English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

In this paper, we examine the use of data from multiple sensing modes, i.e.,accelerometry and global navigation satellite system (GNSS), for classifyinganimal behavior. We extract three new features from the GNSS data, namely,distance from water point, median speed, and median estimated horizontalposition error. We combine the information available from the accelerometry andGNSS data via two approaches. The first approach is based on concatenating thefeatures extracted from both sensor data and feeding the concatenated featurevector into a multi-layer perceptron (MLP) classifier. The second approach isbased on fusing the posterior probabilities predicted by two MLP classifiers.The input to each classifier is the features extracted from the data of onesensing mode. We evaluate the performance of the developed multimodal animalbehavior classification algorithms using two real-world datasets collected viasmart cattle collar tags and ear tags. The leave-one-animal-outcross-validation results show that both approaches improve the classificationperformance appreciably compared with using data of only one sensing mode. Thisis more notable for the infrequent but important behaviors of walking anddrinking. The algorithms developed based on both approaches require littlecomputational and memory resources hence are suitable for implementation onembedded systems of our collar tags and ear tags. However, the multimodalanimal behavior classification algorithm based on posterior probability fusionis preferable to the one based on feature concatenation as it delivers betterclassification accuracy, has less computational and memory complexity, is morerobust to sensor data failure, and enjoys better modularity.

Multimodal sensor data fusion for in-situ classification of animal behavior using accelerometry and GNSS data