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/

Global: Zendy Plus annual

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

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Many applications utilizing Unmanned Aerial Vehicles (UAVs) require the useof computer vision algorithms to analyze the information captured from theiron-board camera. Recent advances in deep learning have made it possible to usesingle-shot Convolutional Neural Network (CNN) detection algorithms thatprocess the input image to detect various objects of interest. To keep thecomputational demands low these neural networks typically operate on smallimage sizes which, however, makes it difficult to detect small objects. This isfurther emphasized when considering UAVs equipped with cameras where due to theviewing range, objects tend to appear relatively small. This paper therefore,explores the trade-offs involved when maintaining the resolution of the objectsof interest by extracting smaller patches (tiles) from the larger input imageand processing them using a neural network. Specifically, we introduce anattention mechanism to focus on detecting objects only in some of the tiles anda memory mechanism to keep track of information for tiles that are notprocessed. Through the analysis of different methods and experiments we showthat by carefully selecting which tiles to process we can considerably improvethe detection accuracy while maintaining comparable performance to CNNs thatresize and process a single image which makes the proposed approach suitablefor UAV applications.

Efficient ConvNet-based Object Detection for Unmanned Aerial Vehicles by Selective Tile Processing