49 Hits in 10.0 sec

Stingray Detection of Aerial Images Using Augmented Training Images Generated by A Conditional Generative Model [article]

Yi-Min Chou, Chien-Hung Chen, Keng-Hao Liu, Chu-Song Chen
2018 arXiv   pre-print
To this end, we use a deep object detection method, faster RCNN, to train a stingray detector based on a limited training set of images.  ...  Experimental results show that satisfiable performance can be obtained by using our approach on stingray detection in aerial images.  ...  Acknowledgements This work is supported in part by the projects MOST 107-2634-F-001-004 and MOST 106-2221-E-110-074.  ... 
arXiv:1805.04262v3 fatcat:r4depknrhfaazemfyqlim4b3aq

Sharkeye: Real-Time Autonomous Personal Shark Alerting via Aerial Surveillance

Robert Gorkin, Kye Adams, Matthew J Berryman, Sam Aubin, Wanqing Li, Andrew R Davis, Johan Barthelemy
2020 Drones  
water, and with analysis capable of detecting shark analogues, rays, and surfers in average beach conditions, and all based on ~1 h of training data in total.  ...  To execute, our team developed a novel detection algorithm trained via machine learning based on aerial footage of real sharks and rays collected at local beaches, hosted and deployed the algorithm in  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/drones4020018 fatcat:3mps6ehedrfdhegofpyoeqttjy

A Survey of Deep Learning-based Object Detection Methods and Datasets for Overhead Imagery

Junhyung Kang, Shahroz Tariq, Han Oh, Simon S. Woo
2022 IEEE Access  
We thank Jin Yong Park for reviewing the earlier version of the draft, and providing helpful and insightful comments.  ...  [60] proposed an interesting approach for detecting stingrays, employing a generative approach called Conditional GLO (C-GLO) in aerial images.  ...  Through training with these augmented images, the baseline model showed the significant performance improvement. Also, Chen et al.  ... 
doi:10.1109/access.2022.3149052 fatcat:iwvyg7qf6jgntgrwk4bmonrd5m

Weakly Supervised Detection of Marine Animals in High Resolution Aerial Images

Paul Berg, Deise Santana Maia, Minh-Tan Pham, Sébastien Lefèvre
2022 Remote Sensing  
using state-of-the-art object detection models.  ...  Since analyzing thousands of images acquired during a flight survey can be a cumbersome and time consuming task, we focus in this article on the weakly supervised detection of marine animals.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/rs14020339 fatcat:uwki27gci5e7hoye5cutpvmeey

Grounded Language-Image Pre-training [article]

Liunian Harold Li, Pengchuan Zhang, Haotian Zhang, Jianwei Yang, Chunyuan Li, Yiwu Zhong, Lijuan Wang, Lu Yuan, Lei Zhang, Jenq-Neng Hwang, Kai-Wei Chang, Jianfeng Gao
2021 arXiv   pre-print
pairs by generating grounding boxes in a self-training fashion, making the learned representation semantic-rich.  ...  This paper presents a grounded language-image pre-training (GLIP) model for learning object-level, language-aware, and semantic-rich visual representations.  ...  Given a good grounding model (teacher), we can augment GLIP pre-training data by automatically generating grounding boxes for massive image-text-paired data, in which noun phrases are detected by an NLP  ... 
arXiv:2112.03857v1 fatcat:fzgqgdqbojhrlpv5dmkxnivyhm

A deep learning-integrated micro-CT image analysis pipeline for quantifying rice lodging resistance-related traits

Di Wu, Dan Wu, Hui Feng, Lingfeng Duan, Guoxing Dai, Xiao Liu, Kang Wang, Peng Yang, Guoxing Chen, Alan P. Gay, John H. Doonan, Zhiyou Niu (+2 others)
2021 Plant Communications  
In this study, we used a high-throughput micro-CT-RGB imaging system and deep learning (SegNet) to develop a high-throughput micro-CT image analysis pipeline that can extract 24 rice culm morphological  ...  We also built models of bending stress using culm traits at the mature and tillering stages, and the R 2 values were 0.722 and 0.544, respectively.  ...  SUPPLEMENTAL INFORMATION DECLARATION OF INTERESTS The authors declare that they have no competing interests.  ... 
doi:10.1016/j.xplc.2021.100165 pmid:33898978 pmcid:PMC8060729 fatcat:46dbeyuy4faerd5bp3exx2pbo4

Unoccupied Aircraft Systems in Marine Science and Conservation

David W. Johnston
2018 Annual Review of Marine Science  
The use of unoccupied aircraft systems (UASs, also known as drones) in science is growing rapidly.  ...  It concludes with details on potential effects of UASs on marine wildlife and a look to the future of UASs in marine science and conservation.  ...  Support for several projects highlighted in the review came from the National Science Foundation, the US National Park Service, the Canadian Department of Fisheries and Oceans, National Geographic, and  ... 
doi:10.1146/annurev-marine-010318-095323 pmid:30020850 fatcat:fh5rbl77kndcbgdwsv3a6ngvka

Detecting Biological Locomotion in Video: A Computational Approach [article]

Soo Min Kang, Richard P. Wildes
2021 arXiv   pre-print
Significantly, the motion exhibited by the body parts of a biological entity to navigate through an environment can be modeled by a combination of an overall positional advance with an overlaid asymmetric  ...  In this report, we refer to the locomotion of general biological species as biolocomotion. We present a computational approach to detect biolocomotion in unprocessed video.  ...  The encoded descriptors that result from a set of training videos are used to train a classifier (e.g., SVM [28] ), which is used to obtain a biolocomotion score (e.g., via Platt scaling [88] ) for encoded  ... 
arXiv:2105.12661v1 fatcat:ag3p6gbn3fhpddgashubjtkase

Grafting acoustic instruments and signal processing: Creative control and augmented expressivity

Dan Overholt
2013 Journal of the Acoustical Society of America  
Beranek, we seek to correlate acoustical measurements of spaces used for Sacred Harp singing with subjective evaluations of those spaces made by the singers themselves.  ...  In the spirit of the text Concert Halls and Opera Houses by L.  ...  The stress wave in the pile is predicted using a force-generator model of the hammer-pile system.  ... 
doi:10.1121/1.4830794 fatcat:2sexsmbypjcozjctnorzoiy7te

Trends in preference, programming and design of concert halls for symphonic music

Anders C. Gade
2008 Journal of the Acoustical Society of America  
for use in training language models for a LID system.  ...  This study shows that automatically generated speakers Acoustics'08 Paris have a great impact on the accuracy of LID models, in this case an initial training set of 15 minutes per language was augmented  ...  By using noise generated by a can of com-pressed air sprayed on the surface of a plexiglass plate, we are able to reconstruct not only the time of flight but the whole waveforms between the sensors.  ... 
doi:10.1121/1.2932471 fatcat:hhka5ml2fvcszklkzau2lw3gka

The structural acoustic properties of stiffened shells

Yu Luan
2008 Journal of the Acoustical Society of America  
for Љa second golden age of aeroacousticsЉ ͑to quote Lighthill himself͒.  ...  by jet aircrafts.  ...  for use in training language models for a LID system.  ... 
doi:10.1121/1.2932806 fatcat:zohbewf2k5h7fly2hoqzfdk42e

Acoustic GIS‐based monitoring of Atlantic cod ecosystems in coastal Newfoundland

George A. Rose
2008 Journal of the Acoustical Society of America  
for Љa second golden age of aeroacousticsЉ ͑to quote Lighthill himself͒.  ...  by jet aircrafts.  ...  for use in training language models for a LID system.  ... 
doi:10.1121/1.2932540 fatcat:v6w3j7upx5duhkunch6oai5jve

Survey on 6G Frontiers: Trends, Applications, Requirements, Technologies and Future Research

Chamitha de Alwis, Anshuman Kalla, Quoc-Viet Pham, Pardeep Kumar, Kapal Dev, Won-Joo Hwang, Madhusanka Liyanage
2021 IEEE Open Journal of the Communications Society  
By 2022, the fleet of small model UAVs (primarily used for recreational purposes by hobbyists) is expected to reach a mark of 1.38 million units whereas small non-model UAVs (primarily used for commercial  ...  M H L M H M AI and FL • Provides a distributed (AI) framework for training models such that every participating device trains the model locally and shares the updated model instead of data.  ... 
doi:10.1109/ojcoms.2021.3071496 fatcat:lpu2cwh6xfebjpbt56svuwxoti

Symposia and Oral Abstracts

2007 Integrative and Comparative Biology  
Chronograms were estimated in BEAST 1.45 using a Bayesian relaxed-clock model with uncorrelated log-normally distributed rates.  ...  Our results show that measurement of fecal cortisol metabolites by an enzyme immunoassay is a useful tool to monitor adrenocortical activity.  ...  Here, we used a finite element model (FEM) of a male moth antenna to examine the 3-D patterns of strain at the base of the vibrating antenna as a moth executes an aerial maneuver.  ... 
doi:10.1093/icb/icm104 fatcat:uqdscbqihbeoxoo2unutfwtmdi


2017 Mirovaya Ekonomika i Mezhdunarodnye Otnosheniya  
ISRO is also developing the Radar Imaging Satellite using synthetic aperture radar that will be able to take three-meter resolution images in all-terrain, all-weather, day/night conditions, which is a  ...  practices can be a successful model, as demonstrated by the space debris mitigation guidelines adopted by COPUOS. 91 Indicating a shift in US policy toward international regulation of space activities  ...  outer space by the present and future generations."  ... 
doi:10.20542/0131-2227-2017-61-10-51-56 fatcat:n5fghvivevgo7ow7pumzepsu6y
« Previous Showing results 1 — 15 out of 49 results