An Adaptive Embedding Network with Spatial Constraints for the Use of Few-Shot Learning in Endangered-Animal Detection

Abstract

Image recording is now ubiquitous in the fields of endangered-animal conservation and GIS. However, endangered animals are rarely seen, and, thus, only a few samples of images of them are available. In particular, the study of endangered-animal detection has a vital spatial component. We propose an adaptive, few-shot learning approach to endangered-animal detection through data augmentation by applying constraints on the mixture of foreground and background images based on species distributions. First, the pre-trained, salient network U2-Net segments the foregrounds and backgrounds of images of endangered animals. Then, the pre-trained image completion network CR-Fill is used to repair the incomplete environment. Furthermore, our approach identifies a foreground–background mixture of different images to produce multiple new image examples, using the relation network to permit a more realistic mixture of foreground and background images. It does not require further supervision, and it is easy to embed into existing networks, which learn to compensate for the uncertainties and nonstationarities of few-shot learning. Our experimental results are in excellent agreement with theoretical predictions by different evaluation metrics, and they unveil the future potential of video surveillance to address endangered-animal detection in studies of their behavior and conservation.

A novel DbneAlexnet with Gazelle Hunting Optimization Algorithm enabled wild animal detection in WMSN data communication in IoT environment

“…The routing evaluation is done between the proposed GHOA method and the routing strategies, such as BOA + ACO, 21 , EDACR, 4 HRDSS, 7 and ACO 21 . Similarly, the GHOA DbneAlexNet performance is evaluated by comparing it with the existing methods, such as SHC, 13 energy reduction method, 14 smart farming technique, 11 and few shot learning 20 …”

“…Unfortunately, the strategy failed to improve system performance by using big data analysis and higher quality hardware components. Feng, J., and Li, J 20 . used few shots learning to detect exotic animals.…”

“…Throughout the Americas, Asia, and Europe, a number of technologies have been made in response to these possibilities, which include smart cities, smart autos, homes, industries, and GPS tracking systems. Because of the introduction of multimedia objects in data transmission, a variety of commercial and military applications have emerged, including remote monitoring of patients in telehealth and telemedicine and traffic control systems augmented by smart video cameras, among others 20 . This necessitates a functional update of IoT network to IoMT.…”

“…Because of the introduction of multimedia objects in data transmission, a variety of commercial and military applications have emerged, including remote monitoring of patients in telehealth and telemedicine and traffic control systems augmented by smart video cameras, among others. 20 This necessitates a functional update of IoT network to IoMT.…”

“…Feng, J., and Li, J. 20 used few shots learning to detect exotic animals. Also, it provided image primitives for the integration of contextual information to create a full collection of instances, which was helpful for boosting learning sample efficiency, notably for endangered animal detection.…”