)

Keywords:

automated vision-based method, color features, Color model, Computer vision, elephants detection, Environmental management, environmental science computing, Feature extraction, frontline conservation issue, human encroachment, Human-Elephant Conflict, human-elephant conflicts, image classification, Image color analysis, image colour analysis, Image processing, Image texture, input image classification, learning (artificial intelligence), Object Detection, Object recognition, Real-time systems, Support Vector Machine, Support vector machines, Surveillance, Texture, Texture features, Vibrations, vision based surveillance method

Abstract:

Human-elephant conflict is a frontline conservation issue in the world. The loss and fragmentation of elephant's habitat owing to the increased human encroachment leads to a notable conservation issue. It raises the need for a non-invasive and efficient solution for the mitigation of human-elephant conflicts. Consequently, deploying a vision based surveillance method in the real time environment can prove to be significantly useful to provide the warnings well in advance thereby reducing the human elephant conflict. In this paper, a method for the identification of elephant as an object using image processing is proposed. The method dynamically learns from the trained images with different backgrounds, lighting conditions. Further, it classifies the input image based on the features of color and texture. The outcomes demonstrate that the proposed method effectively deals with the detection of elephants in near and far distances, cluttered and occluded environment

Cite this Research Publication

G. Ramesh, Dr. Senthil Kumar M., Sini Raj P., Krishnamoorthy, V., and Gowtham, R., “An Automated Vision-based Method to Detect Elephants for Mitigation of Human-elephant Conflicts”, in 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), 2017.