Automated annotation of coral reef survey images

By Oscar Beijbom, Peter J. Edmunds, David I. Kline, B. Greg Mitchell, David J. Kriegman. 2012

340 citations recorded by the source corpus.

Subfields: Ecology, Media Technology, Nature and Landscape Conservation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology

1. John D. Hedley
2. Marc Olano
3. Erle C. Ellis
4. Jonathan P. Dandois
5. Peter J. Mumby
6. Alastair R. Harborne
7. Salman Hameed Khan
8. Thomas Baltzer Moeslund

In Media Technology

1. Munawar Hayat
2. Salman Hameed Khan
3. Peigen Luo
4. Iljung S. Kwak
5. Mohammed Bennamoun
6. Thomas Baltzer Moeslund
7. Trevor Darrell
8. Junaed Sattar

In Nature and Landscape Conservation

1. Peter J. Mumby
2. Scarla J. Weeks
3. Erle C. Ellis
4. Jonathan P. Dandois

Citations in the corpus

• Compact Bilinear Pooling (2015, background): “It is important in many deployment scenarios where labels are expensive or time-consuming to acquire [1].”
• A systematic study of the class imbalance problem in convolutional neural networks (2017, background): “There are plenty of examples in domains like computer vision [6, 7, 8, 9, 10], medical diagnosis [11, 12], fraud detection [13] and others [14, 15, 16] where this issue is highly significant and the frequency of one class (e.g., cancer) can be 1000 times less than another class (e.g., healthy…”
• Best Practices for Fine-Tuning Visual Classifiers to New Domains (2016, methodology): “We evaluated our experiments on 6 datasets: Caltech256, SUN397, MIT Indoor Scene Recognition (MITIndoor) [13], Caltech-UCSD-Birds-200 (CUB) [14], Moorea Labeled Corals (Coral) [15], and Imaging Flow Cytobot Data Plankton (Plankton) [16].”
• Remote Sensing of Coral Reefs for Monitoring and Management: A Review (2016, methodology): “The next generation of field surveys is likely to be autonomous, towed or diver-operated camera surveys, such as the XL Catlin Seaview Survey [25] combined with automatic image analysis [26].”
• Learning Concept Embeddings with Combined Human-Machine Expertise (2015, background): “Unfortunately, asking experts to exhaustively and authoritatively annotate the dataset is not always possible [3].”
• Optimal Altitude, Overlap, and Weather Conditions for Computer Vision UAV Estimates of Forest Structure (2015, methodology): “[79] used image features to automatically classify coral communities.”
• Detection of Marine Animals in a New Underwater Dataset with Varying Visibility (2019, background): “Other annotated coral reef datasets include the Moorea Labeled Corals [4] which is an annotated subset of the Moorea Labeled Corals Long Term Ecological Research project in French Polynesia and the Eilat Fluorescence dataset [5], which experiments with a combination of standard reflectance and fluores”
• Cost-Sensitive Learning of Deep Feature Representations From Imbalanced Data (2015, methodology): “For similar reasons, we used the RGB color space instead of LAB, which was shown to perform better on the MLC dataset [4].”
• Semantic Segmentation of Underwater Imagery: Dataset and Benchmark (2020, methodology): “Several important contributions address the problems of coral-reef classification and segmentation [12], [11], coral-reefs’ coverage estimation [11], fish detection and segmentation [13], [14], [15], etc.”