scholarly journals On Image Classification in Video Analysis of Omnidirectional Apis Mellifera Traffic: Random Reinforced Forests vs. Shallow Convolutional Networks

2021 ◽  
Vol 11 (17) ◽  
pp. 8141
Author(s):  
Vladimir Kulyukin ◽  
Nikhil Ganta ◽  
Anastasiia Tkachenko
Keyword(s):  
Image Classification ◽  
Video Analysis ◽  
Motion Detection ◽  
Health Assessment ◽  
Tier 2 ◽  
Convolutional Networks ◽  
Theoretical Comparison ◽  
Tier 1 ◽  
Close Proximity ◽  
Automated Video Analysis

Omnidirectional honeybee traffic is the number of bees moving in arbitrary directions in close proximity to the landing pad of a beehive over a period of time. Automated video analysis of such traffic is critical for continuous colony health assessment. In our previous research, we proposed a two-tier algorithm to measure omnidirectional bee traffic in videos. Our algorithm combines motion detection with image classification: in tier 1, motion detection functions as class-agnostic object location to generate regions with possible objects; in tier 2, each region from tier 1 is classified by a class-specific classifier. In this article, we present an empirical and theoretical comparison of random reinforced forests and shallow convolutional networks as tier 2 classifiers. A random reinforced forest is a random forest trained on a dataset with reinforcement learning. We present several methods of training random reinforced forests and compare their performance with shallow convolutional networks on seven image datasets. We develop a theoretical framework to assess the complexity of image classification by a image classifier. We formulate and prove three theorems on finding optimal random reinforced forests. Our conclusion is that, despite their limitations, random reinforced forests are a reasonable alternative to convolutional networks when memory footprints and classification and energy efficiencies are important factors. We outline several ways in which the performance of random reinforced forests may be improved.

scholarly journals On Video Analysis of Omnidirectional Bee Traffic: Counting Bee Motions with Motion Detection and Image Classification

2019 ◽  
Vol 9 (18) ◽  
pp. 3743 ◽  
Author(s):  
Vladimir Kulyukin ◽  
Sarbajit Mukherjee
Keyword(s):  
Image Classification ◽  
Motion Detection ◽  
Structural Modification ◽  
Health Assessment ◽  
Support Vector ◽  
Monitoring Systems ◽  
Ensemble Of Classifiers ◽  
Close Proximity ◽  
Honeybee Colony ◽  
Performance Benchmark

Omnidirectional bee traffic is the number of bees moving in arbitrary directions in close proximity to the landing pad of a given hive over a given period of time. Video bee traffic analysis has the potential to automate the assessment of omnidirectional bee traffic levels, which, in turn, may lead to a complete or partial automation of honeybee colony health assessment. In this investigation, we proposed, implemented, and partially evaluated a two-tier method for counting bee motions to estimate levels of omnidirectional bee traffic in bee traffic videos. Our method couples motion detection with image classification so that motion detection acts as a class-agnostic object location method that generates a set of regions with possible objects and each such region is classified by a class-specific classifier such as a convolutional neural network or a support vector machine or an ensemble of classifiers such as a random forest. The method has been, and is being iteratively field tested in BeePi monitors, multi-sensor electronic beehive monitoring systems, installed on live Langstroth beehives in real apiaries. Deployment of a BeePi monitor on top of a beehive does not require any structural modification of the beehive’s woodenware, and is not disruptive to natural beehive cycles. To ensure the replicability of the reported findings and to provide a performance benchmark for interested research communities and citizen scientists, we have made public our curated and labeled image datasets of 167,261 honeybee images and our omnidirectional bee traffic videos used in this investigation.

Optimal Timing for Launching Installation of Tier 2 and 3 Systems of School-Wide Positive Behavioral Interventions and Supports

2021 ◽  
pp. 109830072199608
Author(s):  
Angus Kittelman ◽  
Sterett H. Mercer ◽  
Kent McIntosh ◽  
Robert Hoselton
Keyword(s):  
Longitudinal Study ◽  
Behavioral Interventions ◽  
Structural Equation ◽  
Equation Modeling ◽  
Optimal Timing ◽  
First Year ◽  
Tier 2 ◽  
Positive Behavioral Interventions ◽  
Tier 1 ◽  
Tier 3

The purpose of this longitudinal study was to examine patterns in implementation of Tier 2 and 3 school-wide positive behavioral interventions and supports (SWPBIS) systems to identify timings of installation that led to higher implementation of advanced tiers. Extant data from 776 schools in 27 states reporting on the first 3 years of Tier 2 implementation and 359 schools in 23 states reporting on the first year of Tier 3 implementation were analyzed. Using structural equation modeling, we found that higher Tier 1 implementation predicted subsequent Tier 2 and Tier 3 implementation. In addition, waiting 2 or 3 years after initial Tier 1 implementation to launch Tier 2 systems predicted higher initial Tier 2 implementation (compared with implementing the next year). Finally, we found that launching Tier 3 systems after Tier 2 systems, compared with launching both tiers simultaneously, predicted higher Tier 2 implementation in the second and third year, so long as Tier 3 systems were launched within 3 years of Tier 2 systems. These findings provide empirical guidance for when to launch Tier 2 and 3 systems; however, we emphasize that delays in launching advanced systems should not equate to delays in more intensive supports for students.

scholarly journals Global Inland Capture and Culture Finfisheries Follow Different Trends When Evaluated by the Human Development Index

2021 ◽  
Vol 13 (15) ◽  
pp. 8420
Author(s):  
Peter W. Sorensen ◽  
Maria Lourdes D. Palomares
Keyword(s):  
Socioeconomic Factors ◽  
Human Development ◽  
Human Development Index ◽  
Developed Countries ◽  
Global Scale ◽  
Tier 2 ◽  
Development Index ◽  
Tier 1 ◽  
Catch Reconstruction ◽  
Tier 3

To assess whether and how socioeconomic factors might be influencing global freshwater finfisheries, inland fishery data reported to the FAO between 1950 and 2015 were grouped by capture and culture, country human development index, plotted, and compared. We found that while capture inland finfishes have greatly increased on a global scale, this trend is being driven almost entirely by poorly developed (Tier-3) countries which also identify only 17% of their catch. In contrast, capture finfisheries have recently plateaued in moderately-developed (Tier-2) countries which are also identifying 16% of their catch but are dominated by a single country, China. In contrast, reported capture finfisheries are declining in well-developed (Tier-1) countries which identify nearly all (78%) of their fishes. Simultaneously, aquacultural activity has been increasing rapidly in both Tier-2 and Tier-3 countries, but only slowly in Tier-1 countries; remarkably, nearly all cultured species are being identified by all tier groups. These distinctly different trends suggest that socioeconomic factors influence how countries report and conduct capture finfisheries. Reported rapid increases in capture fisheries are worrisome in poorly developed countries because they cannot be explained and thus these fisheries cannot be managed meaningfully even though they depend on them for food. Our descriptive, proof-of-concept study suggests that socioeconomic factors should be considered in future, more sophisticated efforts to understand global freshwater fisheries which might include catch reconstruction.

scholarly journals Seismic characteristics of polygonal fault systems in the Great South Basin, New Zealand

2020 ◽  
Vol 12 (1) ◽  
pp. 851-865
Author(s):  
Sukonmeth Jitmahantakul ◽  
Piyaphong Chenrai ◽  
Pitsanupong Kanjanapayont ◽  
Waruntorn Kanitpanyacharoen
Keyword(s):  
Three Dimensional ◽  
Late Eocene ◽  
Seismic Survey ◽  
Fault System ◽  
Normal Faults ◽  
Tier 2 ◽  
South Basin ◽  
Fault Systems ◽  
Tier 1 ◽  
Polygonal Fault

AbstractA well-developed multi-tier polygonal fault system is located in the Great South Basin offshore New Zealand’s South Island. The system has been characterised using a high-quality three-dimensional seismic survey tied to available exploration boreholes using regional two-dimensional seismic data. In this study area, two polygonal fault intervals are identified and analysed, Tier 1 and Tier 2. Tier 1 coincides with the Tucker Cove Formation (Late Eocene) with small polygonal faults. Tier 2 is restricted to the Paleocene-to-Late Eocene interval with a great number of large faults. In map view, polygonal fault cells are outlined by a series of conjugate pairs of normal faults. The polygonal faults are demonstrated to be controlled by depositional facies, specifically offshore bathyal deposits characterised by fine-grained clays, marls and muds. Fault throw analysis is used to understand the propagation history of the polygonal faults in this area. Tier 1 and Tier 2 initiate at about Late Eocene and Early Eocene, respectively, based on their maximum fault throws. A set of three-dimensional fault throw images within Tier 2 shows that maximum fault throws of the inner polygonal fault cell occurs at the same age, while the outer polygonal fault cell exhibits maximum fault throws at shallower levels of different ages. The polygonal fault systems are believed to be related to the dewatering of sedimentary formation during the diagenesis process. Interpretation of the polygonal fault in this area is useful in assessing the migration pathway and seal ability of the Eocene mudstone sequence in the Great South Basin.

scholarly journals Use of automated video analysis for the evaluation of bicycle movement and interaction

2014 ◽  
Author(s):  
Heather Twaddle ◽  
Tobias Schendzielorz ◽  
Oliver Fakler ◽  
Sasan Amini
Keyword(s):  
Video Analysis ◽  
Automated Video Analysis

Automated video analysis of pig activity at pen level highly correlates to human observations of behavioural activities

2014 ◽  
Vol 160 ◽  
pp. 132-137 ◽  
Author(s):  
S. Ott ◽  
C.P.H. Moons ◽  
M.A. Kashiha ◽  
C. Bahr ◽  
F.A.M. Tuyttens ◽  
...  
Keyword(s):  
Video Analysis ◽  
Automated Video Analysis

Effects of acute methamphetamine administration on spacing in paired rats: Investigation with an automated video-analysis method

1996 ◽  
Vol 20 (6) ◽  
pp. 1037-1049 ◽  
Author(s):  
Toshikazu Shinba ◽  
Ken-Ichi Yamamoto ◽  
Gong-Min Cao ◽  
G.O. Mugishima ◽  
Yoshinori Andow ◽  
...  
Keyword(s):  
Video Analysis ◽  
Analysis Method ◽  
Automated Video Analysis

Incorporation of plasma-based next-generation sequencing to improve guideline-concordant molecular testing in patients with newly diagnosed metastatic nonsquamous non-small cell lung cancer.

2021 ◽  
Vol 39 (28_suppl) ◽  
pp. 14-14
Author(s):  
Charu Aggarwal ◽  
Melina Elpi Marmarelis ◽  
Wei-Ting Hwang ◽  
Dylan G. Scholes ◽  
Aditi Puri Singh ◽  
...  
Keyword(s):  
Clinical Practice ◽  
Practice Change ◽  
Molecular Testing ◽  
Tier 2 ◽  
Newly Diagnosed ◽  
Next Generation ◽  
First Line ◽  
Tier 1 ◽  
Tier 3 ◽  
The Impact

14 Background: Current NCCN guidelines recommend comprehensive molecular profiling for all newly diagnosed patients with metastatic non-squamous NSCLC to enable the delivery of personalized medicine. We have previously demonstrated that incorporation of plasma based next-generation gene sequencing (NGS) improves detection of clinically actionable mutations in patients with advanced NSCLC (Aggarwal et al, JAMA Oncology, 2018). To increase rates of comprehensive molecular testing at our institution, we adapted our clinical practice to include concurrent use of plasma (P) and tissue (T) based NGS upon initial diagnosis. P NGS testing was performed using a commercial 74 gene assay. We analyzed the impact of this practice change on guideline concordant molecular testing at our institution. Methods: A retrospective cohort study of patients with newly diagnosed metastatic non-squamous NSCLC following the implementation of this practice change in 12/2018 was performed. Tiers of NCCN guideline concordant testing were defined, Tier 1: complete EGFR, ALK, BRAF, ROS1, MET, RET, NTRK testing, Tier 2: included above, but with incomplete NTRK testing, Tier 3: > 2 genes tested, Tier 4: single gene testing, Tier 5: no testing. Proportion of patients with comprehensive molecular testing by modality (T NGS vs. T+P NGS) were compared using one-sided Fisher’s exact test. Results: Between 01/2019, and 12/2019, 170 patients with newly diagnosed metastatic non-Sq NSCLC were treated at our institution. Overall, 98.2% (167/170) patients underwent molecular testing, Tier 1: n = 100 (59%), Tier 2: n = 39 (23%), Tier 3/4: n = 28 (16.5%), Tier 5: n = 3 (2%). Amongst these patients, 43.1% (72/167) were tested with T NGS alone, 8% (15/167) with P NGS alone, and 47.9% (80/167) with T+P NGS. A higher proportion of patients underwent comprehensive molecular testing (Tiers 1+2) using T+P NGS: 95.7% (79/80) compared to T alone: 62.5% (45/72), p < 0.0005. Prior to the initiation of first line treatment, 72.4% (123/170) patients underwent molecular testing, Tier 1: n = 73 (59%), Tier 2: n = 27 (22%) and Tier 3/4: n = 23 (18%). Amongst these, 39% (48/123) were tested with T NGS alone, 7% (9/123) with P NGS alone and 53.6% (66/123) with T+P NGS. A higher proportion of patients underwent comprehensive molecular testing (Tiers 1+2) using T+P NGS, 100% (66/66) compared to 52% (25/48) with T NGS alone (p < 0.0005). Conclusions: Incorporation of concurrent T+P NGS testing in treatment naïve metastatic non-Sq NSCLC significantly increased the proportion of patients undergoing guideline concordant molecular testing, including prior to initiation of first-line therapy at our institution. Concurrent T+P NGS should be adopted into institutional pathways and routine clinical practice.

Sign in / Sign up

Export Citation Format

Share Document